====================================================================================================
#loader.py
### canonically 
### torchrun --standalone --nproc_per_node=8 loader.py
### but for us, probably 
### set USE_LIBUV=0
### set RANK 
### set TORCH_CUDNN_SDPA_ENABLED=1
### torchrun --standalone --nproc_per_node=1 loader.py
import os
import sys
with open(sys.argv[0]) as f:
    code = f.read() # read the code of this file ASAP, for logging
import uuid
import glob
import time
from dataclasses import dataclass

import numpy as np
import torch
import bitsandbytes as bnb
from torch import nn
import torch.nn.functional as F
import torch.distributed as dist
import torch._inductor.config as config
from torch.nn.parallel import DistributedDataParallel as DDP

import pgptlformer

#wacky env stuff:
#import tritonpathsetter
#tritonpathsetter.set_cuda_paths()
#tritonpathsetter.add_cuda_files()

### modded-nanogpt distributed dataset loader
# -----------------------------------------------------------------------------
# their simple Distributed Data Loader

def _peek_data_shard(filename):
    # only reads the header, returns header data
    with open(filename, "rb") as f:
        # first read the header, which is 256 int32 integers (4 bytes each)
        header = np.frombuffer(f.read(256*4), dtype=np.int32)
    if header[0] != 20240520:
        print("ERROR: magic number mismatch in the data .bin file!")
        print("---> HINT: Are you passing in a correct file with --input_bin?")
        print("---> HINT: Dataset encoding changed recently, re-run data prepro or refer again to README")
        print("---> HINT: For example re-run: `python dev/data/tinyshakespeare.py`, then re-try")
        exit(1)
    assert header[1] == 1, "unsupported version"
    ntok = header[2] # number of tokens (claimed)
    return ntok # for now just return the number of tokens

def _load_data_shard(filename):
    with open(filename, "rb") as f:
        # first read the header, which is 256 int32 integers (4 bytes each)
        header = np.frombuffer(f.read(256*4), dtype=np.int32)
        assert header[0] == 20240520, "magic number mismatch in the data .bin file"
        assert header[1] == 1, "unsupported version"
        ntok = header[2] # number of tokens (claimed)
        # the rest of it are tokens, stored as uint16
        tokens = np.frombuffer(f.read(), dtype=np.uint16)
    assert len(tokens) == ntok, "number of tokens read does not match header?"
    return tokens

class DistributedDataLoader:
    def __init__(self, filename_pattern, B, T, process_rank, num_processes):
        self.process_rank = process_rank
        self.num_processes = num_processes
        self.B = B
        self.T = T

        # glob files that match the pattern
        self.files = sorted(glob.glob(filename_pattern))
        assert len(self.files) > 0, f"did not find any files that match the pattern {filename_pattern}"

        # load and validate all data shards, count number of tokens in total
        ntok_total = 0
        for fname in self.files:
            shard_ntok = _peek_data_shard(fname)
            assert shard_ntok >= num_processes * B * T + 1
            ntok_total += int(shard_ntok)
        self.ntok_total = ntok_total

        # kick things off
        self.reset()

    def reset(self):
        self.current_shard = 0
        self.current_position = self.process_rank * self.B * self.T
        self.tokens = _load_data_shard(self.files[self.current_shard])

    def advance(self): # advance to next data shard
        self.current_shard = (self.current_shard + 1) % len(self.files)
        self.current_position = self.process_rank * self.B * self.T
        self.tokens = _load_data_shard(self.files[self.current_shard])

    def next_batch(self):
        B = self.B
        T = self.T
        buf = self.tokens[self.current_position : self.current_position+B*T+1]
        buf = torch.tensor(buf.astype(np.int32), dtype=torch.long)
        x = (buf[:-1]).view(B, T) # inputs
        y = (buf[1:]).view(B, T) # targets
        # advance current position and load next shard if necessary
        self.current_position += B * T * self.num_processes
        if self.current_position + (B * T * self.num_processes + 1) > len(self.tokens):
            self.advance()
        return x.cuda(), y.cuda()

# -----------------------------------------------------------------------------

# downgrade to poor man's data loader:
# maybe superfluous bc distributed data loader started working
# delete? [ ]
def get_batch(split):
    # We recreate np.memmap every batch to avoid a memory leak, as per
    # https://stackoverflow.com/questions/45132940/numpy-memmap-memory-usage-want-to-iterate-once/61472122#61472122
    block_size = args.sequence_length
    batch_size = args.batch_size
    if split == 'train':
        data = np.memmap(os.path.join(data_dir, 'train.bin'), dtype=np.uint16, mode='r')
    else:
        data = np.memmap(os.path.join(data_dir, 'val.bin'), dtype=np.uint16, mode='r')
    ix = torch.randint(len(data) - block_size, (batch_size,))
    x = torch.stack([torch.from_numpy((data[i:i+block_size]).astype(np.int64)) for i in ix])
    y = torch.stack([torch.from_numpy((data[i+1:i+1+block_size]).astype(np.int64)) for i in ix])
    if device_type == 'cuda':
        # pin arrays x,y, which allows us to move them to GPU asynchronously (non_blocking=True)
        x, y = x.pin_memory().to(device, non_blocking=True), y.pin_memory().to(device, non_blocking=True)
    else:
        x, y = x.to(device), y.to(device)
    return x, y

### modded-nanogpt
### either 24/16*20=30 batches per 4090 or 24/32*20=15 batches per 4090, 
### depending on what kind of v100 tinystories used. 
@dataclass
class Hyperparameters:
    # data hyperparams
    input_bin : str = 'data/tinystories-pqt/tinystories-pqt_train_*.bin' # input .bin to train on
    input_val_bin : str = 'data/tinystories-pqt/tinystories-pqt_val_*.bin' # input .bin to eval validation loss on
    # optimization hyperparams
    batch_size : int = 4*32 # macrobatch size, in sequences, across all devices
    device_batch_size : int = 32 # batch size, in sequences, per device. try to increase/decrease by powers of 2
    sequence_length : int = 512 # sequence length, in tokens
    num_iterations : int = 6250 # number of iterations to run
    attack : int = 40 # 2*(1-betas)^-1
    release : int = 256 # number of iterations of linear warmup/warmdown for triangular or trapezoidal schedule
    weight_decay : float = 0
    # evaluation and logging hyperparams
    val_loss_every : int = 2000 # every how many steps to evaluate val loss? 0 for only at the end
    val_tokens : int = 5242880 # how many tokens of validation data? it's important to keep this fixed for consistent comparisons
    save_every : int = 0 # every how many steps to save the checkpoint? 0 for only at the end
    run_name : str = "re-pqt-rmsXrms-ATTNII_rev2"
    # supercompute boilerplate
    ddp_run : bool = False #this stuff is so nyannoying
    device = "cuda" # examples: 'cpu', 'cuda', 'cuda:0', 'cuda:1' etc., or try 'mps' on macbooks
    torch_compile = True   #hahahaha
    use_z_loss = True
    z_loss_coefficient = 1e-4
args = Hyperparameters()

# convenience variables
B, T = args.device_batch_size, args.sequence_length
# set up DDP (distributed data parallel). torchrun sets this env variable
assert torch.cuda.is_available()

if args.ddp_run == True:
    dist.init_process_group(backend='nccl')
    ddp_rank = int(os.environ['RANK'])
    ddp_local_rank = int(os.environ['LOCAL_RANK'])
    ddp_world_size = int(os.environ['WORLD_SIZE'])
    device = f'cuda:{ddp_local_rank}'
    torch.cuda.set_device(device)
    print(f"using device: {device}")
    master_process = (ddp_rank == 0) # this process will do logging, checkpointing etc.
else:
    # if not ddp, we are running on a single gpu, and one process
    master_process = True
    seed_offset = 0
    ddp_world_size = 1
    ddp_rank = 0
    device = args.device
#tokens_per_iter = train_accumulation_steps * ddp_world_size * batch_size * block_size
#print(f"tokens per iteration will be: {tokens_per_iter:,}")


# calculate the number of steps to take in the val loop.
assert args.val_tokens % (B * T * ddp_world_size) == 0
val_steps = args.val_tokens // (B * T * ddp_world_size)
# calculate the steps of gradient accumulation required to attain the desired global batch size.
assert args.batch_size % (B * ddp_world_size) == 0
train_accumulation_steps = args.batch_size // (B * ddp_world_size)

# load tokens 
train_loader = DistributedDataLoader(args.input_bin, B, T, ddp_rank, ddp_world_size)
val_loader = DistributedDataLoader(args.input_val_bin, B, T, ddp_rank, ddp_world_size)
if master_process:
    print(f"Training DataLoader: total number of tokens: {train_loader.ntok_total} across {len(train_loader.files)} files")
    print(f"Validation DataLoader: total number of tokens: {val_loader.ntok_total} across {len(val_loader.files)} files")
x, y = train_loader.next_batch()

if master_process:
    print("Building model...")

#tinystories
#num_vocab=50304 for non-tinystories models
#qknorm="identitynorm" for nonqknorm models
layer_prefab = {"dim":256,"dim_head":32,"headcount":8,"ff_mult":4, 
"lambda":True,"layerwisenorm":"rmsnorm","qknorm":"dynamic_shape_rmsnorm", 
"attention_deux":True, "training_seqlen":args.sequence_length}
#global_prefab = {"vocab_size":8192, "num_layers":4}
#weird errors
global_prefab = {"vocab_size":50304, "num_layers":4}
config = {}
config.update(layer_prefab)
config.update(global_prefab)

model = pgptlformer.PGPT_Lformer(config)
if hasattr(config, "coordinate_descent_tuning"):
    config.coordinate_descent_tuning = True # suggested by @Chillee
model = model.to(device)
if args.torch_compile:
    model = torch.compile(model)

# here we wrap model into DDP container
if args.ddp_run:
    model = DDP(model, device_ids=[ddp_local_rank])
#raw_model = model.modules() # always contains the "raw" unwrapped model
ctx = torch.amp.autocast(device_type='cuda', dtype=torch.bfloat16)

if master_process:
    print("Model built.")

# CUDNN attention is ~4ms faster than Flash, but doesn't get selected by default in PyTorch 2.5.1
from torch.backends.cuda import enable_cudnn_sdp, enable_flash_sdp, enable_math_sdp, enable_mem_efficient_sdp
enable_cudnn_sdp(True)
enable_flash_sdp(True)
enable_mem_efficient_sdp(True)
enable_math_sdp(False)

# modded-nanogpt optimizer inits
adam1 = torch.optim.Adam([model.lambdaformer.what_the_embedder_doin.weight], lr=0.3,    betas=(0.9, 0.95) )
adam2 = torch.optim.Adam([model.tokenpicker_head.weight],                    lr=0.002,  betas=(0.9, 0.95) )
params = list(model.lambdaformer.blocks.parameters())
matrix_params = [p for p in params if p.ndim == 2]
scalar_params = [p for p in params if p.ndim < 2]
adam3 = bnb.optim.Adam8bit(matrix_params, lr=0.02, betas=(0.9, 0.95) ) #tune this, sensitive
adam4 = bnb.optim.Adam8bit(scalar_params, lr=0.02, betas=(0.9, 0.95) ) #???, less sensitive
optim_ensemble = [adam1, adam2, adam3, adam4]

# lr scheduler
def get_ASR_env(it):
    assert it <= args.num_iterations
    # A) the famous linear warmup back at it again
    if it < args.attack:
        return (it+1) / args.attack 
    # S) constant sustain
    elif it < args.num_iterations - args.release:
        return 1.0
    # R) release
    else:
        release_ratio = (args.num_iterations - it) / args.release
        return release_ratio

schedulers = [torch.optim.lr_scheduler.LambdaLR(opt, get_ASR_env) for opt in optim_ensemble]

# begin logging
if master_process:
    run_id = str(uuid.uuid4())
    if args.run_name is not None:
        sep="-"
        run_id = sep.join([args.run_name, run_id])
    
    logdir = 'logs/%s/' % run_id
    os.makedirs(logdir, exist_ok=True)
    logfile = 'logs/%s.txt' % run_id
    # create the log file
    with open(logfile, "w") as f:
        # begin the log by printing this file (the Python code)
        f.write('='*100 + '\n')
        f.write(code)
        f.write('='*100 + '\n')
        # log information about the hardware/software environment this is running on
        # and print the full `nvidia-smi` to file
        f.write(f"Running pytorch {torch.version.__version__} compiled for CUDA {torch.version.cuda}\nnvidia-smi:\n")
        import subprocess
        result = subprocess.run(['nvidia-smi'], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
        f.write(f'{result.stdout}\n')
        f.write('='*100 + '\n')

training_time_ms = 0
# start the clock
torch.cuda.synchronize()
t0 = time.time()
# begin training
train_loader.reset()

for step in range(args.num_iterations + 1):
    last_step = (step == args.num_iterations)
    # This effectively ignores timing first 10 steps, which are slower for weird reasons.
    # Alternately, and slightly more correctly in terms of benchmarking, we could do 10
    # steps with dummy data first, and then re-initialize the model and reset the loader.
    if step == 10:
        training_time_ms = 0
        t0 = time.time()
    timed_steps = float('nan') if step <= 11 else (step - 10) + 1 # <= 11 to avoid bug in val

    # once in a while evaluate the validation dataset
    if (last_step or (args.val_loss_every > 0 and step % args.val_loss_every == 0)):
        # stop the clock
        torch.cuda.synchronize()
        training_time_ms += 1000 * (time.time() - t0)
        # run validation batches
        model.eval()
        val_loader.reset()
        val_loss = 0.0
        val_aux_loss = 0.0
        for _ in range(val_steps):
            x_val, y_val = val_loader.next_batch()
            with ctx: # of course, we'd like to use no_grad() here too, but that creates a torch.compile error for some reason
                _, loss, z_loss = model(x_val, y_val, return_logits=False, return_zloss=args.use_z_loss)
                val_loss += loss.detach()
                if z_loss is not None:
                    val_aux_loss += z_loss.detach()*args.z_loss_coefficient
                del loss, z_loss
        if args.ddp_run:
            dist.all_reduce(val_loss, op=dist.ReduceOp.AVG)
            dist.all_reduce(val_aux_loss, op=dist.ReduceOp.AVG)
        val_loss /= val_steps
        val_aux_loss /= val_steps
        # log val loss to console and to logfile
        if master_process:
            print(f'step:{step}/{args.num_iterations} val_loss:{val_loss:.4f} val_aux_loss:{val_aux_loss:.4f} train_time:{training_time_ms:.0f}ms step_avg:{training_time_ms/(timed_steps-1):.2f}ms')
            with open(logfile, "a") as f:
                f.write(f'step:{step}/{args.num_iterations} val_loss:{val_loss:.4f} val_aux_loss:{val_aux_loss:.4f} train_time:{training_time_ms:.0f}ms step_avg:{training_time_ms/(timed_steps-1):.2f}ms\n')
        # start the clock again
        torch.cuda.synchronize()
        t0 = time.time()

        if master_process and (last_step or (args.save_every > 0 and step % args.save_every == 0)):
            # stop the clock
            torch.cuda.synchronize()
            training_time_ms += 1000 * (time.time() - t0)
            # save the state of the training process
            log = dict(step=step, code=code, model=model.state_dict(), model_args=config, optim_ensemble=[opt.state_dict() for opt in optim_ensemble])
            torch.save(log, 'logs/%s/state_step%06d.pt' % (run_id, step))
            # start the clock again
            torch.cuda.synchronize()
            t0 = time.time()

    # bit confusing: we want to make sure to eval on 0th iteration
    # but also after the very last iteration. so we loop for step <= num_iterations
    # instead of just < num_iterations (one extra due to <=), only to do
    # the validation/sampling one last time, and then we break right here as we're done.
    if last_step:
        break

    # --- train time ---
    model.train()
    for i in range(1, train_accumulation_steps+1):
        # forward pass
        with ctx:
            _, loss, z_loss = model(x, y, return_logits=False, return_zloss=args.use_z_loss)
            train_loss = loss.detach()
            if z_loss is not None:
                train_aux_loss = z_loss.detach()*args.z_loss_coefficient
                loss = loss+z_loss*args.z_loss_coefficient
            else:
                train_aux_loss = 0
        # advance the dataset for the next batch
        x, y = train_loader.next_batch()
        # backward pass
        if args.ddp_run:
            if i < train_accumulation_steps:
                with model.no_sync(): # there's no need to sync gradients every accumulation step
                    loss.backward()
        else:
            loss.backward() # just sync on the last step

    for p in model.parameters():    #grad accum normalization?
        p.grad /= train_accumulation_steps
    # skip muon momentum warmup since we're adaming it
    #...
    # step the optimizers and schedulers
    for opt, sched in zip(optim_ensemble, schedulers):
        opt.step()
        sched.step()
    # null the gradients
    model.zero_grad(set_to_none=True)
    # --- train time is already over ---

     #dist.all_reduce(train_loss, op=dist.ReduceOp.AVG) # all-reducing the training loss would be more correct in terms of logging, but slower
    if master_process:
        approx_time = training_time_ms + 1000 * (time.time() - t0)
        print(f"step:{step+1}/{args.num_iterations} train_loss:{train_loss.item():.4f} aux_loss:{train_aux_loss.item():.4f} train_time:{approx_time:.0f}ms step_avg:{approx_time/timed_steps:.2f}ms")
        with open(logfile, "a") as f:
            f.write(f"step:{step+1}/{args.num_iterations} train_loss:{train_loss.item():.4f} aux_loss:{train_aux_loss.item():.4f} train_time:{approx_time:.0f}ms step_avg:{approx_time/timed_steps:.2f}ms\n")
if master_process:
    print(f"peak memory consumption: {torch.cuda.max_memory_allocated() // 1024 // 1024} MiB")

# clean up nice
if args.ddp_run:
    dist.destroy_process_group()====================================================================================================
Running pytorch 2.5.1+cu124 compiled for CUDA 12.4
nvidia-smi:
Sun Feb  2 02:21:32 2025       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 551.61                 Driver Version: 551.61         CUDA Version: 12.4     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                     TCC/WDDM  | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 4090      WDDM  |   00000000:01:00.0  On |                  Off |
|  0%   47C    P2             67W /  350W |    1483MiB /  24564MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
                                                                                         
+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A      1496    C+G   C:\Windows\System32\dwm.exe                 N/A      |
|    0   N/A  N/A      2624    C+G   ...-desktop\RemoteDesktopCompanion.exe      N/A      |
|    0   N/A  N/A      2976    C+G   ...B\system_tray\lghub_system_tray.exe      N/A      |
|    0   N/A  N/A      3556    C+G   C:\Windows\explorer.exe                     N/A      |
|    0   N/A  N/A      4288    C+G   ...CBS_cw5n1h2txyewy\TextInputHost.exe      N/A      |
|    0   N/A  N/A      6564    C+G   ...ekyb3d8bbwe\PhoneExperienceHost.exe      N/A      |
|    0   N/A  N/A      7444    C+G   C:\Program Files\VSCodium\VSCodium.exe      N/A      |
|    0   N/A  N/A      8884    C+G   ...1.0_x64__8wekyb3d8bbwe\Video.UI.exe      N/A      |
|    0   N/A  N/A      9636    C+G   ....Search_cw5n1h2txyewy\SearchApp.exe      N/A      |
|    0   N/A  N/A     12332    C+G   ...al\Discord\app-1.0.9180\Discord.exe      N/A      |
|    0   N/A  N/A     12688    C+G   ...cal\Microsoft\OneDrive\OneDrive.exe      N/A      |
|    0   N/A  N/A     13000    C+G   ...64__8wekyb3d8bbwe\CalculatorApp.exe      N/A      |
|    0   N/A  N/A     14168      C   ...rograms\Python\Python310\python.exe      N/A      |
|    0   N/A  N/A     16068    C+G   ...\cef\cef.win7x64\steamwebhelper.exe      N/A      |
|    0   N/A  N/A     18804    C+G   ...a\Local\slack\app-4.41.98\slack.exe      N/A      |
|    0   N/A  N/A     19876    C+G   ...a\Local\slack\app-4.41.98\slack.exe      N/A      |
|    0   N/A  N/A     24308    C+G   ...5n1h2txyewy\ShellExperienceHost.exe      N/A      |
+-----------------------------------------------------------------------------------------+

====================================================================================================
step:0/6250 val_loss:10.8259 val_aux_loss:0.0421 train_time:55ms step_avg:nanms
step:1/6250 train_loss:10.8258 aux_loss:0.0421 train_time:9629ms step_avg:nanms
step:2/6250 train_loss:10.8239 aux_loss:0.0421 train_time:9761ms step_avg:nanms
step:3/6250 train_loss:10.8120 aux_loss:0.0421 train_time:9891ms step_avg:nanms
step:4/6250 train_loss:10.8025 aux_loss:0.0421 train_time:10020ms step_avg:nanms
step:5/6250 train_loss:10.7488 aux_loss:0.0420 train_time:10149ms step_avg:nanms
step:6/6250 train_loss:10.6739 aux_loss:0.0418 train_time:10279ms step_avg:nanms
step:7/6250 train_loss:10.5698 aux_loss:0.0416 train_time:10408ms step_avg:nanms
step:8/6250 train_loss:10.4720 aux_loss:0.0414 train_time:10536ms step_avg:nanms
step:9/6250 train_loss:10.3021 aux_loss:0.0410 train_time:10667ms step_avg:nanms
step:10/6250 train_loss:10.1203 aux_loss:0.0406 train_time:10797ms step_avg:nanms
step:11/6250 train_loss:9.9528 aux_loss:0.0403 train_time:131ms step_avg:nanms
step:12/6250 train_loss:9.7262 aux_loss:0.0399 train_time:265ms step_avg:nanms
step:13/6250 train_loss:9.4927 aux_loss:0.0395 train_time:398ms step_avg:132.71ms
step:14/6250 train_loss:9.2378 aux_loss:0.0390 train_time:530ms step_avg:132.53ms
step:15/6250 train_loss:8.9773 aux_loss:0.0386 train_time:659ms step_avg:131.82ms
step:16/6250 train_loss:8.7440 aux_loss:0.0384 train_time:793ms step_avg:132.19ms
step:17/6250 train_loss:8.5183 aux_loss:0.0386 train_time:926ms step_avg:132.30ms
step:18/6250 train_loss:8.2471 aux_loss:0.0384 train_time:1055ms step_avg:131.89ms
step:19/6250 train_loss:7.8890 aux_loss:0.0373 train_time:1189ms step_avg:132.12ms
step:20/6250 train_loss:7.6552 aux_loss:0.0374 train_time:1322ms step_avg:132.21ms
step:21/6250 train_loss:7.4059 aux_loss:0.0372 train_time:1459ms step_avg:132.65ms
step:22/6250 train_loss:7.1355 aux_loss:0.0373 train_time:1598ms step_avg:133.19ms
step:23/6250 train_loss:7.0230 aux_loss:0.0378 train_time:1738ms step_avg:133.72ms
step:24/6250 train_loss:6.7303 aux_loss:0.0375 train_time:1879ms step_avg:134.24ms
step:25/6250 train_loss:6.6080 aux_loss:0.0377 train_time:2020ms step_avg:134.69ms
step:26/6250 train_loss:6.5241 aux_loss:0.0399 train_time:2160ms step_avg:135.02ms
step:27/6250 train_loss:6.3658 aux_loss:0.0394 train_time:2299ms step_avg:135.26ms
step:28/6250 train_loss:6.2951 aux_loss:0.0391 train_time:2440ms step_avg:135.58ms
step:29/6250 train_loss:6.1911 aux_loss:0.0397 train_time:2580ms step_avg:135.81ms
step:30/6250 train_loss:6.1837 aux_loss:0.0390 train_time:2720ms step_avg:136.02ms
step:31/6250 train_loss:6.0650 aux_loss:0.0371 train_time:2862ms step_avg:136.30ms
step:32/6250 train_loss:6.0454 aux_loss:0.0379 train_time:3003ms step_avg:136.51ms
step:33/6250 train_loss:6.0424 aux_loss:0.0381 train_time:3145ms step_avg:136.75ms
step:34/6250 train_loss:5.9355 aux_loss:0.0377 train_time:3286ms step_avg:136.91ms
step:35/6250 train_loss:12.9989 aux_loss:0.0541 train_time:3427ms step_avg:137.08ms
step:36/6250 train_loss:10.5944 aux_loss:0.0429 train_time:3567ms step_avg:137.19ms
step:37/6250 train_loss:7.3546 aux_loss:0.0372 train_time:3708ms step_avg:137.33ms
step:38/6250 train_loss:6.3961 aux_loss:0.0423 train_time:3850ms step_avg:137.50ms
step:39/6250 train_loss:6.0497 aux_loss:0.0368 train_time:3991ms step_avg:137.62ms
step:40/6250 train_loss:5.9997 aux_loss:0.0392 train_time:4132ms step_avg:137.73ms
step:41/6250 train_loss:6.1400 aux_loss:0.0396 train_time:4273ms step_avg:137.84ms
step:42/6250 train_loss:6.0656 aux_loss:0.0412 train_time:4415ms step_avg:137.96ms
step:43/6250 train_loss:5.9784 aux_loss:0.0405 train_time:4555ms step_avg:138.02ms
step:44/6250 train_loss:5.9702 aux_loss:0.0382 train_time:4696ms step_avg:138.11ms
step:45/6250 train_loss:5.9038 aux_loss:0.0380 train_time:4838ms step_avg:138.22ms
step:46/6250 train_loss:5.8988 aux_loss:0.0366 train_time:4979ms step_avg:138.30ms
step:47/6250 train_loss:5.8939 aux_loss:0.0372 train_time:5120ms step_avg:138.39ms
step:48/6250 train_loss:5.9546 aux_loss:0.0368 train_time:5261ms step_avg:138.46ms
step:49/6250 train_loss:5.9193 aux_loss:0.0379 train_time:5402ms step_avg:138.52ms
step:50/6250 train_loss:5.9416 aux_loss:0.0375 train_time:5543ms step_avg:138.58ms
step:51/6250 train_loss:5.9372 aux_loss:0.0385 train_time:5684ms step_avg:138.65ms
step:52/6250 train_loss:5.8653 aux_loss:0.0384 train_time:5825ms step_avg:138.70ms
step:53/6250 train_loss:5.9014 aux_loss:0.0391 train_time:5972ms step_avg:138.90ms
step:54/6250 train_loss:5.8766 aux_loss:0.0393 train_time:6111ms step_avg:138.90ms
step:55/6250 train_loss:5.8818 aux_loss:0.0396 train_time:6252ms step_avg:138.94ms
step:56/6250 train_loss:5.9232 aux_loss:0.0393 train_time:6395ms step_avg:139.03ms
step:57/6250 train_loss:5.9100 aux_loss:0.0395 train_time:6533ms step_avg:139.01ms
step:58/6250 train_loss:5.8922 aux_loss:0.0390 train_time:6670ms step_avg:138.97ms
step:59/6250 train_loss:5.8875 aux_loss:0.0388 train_time:6800ms step_avg:138.79ms
step:60/6250 train_loss:5.8547 aux_loss:0.0387 train_time:6931ms step_avg:138.63ms
step:61/6250 train_loss:5.8666 aux_loss:0.0385 train_time:7064ms step_avg:138.51ms
step:62/6250 train_loss:5.9603 aux_loss:0.0384 train_time:7203ms step_avg:138.52ms
step:63/6250 train_loss:5.8821 aux_loss:0.0383 train_time:7334ms step_avg:138.38ms
step:64/6250 train_loss:5.8092 aux_loss:0.0382 train_time:7464ms step_avg:138.22ms
step:65/6250 train_loss:5.9189 aux_loss:0.0382 train_time:7599ms step_avg:138.16ms
step:66/6250 train_loss:5.8861 aux_loss:0.0383 train_time:7733ms step_avg:138.09ms
step:67/6250 train_loss:5.9101 aux_loss:0.0384 train_time:7871ms step_avg:138.09ms
step:68/6250 train_loss:5.8838 aux_loss:0.0385 train_time:8008ms step_avg:138.07ms
step:69/6250 train_loss:5.9100 aux_loss:0.0386 train_time:8146ms step_avg:138.07ms
step:70/6250 train_loss:5.8773 aux_loss:0.0386 train_time:8282ms step_avg:138.04ms
step:71/6250 train_loss:5.8783 aux_loss:0.0387 train_time:8415ms step_avg:137.95ms
step:72/6250 train_loss:5.8890 aux_loss:0.0387 train_time:8545ms step_avg:137.82ms
step:73/6250 train_loss:5.8704 aux_loss:0.0387 train_time:8678ms step_avg:137.74ms
step:74/6250 train_loss:5.9233 aux_loss:0.0387 train_time:8813ms step_avg:137.70ms
step:75/6250 train_loss:5.9103 aux_loss:0.0386 train_time:8948ms step_avg:137.66ms
step:76/6250 train_loss:5.9569 aux_loss:0.0386 train_time:9082ms step_avg:137.61ms
step:77/6250 train_loss:5.8366 aux_loss:0.0385 train_time:9213ms step_avg:137.51ms
step:78/6250 train_loss:5.8977 aux_loss:0.0384 train_time:9343ms step_avg:137.40ms
step:79/6250 train_loss:5.8869 aux_loss:0.0384 train_time:9476ms step_avg:137.34ms
step:80/6250 train_loss:5.8443 aux_loss:0.0383 train_time:9615ms step_avg:137.36ms
step:81/6250 train_loss:5.8733 aux_loss:0.0384 train_time:9746ms step_avg:137.27ms
step:82/6250 train_loss:5.8726 aux_loss:0.0384 train_time:9877ms step_avg:137.19ms
step:83/6250 train_loss:5.8389 aux_loss:0.0384 train_time:10007ms step_avg:137.09ms
step:84/6250 train_loss:5.8672 aux_loss:0.0385 train_time:10143ms step_avg:137.07ms
step:85/6250 train_loss:5.9401 aux_loss:0.0385 train_time:10278ms step_avg:137.04ms
step:86/6250 train_loss:5.8412 aux_loss:0.0388 train_time:10411ms step_avg:136.99ms
step:87/6250 train_loss:5.8835 aux_loss:0.0385 train_time:10542ms step_avg:136.90ms
step:88/6250 train_loss:5.8715 aux_loss:0.0384 train_time:10672ms step_avg:136.81ms
step:89/6250 train_loss:5.8867 aux_loss:0.0384 train_time:10803ms step_avg:136.74ms
step:90/6250 train_loss:5.8251 aux_loss:0.0383 train_time:10933ms step_avg:136.66ms
step:91/6250 train_loss:5.7956 aux_loss:0.0383 train_time:11062ms step_avg:136.56ms
step:92/6250 train_loss:5.7817 aux_loss:0.0387 train_time:11193ms step_avg:136.50ms
step:93/6250 train_loss:5.8162 aux_loss:0.0384 train_time:11322ms step_avg:136.41ms
step:94/6250 train_loss:5.7899 aux_loss:0.0381 train_time:11452ms step_avg:136.33ms
step:95/6250 train_loss:5.6707 aux_loss:0.0382 train_time:11583ms step_avg:136.27ms
step:96/6250 train_loss:5.7278 aux_loss:0.0380 train_time:11713ms step_avg:136.19ms
step:97/6250 train_loss:5.7759 aux_loss:0.0384 train_time:11843ms step_avg:136.12ms
step:98/6250 train_loss:5.6770 aux_loss:0.0388 train_time:11974ms step_avg:136.06ms
step:99/6250 train_loss:5.6646 aux_loss:0.0389 train_time:12106ms step_avg:136.02ms
step:100/6250 train_loss:5.6833 aux_loss:0.0388 train_time:12237ms step_avg:135.96ms
step:101/6250 train_loss:5.6469 aux_loss:0.0387 train_time:12368ms step_avg:135.91ms
step:102/6250 train_loss:5.6562 aux_loss:0.0386 train_time:12500ms step_avg:135.87ms
step:103/6250 train_loss:5.6225 aux_loss:0.0388 train_time:12631ms step_avg:135.82ms
step:104/6250 train_loss:5.6395 aux_loss:0.0390 train_time:12764ms step_avg:135.79ms
step:105/6250 train_loss:5.5740 aux_loss:0.0393 train_time:12895ms step_avg:135.74ms
step:106/6250 train_loss:5.6328 aux_loss:0.0394 train_time:13028ms step_avg:135.71ms
step:107/6250 train_loss:5.5673 aux_loss:0.0393 train_time:13159ms step_avg:135.66ms
step:108/6250 train_loss:5.4874 aux_loss:0.0392 train_time:13291ms step_avg:135.63ms
step:109/6250 train_loss:5.4741 aux_loss:0.0392 train_time:13422ms step_avg:135.58ms
step:110/6250 train_loss:5.3949 aux_loss:0.0394 train_time:13553ms step_avg:135.53ms
step:111/6250 train_loss:5.4279 aux_loss:0.0395 train_time:13684ms step_avg:135.49ms
step:112/6250 train_loss:5.4245 aux_loss:0.0396 train_time:13815ms step_avg:135.44ms
step:113/6250 train_loss:5.3276 aux_loss:0.0400 train_time:13955ms step_avg:135.49ms
step:114/6250 train_loss:5.2533 aux_loss:0.0401 train_time:14096ms step_avg:135.54ms
step:115/6250 train_loss:5.2957 aux_loss:0.0402 train_time:14227ms step_avg:135.49ms
step:116/6250 train_loss:5.2519 aux_loss:0.0406 train_time:14357ms step_avg:135.44ms
step:117/6250 train_loss:5.1907 aux_loss:0.0405 train_time:14487ms step_avg:135.39ms
step:118/6250 train_loss:5.1511 aux_loss:0.0404 train_time:14627ms step_avg:135.43ms
step:119/6250 train_loss:5.2115 aux_loss:0.0408 train_time:14770ms step_avg:135.50ms
step:120/6250 train_loss:5.1255 aux_loss:0.0411 train_time:14902ms step_avg:135.47ms
step:121/6250 train_loss:5.0767 aux_loss:0.0411 train_time:15032ms step_avg:135.42ms
step:122/6250 train_loss:5.0682 aux_loss:0.0412 train_time:15163ms step_avg:135.38ms
step:123/6250 train_loss:4.9806 aux_loss:0.0414 train_time:15294ms step_avg:135.35ms
step:124/6250 train_loss:4.9670 aux_loss:0.0416 train_time:15436ms step_avg:135.40ms
step:125/6250 train_loss:4.9340 aux_loss:0.0417 train_time:15578ms step_avg:135.46ms
step:126/6250 train_loss:5.0706 aux_loss:0.0418 train_time:15711ms step_avg:135.44ms
step:127/6250 train_loss:4.8906 aux_loss:0.0421 train_time:15853ms step_avg:135.49ms
step:128/6250 train_loss:4.9020 aux_loss:0.0426 train_time:15986ms step_avg:135.47ms
step:129/6250 train_loss:4.8228 aux_loss:0.0427 train_time:16128ms step_avg:135.53ms
step:130/6250 train_loss:4.8081 aux_loss:0.0427 train_time:16269ms step_avg:135.57ms
step:131/6250 train_loss:4.7898 aux_loss:0.0430 train_time:16404ms step_avg:135.57ms
step:132/6250 train_loss:4.8331 aux_loss:0.0429 train_time:16546ms step_avg:135.62ms
step:133/6250 train_loss:4.7634 aux_loss:0.0435 train_time:16682ms step_avg:135.63ms
step:134/6250 train_loss:4.8317 aux_loss:0.0433 train_time:16814ms step_avg:135.60ms
step:135/6250 train_loss:4.6142 aux_loss:0.0434 train_time:16946ms step_avg:135.57ms
step:136/6250 train_loss:4.6322 aux_loss:0.0439 train_time:17083ms step_avg:135.58ms
step:137/6250 train_loss:4.6407 aux_loss:0.0440 train_time:17218ms step_avg:135.58ms
step:138/6250 train_loss:4.5743 aux_loss:0.0440 train_time:17357ms step_avg:135.60ms
step:139/6250 train_loss:4.6063 aux_loss:0.0444 train_time:17489ms step_avg:135.58ms
step:140/6250 train_loss:4.6150 aux_loss:0.0443 train_time:17619ms step_avg:135.53ms
step:141/6250 train_loss:4.5034 aux_loss:0.0446 train_time:17753ms step_avg:135.52ms
step:142/6250 train_loss:4.4995 aux_loss:0.0447 train_time:17892ms step_avg:135.55ms
step:143/6250 train_loss:4.5069 aux_loss:0.0447 train_time:18024ms step_avg:135.52ms
step:144/6250 train_loss:4.4703 aux_loss:0.0452 train_time:18154ms step_avg:135.48ms
step:145/6250 train_loss:4.5203 aux_loss:0.0453 train_time:18285ms step_avg:135.44ms
step:146/6250 train_loss:4.4667 aux_loss:0.0454 train_time:18415ms step_avg:135.40ms
step:147/6250 train_loss:4.4026 aux_loss:0.0459 train_time:18546ms step_avg:135.37ms
step:148/6250 train_loss:4.4245 aux_loss:0.0458 train_time:18682ms step_avg:135.38ms
step:149/6250 train_loss:4.3743 aux_loss:0.0461 train_time:18819ms step_avg:135.39ms
step:150/6250 train_loss:4.3211 aux_loss:0.0464 train_time:18957ms step_avg:135.41ms
step:151/6250 train_loss:4.4053 aux_loss:0.0461 train_time:19095ms step_avg:135.43ms
step:152/6250 train_loss:4.3320 aux_loss:0.0466 train_time:19228ms step_avg:135.41ms
step:153/6250 train_loss:4.3041 aux_loss:0.0469 train_time:19358ms step_avg:135.37ms
step:154/6250 train_loss:4.3132 aux_loss:0.0468 train_time:19494ms step_avg:135.38ms
step:155/6250 train_loss:4.2537 aux_loss:0.0471 train_time:19631ms step_avg:135.39ms
step:156/6250 train_loss:4.2057 aux_loss:0.0471 train_time:19772ms step_avg:135.43ms
step:157/6250 train_loss:4.2493 aux_loss:0.0470 train_time:19910ms step_avg:135.44ms
step:158/6250 train_loss:4.2628 aux_loss:0.0475 train_time:20048ms step_avg:135.46ms
step:159/6250 train_loss:4.2400 aux_loss:0.0479 train_time:20185ms step_avg:135.47ms
step:160/6250 train_loss:4.2518 aux_loss:0.0476 train_time:20317ms step_avg:135.45ms
step:161/6250 train_loss:4.2902 aux_loss:0.0478 train_time:20448ms step_avg:135.42ms
step:162/6250 train_loss:4.1632 aux_loss:0.0482 train_time:20578ms step_avg:135.38ms
step:163/6250 train_loss:4.1414 aux_loss:0.0485 train_time:20715ms step_avg:135.39ms
step:164/6250 train_loss:4.2104 aux_loss:0.0487 train_time:20847ms step_avg:135.37ms
step:165/6250 train_loss:4.0959 aux_loss:0.0488 train_time:20977ms step_avg:135.33ms
step:166/6250 train_loss:4.1172 aux_loss:0.0488 train_time:21107ms step_avg:135.30ms
step:167/6250 train_loss:4.1600 aux_loss:0.0486 train_time:21237ms step_avg:135.27ms
step:168/6250 train_loss:4.2874 aux_loss:0.0486 train_time:21368ms step_avg:135.24ms
step:169/6250 train_loss:4.1598 aux_loss:0.0487 train_time:21498ms step_avg:135.21ms
step:170/6250 train_loss:4.1307 aux_loss:0.0483 train_time:21629ms step_avg:135.18ms
step:171/6250 train_loss:4.0606 aux_loss:0.0487 train_time:21760ms step_avg:135.16ms
step:172/6250 train_loss:4.1056 aux_loss:0.0490 train_time:21891ms step_avg:135.13ms
step:173/6250 train_loss:4.0859 aux_loss:0.0487 train_time:22024ms step_avg:135.12ms
step:174/6250 train_loss:4.1365 aux_loss:0.0485 train_time:22165ms step_avg:135.15ms
step:175/6250 train_loss:4.0331 aux_loss:0.0487 train_time:22304ms step_avg:135.18ms
step:176/6250 train_loss:4.0761 aux_loss:0.0489 train_time:22444ms step_avg:135.21ms
step:177/6250 train_loss:4.0191 aux_loss:0.0491 train_time:22583ms step_avg:135.23ms
step:178/6250 train_loss:4.0401 aux_loss:0.0493 train_time:22722ms step_avg:135.25ms
step:179/6250 train_loss:3.9878 aux_loss:0.0489 train_time:22857ms step_avg:135.25ms
step:180/6250 train_loss:3.9655 aux_loss:0.0488 train_time:22988ms step_avg:135.22ms
step:181/6250 train_loss:3.9814 aux_loss:0.0490 train_time:23118ms step_avg:135.19ms
step:182/6250 train_loss:3.9360 aux_loss:0.0494 train_time:23248ms step_avg:135.16ms
step:183/6250 train_loss:3.9600 aux_loss:0.0497 train_time:23379ms step_avg:135.14ms
step:184/6250 train_loss:4.0056 aux_loss:0.0497 train_time:23509ms step_avg:135.11ms
step:185/6250 train_loss:3.9186 aux_loss:0.0498 train_time:23640ms step_avg:135.09ms
step:186/6250 train_loss:3.9554 aux_loss:0.0496 train_time:23772ms step_avg:135.07ms
step:187/6250 train_loss:3.9025 aux_loss:0.0495 train_time:23903ms step_avg:135.04ms
step:188/6250 train_loss:3.9602 aux_loss:0.0494 train_time:24034ms step_avg:135.02ms
step:189/6250 train_loss:3.9005 aux_loss:0.0495 train_time:24165ms step_avg:135.00ms
step:190/6250 train_loss:3.8958 aux_loss:0.0500 train_time:24296ms step_avg:134.98ms
step:191/6250 train_loss:3.9425 aux_loss:0.0499 train_time:24427ms step_avg:134.96ms
step:192/6250 train_loss:3.8730 aux_loss:0.0501 train_time:24557ms step_avg:134.93ms
step:193/6250 train_loss:3.9754 aux_loss:0.0500 train_time:24688ms step_avg:134.91ms
step:194/6250 train_loss:3.8843 aux_loss:0.0506 train_time:24819ms step_avg:134.89ms
step:195/6250 train_loss:3.9034 aux_loss:0.0500 train_time:24949ms step_avg:134.86ms
step:196/6250 train_loss:3.9200 aux_loss:0.0503 train_time:25080ms step_avg:134.84ms
step:197/6250 train_loss:3.7997 aux_loss:0.0502 train_time:25211ms step_avg:134.82ms
step:198/6250 train_loss:3.8402 aux_loss:0.0500 train_time:25342ms step_avg:134.80ms
step:199/6250 train_loss:3.8455 aux_loss:0.0507 train_time:25472ms step_avg:134.77ms
step:200/6250 train_loss:3.8667 aux_loss:0.0503 train_time:25607ms step_avg:134.77ms
step:201/6250 train_loss:3.8445 aux_loss:0.0507 train_time:25740ms step_avg:134.76ms
step:202/6250 train_loss:3.8051 aux_loss:0.0506 train_time:25872ms step_avg:134.75ms
step:203/6250 train_loss:3.8238 aux_loss:0.0504 train_time:26005ms step_avg:134.74ms
step:204/6250 train_loss:3.9173 aux_loss:0.0511 train_time:26137ms step_avg:134.73ms
step:205/6250 train_loss:3.8770 aux_loss:0.0505 train_time:26269ms step_avg:134.71ms
step:206/6250 train_loss:3.9096 aux_loss:0.0505 train_time:26401ms step_avg:134.70ms
step:207/6250 train_loss:3.7681 aux_loss:0.0511 train_time:26534ms step_avg:134.69ms
step:208/6250 train_loss:3.7987 aux_loss:0.0502 train_time:26685ms step_avg:134.77ms
step:209/6250 train_loss:3.8330 aux_loss:0.0503 train_time:26826ms step_avg:134.80ms
step:210/6250 train_loss:3.8075 aux_loss:0.0507 train_time:26965ms step_avg:134.83ms
step:211/6250 train_loss:3.7734 aux_loss:0.0507 train_time:27105ms step_avg:134.85ms
step:212/6250 train_loss:3.7872 aux_loss:0.0507 train_time:27246ms step_avg:134.88ms
step:213/6250 train_loss:3.7682 aux_loss:0.0509 train_time:27383ms step_avg:134.89ms
step:214/6250 train_loss:3.8338 aux_loss:0.0505 train_time:27513ms step_avg:134.87ms
step:215/6250 train_loss:3.7377 aux_loss:0.0505 train_time:27646ms step_avg:134.86ms
step:216/6250 train_loss:3.6915 aux_loss:0.0505 train_time:27778ms step_avg:134.84ms
step:217/6250 train_loss:3.7957 aux_loss:0.0505 train_time:27912ms step_avg:134.84ms
step:218/6250 train_loss:3.7846 aux_loss:0.0502 train_time:28052ms step_avg:134.86ms
step:219/6250 train_loss:3.7393 aux_loss:0.0507 train_time:28196ms step_avg:134.91ms
step:220/6250 train_loss:3.7489 aux_loss:0.0503 train_time:28337ms step_avg:134.94ms
step:221/6250 train_loss:3.7676 aux_loss:0.0502 train_time:28476ms step_avg:134.96ms
step:222/6250 train_loss:3.7763 aux_loss:0.0507 train_time:28616ms step_avg:134.98ms
step:223/6250 train_loss:3.7323 aux_loss:0.0507 train_time:28754ms step_avg:134.99ms
step:224/6250 train_loss:3.7605 aux_loss:0.0508 train_time:28893ms step_avg:135.02ms
step:225/6250 train_loss:3.7648 aux_loss:0.0506 train_time:29031ms step_avg:135.03ms
step:226/6250 train_loss:3.7536 aux_loss:0.0505 train_time:29163ms step_avg:135.01ms
step:227/6250 train_loss:3.7412 aux_loss:0.0503 train_time:29302ms step_avg:135.03ms
step:228/6250 train_loss:3.7129 aux_loss:0.0501 train_time:29442ms step_avg:135.05ms
step:229/6250 train_loss:3.6900 aux_loss:0.0499 train_time:29581ms step_avg:135.07ms
step:230/6250 train_loss:3.6918 aux_loss:0.0504 train_time:29720ms step_avg:135.09ms
step:231/6250 train_loss:3.7713 aux_loss:0.0496 train_time:29859ms step_avg:135.11ms
step:232/6250 train_loss:3.7234 aux_loss:0.0515 train_time:29999ms step_avg:135.13ms
step:233/6250 train_loss:3.7658 aux_loss:0.0506 train_time:30132ms step_avg:135.12ms
step:234/6250 train_loss:3.7641 aux_loss:0.0502 train_time:30263ms step_avg:135.10ms
step:235/6250 train_loss:3.5813 aux_loss:0.0511 train_time:30393ms step_avg:135.08ms
step:236/6250 train_loss:3.6714 aux_loss:0.0510 train_time:30524ms step_avg:135.06ms
step:237/6250 train_loss:3.7466 aux_loss:0.0496 train_time:30655ms step_avg:135.04ms
step:238/6250 train_loss:3.6340 aux_loss:0.0504 train_time:30787ms step_avg:135.03ms
step:239/6250 train_loss:3.6063 aux_loss:0.0512 train_time:30917ms step_avg:135.01ms
step:240/6250 train_loss:3.7820 aux_loss:0.0508 train_time:31047ms step_avg:134.99ms
step:241/6250 train_loss:3.7553 aux_loss:0.0500 train_time:31183ms step_avg:134.99ms
step:242/6250 train_loss:3.6533 aux_loss:0.0502 train_time:31318ms step_avg:134.99ms
step:243/6250 train_loss:3.6342 aux_loss:0.0506 train_time:31452ms step_avg:134.99ms
step:244/6250 train_loss:3.6929 aux_loss:0.0502 train_time:31590ms step_avg:135.00ms
step:245/6250 train_loss:3.6769 aux_loss:0.0499 train_time:31725ms step_avg:135.00ms
step:246/6250 train_loss:3.6703 aux_loss:0.0499 train_time:31857ms step_avg:134.99ms
step:247/6250 train_loss:3.5492 aux_loss:0.0501 train_time:31988ms step_avg:134.97ms
step:248/6250 train_loss:3.6183 aux_loss:0.0500 train_time:32119ms step_avg:134.95ms
step:249/6250 train_loss:3.6350 aux_loss:0.0502 train_time:32250ms step_avg:134.94ms
step:250/6250 train_loss:3.5753 aux_loss:0.0505 train_time:32380ms step_avg:134.91ms
step:251/6250 train_loss:3.6514 aux_loss:0.0509 train_time:32511ms step_avg:134.90ms
step:252/6250 train_loss:3.6309 aux_loss:0.0503 train_time:32642ms step_avg:134.88ms
step:253/6250 train_loss:3.6499 aux_loss:0.0503 train_time:32773ms step_avg:134.87ms
step:254/6250 train_loss:3.6509 aux_loss:0.0504 train_time:32904ms step_avg:134.85ms
step:255/6250 train_loss:3.6222 aux_loss:0.0501 train_time:33036ms step_avg:134.84ms
step:256/6250 train_loss:3.5627 aux_loss:0.0498 train_time:33169ms step_avg:134.83ms
step:257/6250 train_loss:3.6244 aux_loss:0.0501 train_time:33299ms step_avg:134.82ms
step:258/6250 train_loss:3.6468 aux_loss:0.0501 train_time:33429ms step_avg:134.80ms
step:259/6250 train_loss:3.6624 aux_loss:0.0494 train_time:33560ms step_avg:134.78ms
step:260/6250 train_loss:3.6377 aux_loss:0.0504 train_time:33689ms step_avg:134.76ms
step:261/6250 train_loss:3.6494 aux_loss:0.0504 train_time:33822ms step_avg:134.75ms
step:262/6250 train_loss:3.6072 aux_loss:0.0503 train_time:33952ms step_avg:134.73ms
step:263/6250 train_loss:3.4608 aux_loss:0.0501 train_time:34085ms step_avg:134.73ms
step:264/6250 train_loss:3.5883 aux_loss:0.0502 train_time:34217ms step_avg:134.71ms
step:265/6250 train_loss:3.5339 aux_loss:0.0501 train_time:34347ms step_avg:134.69ms
step:266/6250 train_loss:3.6092 aux_loss:0.0502 train_time:34478ms step_avg:134.68ms
step:267/6250 train_loss:3.5023 aux_loss:0.0501 train_time:34609ms step_avg:134.67ms
step:268/6250 train_loss:3.5869 aux_loss:0.0498 train_time:34740ms step_avg:134.65ms
step:269/6250 train_loss:3.5096 aux_loss:0.0499 train_time:34871ms step_avg:134.64ms
step:270/6250 train_loss:3.4987 aux_loss:0.0503 train_time:35006ms step_avg:134.64ms
step:271/6250 train_loss:3.5331 aux_loss:0.0501 train_time:35138ms step_avg:134.63ms
step:272/6250 train_loss:3.5506 aux_loss:0.0499 train_time:35269ms step_avg:134.61ms
step:273/6250 train_loss:3.4870 aux_loss:0.0497 train_time:35400ms step_avg:134.60ms
step:274/6250 train_loss:3.5514 aux_loss:0.0498 train_time:35530ms step_avg:134.58ms
step:275/6250 train_loss:3.4704 aux_loss:0.0501 train_time:35662ms step_avg:134.57ms
step:276/6250 train_loss:3.5276 aux_loss:0.0497 train_time:35793ms step_avg:134.56ms
step:277/6250 train_loss:3.5790 aux_loss:0.0498 train_time:35925ms step_avg:134.55ms
step:278/6250 train_loss:3.5533 aux_loss:0.0501 train_time:36056ms step_avg:134.54ms
step:279/6250 train_loss:3.4405 aux_loss:0.0499 train_time:36186ms step_avg:134.52ms
step:280/6250 train_loss:3.3890 aux_loss:0.0502 train_time:36317ms step_avg:134.51ms
step:281/6250 train_loss:3.5472 aux_loss:0.0503 train_time:36450ms step_avg:134.50ms
step:282/6250 train_loss:3.4089 aux_loss:0.0501 train_time:36581ms step_avg:134.49ms
step:283/6250 train_loss:3.4603 aux_loss:0.0502 train_time:36711ms step_avg:134.47ms
step:284/6250 train_loss:3.5246 aux_loss:0.0493 train_time:36845ms step_avg:134.47ms
step:285/6250 train_loss:3.5178 aux_loss:0.0497 train_time:36977ms step_avg:134.46ms
step:286/6250 train_loss:3.4807 aux_loss:0.0495 train_time:37108ms step_avg:134.45ms
step:287/6250 train_loss:3.5207 aux_loss:0.0501 train_time:37238ms step_avg:134.43ms
step:288/6250 train_loss:3.5397 aux_loss:0.0496 train_time:37368ms step_avg:134.42ms
step:289/6250 train_loss:3.4592 aux_loss:0.0495 train_time:37499ms step_avg:134.41ms
step:290/6250 train_loss:3.4958 aux_loss:0.0498 train_time:37631ms step_avg:134.40ms
step:291/6250 train_loss:3.5257 aux_loss:0.0496 train_time:37763ms step_avg:134.39ms
step:292/6250 train_loss:3.4365 aux_loss:0.0497 train_time:37895ms step_avg:134.38ms
step:293/6250 train_loss:3.4519 aux_loss:0.0497 train_time:38025ms step_avg:134.36ms
step:294/6250 train_loss:3.3961 aux_loss:0.0500 train_time:38156ms step_avg:134.35ms
step:295/6250 train_loss:3.4766 aux_loss:0.0498 train_time:38287ms step_avg:134.34ms
step:296/6250 train_loss:3.4961 aux_loss:0.0495 train_time:38417ms step_avg:134.33ms
step:297/6250 train_loss:3.3975 aux_loss:0.0499 train_time:38548ms step_avg:134.31ms
step:298/6250 train_loss:3.4615 aux_loss:0.0498 train_time:38679ms step_avg:134.30ms
step:299/6250 train_loss:3.4638 aux_loss:0.0496 train_time:38811ms step_avg:134.30ms
step:300/6250 train_loss:3.3395 aux_loss:0.0500 train_time:38944ms step_avg:134.29ms
step:301/6250 train_loss:3.4567 aux_loss:0.0494 train_time:39075ms step_avg:134.28ms
step:302/6250 train_loss:3.4818 aux_loss:0.0493 train_time:39206ms step_avg:134.27ms
step:303/6250 train_loss:3.4772 aux_loss:0.0494 train_time:39337ms step_avg:134.25ms
step:304/6250 train_loss:3.4120 aux_loss:0.0487 train_time:39467ms step_avg:134.24ms
step:305/6250 train_loss:3.3054 aux_loss:0.0499 train_time:39598ms step_avg:134.23ms
step:306/6250 train_loss:3.4170 aux_loss:0.0493 train_time:39729ms step_avg:134.22ms
step:307/6250 train_loss:3.4214 aux_loss:0.0493 train_time:39863ms step_avg:134.22ms
step:308/6250 train_loss:3.5263 aux_loss:0.0498 train_time:39996ms step_avg:134.22ms
step:309/6250 train_loss:3.3029 aux_loss:0.0496 train_time:40127ms step_avg:134.21ms
step:310/6250 train_loss:3.3907 aux_loss:0.0496 train_time:40257ms step_avg:134.19ms
step:311/6250 train_loss:3.3579 aux_loss:0.0494 train_time:40388ms step_avg:134.18ms
step:312/6250 train_loss:3.4173 aux_loss:0.0492 train_time:40518ms step_avg:134.17ms
step:313/6250 train_loss:3.5633 aux_loss:0.0494 train_time:40650ms step_avg:134.16ms
step:314/6250 train_loss:3.3812 aux_loss:0.0491 train_time:40782ms step_avg:134.15ms
step:315/6250 train_loss:3.5071 aux_loss:0.0490 train_time:40913ms step_avg:134.14ms
step:316/6250 train_loss:3.3191 aux_loss:0.0495 train_time:41045ms step_avg:134.14ms
step:317/6250 train_loss:3.4174 aux_loss:0.0492 train_time:41176ms step_avg:134.13ms
step:318/6250 train_loss:3.3679 aux_loss:0.0491 train_time:41306ms step_avg:134.11ms
step:319/6250 train_loss:3.3604 aux_loss:0.0494 train_time:41437ms step_avg:134.10ms
step:320/6250 train_loss:3.3869 aux_loss:0.0493 train_time:41567ms step_avg:134.09ms
step:321/6250 train_loss:3.3124 aux_loss:0.0489 train_time:41698ms step_avg:134.08ms
step:322/6250 train_loss:3.3719 aux_loss:0.0491 train_time:41829ms step_avg:134.07ms
step:323/6250 train_loss:3.3386 aux_loss:0.0493 train_time:41961ms step_avg:134.06ms
step:324/6250 train_loss:3.4407 aux_loss:0.0491 train_time:42093ms step_avg:134.06ms
step:325/6250 train_loss:3.3856 aux_loss:0.0490 train_time:42223ms step_avg:134.04ms
step:326/6250 train_loss:3.4170 aux_loss:0.0488 train_time:42356ms step_avg:134.04ms
step:327/6250 train_loss:3.3316 aux_loss:0.0489 train_time:42490ms step_avg:134.04ms
step:328/6250 train_loss:3.4090 aux_loss:0.0488 train_time:42621ms step_avg:134.03ms
step:329/6250 train_loss:3.3431 aux_loss:0.0491 train_time:42752ms step_avg:134.02ms
step:330/6250 train_loss:3.3918 aux_loss:0.0492 train_time:42883ms step_avg:134.01ms
step:331/6250 train_loss:3.2492 aux_loss:0.0496 train_time:43015ms step_avg:134.00ms
step:332/6250 train_loss:3.4058 aux_loss:0.0492 train_time:43145ms step_avg:133.99ms
step:333/6250 train_loss:3.3829 aux_loss:0.0489 train_time:43276ms step_avg:133.98ms
step:334/6250 train_loss:3.3603 aux_loss:0.0489 train_time:43406ms step_avg:133.97ms
step:335/6250 train_loss:3.2681 aux_loss:0.0490 train_time:43537ms step_avg:133.96ms
step:336/6250 train_loss:3.3260 aux_loss:0.0487 train_time:43668ms step_avg:133.95ms
step:337/6250 train_loss:3.4021 aux_loss:0.0485 train_time:43800ms step_avg:133.94ms
step:338/6250 train_loss:3.2417 aux_loss:0.0491 train_time:43930ms step_avg:133.93ms
step:339/6250 train_loss:3.4059 aux_loss:0.0494 train_time:44061ms step_avg:133.93ms
step:340/6250 train_loss:3.2820 aux_loss:0.0490 train_time:44192ms step_avg:133.92ms
step:341/6250 train_loss:3.3212 aux_loss:0.0493 train_time:44323ms step_avg:133.91ms
step:342/6250 train_loss:3.2277 aux_loss:0.0491 train_time:44453ms step_avg:133.90ms
step:343/6250 train_loss:3.3220 aux_loss:0.0491 train_time:44584ms step_avg:133.89ms
step:344/6250 train_loss:3.4147 aux_loss:0.0490 train_time:44714ms step_avg:133.88ms
step:345/6250 train_loss:3.3139 aux_loss:0.0488 train_time:44845ms step_avg:133.87ms
step:346/6250 train_loss:3.2185 aux_loss:0.0488 train_time:44977ms step_avg:133.86ms
step:347/6250 train_loss:3.2855 aux_loss:0.0487 train_time:45108ms step_avg:133.85ms
step:348/6250 train_loss:3.3224 aux_loss:0.0488 train_time:45239ms step_avg:133.84ms
step:349/6250 train_loss:3.1932 aux_loss:0.0489 train_time:45371ms step_avg:133.84ms
step:350/6250 train_loss:3.3435 aux_loss:0.0486 train_time:45501ms step_avg:133.83ms
step:351/6250 train_loss:3.2327 aux_loss:0.0492 train_time:45631ms step_avg:133.81ms
step:352/6250 train_loss:3.2685 aux_loss:0.0494 train_time:45762ms step_avg:133.81ms
step:353/6250 train_loss:3.2756 aux_loss:0.0490 train_time:45894ms step_avg:133.80ms
step:354/6250 train_loss:3.1400 aux_loss:0.0492 train_time:46024ms step_avg:133.79ms
step:355/6250 train_loss:3.3779 aux_loss:0.0488 train_time:46155ms step_avg:133.78ms
step:356/6250 train_loss:3.2511 aux_loss:0.0489 train_time:46286ms step_avg:133.77ms
step:357/6250 train_loss:3.3207 aux_loss:0.0487 train_time:46419ms step_avg:133.77ms
step:358/6250 train_loss:3.2920 aux_loss:0.0487 train_time:46550ms step_avg:133.76ms
step:359/6250 train_loss:3.1939 aux_loss:0.0490 train_time:46683ms step_avg:133.76ms
step:360/6250 train_loss:3.3340 aux_loss:0.0490 train_time:46814ms step_avg:133.75ms
step:361/6250 train_loss:3.1810 aux_loss:0.0493 train_time:46945ms step_avg:133.75ms
step:362/6250 train_loss:3.3111 aux_loss:0.0490 train_time:47075ms step_avg:133.73ms
step:363/6250 train_loss:3.2639 aux_loss:0.0486 train_time:47206ms step_avg:133.73ms
step:364/6250 train_loss:3.2030 aux_loss:0.0488 train_time:47337ms step_avg:133.72ms
step:365/6250 train_loss:3.1417 aux_loss:0.0491 train_time:47467ms step_avg:133.71ms
step:366/6250 train_loss:3.1549 aux_loss:0.0487 train_time:47598ms step_avg:133.70ms
step:367/6250 train_loss:3.2562 aux_loss:0.0485 train_time:47731ms step_avg:133.70ms
step:368/6250 train_loss:3.2331 aux_loss:0.0487 train_time:47863ms step_avg:133.69ms
step:369/6250 train_loss:3.2061 aux_loss:0.0488 train_time:47994ms step_avg:133.69ms
step:370/6250 train_loss:3.2076 aux_loss:0.0492 train_time:48125ms step_avg:133.68ms
step:371/6250 train_loss:3.1723 aux_loss:0.0487 train_time:48256ms step_avg:133.67ms
step:372/6250 train_loss:3.2317 aux_loss:0.0489 train_time:48387ms step_avg:133.67ms
step:373/6250 train_loss:3.2020 aux_loss:0.0490 train_time:48518ms step_avg:133.66ms
step:374/6250 train_loss:3.1237 aux_loss:0.0492 train_time:48649ms step_avg:133.65ms
step:375/6250 train_loss:3.2126 aux_loss:0.0493 train_time:48780ms step_avg:133.64ms
step:376/6250 train_loss:3.1801 aux_loss:0.0491 train_time:48911ms step_avg:133.64ms
step:377/6250 train_loss:3.1795 aux_loss:0.0487 train_time:49043ms step_avg:133.63ms
step:378/6250 train_loss:3.2783 aux_loss:0.0484 train_time:49173ms step_avg:133.62ms
step:379/6250 train_loss:3.2449 aux_loss:0.0486 train_time:49303ms step_avg:133.61ms
step:380/6250 train_loss:3.2796 aux_loss:0.0489 train_time:49434ms step_avg:133.61ms
step:381/6250 train_loss:3.2060 aux_loss:0.0487 train_time:49564ms step_avg:133.60ms
step:382/6250 train_loss:3.2873 aux_loss:0.0484 train_time:49696ms step_avg:133.59ms
step:383/6250 train_loss:3.1802 aux_loss:0.0489 train_time:49828ms step_avg:133.59ms
step:384/6250 train_loss:3.2268 aux_loss:0.0487 train_time:49959ms step_avg:133.58ms
step:385/6250 train_loss:3.2929 aux_loss:0.0485 train_time:50091ms step_avg:133.58ms
step:386/6250 train_loss:3.1876 aux_loss:0.0488 train_time:50221ms step_avg:133.57ms
step:387/6250 train_loss:3.0750 aux_loss:0.0489 train_time:50352ms step_avg:133.56ms
step:388/6250 train_loss:3.1760 aux_loss:0.0491 train_time:50482ms step_avg:133.55ms
step:389/6250 train_loss:3.1482 aux_loss:0.0487 train_time:50613ms step_avg:133.54ms
step:390/6250 train_loss:3.1360 aux_loss:0.0485 train_time:50744ms step_avg:133.54ms
step:391/6250 train_loss:3.1858 aux_loss:0.0486 train_time:50876ms step_avg:133.53ms
step:392/6250 train_loss:3.1591 aux_loss:0.0483 train_time:51007ms step_avg:133.53ms
step:393/6250 train_loss:3.1800 aux_loss:0.0485 train_time:51139ms step_avg:133.52ms
step:394/6250 train_loss:3.1562 aux_loss:0.0493 train_time:51270ms step_avg:133.52ms
step:395/6250 train_loss:3.1653 aux_loss:0.0492 train_time:51402ms step_avg:133.51ms
step:396/6250 train_loss:3.0628 aux_loss:0.0489 train_time:51533ms step_avg:133.51ms
step:397/6250 train_loss:3.2352 aux_loss:0.0488 train_time:51664ms step_avg:133.50ms
step:398/6250 train_loss:3.0915 aux_loss:0.0488 train_time:51794ms step_avg:133.49ms
step:399/6250 train_loss:3.1452 aux_loss:0.0484 train_time:51925ms step_avg:133.48ms
step:400/6250 train_loss:3.1142 aux_loss:0.0488 train_time:52057ms step_avg:133.48ms
step:401/6250 train_loss:3.1529 aux_loss:0.0486 train_time:52188ms step_avg:133.47ms
step:402/6250 train_loss:3.1614 aux_loss:0.0487 train_time:52319ms step_avg:133.47ms
step:403/6250 train_loss:3.0868 aux_loss:0.0490 train_time:52450ms step_avg:133.46ms
step:404/6250 train_loss:3.2476 aux_loss:0.0483 train_time:52581ms step_avg:133.45ms
step:405/6250 train_loss:3.1061 aux_loss:0.0482 train_time:52712ms step_avg:133.45ms
step:406/6250 train_loss:3.1164 aux_loss:0.0487 train_time:52844ms step_avg:133.44ms
step:407/6250 train_loss:3.1227 aux_loss:0.0489 train_time:52977ms step_avg:133.44ms
step:408/6250 train_loss:3.0785 aux_loss:0.0488 train_time:53108ms step_avg:133.44ms
step:409/6250 train_loss:3.1172 aux_loss:0.0487 train_time:53239ms step_avg:133.43ms
step:410/6250 train_loss:3.0911 aux_loss:0.0485 train_time:53369ms step_avg:133.42ms
step:411/6250 train_loss:3.2060 aux_loss:0.0484 train_time:53501ms step_avg:133.42ms
step:412/6250 train_loss:3.2010 aux_loss:0.0483 train_time:53632ms step_avg:133.41ms
step:413/6250 train_loss:3.1770 aux_loss:0.0485 train_time:53762ms step_avg:133.40ms
step:414/6250 train_loss:3.1020 aux_loss:0.0487 train_time:53893ms step_avg:133.40ms
step:415/6250 train_loss:3.1807 aux_loss:0.0486 train_time:54023ms step_avg:133.39ms
step:416/6250 train_loss:3.0565 aux_loss:0.0487 train_time:54155ms step_avg:133.39ms
step:417/6250 train_loss:3.1249 aux_loss:0.0484 train_time:54285ms step_avg:133.38ms
step:418/6250 train_loss:3.1532 aux_loss:0.0485 train_time:54417ms step_avg:133.37ms
step:419/6250 train_loss:3.1707 aux_loss:0.0481 train_time:54548ms step_avg:133.37ms
step:420/6250 train_loss:3.1145 aux_loss:0.0485 train_time:54680ms step_avg:133.37ms
step:421/6250 train_loss:3.0305 aux_loss:0.0487 train_time:54812ms step_avg:133.36ms
step:422/6250 train_loss:3.0215 aux_loss:0.0487 train_time:54947ms step_avg:133.37ms
step:423/6250 train_loss:3.1690 aux_loss:0.0484 train_time:55083ms step_avg:133.37ms
step:424/6250 train_loss:3.1309 aux_loss:0.0485 train_time:55217ms step_avg:133.37ms
step:425/6250 train_loss:3.1233 aux_loss:0.0482 train_time:55348ms step_avg:133.37ms
step:426/6250 train_loss:3.1641 aux_loss:0.0482 train_time:55479ms step_avg:133.36ms
step:427/6250 train_loss:3.1420 aux_loss:0.0485 train_time:55610ms step_avg:133.36ms
step:428/6250 train_loss:3.1395 aux_loss:0.0486 train_time:55741ms step_avg:133.35ms
step:429/6250 train_loss:3.0518 aux_loss:0.0486 train_time:55872ms step_avg:133.35ms
step:430/6250 train_loss:3.0440 aux_loss:0.0486 train_time:56002ms step_avg:133.34ms
step:431/6250 train_loss:3.0377 aux_loss:0.0484 train_time:56133ms step_avg:133.33ms
step:432/6250 train_loss:3.0078 aux_loss:0.0486 train_time:56263ms step_avg:133.32ms
step:433/6250 train_loss:3.0944 aux_loss:0.0484 train_time:56395ms step_avg:133.32ms
step:434/6250 train_loss:3.0814 aux_loss:0.0480 train_time:56525ms step_avg:133.31ms
step:435/6250 train_loss:3.1118 aux_loss:0.0482 train_time:56657ms step_avg:133.31ms
step:436/6250 train_loss:3.0296 aux_loss:0.0485 train_time:56790ms step_avg:133.31ms
step:437/6250 train_loss:3.1441 aux_loss:0.0483 train_time:56923ms step_avg:133.31ms
step:438/6250 train_loss:3.0721 aux_loss:0.0483 train_time:57055ms step_avg:133.31ms
step:439/6250 train_loss:3.0994 aux_loss:0.0484 train_time:57185ms step_avg:133.30ms
step:440/6250 train_loss:3.0370 aux_loss:0.0484 train_time:57316ms step_avg:133.29ms
step:441/6250 train_loss:3.0561 aux_loss:0.0482 train_time:57447ms step_avg:133.29ms
step:442/6250 train_loss:3.1909 aux_loss:0.0478 train_time:57579ms step_avg:133.29ms
step:443/6250 train_loss:3.0931 aux_loss:0.0479 train_time:57710ms step_avg:133.28ms
step:444/6250 train_loss:3.1227 aux_loss:0.0483 train_time:57842ms step_avg:133.28ms
step:445/6250 train_loss:2.9577 aux_loss:0.0487 train_time:57973ms step_avg:133.27ms
step:446/6250 train_loss:3.1266 aux_loss:0.0483 train_time:58103ms step_avg:133.26ms
step:447/6250 train_loss:3.0852 aux_loss:0.0482 train_time:58234ms step_avg:133.26ms
step:448/6250 train_loss:3.0540 aux_loss:0.0483 train_time:58365ms step_avg:133.25ms
step:449/6250 train_loss:3.1104 aux_loss:0.0485 train_time:58497ms step_avg:133.25ms
step:450/6250 train_loss:3.0531 aux_loss:0.0482 train_time:58628ms step_avg:133.25ms
step:451/6250 train_loss:2.9680 aux_loss:0.0480 train_time:58759ms step_avg:133.24ms
step:452/6250 train_loss:2.9877 aux_loss:0.0485 train_time:58891ms step_avg:133.24ms
step:453/6250 train_loss:3.0680 aux_loss:0.0486 train_time:59021ms step_avg:133.23ms
step:454/6250 train_loss:2.9816 aux_loss:0.0486 train_time:59152ms step_avg:133.22ms
step:455/6250 train_loss:3.0993 aux_loss:0.0485 train_time:59283ms step_avg:133.22ms
step:456/6250 train_loss:3.1169 aux_loss:0.0483 train_time:59415ms step_avg:133.22ms
step:457/6250 train_loss:2.9498 aux_loss:0.0483 train_time:59546ms step_avg:133.21ms
step:458/6250 train_loss:3.0633 aux_loss:0.0481 train_time:59678ms step_avg:133.21ms
step:459/6250 train_loss:3.0046 aux_loss:0.0480 train_time:59809ms step_avg:133.20ms
step:460/6250 train_loss:3.1028 aux_loss:0.0484 train_time:59940ms step_avg:133.20ms
step:461/6250 train_loss:3.0184 aux_loss:0.0485 train_time:60071ms step_avg:133.19ms
step:462/6250 train_loss:3.0364 aux_loss:0.0484 train_time:60201ms step_avg:133.19ms
step:463/6250 train_loss:3.0769 aux_loss:0.0481 train_time:60333ms step_avg:133.19ms
step:464/6250 train_loss:3.0428 aux_loss:0.0482 train_time:60464ms step_avg:133.18ms
step:465/6250 train_loss:3.0525 aux_loss:0.0485 train_time:60596ms step_avg:133.18ms
step:466/6250 train_loss:2.9456 aux_loss:0.0483 train_time:60728ms step_avg:133.18ms
step:467/6250 train_loss:2.9226 aux_loss:0.0479 train_time:60859ms step_avg:133.17ms
step:468/6250 train_loss:3.0090 aux_loss:0.0480 train_time:60990ms step_avg:133.17ms
step:469/6250 train_loss:3.0163 aux_loss:0.0481 train_time:61120ms step_avg:133.16ms
step:470/6250 train_loss:3.0783 aux_loss:0.0482 train_time:61250ms step_avg:133.15ms
step:471/6250 train_loss:3.0261 aux_loss:0.0482 train_time:61381ms step_avg:133.15ms
step:472/6250 train_loss:2.9485 aux_loss:0.0481 train_time:61513ms step_avg:133.15ms
step:473/6250 train_loss:3.0223 aux_loss:0.0483 train_time:61644ms step_avg:133.14ms
step:474/6250 train_loss:3.0217 aux_loss:0.0482 train_time:61776ms step_avg:133.14ms
step:475/6250 train_loss:3.0287 aux_loss:0.0480 train_time:61907ms step_avg:133.13ms
step:476/6250 train_loss:2.9927 aux_loss:0.0479 train_time:62038ms step_avg:133.13ms
step:477/6250 train_loss:2.9986 aux_loss:0.0478 train_time:62168ms step_avg:133.12ms
step:478/6250 train_loss:2.9481 aux_loss:0.0481 train_time:62299ms step_avg:133.12ms
step:479/6250 train_loss:2.9710 aux_loss:0.0483 train_time:62430ms step_avg:133.11ms
step:480/6250 train_loss:3.0188 aux_loss:0.0483 train_time:62561ms step_avg:133.11ms
step:481/6250 train_loss:2.9737 aux_loss:0.0486 train_time:62693ms step_avg:133.11ms
step:482/6250 train_loss:3.0093 aux_loss:0.0482 train_time:62825ms step_avg:133.10ms
step:483/6250 train_loss:3.0853 aux_loss:0.0486 train_time:62958ms step_avg:133.10ms
step:484/6250 train_loss:3.0651 aux_loss:0.0483 train_time:63090ms step_avg:133.10ms
step:485/6250 train_loss:2.9305 aux_loss:0.0477 train_time:63220ms step_avg:133.09ms
step:486/6250 train_loss:3.0737 aux_loss:0.0475 train_time:63351ms step_avg:133.09ms
step:487/6250 train_loss:2.9883 aux_loss:0.0479 train_time:63482ms step_avg:133.09ms
step:488/6250 train_loss:3.0067 aux_loss:0.0480 train_time:63613ms step_avg:133.08ms
step:489/6250 train_loss:2.9907 aux_loss:0.0482 train_time:63744ms step_avg:133.08ms
step:490/6250 train_loss:2.9843 aux_loss:0.0481 train_time:63877ms step_avg:133.08ms
step:491/6250 train_loss:2.9482 aux_loss:0.0481 train_time:64008ms step_avg:133.07ms
step:492/6250 train_loss:3.0230 aux_loss:0.0478 train_time:64138ms step_avg:133.07ms
step:493/6250 train_loss:2.9971 aux_loss:0.0481 train_time:64269ms step_avg:133.06ms
step:494/6250 train_loss:2.9409 aux_loss:0.0479 train_time:64402ms step_avg:133.06ms
step:495/6250 train_loss:2.9959 aux_loss:0.0478 train_time:64532ms step_avg:133.06ms
step:496/6250 train_loss:2.9193 aux_loss:0.0481 train_time:64662ms step_avg:133.05ms
step:497/6250 train_loss:2.9838 aux_loss:0.0480 train_time:64795ms step_avg:133.05ms
step:498/6250 train_loss:2.9002 aux_loss:0.0482 train_time:64928ms step_avg:133.05ms
step:499/6250 train_loss:3.0159 aux_loss:0.0483 train_time:65058ms step_avg:133.04ms
step:500/6250 train_loss:2.9605 aux_loss:0.0480 train_time:65189ms step_avg:133.04ms
step:501/6250 train_loss:2.9036 aux_loss:0.0476 train_time:65320ms step_avg:133.03ms
step:502/6250 train_loss:3.0825 aux_loss:0.0474 train_time:65451ms step_avg:133.03ms
step:503/6250 train_loss:2.9792 aux_loss:0.0478 train_time:65581ms step_avg:133.02ms
step:504/6250 train_loss:3.0256 aux_loss:0.0477 train_time:65711ms step_avg:133.02ms
step:505/6250 train_loss:2.9948 aux_loss:0.0476 train_time:65843ms step_avg:133.02ms
step:506/6250 train_loss:3.0077 aux_loss:0.0478 train_time:65976ms step_avg:133.02ms
step:507/6250 train_loss:2.9602 aux_loss:0.0479 train_time:66107ms step_avg:133.01ms
step:508/6250 train_loss:3.0063 aux_loss:0.0478 train_time:66238ms step_avg:133.01ms
step:509/6250 train_loss:3.0700 aux_loss:0.0477 train_time:66370ms step_avg:133.01ms
step:510/6250 train_loss:3.0570 aux_loss:0.0478 train_time:66500ms step_avg:133.00ms
step:511/6250 train_loss:2.9394 aux_loss:0.0480 train_time:66631ms step_avg:133.00ms
step:512/6250 train_loss:3.0614 aux_loss:0.0475 train_time:66761ms step_avg:132.99ms
step:513/6250 train_loss:2.9457 aux_loss:0.0479 train_time:66894ms step_avg:132.99ms
step:514/6250 train_loss:2.9995 aux_loss:0.0479 train_time:67025ms step_avg:132.99ms
step:515/6250 train_loss:3.0166 aux_loss:0.0477 train_time:67156ms step_avg:132.98ms
step:516/6250 train_loss:3.0057 aux_loss:0.0478 train_time:67287ms step_avg:132.98ms
step:517/6250 train_loss:2.8942 aux_loss:0.0480 train_time:67419ms step_avg:132.98ms
step:518/6250 train_loss:2.9932 aux_loss:0.0477 train_time:67550ms step_avg:132.97ms
step:519/6250 train_loss:2.9947 aux_loss:0.0479 train_time:67681ms step_avg:132.97ms
step:520/6250 train_loss:2.9928 aux_loss:0.0478 train_time:67813ms step_avg:132.97ms
step:521/6250 train_loss:2.9825 aux_loss:0.0479 train_time:67946ms step_avg:132.97ms
step:522/6250 train_loss:2.9188 aux_loss:0.0481 train_time:68078ms step_avg:132.96ms
step:523/6250 train_loss:2.9734 aux_loss:0.0479 train_time:68209ms step_avg:132.96ms
step:524/6250 train_loss:2.9692 aux_loss:0.0476 train_time:68339ms step_avg:132.95ms
step:525/6250 train_loss:2.9141 aux_loss:0.0476 train_time:68470ms step_avg:132.95ms
step:526/6250 train_loss:2.9215 aux_loss:0.0476 train_time:68600ms step_avg:132.95ms
step:527/6250 train_loss:2.9390 aux_loss:0.0475 train_time:68730ms step_avg:132.94ms
step:528/6250 train_loss:2.9778 aux_loss:0.0474 train_time:68863ms step_avg:132.94ms
step:529/6250 train_loss:3.0005 aux_loss:0.0478 train_time:68995ms step_avg:132.94ms
step:530/6250 train_loss:2.9483 aux_loss:0.0480 train_time:69125ms step_avg:132.93ms
step:531/6250 train_loss:2.9265 aux_loss:0.0480 train_time:69256ms step_avg:132.93ms
step:532/6250 train_loss:2.8482 aux_loss:0.0481 train_time:69389ms step_avg:132.93ms
step:533/6250 train_loss:2.8974 aux_loss:0.0478 train_time:69518ms step_avg:132.92ms
step:534/6250 train_loss:2.8946 aux_loss:0.0481 train_time:69648ms step_avg:132.92ms
step:535/6250 train_loss:2.9558 aux_loss:0.0479 train_time:69779ms step_avg:132.91ms
step:536/6250 train_loss:2.8728 aux_loss:0.0478 train_time:69912ms step_avg:132.91ms
step:537/6250 train_loss:2.8834 aux_loss:0.0480 train_time:70043ms step_avg:132.91ms
step:538/6250 train_loss:2.9170 aux_loss:0.0477 train_time:70175ms step_avg:132.91ms
step:539/6250 train_loss:2.9588 aux_loss:0.0479 train_time:70306ms step_avg:132.90ms
step:540/6250 train_loss:2.9529 aux_loss:0.0478 train_time:70437ms step_avg:132.90ms
step:541/6250 train_loss:2.9081 aux_loss:0.0480 train_time:70567ms step_avg:132.90ms
step:542/6250 train_loss:2.8992 aux_loss:0.0479 train_time:70698ms step_avg:132.89ms
step:543/6250 train_loss:3.0055 aux_loss:0.0478 train_time:70830ms step_avg:132.89ms
step:544/6250 train_loss:2.8879 aux_loss:0.0479 train_time:70961ms step_avg:132.89ms
step:545/6250 train_loss:2.9881 aux_loss:0.0476 train_time:71094ms step_avg:132.89ms
step:546/6250 train_loss:2.7838 aux_loss:0.0477 train_time:71225ms step_avg:132.88ms
step:547/6250 train_loss:2.8789 aux_loss:0.0478 train_time:71357ms step_avg:132.88ms
step:548/6250 train_loss:2.9601 aux_loss:0.0476 train_time:71487ms step_avg:132.87ms
step:549/6250 train_loss:2.9870 aux_loss:0.0474 train_time:71617ms step_avg:132.87ms
step:550/6250 train_loss:2.9350 aux_loss:0.0477 train_time:71748ms step_avg:132.87ms
step:551/6250 train_loss:2.9832 aux_loss:0.0477 train_time:71880ms step_avg:132.86ms
step:552/6250 train_loss:2.8696 aux_loss:0.0477 train_time:72011ms step_avg:132.86ms
step:553/6250 train_loss:2.9409 aux_loss:0.0476 train_time:72142ms step_avg:132.86ms
step:554/6250 train_loss:3.0000 aux_loss:0.0477 train_time:72274ms step_avg:132.86ms
step:555/6250 train_loss:2.8950 aux_loss:0.0478 train_time:72406ms step_avg:132.85ms
step:556/6250 train_loss:2.8959 aux_loss:0.0476 train_time:72537ms step_avg:132.85ms
step:557/6250 train_loss:2.9884 aux_loss:0.0472 train_time:72668ms step_avg:132.85ms
step:558/6250 train_loss:2.8369 aux_loss:0.0476 train_time:72798ms step_avg:132.84ms
step:559/6250 train_loss:2.8986 aux_loss:0.0478 train_time:72931ms step_avg:132.84ms
step:560/6250 train_loss:2.9451 aux_loss:0.0474 train_time:73062ms step_avg:132.84ms
step:561/6250 train_loss:2.9821 aux_loss:0.0478 train_time:73194ms step_avg:132.84ms
step:562/6250 train_loss:2.9700 aux_loss:0.0477 train_time:73324ms step_avg:132.83ms
step:563/6250 train_loss:2.8873 aux_loss:0.0477 train_time:73456ms step_avg:132.83ms
step:564/6250 train_loss:2.8752 aux_loss:0.0478 train_time:73586ms step_avg:132.83ms
step:565/6250 train_loss:2.9035 aux_loss:0.0478 train_time:73716ms step_avg:132.82ms
step:566/6250 train_loss:2.8205 aux_loss:0.0476 train_time:73848ms step_avg:132.82ms
step:567/6250 train_loss:2.9442 aux_loss:0.0474 train_time:73978ms step_avg:132.82ms
step:568/6250 train_loss:2.8979 aux_loss:0.0476 train_time:74109ms step_avg:132.81ms
step:569/6250 train_loss:2.9038 aux_loss:0.0476 train_time:74240ms step_avg:132.81ms
step:570/6250 train_loss:2.9061 aux_loss:0.0480 train_time:74372ms step_avg:132.81ms
step:571/6250 train_loss:2.9271 aux_loss:0.0478 train_time:74503ms step_avg:132.80ms
step:572/6250 train_loss:2.9233 aux_loss:0.0475 train_time:74634ms step_avg:132.80ms
step:573/6250 train_loss:2.8288 aux_loss:0.0477 train_time:74764ms step_avg:132.80ms
step:574/6250 train_loss:2.8644 aux_loss:0.0476 train_time:74896ms step_avg:132.80ms
step:575/6250 train_loss:2.8834 aux_loss:0.0475 train_time:75027ms step_avg:132.79ms
step:576/6250 train_loss:2.8308 aux_loss:0.0474 train_time:75159ms step_avg:132.79ms
step:577/6250 train_loss:2.8628 aux_loss:0.0474 train_time:75294ms step_avg:132.79ms
step:578/6250 train_loss:2.8262 aux_loss:0.0475 train_time:75426ms step_avg:132.79ms
step:579/6250 train_loss:2.7993 aux_loss:0.0477 train_time:75557ms step_avg:132.79ms
step:580/6250 train_loss:2.8181 aux_loss:0.0478 train_time:75688ms step_avg:132.79ms
step:581/6250 train_loss:2.8493 aux_loss:0.0476 train_time:75823ms step_avg:132.79ms
step:582/6250 train_loss:2.9241 aux_loss:0.0476 train_time:75960ms step_avg:132.80ms
step:583/6250 train_loss:2.9220 aux_loss:0.0474 train_time:76091ms step_avg:132.79ms
step:584/6250 train_loss:2.9235 aux_loss:0.0473 train_time:76223ms step_avg:132.79ms
step:585/6250 train_loss:2.7899 aux_loss:0.0473 train_time:76362ms step_avg:132.80ms
step:586/6250 train_loss:2.9178 aux_loss:0.0473 train_time:76501ms step_avg:132.81ms
step:587/6250 train_loss:2.9542 aux_loss:0.0475 train_time:76639ms step_avg:132.82ms
step:588/6250 train_loss:2.9219 aux_loss:0.0475 train_time:76779ms step_avg:132.84ms
step:589/6250 train_loss:2.8827 aux_loss:0.0478 train_time:76930ms step_avg:132.87ms
step:590/6250 train_loss:2.8935 aux_loss:0.0474 train_time:77090ms step_avg:132.91ms
step:591/6250 train_loss:2.8547 aux_loss:0.0478 train_time:77227ms step_avg:132.92ms
step:592/6250 train_loss:2.9317 aux_loss:0.0477 train_time:77359ms step_avg:132.92ms
step:593/6250 train_loss:2.8762 aux_loss:0.0475 train_time:77491ms step_avg:132.92ms
step:594/6250 train_loss:2.8046 aux_loss:0.0476 train_time:77621ms step_avg:132.91ms
step:595/6250 train_loss:2.8570 aux_loss:0.0476 train_time:77753ms step_avg:132.91ms
step:596/6250 train_loss:2.9484 aux_loss:0.0472 train_time:77887ms step_avg:132.91ms
step:597/6250 train_loss:2.9764 aux_loss:0.0470 train_time:78038ms step_avg:132.94ms
step:598/6250 train_loss:2.8633 aux_loss:0.0472 train_time:78187ms step_avg:132.97ms
step:599/6250 train_loss:2.7863 aux_loss:0.0473 train_time:78341ms step_avg:133.01ms
step:600/6250 train_loss:2.7411 aux_loss:0.0475 train_time:78496ms step_avg:133.04ms
step:601/6250 train_loss:2.8574 aux_loss:0.0470 train_time:78649ms step_avg:133.08ms
step:602/6250 train_loss:2.8701 aux_loss:0.0474 train_time:78782ms step_avg:133.08ms
step:603/6250 train_loss:2.8722 aux_loss:0.0476 train_time:78915ms step_avg:133.08ms
step:604/6250 train_loss:2.8852 aux_loss:0.0475 train_time:79046ms step_avg:133.07ms
step:605/6250 train_loss:2.7895 aux_loss:0.0478 train_time:79180ms step_avg:133.08ms
step:606/6250 train_loss:2.9060 aux_loss:0.0474 train_time:79316ms step_avg:133.08ms
step:607/6250 train_loss:2.8684 aux_loss:0.0471 train_time:79450ms step_avg:133.08ms
step:608/6250 train_loss:2.7832 aux_loss:0.0474 train_time:79589ms step_avg:133.09ms
step:609/6250 train_loss:2.8827 aux_loss:0.0471 train_time:79720ms step_avg:133.09ms
step:610/6250 train_loss:2.7713 aux_loss:0.0473 train_time:79857ms step_avg:133.10ms
step:611/6250 train_loss:2.8402 aux_loss:0.0472 train_time:79996ms step_avg:133.11ms
step:612/6250 train_loss:2.8786 aux_loss:0.0472 train_time:80136ms step_avg:133.12ms
step:613/6250 train_loss:2.9157 aux_loss:0.0472 train_time:80277ms step_avg:133.13ms
step:614/6250 train_loss:2.8338 aux_loss:0.0475 train_time:80418ms step_avg:133.14ms
step:615/6250 train_loss:2.8539 aux_loss:0.0474 train_time:80559ms step_avg:133.16ms
step:616/6250 train_loss:2.8092 aux_loss:0.0478 train_time:80694ms step_avg:133.16ms
step:617/6250 train_loss:2.8902 aux_loss:0.0473 train_time:80831ms step_avg:133.17ms
step:618/6250 train_loss:2.7667 aux_loss:0.0475 train_time:80973ms step_avg:133.18ms
step:619/6250 train_loss:2.7867 aux_loss:0.0474 train_time:81108ms step_avg:133.18ms
step:620/6250 train_loss:2.8586 aux_loss:0.0471 train_time:81245ms step_avg:133.19ms
step:621/6250 train_loss:2.9337 aux_loss:0.0472 train_time:81386ms step_avg:133.20ms
step:622/6250 train_loss:2.7507 aux_loss:0.0474 train_time:81523ms step_avg:133.21ms
step:623/6250 train_loss:2.8761 aux_loss:0.0471 train_time:81665ms step_avg:133.22ms
step:624/6250 train_loss:2.8665 aux_loss:0.0472 train_time:81802ms step_avg:133.23ms
step:625/6250 train_loss:2.8237 aux_loss:0.0474 train_time:81939ms step_avg:133.23ms
step:626/6250 train_loss:2.8638 aux_loss:0.0473 train_time:82074ms step_avg:133.24ms
step:627/6250 train_loss:2.8493 aux_loss:0.0473 train_time:82205ms step_avg:133.23ms
step:628/6250 train_loss:2.8217 aux_loss:0.0473 train_time:82338ms step_avg:133.23ms
step:629/6250 train_loss:2.8775 aux_loss:0.0472 train_time:82478ms step_avg:133.24ms
step:630/6250 train_loss:2.8811 aux_loss:0.0473 train_time:82620ms step_avg:133.26ms
step:631/6250 train_loss:2.8653 aux_loss:0.0472 train_time:82759ms step_avg:133.27ms
step:632/6250 train_loss:2.8616 aux_loss:0.0470 train_time:82892ms step_avg:133.27ms
step:633/6250 train_loss:2.9327 aux_loss:0.0467 train_time:83029ms step_avg:133.27ms
step:634/6250 train_loss:2.8472 aux_loss:0.0470 train_time:83173ms step_avg:133.29ms
step:635/6250 train_loss:2.8198 aux_loss:0.0472 train_time:83307ms step_avg:133.29ms
step:636/6250 train_loss:2.8735 aux_loss:0.0469 train_time:83442ms step_avg:133.29ms
step:637/6250 train_loss:2.7018 aux_loss:0.0471 train_time:83580ms step_avg:133.30ms
step:638/6250 train_loss:2.8587 aux_loss:0.0470 train_time:83711ms step_avg:133.30ms
step:639/6250 train_loss:2.8115 aux_loss:0.0473 train_time:83843ms step_avg:133.30ms
step:640/6250 train_loss:2.6814 aux_loss:0.0474 train_time:83976ms step_avg:133.30ms
step:641/6250 train_loss:2.8934 aux_loss:0.0472 train_time:84106ms step_avg:133.29ms
step:642/6250 train_loss:2.8128 aux_loss:0.0473 train_time:84237ms step_avg:133.29ms
step:643/6250 train_loss:2.8289 aux_loss:0.0472 train_time:84370ms step_avg:133.29ms
step:644/6250 train_loss:2.7583 aux_loss:0.0472 train_time:84502ms step_avg:133.28ms
step:645/6250 train_loss:2.7839 aux_loss:0.0468 train_time:84635ms step_avg:133.28ms
step:646/6250 train_loss:2.8679 aux_loss:0.0472 train_time:84766ms step_avg:133.28ms
step:647/6250 train_loss:2.6935 aux_loss:0.0474 train_time:84899ms step_avg:133.28ms
step:648/6250 train_loss:2.7994 aux_loss:0.0472 train_time:85031ms step_avg:133.28ms
step:649/6250 train_loss:2.7310 aux_loss:0.0474 train_time:85163ms step_avg:133.28ms
step:650/6250 train_loss:2.7167 aux_loss:0.0473 train_time:85295ms step_avg:133.27ms
step:651/6250 train_loss:2.8479 aux_loss:0.0469 train_time:85428ms step_avg:133.27ms
step:652/6250 train_loss:2.8749 aux_loss:0.0469 train_time:85559ms step_avg:133.27ms
step:653/6250 train_loss:2.7614 aux_loss:0.0471 train_time:85690ms step_avg:133.27ms
step:654/6250 train_loss:2.7175 aux_loss:0.0473 train_time:85823ms step_avg:133.27ms
step:655/6250 train_loss:2.8080 aux_loss:0.0471 train_time:85956ms step_avg:133.27ms
step:656/6250 train_loss:2.8318 aux_loss:0.0472 train_time:86088ms step_avg:133.26ms
step:657/6250 train_loss:2.7546 aux_loss:0.0473 train_time:86221ms step_avg:133.26ms
step:658/6250 train_loss:2.8167 aux_loss:0.0469 train_time:86354ms step_avg:133.26ms
step:659/6250 train_loss:2.8090 aux_loss:0.0469 train_time:86485ms step_avg:133.26ms
step:660/6250 train_loss:2.7316 aux_loss:0.0471 train_time:86617ms step_avg:133.26ms
step:661/6250 train_loss:2.8647 aux_loss:0.0472 train_time:86750ms step_avg:133.26ms
step:662/6250 train_loss:2.7403 aux_loss:0.0469 train_time:86882ms step_avg:133.25ms
step:663/6250 train_loss:2.7873 aux_loss:0.0468 train_time:87014ms step_avg:133.25ms
step:664/6250 train_loss:2.7709 aux_loss:0.0469 train_time:87145ms step_avg:133.25ms
step:665/6250 train_loss:2.8700 aux_loss:0.0468 train_time:87277ms step_avg:133.25ms
step:666/6250 train_loss:2.8416 aux_loss:0.0467 train_time:87411ms step_avg:133.25ms
step:667/6250 train_loss:2.8801 aux_loss:0.0470 train_time:87543ms step_avg:133.25ms
step:668/6250 train_loss:2.8477 aux_loss:0.0473 train_time:87675ms step_avg:133.24ms
step:669/6250 train_loss:2.7659 aux_loss:0.0471 train_time:87807ms step_avg:133.24ms
step:670/6250 train_loss:2.8184 aux_loss:0.0470 train_time:87939ms step_avg:133.24ms
step:671/6250 train_loss:2.6871 aux_loss:0.0472 train_time:88070ms step_avg:133.24ms
step:672/6250 train_loss:2.7357 aux_loss:0.0471 train_time:88201ms step_avg:133.23ms
step:673/6250 train_loss:2.7449 aux_loss:0.0473 train_time:88333ms step_avg:133.23ms
step:674/6250 train_loss:2.8356 aux_loss:0.0472 train_time:88465ms step_avg:133.23ms
step:675/6250 train_loss:2.7566 aux_loss:0.0471 train_time:88596ms step_avg:133.23ms
step:676/6250 train_loss:2.8083 aux_loss:0.0468 train_time:88728ms step_avg:133.22ms
step:677/6250 train_loss:2.7867 aux_loss:0.0469 train_time:88860ms step_avg:133.22ms
step:678/6250 train_loss:2.7927 aux_loss:0.0469 train_time:88992ms step_avg:133.22ms
step:679/6250 train_loss:2.7971 aux_loss:0.0467 train_time:89123ms step_avg:133.22ms
step:680/6250 train_loss:2.8791 aux_loss:0.0466 train_time:89253ms step_avg:133.21ms
step:681/6250 train_loss:2.7692 aux_loss:0.0465 train_time:89385ms step_avg:133.21ms
step:682/6250 train_loss:2.7816 aux_loss:0.0468 train_time:89517ms step_avg:133.21ms
step:683/6250 train_loss:2.8590 aux_loss:0.0468 train_time:89649ms step_avg:133.21ms
step:684/6250 train_loss:2.8465 aux_loss:0.0471 train_time:89780ms step_avg:133.20ms
step:685/6250 train_loss:2.7621 aux_loss:0.0469 train_time:89914ms step_avg:133.21ms
step:686/6250 train_loss:2.8105 aux_loss:0.0469 train_time:90046ms step_avg:133.20ms
step:687/6250 train_loss:2.8109 aux_loss:0.0469 train_time:90178ms step_avg:133.20ms
step:688/6250 train_loss:2.8606 aux_loss:0.0466 train_time:90310ms step_avg:133.20ms
step:689/6250 train_loss:2.8921 aux_loss:0.0466 train_time:90443ms step_avg:133.20ms
step:690/6250 train_loss:2.8172 aux_loss:0.0463 train_time:90575ms step_avg:133.20ms
step:691/6250 train_loss:2.7876 aux_loss:0.0465 train_time:90707ms step_avg:133.20ms
step:692/6250 train_loss:2.8627 aux_loss:0.0470 train_time:90838ms step_avg:133.19ms
step:693/6250 train_loss:2.8566 aux_loss:0.0469 train_time:90970ms step_avg:133.19ms
step:694/6250 train_loss:2.7708 aux_loss:0.0468 train_time:91101ms step_avg:133.19ms
step:695/6250 train_loss:2.7418 aux_loss:0.0468 train_time:91233ms step_avg:133.19ms
step:696/6250 train_loss:2.8495 aux_loss:0.0466 train_time:91364ms step_avg:133.18ms
step:697/6250 train_loss:2.8348 aux_loss:0.0466 train_time:91495ms step_avg:133.18ms
step:698/6250 train_loss:2.8555 aux_loss:0.0465 train_time:91627ms step_avg:133.18ms
step:699/6250 train_loss:2.7758 aux_loss:0.0466 train_time:91757ms step_avg:133.17ms
step:700/6250 train_loss:2.8494 aux_loss:0.0467 train_time:91890ms step_avg:133.17ms
step:701/6250 train_loss:2.8691 aux_loss:0.0467 train_time:92022ms step_avg:133.17ms
step:702/6250 train_loss:2.7500 aux_loss:0.0469 train_time:92153ms step_avg:133.17ms
step:703/6250 train_loss:2.7459 aux_loss:0.0469 train_time:92284ms step_avg:133.17ms
step:704/6250 train_loss:2.8428 aux_loss:0.0465 train_time:92416ms step_avg:133.16ms
step:705/6250 train_loss:2.7400 aux_loss:0.0466 train_time:92548ms step_avg:133.16ms
step:706/6250 train_loss:2.6968 aux_loss:0.0468 train_time:92680ms step_avg:133.16ms
step:707/6250 train_loss:2.7596 aux_loss:0.0469 train_time:92813ms step_avg:133.16ms
step:708/6250 train_loss:2.7792 aux_loss:0.0468 train_time:92945ms step_avg:133.16ms
step:709/6250 train_loss:2.7591 aux_loss:0.0468 train_time:93076ms step_avg:133.16ms
step:710/6250 train_loss:2.7244 aux_loss:0.0466 train_time:93207ms step_avg:133.15ms
step:711/6250 train_loss:2.7365 aux_loss:0.0465 train_time:93338ms step_avg:133.15ms
step:712/6250 train_loss:2.7296 aux_loss:0.0469 train_time:93470ms step_avg:133.15ms
step:713/6250 train_loss:2.7039 aux_loss:0.0467 train_time:93601ms step_avg:133.14ms
step:714/6250 train_loss:2.7980 aux_loss:0.0466 train_time:93731ms step_avg:133.14ms
step:715/6250 train_loss:2.7906 aux_loss:0.0466 train_time:93863ms step_avg:133.14ms
step:716/6250 train_loss:2.7459 aux_loss:0.0465 train_time:93994ms step_avg:133.14ms
step:717/6250 train_loss:2.6962 aux_loss:0.0469 train_time:94126ms step_avg:133.13ms
step:718/6250 train_loss:2.8336 aux_loss:0.0468 train_time:94256ms step_avg:133.13ms
step:719/6250 train_loss:2.7731 aux_loss:0.0468 train_time:94389ms step_avg:133.13ms
step:720/6250 train_loss:2.8207 aux_loss:0.0466 train_time:94519ms step_avg:133.13ms
step:721/6250 train_loss:2.6909 aux_loss:0.0466 train_time:94650ms step_avg:133.12ms
step:722/6250 train_loss:2.7500 aux_loss:0.0466 train_time:94781ms step_avg:133.12ms
step:723/6250 train_loss:2.7306 aux_loss:0.0467 train_time:94913ms step_avg:133.12ms
step:724/6250 train_loss:2.7492 aux_loss:0.0467 train_time:95044ms step_avg:133.11ms
step:725/6250 train_loss:2.7798 aux_loss:0.0467 train_time:95174ms step_avg:133.11ms
step:726/6250 train_loss:2.6514 aux_loss:0.0466 train_time:95305ms step_avg:133.11ms
step:727/6250 train_loss:2.8082 aux_loss:0.0465 train_time:95437ms step_avg:133.11ms
step:728/6250 train_loss:2.6045 aux_loss:0.0469 train_time:95568ms step_avg:133.10ms
step:729/6250 train_loss:2.7241 aux_loss:0.0466 train_time:95699ms step_avg:133.10ms
step:730/6250 train_loss:2.7987 aux_loss:0.0469 train_time:95830ms step_avg:133.10ms
step:731/6250 train_loss:2.8371 aux_loss:0.0467 train_time:95962ms step_avg:133.10ms
step:732/6250 train_loss:2.7665 aux_loss:0.0468 train_time:96092ms step_avg:133.09ms
step:733/6250 train_loss:2.7518 aux_loss:0.0467 train_time:96223ms step_avg:133.09ms
step:734/6250 train_loss:2.6789 aux_loss:0.0468 train_time:96354ms step_avg:133.09ms
step:735/6250 train_loss:2.7423 aux_loss:0.0466 train_time:96486ms step_avg:133.08ms
step:736/6250 train_loss:2.7811 aux_loss:0.0464 train_time:96617ms step_avg:133.08ms
step:737/6250 train_loss:2.7756 aux_loss:0.0465 train_time:96748ms step_avg:133.08ms
step:738/6250 train_loss:2.7366 aux_loss:0.0466 train_time:96880ms step_avg:133.08ms
step:739/6250 train_loss:2.7099 aux_loss:0.0467 train_time:97011ms step_avg:133.07ms
step:740/6250 train_loss:2.7911 aux_loss:0.0465 train_time:97141ms step_avg:133.07ms
step:741/6250 train_loss:2.6772 aux_loss:0.0465 train_time:97272ms step_avg:133.07ms
step:742/6250 train_loss:2.7504 aux_loss:0.0465 train_time:97404ms step_avg:133.07ms
step:743/6250 train_loss:2.6649 aux_loss:0.0466 train_time:97536ms step_avg:133.06ms
step:744/6250 train_loss:2.6986 aux_loss:0.0467 train_time:97667ms step_avg:133.06ms
step:745/6250 train_loss:2.7560 aux_loss:0.0466 train_time:97799ms step_avg:133.06ms
step:746/6250 train_loss:2.6671 aux_loss:0.0466 train_time:97930ms step_avg:133.06ms
step:747/6250 train_loss:2.8136 aux_loss:0.0465 train_time:98061ms step_avg:133.05ms
step:748/6250 train_loss:2.8261 aux_loss:0.0466 train_time:98191ms step_avg:133.05ms
step:749/6250 train_loss:2.6295 aux_loss:0.0465 train_time:98323ms step_avg:133.05ms
step:750/6250 train_loss:2.6653 aux_loss:0.0467 train_time:98456ms step_avg:133.05ms
step:751/6250 train_loss:2.6695 aux_loss:0.0464 train_time:98589ms step_avg:133.05ms
step:752/6250 train_loss:2.7274 aux_loss:0.0462 train_time:98721ms step_avg:133.05ms
step:753/6250 train_loss:2.8409 aux_loss:0.0462 train_time:98853ms step_avg:133.05ms
step:754/6250 train_loss:2.7146 aux_loss:0.0465 train_time:98985ms step_avg:133.04ms
step:755/6250 train_loss:2.7376 aux_loss:0.0464 train_time:99117ms step_avg:133.04ms
step:756/6250 train_loss:2.6990 aux_loss:0.0464 train_time:99249ms step_avg:133.04ms
step:757/6250 train_loss:2.7413 aux_loss:0.0463 train_time:99381ms step_avg:133.04ms
step:758/6250 train_loss:2.6368 aux_loss:0.0464 train_time:99513ms step_avg:133.04ms
step:759/6250 train_loss:2.6860 aux_loss:0.0463 train_time:99646ms step_avg:133.04ms
step:760/6250 train_loss:2.7231 aux_loss:0.0465 train_time:99779ms step_avg:133.04ms
step:761/6250 train_loss:2.7475 aux_loss:0.0463 train_time:99911ms step_avg:133.04ms
step:762/6250 train_loss:2.7414 aux_loss:0.0462 train_time:100042ms step_avg:133.03ms
step:763/6250 train_loss:2.8278 aux_loss:0.0462 train_time:100174ms step_avg:133.03ms
step:764/6250 train_loss:2.6149 aux_loss:0.0464 train_time:100305ms step_avg:133.03ms
step:765/6250 train_loss:2.7881 aux_loss:0.0463 train_time:100439ms step_avg:133.03ms
step:766/6250 train_loss:2.7604 aux_loss:0.0463 train_time:100571ms step_avg:133.03ms
step:767/6250 train_loss:2.6839 aux_loss:0.0464 train_time:100703ms step_avg:133.03ms
step:768/6250 train_loss:2.6308 aux_loss:0.0465 train_time:100835ms step_avg:133.03ms
step:769/6250 train_loss:2.7081 aux_loss:0.0462 train_time:100967ms step_avg:133.03ms
step:770/6250 train_loss:2.6481 aux_loss:0.0465 train_time:101097ms step_avg:133.02ms
step:771/6250 train_loss:2.7892 aux_loss:0.0461 train_time:101228ms step_avg:133.02ms
step:772/6250 train_loss:2.7876 aux_loss:0.0464 train_time:101359ms step_avg:133.02ms
step:773/6250 train_loss:2.7728 aux_loss:0.0463 train_time:101490ms step_avg:133.01ms
step:774/6250 train_loss:2.6615 aux_loss:0.0464 train_time:101621ms step_avg:133.01ms
step:775/6250 train_loss:2.7940 aux_loss:0.0463 train_time:101752ms step_avg:133.01ms
step:776/6250 train_loss:2.6545 aux_loss:0.0463 train_time:101885ms step_avg:133.01ms
step:777/6250 train_loss:2.7459 aux_loss:0.0462 train_time:102016ms step_avg:133.01ms
step:778/6250 train_loss:2.7071 aux_loss:0.0461 train_time:102148ms step_avg:133.01ms
step:779/6250 train_loss:2.7071 aux_loss:0.0463 train_time:102279ms step_avg:133.00ms
step:780/6250 train_loss:2.7436 aux_loss:0.0461 train_time:102412ms step_avg:133.00ms
step:781/6250 train_loss:2.7612 aux_loss:0.0461 train_time:102545ms step_avg:133.00ms
step:782/6250 train_loss:2.7691 aux_loss:0.0463 train_time:102676ms step_avg:133.00ms
step:783/6250 train_loss:2.7299 aux_loss:0.0461 train_time:102810ms step_avg:133.00ms
step:784/6250 train_loss:2.6690 aux_loss:0.0463 train_time:102944ms step_avg:133.00ms
step:785/6250 train_loss:2.7619 aux_loss:0.0463 train_time:103075ms step_avg:133.00ms
step:786/6250 train_loss:2.7299 aux_loss:0.0461 train_time:103207ms step_avg:133.00ms
step:787/6250 train_loss:2.7478 aux_loss:0.0461 train_time:103339ms step_avg:133.00ms
step:788/6250 train_loss:2.6893 aux_loss:0.0461 train_time:103470ms step_avg:133.00ms
step:789/6250 train_loss:2.5912 aux_loss:0.0464 train_time:103601ms step_avg:132.99ms
step:790/6250 train_loss:2.6777 aux_loss:0.0462 train_time:103732ms step_avg:132.99ms
step:791/6250 train_loss:2.6970 aux_loss:0.0463 train_time:103865ms step_avg:132.99ms
step:792/6250 train_loss:2.7241 aux_loss:0.0461 train_time:103996ms step_avg:132.99ms
step:793/6250 train_loss:2.6926 aux_loss:0.0462 train_time:104126ms step_avg:132.98ms
step:794/6250 train_loss:2.7006 aux_loss:0.0462 train_time:104258ms step_avg:132.98ms
step:795/6250 train_loss:2.7321 aux_loss:0.0462 train_time:104389ms step_avg:132.98ms
step:796/6250 train_loss:2.6416 aux_loss:0.0461 train_time:104520ms step_avg:132.98ms
step:797/6250 train_loss:2.6689 aux_loss:0.0461 train_time:104650ms step_avg:132.97ms
step:798/6250 train_loss:2.7381 aux_loss:0.0459 train_time:104782ms step_avg:132.97ms
step:799/6250 train_loss:2.6413 aux_loss:0.0463 train_time:104915ms step_avg:132.97ms
step:800/6250 train_loss:2.7025 aux_loss:0.0462 train_time:105046ms step_avg:132.97ms
step:801/6250 train_loss:2.7497 aux_loss:0.0464 train_time:105178ms step_avg:132.97ms
step:802/6250 train_loss:2.6706 aux_loss:0.0463 train_time:105309ms step_avg:132.97ms
step:803/6250 train_loss:2.7505 aux_loss:0.0458 train_time:105442ms step_avg:132.97ms
step:804/6250 train_loss:2.7162 aux_loss:0.0458 train_time:105573ms step_avg:132.96ms
step:805/6250 train_loss:2.6947 aux_loss:0.0458 train_time:105705ms step_avg:132.96ms
step:806/6250 train_loss:2.6592 aux_loss:0.0460 train_time:105837ms step_avg:132.96ms
step:807/6250 train_loss:2.7683 aux_loss:0.0460 train_time:105968ms step_avg:132.96ms
step:808/6250 train_loss:2.6861 aux_loss:0.0460 train_time:106099ms step_avg:132.96ms
step:809/6250 train_loss:2.6971 aux_loss:0.0459 train_time:106230ms step_avg:132.95ms
step:810/6250 train_loss:2.7487 aux_loss:0.0460 train_time:106362ms step_avg:132.95ms
step:811/6250 train_loss:2.6848 aux_loss:0.0459 train_time:106493ms step_avg:132.95ms
step:812/6250 train_loss:2.7711 aux_loss:0.0458 train_time:106624ms step_avg:132.95ms
step:813/6250 train_loss:2.7749 aux_loss:0.0460 train_time:106755ms step_avg:132.95ms
step:814/6250 train_loss:2.6720 aux_loss:0.0461 train_time:106887ms step_avg:132.94ms
step:815/6250 train_loss:2.7443 aux_loss:0.0462 train_time:107019ms step_avg:132.94ms
step:816/6250 train_loss:2.7288 aux_loss:0.0460 train_time:107150ms step_avg:132.94ms
step:817/6250 train_loss:2.5718 aux_loss:0.0464 train_time:107281ms step_avg:132.94ms
step:818/6250 train_loss:2.6668 aux_loss:0.0458 train_time:107412ms step_avg:132.94ms
step:819/6250 train_loss:2.7036 aux_loss:0.0457 train_time:107544ms step_avg:132.93ms
step:820/6250 train_loss:2.5747 aux_loss:0.0461 train_time:107675ms step_avg:132.93ms
step:821/6250 train_loss:2.7770 aux_loss:0.0459 train_time:107806ms step_avg:132.93ms
step:822/6250 train_loss:2.6575 aux_loss:0.0456 train_time:107940ms step_avg:132.93ms
step:823/6250 train_loss:2.6672 aux_loss:0.0456 train_time:108071ms step_avg:132.93ms
step:824/6250 train_loss:2.6312 aux_loss:0.0457 train_time:108203ms step_avg:132.93ms
step:825/6250 train_loss:2.6329 aux_loss:0.0461 train_time:108336ms step_avg:132.93ms
step:826/6250 train_loss:2.6966 aux_loss:0.0461 train_time:108468ms step_avg:132.93ms
step:827/6250 train_loss:2.6821 aux_loss:0.0461 train_time:108599ms step_avg:132.92ms
step:828/6250 train_loss:2.6991 aux_loss:0.0458 train_time:108730ms step_avg:132.92ms
step:829/6250 train_loss:2.6223 aux_loss:0.0458 train_time:108864ms step_avg:132.92ms
step:830/6250 train_loss:2.6296 aux_loss:0.0461 train_time:108995ms step_avg:132.92ms
step:831/6250 train_loss:2.7206 aux_loss:0.0461 train_time:109127ms step_avg:132.92ms
step:832/6250 train_loss:2.6346 aux_loss:0.0463 train_time:109259ms step_avg:132.92ms
step:833/6250 train_loss:2.6678 aux_loss:0.0461 train_time:109392ms step_avg:132.92ms
step:834/6250 train_loss:2.6950 aux_loss:0.0459 train_time:109525ms step_avg:132.92ms
step:835/6250 train_loss:2.7835 aux_loss:0.0459 train_time:109655ms step_avg:132.91ms
step:836/6250 train_loss:2.7037 aux_loss:0.0457 train_time:109786ms step_avg:132.91ms
step:837/6250 train_loss:2.5621 aux_loss:0.0461 train_time:109919ms step_avg:132.91ms
step:838/6250 train_loss:2.6579 aux_loss:0.0458 train_time:110049ms step_avg:132.91ms
step:839/6250 train_loss:2.5861 aux_loss:0.0460 train_time:110180ms step_avg:132.91ms
step:840/6250 train_loss:2.6545 aux_loss:0.0462 train_time:110311ms step_avg:132.90ms
step:841/6250 train_loss:2.7359 aux_loss:0.0459 train_time:110444ms step_avg:132.90ms
step:842/6250 train_loss:2.6578 aux_loss:0.0459 train_time:110575ms step_avg:132.90ms
step:843/6250 train_loss:2.7299 aux_loss:0.0460 train_time:110707ms step_avg:132.90ms
step:844/6250 train_loss:2.6227 aux_loss:0.0462 train_time:110839ms step_avg:132.90ms
step:845/6250 train_loss:2.6137 aux_loss:0.0459 train_time:110971ms step_avg:132.90ms
step:846/6250 train_loss:2.6511 aux_loss:0.0459 train_time:111103ms step_avg:132.90ms
step:847/6250 train_loss:2.6391 aux_loss:0.0458 train_time:111233ms step_avg:132.89ms
step:848/6250 train_loss:2.7347 aux_loss:0.0458 train_time:111365ms step_avg:132.89ms
step:849/6250 train_loss:2.7347 aux_loss:0.0456 train_time:111496ms step_avg:132.89ms
step:850/6250 train_loss:2.7324 aux_loss:0.0458 train_time:111628ms step_avg:132.89ms
step:851/6250 train_loss:2.6774 aux_loss:0.0462 train_time:111760ms step_avg:132.89ms
step:852/6250 train_loss:2.6934 aux_loss:0.0454 train_time:111891ms step_avg:132.89ms
step:853/6250 train_loss:2.6804 aux_loss:0.0454 train_time:112024ms step_avg:132.89ms
step:854/6250 train_loss:2.7845 aux_loss:0.0454 train_time:112155ms step_avg:132.89ms
step:855/6250 train_loss:2.7169 aux_loss:0.0458 train_time:112287ms step_avg:132.88ms
step:856/6250 train_loss:2.6194 aux_loss:0.0457 train_time:112420ms step_avg:132.88ms
step:857/6250 train_loss:2.5937 aux_loss:0.0456 train_time:112552ms step_avg:132.88ms
step:858/6250 train_loss:2.6525 aux_loss:0.0455 train_time:112684ms step_avg:132.88ms
step:859/6250 train_loss:2.7133 aux_loss:0.0455 train_time:112816ms step_avg:132.88ms
step:860/6250 train_loss:2.7596 aux_loss:0.0456 train_time:112948ms step_avg:132.88ms
step:861/6250 train_loss:2.7010 aux_loss:0.0457 train_time:113080ms step_avg:132.88ms
step:862/6250 train_loss:2.7192 aux_loss:0.0458 train_time:113211ms step_avg:132.88ms
step:863/6250 train_loss:2.6706 aux_loss:0.0459 train_time:113342ms step_avg:132.88ms
step:864/6250 train_loss:2.6890 aux_loss:0.0458 train_time:113474ms step_avg:132.87ms
step:865/6250 train_loss:2.7188 aux_loss:0.0454 train_time:113605ms step_avg:132.87ms
step:866/6250 train_loss:2.6636 aux_loss:0.0455 train_time:113736ms step_avg:132.87ms
step:867/6250 train_loss:2.6374 aux_loss:0.0454 train_time:113868ms step_avg:132.87ms
step:868/6250 train_loss:2.6397 aux_loss:0.0455 train_time:114001ms step_avg:132.87ms
step:869/6250 train_loss:2.6796 aux_loss:0.0455 train_time:114134ms step_avg:132.87ms
step:870/6250 train_loss:2.6939 aux_loss:0.0454 train_time:114267ms step_avg:132.87ms
step:871/6250 train_loss:2.6389 aux_loss:0.0456 train_time:114398ms step_avg:132.87ms
step:872/6250 train_loss:2.7041 aux_loss:0.0455 train_time:114529ms step_avg:132.86ms
step:873/6250 train_loss:2.6352 aux_loss:0.0457 train_time:114661ms step_avg:132.86ms
step:874/6250 train_loss:2.6138 aux_loss:0.0457 train_time:114791ms step_avg:132.86ms
step:875/6250 train_loss:2.5902 aux_loss:0.0455 train_time:114923ms step_avg:132.86ms
step:876/6250 train_loss:2.7473 aux_loss:0.0455 train_time:115054ms step_avg:132.86ms
step:877/6250 train_loss:2.6934 aux_loss:0.0453 train_time:115185ms step_avg:132.85ms
step:878/6250 train_loss:2.6282 aux_loss:0.0454 train_time:115316ms step_avg:132.85ms
step:879/6250 train_loss:2.7306 aux_loss:0.0453 train_time:115447ms step_avg:132.85ms
step:880/6250 train_loss:2.7051 aux_loss:0.0453 train_time:115578ms step_avg:132.85ms
step:881/6250 train_loss:2.6339 aux_loss:0.0453 train_time:115711ms step_avg:132.85ms
step:882/6250 train_loss:2.6270 aux_loss:0.0457 train_time:115844ms step_avg:132.85ms
step:883/6250 train_loss:2.6876 aux_loss:0.0457 train_time:115977ms step_avg:132.85ms
step:884/6250 train_loss:2.6822 aux_loss:0.0455 train_time:116109ms step_avg:132.85ms
step:885/6250 train_loss:2.5805 aux_loss:0.0457 train_time:116241ms step_avg:132.85ms
step:886/6250 train_loss:2.5913 aux_loss:0.0456 train_time:116372ms step_avg:132.84ms
step:887/6250 train_loss:2.5491 aux_loss:0.0457 train_time:116505ms step_avg:132.84ms
step:888/6250 train_loss:2.5585 aux_loss:0.0457 train_time:116636ms step_avg:132.84ms
step:889/6250 train_loss:2.6275 aux_loss:0.0455 train_time:116767ms step_avg:132.84ms
step:890/6250 train_loss:2.7104 aux_loss:0.0454 train_time:116901ms step_avg:132.84ms
step:891/6250 train_loss:2.6285 aux_loss:0.0457 train_time:117033ms step_avg:132.84ms
step:892/6250 train_loss:2.5416 aux_loss:0.0457 train_time:117165ms step_avg:132.84ms
step:893/6250 train_loss:2.7232 aux_loss:0.0453 train_time:117298ms step_avg:132.84ms
step:894/6250 train_loss:2.6567 aux_loss:0.0453 train_time:117429ms step_avg:132.84ms
step:895/6250 train_loss:2.7191 aux_loss:0.0454 train_time:117561ms step_avg:132.84ms
step:896/6250 train_loss:2.6726 aux_loss:0.0453 train_time:117693ms step_avg:132.84ms
step:897/6250 train_loss:2.6479 aux_loss:0.0454 train_time:117825ms step_avg:132.84ms
step:898/6250 train_loss:2.5931 aux_loss:0.0456 train_time:117956ms step_avg:132.83ms
step:899/6250 train_loss:2.5626 aux_loss:0.0457 train_time:118087ms step_avg:132.83ms
step:900/6250 train_loss:2.5898 aux_loss:0.0455 train_time:118218ms step_avg:132.83ms
step:901/6250 train_loss:2.6792 aux_loss:0.0452 train_time:118349ms step_avg:132.83ms
step:902/6250 train_loss:2.6921 aux_loss:0.0451 train_time:118482ms step_avg:132.83ms
step:903/6250 train_loss:2.5975 aux_loss:0.0453 train_time:118613ms step_avg:132.83ms
step:904/6250 train_loss:2.6013 aux_loss:0.0453 train_time:118744ms step_avg:132.82ms
step:905/6250 train_loss:2.6331 aux_loss:0.0455 train_time:118876ms step_avg:132.82ms
step:906/6250 train_loss:2.6145 aux_loss:0.0453 train_time:119007ms step_avg:132.82ms
step:907/6250 train_loss:2.7387 aux_loss:0.0452 train_time:119138ms step_avg:132.82ms
step:908/6250 train_loss:2.5841 aux_loss:0.0454 train_time:119269ms step_avg:132.82ms
step:909/6250 train_loss:2.6577 aux_loss:0.0450 train_time:119401ms step_avg:132.81ms
step:910/6250 train_loss:2.6478 aux_loss:0.0452 train_time:119533ms step_avg:132.81ms
step:911/6250 train_loss:2.6157 aux_loss:0.0451 train_time:119664ms step_avg:132.81ms
step:912/6250 train_loss:2.5922 aux_loss:0.0454 train_time:119795ms step_avg:132.81ms
step:913/6250 train_loss:2.6777 aux_loss:0.0452 train_time:119927ms step_avg:132.81ms
step:914/6250 train_loss:2.6899 aux_loss:0.0452 train_time:120059ms step_avg:132.81ms
step:915/6250 train_loss:2.6179 aux_loss:0.0452 train_time:120191ms step_avg:132.81ms
step:916/6250 train_loss:2.5890 aux_loss:0.0453 train_time:120323ms step_avg:132.81ms
step:917/6250 train_loss:2.5649 aux_loss:0.0453 train_time:120454ms step_avg:132.80ms
step:918/6250 train_loss:2.6835 aux_loss:0.0455 train_time:120585ms step_avg:132.80ms
step:919/6250 train_loss:2.5718 aux_loss:0.0454 train_time:120716ms step_avg:132.80ms
step:920/6250 train_loss:2.6147 aux_loss:0.0452 train_time:120847ms step_avg:132.80ms
step:921/6250 train_loss:2.5670 aux_loss:0.0452 train_time:120980ms step_avg:132.80ms
step:922/6250 train_loss:2.6433 aux_loss:0.0453 train_time:121112ms step_avg:132.80ms
step:923/6250 train_loss:2.6096 aux_loss:0.0453 train_time:121244ms step_avg:132.80ms
step:924/6250 train_loss:2.6324 aux_loss:0.0450 train_time:121375ms step_avg:132.80ms
step:925/6250 train_loss:2.6713 aux_loss:0.0449 train_time:121506ms step_avg:132.79ms
step:926/6250 train_loss:2.5520 aux_loss:0.0453 train_time:121637ms step_avg:132.79ms
step:927/6250 train_loss:2.5222 aux_loss:0.0450 train_time:121769ms step_avg:132.79ms
step:928/6250 train_loss:2.7254 aux_loss:0.0451 train_time:121900ms step_avg:132.79ms
step:929/6250 train_loss:2.6327 aux_loss:0.0450 train_time:122032ms step_avg:132.79ms
step:930/6250 train_loss:2.6321 aux_loss:0.0451 train_time:122163ms step_avg:132.79ms
step:931/6250 train_loss:2.6995 aux_loss:0.0451 train_time:122295ms step_avg:132.79ms
step:932/6250 train_loss:2.6876 aux_loss:0.0448 train_time:122426ms step_avg:132.78ms
step:933/6250 train_loss:2.6124 aux_loss:0.0450 train_time:122558ms step_avg:132.78ms
step:934/6250 train_loss:2.5807 aux_loss:0.0451 train_time:122689ms step_avg:132.78ms
step:935/6250 train_loss:2.7004 aux_loss:0.0452 train_time:122821ms step_avg:132.78ms
step:936/6250 train_loss:2.6993 aux_loss:0.0451 train_time:122954ms step_avg:132.78ms
step:937/6250 train_loss:2.5677 aux_loss:0.0452 train_time:123085ms step_avg:132.78ms
step:938/6250 train_loss:2.6532 aux_loss:0.0451 train_time:123216ms step_avg:132.78ms
step:939/6250 train_loss:2.6306 aux_loss:0.0453 train_time:123348ms step_avg:132.77ms
step:940/6250 train_loss:2.6333 aux_loss:0.0452 train_time:123481ms step_avg:132.78ms
step:941/6250 train_loss:2.6070 aux_loss:0.0452 train_time:123612ms step_avg:132.77ms
step:942/6250 train_loss:2.5881 aux_loss:0.0452 train_time:123743ms step_avg:132.77ms
step:943/6250 train_loss:2.5829 aux_loss:0.0451 train_time:123874ms step_avg:132.77ms
step:944/6250 train_loss:2.6122 aux_loss:0.0450 train_time:124005ms step_avg:132.77ms
step:945/6250 train_loss:2.6222 aux_loss:0.0448 train_time:124137ms step_avg:132.77ms
step:946/6250 train_loss:2.5392 aux_loss:0.0450 train_time:124267ms step_avg:132.76ms
step:947/6250 train_loss:2.5579 aux_loss:0.0451 train_time:124398ms step_avg:132.76ms
step:948/6250 train_loss:2.5559 aux_loss:0.0450 train_time:124530ms step_avg:132.76ms
step:949/6250 train_loss:2.7765 aux_loss:0.0449 train_time:124661ms step_avg:132.76ms
step:950/6250 train_loss:2.6294 aux_loss:0.0452 train_time:124793ms step_avg:132.76ms
step:951/6250 train_loss:2.6324 aux_loss:0.0450 train_time:124926ms step_avg:132.76ms
step:952/6250 train_loss:2.5813 aux_loss:0.0449 train_time:125058ms step_avg:132.76ms
step:953/6250 train_loss:2.5902 aux_loss:0.0451 train_time:125189ms step_avg:132.76ms
step:954/6250 train_loss:2.6319 aux_loss:0.0450 train_time:125322ms step_avg:132.76ms
step:955/6250 train_loss:2.5880 aux_loss:0.0451 train_time:125454ms step_avg:132.76ms
step:956/6250 train_loss:2.7526 aux_loss:0.0446 train_time:125586ms step_avg:132.75ms
step:957/6250 train_loss:2.5758 aux_loss:0.0449 train_time:125718ms step_avg:132.75ms
step:958/6250 train_loss:2.6982 aux_loss:0.0449 train_time:125849ms step_avg:132.75ms
step:959/6250 train_loss:2.6148 aux_loss:0.0450 train_time:125982ms step_avg:132.75ms
step:960/6250 train_loss:2.6801 aux_loss:0.0449 train_time:126114ms step_avg:132.75ms
step:961/6250 train_loss:2.5925 aux_loss:0.0450 train_time:126244ms step_avg:132.75ms
step:962/6250 train_loss:2.6216 aux_loss:0.0451 train_time:126376ms step_avg:132.75ms
step:963/6250 train_loss:2.6105 aux_loss:0.0451 train_time:126507ms step_avg:132.75ms
step:964/6250 train_loss:2.6725 aux_loss:0.0449 train_time:126639ms step_avg:132.75ms
step:965/6250 train_loss:2.5067 aux_loss:0.0453 train_time:126770ms step_avg:132.74ms
step:966/6250 train_loss:2.5522 aux_loss:0.0451 train_time:126901ms step_avg:132.74ms
step:967/6250 train_loss:2.5952 aux_loss:0.0447 train_time:127033ms step_avg:132.74ms
step:968/6250 train_loss:2.6500 aux_loss:0.0448 train_time:127164ms step_avg:132.74ms
step:969/6250 train_loss:2.6190 aux_loss:0.0449 train_time:127294ms step_avg:132.74ms
step:970/6250 train_loss:2.6083 aux_loss:0.0447 train_time:127426ms step_avg:132.74ms
step:971/6250 train_loss:2.5630 aux_loss:0.0448 train_time:127559ms step_avg:132.74ms
step:972/6250 train_loss:2.6275 aux_loss:0.0447 train_time:127690ms step_avg:132.73ms
step:973/6250 train_loss:2.6367 aux_loss:0.0446 train_time:127822ms step_avg:132.73ms
step:974/6250 train_loss:2.5473 aux_loss:0.0448 train_time:127953ms step_avg:132.73ms
step:975/6250 train_loss:2.6262 aux_loss:0.0445 train_time:128083ms step_avg:132.73ms
step:976/6250 train_loss:2.7007 aux_loss:0.0445 train_time:128214ms step_avg:132.73ms
step:977/6250 train_loss:2.6461 aux_loss:0.0448 train_time:128345ms step_avg:132.73ms
step:978/6250 train_loss:2.5486 aux_loss:0.0450 train_time:128479ms step_avg:132.73ms
step:979/6250 train_loss:2.5850 aux_loss:0.0449 train_time:128621ms step_avg:132.74ms
step:980/6250 train_loss:2.6114 aux_loss:0.0446 train_time:128764ms step_avg:132.75ms
step:981/6250 train_loss:2.5961 aux_loss:0.0445 train_time:128901ms step_avg:132.75ms
step:982/6250 train_loss:2.6224 aux_loss:0.0445 train_time:129038ms step_avg:132.75ms
step:983/6250 train_loss:2.5449 aux_loss:0.0448 train_time:129172ms step_avg:132.76ms
step:984/6250 train_loss:2.6736 aux_loss:0.0444 train_time:129315ms step_avg:132.77ms
step:985/6250 train_loss:2.6747 aux_loss:0.0446 train_time:129452ms step_avg:132.77ms
step:986/6250 train_loss:2.5949 aux_loss:0.0447 train_time:129584ms step_avg:132.77ms
step:987/6250 train_loss:2.7797 aux_loss:0.0447 train_time:129715ms step_avg:132.77ms
step:988/6250 train_loss:2.5313 aux_loss:0.0446 train_time:129847ms step_avg:132.77ms
step:989/6250 train_loss:2.6765 aux_loss:0.0446 train_time:129979ms step_avg:132.77ms
step:990/6250 train_loss:2.6721 aux_loss:0.0447 train_time:130110ms step_avg:132.77ms
step:991/6250 train_loss:2.6029 aux_loss:0.0449 train_time:130241ms step_avg:132.76ms
step:992/6250 train_loss:2.6107 aux_loss:0.0448 train_time:130372ms step_avg:132.76ms
step:993/6250 train_loss:2.6061 aux_loss:0.0446 train_time:130503ms step_avg:132.76ms
step:994/6250 train_loss:2.5456 aux_loss:0.0446 train_time:130635ms step_avg:132.76ms
step:995/6250 train_loss:2.5160 aux_loss:0.0445 train_time:130765ms step_avg:132.76ms
step:996/6250 train_loss:2.6593 aux_loss:0.0445 train_time:130897ms step_avg:132.76ms
step:997/6250 train_loss:2.4921 aux_loss:0.0445 train_time:131029ms step_avg:132.75ms
step:998/6250 train_loss:2.6708 aux_loss:0.0444 train_time:131161ms step_avg:132.75ms
step:999/6250 train_loss:2.5632 aux_loss:0.0445 train_time:131292ms step_avg:132.75ms
step:1000/6250 train_loss:2.5526 aux_loss:0.0443 train_time:131422ms step_avg:132.75ms
step:1001/6250 train_loss:2.5520 aux_loss:0.0446 train_time:131554ms step_avg:132.75ms
step:1002/6250 train_loss:2.5732 aux_loss:0.0449 train_time:131685ms step_avg:132.75ms
step:1003/6250 train_loss:2.6452 aux_loss:0.0448 train_time:131817ms step_avg:132.75ms
step:1004/6250 train_loss:2.6091 aux_loss:0.0447 train_time:131948ms step_avg:132.74ms
step:1005/6250 train_loss:2.6535 aux_loss:0.0446 train_time:132079ms step_avg:132.74ms
step:1006/6250 train_loss:2.5558 aux_loss:0.0447 train_time:132211ms step_avg:132.74ms
step:1007/6250 train_loss:2.5440 aux_loss:0.0446 train_time:132341ms step_avg:132.74ms
step:1008/6250 train_loss:2.6025 aux_loss:0.0444 train_time:132473ms step_avg:132.74ms
step:1009/6250 train_loss:2.4824 aux_loss:0.0445 train_time:132604ms step_avg:132.74ms
step:1010/6250 train_loss:2.6265 aux_loss:0.0446 train_time:132736ms step_avg:132.74ms
step:1011/6250 train_loss:2.5097 aux_loss:0.0451 train_time:132868ms step_avg:132.74ms
step:1012/6250 train_loss:2.6429 aux_loss:0.0448 train_time:133001ms step_avg:132.74ms
step:1013/6250 train_loss:2.5114 aux_loss:0.0447 train_time:133131ms step_avg:132.73ms
step:1014/6250 train_loss:2.6154 aux_loss:0.0446 train_time:133262ms step_avg:132.73ms
step:1015/6250 train_loss:2.5205 aux_loss:0.0449 train_time:133406ms step_avg:132.74ms
step:1016/6250 train_loss:2.5539 aux_loss:0.0445 train_time:133549ms step_avg:132.75ms
step:1017/6250 train_loss:2.6202 aux_loss:0.0444 train_time:133691ms step_avg:132.76ms
step:1018/6250 train_loss:2.6687 aux_loss:0.0442 train_time:133836ms step_avg:132.77ms
step:1019/6250 train_loss:2.4972 aux_loss:0.0442 train_time:133980ms step_avg:132.78ms
step:1020/6250 train_loss:2.5407 aux_loss:0.0446 train_time:134125ms step_avg:132.80ms
step:1021/6250 train_loss:2.6413 aux_loss:0.0447 train_time:134268ms step_avg:132.81ms
step:1022/6250 train_loss:2.7003 aux_loss:0.0445 train_time:134411ms step_avg:132.82ms
step:1023/6250 train_loss:2.5869 aux_loss:0.0444 train_time:134555ms step_avg:132.83ms
step:1024/6250 train_loss:2.5414 aux_loss:0.0445 train_time:134691ms step_avg:132.83ms
step:1025/6250 train_loss:2.5509 aux_loss:0.0443 train_time:134831ms step_avg:132.84ms
step:1026/6250 train_loss:2.5420 aux_loss:0.0444 train_time:134977ms step_avg:132.85ms
step:1027/6250 train_loss:2.5151 aux_loss:0.0446 train_time:135120ms step_avg:132.86ms
step:1028/6250 train_loss:2.6013 aux_loss:0.0445 train_time:135262ms step_avg:132.87ms
step:1029/6250 train_loss:2.5987 aux_loss:0.0443 train_time:135400ms step_avg:132.88ms
step:1030/6250 train_loss:2.6605 aux_loss:0.0445 train_time:135533ms step_avg:132.88ms
step:1031/6250 train_loss:2.5097 aux_loss:0.0445 train_time:135675ms step_avg:132.88ms
step:1032/6250 train_loss:2.5142 aux_loss:0.0445 train_time:135814ms step_avg:132.89ms
step:1033/6250 train_loss:2.5000 aux_loss:0.0445 train_time:135947ms step_avg:132.89ms
step:1034/6250 train_loss:2.5909 aux_loss:0.0444 train_time:136079ms step_avg:132.89ms
step:1035/6250 train_loss:2.5760 aux_loss:0.0444 train_time:136217ms step_avg:132.89ms
step:1036/6250 train_loss:2.5662 aux_loss:0.0443 train_time:136362ms step_avg:132.91ms
step:1037/6250 train_loss:2.5621 aux_loss:0.0442 train_time:136499ms step_avg:132.91ms
step:1038/6250 train_loss:2.5813 aux_loss:0.0441 train_time:136640ms step_avg:132.92ms
step:1039/6250 train_loss:2.5707 aux_loss:0.0442 train_time:136784ms step_avg:132.93ms
step:1040/6250 train_loss:2.5314 aux_loss:0.0444 train_time:136922ms step_avg:132.93ms
step:1041/6250 train_loss:2.6409 aux_loss:0.0443 train_time:137059ms step_avg:132.94ms
step:1042/6250 train_loss:2.6164 aux_loss:0.0445 train_time:137203ms step_avg:132.95ms
step:1043/6250 train_loss:2.5535 aux_loss:0.0444 train_time:137340ms step_avg:132.95ms
step:1044/6250 train_loss:2.6382 aux_loss:0.0444 train_time:137477ms step_avg:132.96ms
step:1045/6250 train_loss:2.5398 aux_loss:0.0445 train_time:137617ms step_avg:132.96ms
step:1046/6250 train_loss:2.5593 aux_loss:0.0443 train_time:137759ms step_avg:132.97ms
step:1047/6250 train_loss:2.5541 aux_loss:0.0448 train_time:137899ms step_avg:132.98ms
step:1048/6250 train_loss:2.5574 aux_loss:0.0443 train_time:138033ms step_avg:132.98ms
step:1049/6250 train_loss:2.5758 aux_loss:0.0443 train_time:138165ms step_avg:132.98ms
step:1050/6250 train_loss:2.4854 aux_loss:0.0443 train_time:138308ms step_avg:132.99ms
step:1051/6250 train_loss:2.6022 aux_loss:0.0441 train_time:138450ms step_avg:133.00ms
step:1052/6250 train_loss:2.5639 aux_loss:0.0442 train_time:138593ms step_avg:133.01ms
step:1053/6250 train_loss:2.5037 aux_loss:0.0443 train_time:138726ms step_avg:133.01ms
step:1054/6250 train_loss:2.6162 aux_loss:0.0444 train_time:138863ms step_avg:133.01ms
step:1055/6250 train_loss:2.5960 aux_loss:0.0444 train_time:139005ms step_avg:133.02ms
step:1056/6250 train_loss:2.5715 aux_loss:0.0446 train_time:139137ms step_avg:133.02ms
step:1057/6250 train_loss:2.6471 aux_loss:0.0442 train_time:139272ms step_avg:133.02ms
step:1058/6250 train_loss:2.5478 aux_loss:0.0441 train_time:139410ms step_avg:133.02ms
step:1059/6250 train_loss:2.6176 aux_loss:0.0442 train_time:139541ms step_avg:133.02ms
step:1060/6250 train_loss:2.5855 aux_loss:0.0443 train_time:139673ms step_avg:133.02ms
step:1061/6250 train_loss:2.6234 aux_loss:0.0441 train_time:139805ms step_avg:133.02ms
step:1062/6250 train_loss:2.5145 aux_loss:0.0442 train_time:139949ms step_avg:133.03ms
step:1063/6250 train_loss:2.5839 aux_loss:0.0441 train_time:140092ms step_avg:133.04ms
step:1064/6250 train_loss:2.5875 aux_loss:0.0440 train_time:140222ms step_avg:133.04ms
step:1065/6250 train_loss:2.5430 aux_loss:0.0440 train_time:140353ms step_avg:133.04ms
step:1066/6250 train_loss:2.6437 aux_loss:0.0441 train_time:140485ms step_avg:133.03ms
step:1067/6250 train_loss:2.5462 aux_loss:0.0443 train_time:140615ms step_avg:133.03ms
step:1068/6250 train_loss:2.5684 aux_loss:0.0445 train_time:140746ms step_avg:133.03ms
step:1069/6250 train_loss:2.6572 aux_loss:0.0444 train_time:140877ms step_avg:133.03ms
step:1070/6250 train_loss:2.5569 aux_loss:0.0442 train_time:141009ms step_avg:133.03ms
step:1071/6250 train_loss:2.5635 aux_loss:0.0442 train_time:141139ms step_avg:133.02ms
step:1072/6250 train_loss:2.5728 aux_loss:0.0441 train_time:141277ms step_avg:133.03ms
step:1073/6250 train_loss:2.4885 aux_loss:0.0441 train_time:141417ms step_avg:133.04ms
step:1074/6250 train_loss:2.4949 aux_loss:0.0442 train_time:141549ms step_avg:133.03ms
step:1075/6250 train_loss:2.5674 aux_loss:0.0443 train_time:141686ms step_avg:133.04ms
step:1076/6250 train_loss:2.6008 aux_loss:0.0443 train_time:141827ms step_avg:133.05ms
step:1077/6250 train_loss:2.5517 aux_loss:0.0442 train_time:141970ms step_avg:133.06ms
step:1078/6250 train_loss:2.5251 aux_loss:0.0439 train_time:142107ms step_avg:133.06ms
step:1079/6250 train_loss:2.5860 aux_loss:0.0438 train_time:142239ms step_avg:133.06ms
step:1080/6250 train_loss:2.5563 aux_loss:0.0442 train_time:142375ms step_avg:133.06ms
step:1081/6250 train_loss:2.4885 aux_loss:0.0444 train_time:142510ms step_avg:133.06ms
step:1082/6250 train_loss:2.6349 aux_loss:0.0440 train_time:142642ms step_avg:133.06ms
step:1083/6250 train_loss:2.5591 aux_loss:0.0440 train_time:142786ms step_avg:133.07ms
step:1084/6250 train_loss:2.5249 aux_loss:0.0441 train_time:142922ms step_avg:133.07ms
step:1085/6250 train_loss:2.5943 aux_loss:0.0441 train_time:143054ms step_avg:133.07ms
step:1086/6250 train_loss:2.5369 aux_loss:0.0444 train_time:143187ms step_avg:133.07ms
step:1087/6250 train_loss:2.5563 aux_loss:0.0443 train_time:143325ms step_avg:133.08ms
step:1088/6250 train_loss:2.6336 aux_loss:0.0440 train_time:143458ms step_avg:133.08ms
step:1089/6250 train_loss:2.6407 aux_loss:0.0440 train_time:143589ms step_avg:133.08ms
step:1090/6250 train_loss:2.4584 aux_loss:0.0437 train_time:143719ms step_avg:133.07ms
step:1091/6250 train_loss:2.5435 aux_loss:0.0437 train_time:143851ms step_avg:133.07ms
step:1092/6250 train_loss:2.5072 aux_loss:0.0438 train_time:143982ms step_avg:133.07ms
step:1093/6250 train_loss:2.5699 aux_loss:0.0438 train_time:144118ms step_avg:133.07ms
step:1094/6250 train_loss:2.5415 aux_loss:0.0441 train_time:144261ms step_avg:133.08ms
step:1095/6250 train_loss:2.6039 aux_loss:0.0440 train_time:144395ms step_avg:133.08ms
step:1096/6250 train_loss:2.5742 aux_loss:0.0440 train_time:144528ms step_avg:133.08ms
step:1097/6250 train_loss:2.5981 aux_loss:0.0438 train_time:144667ms step_avg:133.09ms
step:1098/6250 train_loss:2.5859 aux_loss:0.0437 train_time:144799ms step_avg:133.09ms
step:1099/6250 train_loss:2.6164 aux_loss:0.0439 train_time:144931ms step_avg:133.09ms
step:1100/6250 train_loss:2.5615 aux_loss:0.0438 train_time:145064ms step_avg:133.09ms
step:1101/6250 train_loss:2.4937 aux_loss:0.0442 train_time:145195ms step_avg:133.08ms
step:1102/6250 train_loss:2.5268 aux_loss:0.0440 train_time:145326ms step_avg:133.08ms
step:1103/6250 train_loss:2.5402 aux_loss:0.0439 train_time:145457ms step_avg:133.08ms
step:1104/6250 train_loss:2.6590 aux_loss:0.0438 train_time:145594ms step_avg:133.08ms
step:1105/6250 train_loss:2.6484 aux_loss:0.0438 train_time:145737ms step_avg:133.09ms
step:1106/6250 train_loss:2.5325 aux_loss:0.0438 train_time:145879ms step_avg:133.10ms
step:1107/6250 train_loss:2.6059 aux_loss:0.0439 train_time:146013ms step_avg:133.10ms
step:1108/6250 train_loss:2.5086 aux_loss:0.0442 train_time:146145ms step_avg:133.10ms
step:1109/6250 train_loss:2.5656 aux_loss:0.0440 train_time:146283ms step_avg:133.11ms
step:1110/6250 train_loss:2.6773 aux_loss:0.0438 train_time:146421ms step_avg:133.11ms
step:1111/6250 train_loss:2.5390 aux_loss:0.0437 train_time:146556ms step_avg:133.11ms
step:1112/6250 train_loss:2.5522 aux_loss:0.0437 train_time:146686ms step_avg:133.11ms
step:1113/6250 train_loss:2.5201 aux_loss:0.0439 train_time:146818ms step_avg:133.11ms
step:1114/6250 train_loss:2.3989 aux_loss:0.0440 train_time:146949ms step_avg:133.11ms
step:1115/6250 train_loss:2.5592 aux_loss:0.0442 train_time:147080ms step_avg:133.10ms
step:1116/6250 train_loss:2.4835 aux_loss:0.0441 train_time:147220ms step_avg:133.11ms
step:1117/6250 train_loss:2.4871 aux_loss:0.0437 train_time:147361ms step_avg:133.12ms
step:1118/6250 train_loss:2.4917 aux_loss:0.0434 train_time:147500ms step_avg:133.12ms
step:1119/6250 train_loss:2.5743 aux_loss:0.0438 train_time:147633ms step_avg:133.12ms
step:1120/6250 train_loss:2.5098 aux_loss:0.0440 train_time:147764ms step_avg:133.12ms
step:1121/6250 train_loss:2.3887 aux_loss:0.0438 train_time:147895ms step_avg:133.12ms
step:1122/6250 train_loss:2.5340 aux_loss:0.0435 train_time:148027ms step_avg:133.12ms
step:1123/6250 train_loss:2.6122 aux_loss:0.0434 train_time:148158ms step_avg:133.12ms
step:1124/6250 train_loss:2.5186 aux_loss:0.0439 train_time:148289ms step_avg:133.11ms
step:1125/6250 train_loss:2.5711 aux_loss:0.0440 train_time:148421ms step_avg:133.11ms
step:1126/6250 train_loss:2.4953 aux_loss:0.0441 train_time:148559ms step_avg:133.12ms
step:1127/6250 train_loss:2.6275 aux_loss:0.0439 train_time:148691ms step_avg:133.12ms
step:1128/6250 train_loss:2.5365 aux_loss:0.0439 train_time:148823ms step_avg:133.11ms
step:1129/6250 train_loss:2.5431 aux_loss:0.0437 train_time:148954ms step_avg:133.11ms
step:1130/6250 train_loss:2.6109 aux_loss:0.0436 train_time:149086ms step_avg:133.11ms
step:1131/6250 train_loss:2.5370 aux_loss:0.0439 train_time:149220ms step_avg:133.11ms
step:1132/6250 train_loss:2.5300 aux_loss:0.0440 train_time:149358ms step_avg:133.12ms
step:1133/6250 train_loss:2.6210 aux_loss:0.0437 train_time:149497ms step_avg:133.12ms
step:1134/6250 train_loss:2.5210 aux_loss:0.0438 train_time:149634ms step_avg:133.13ms
step:1135/6250 train_loss:2.5717 aux_loss:0.0436 train_time:149769ms step_avg:133.13ms
step:1136/6250 train_loss:2.5797 aux_loss:0.0434 train_time:149902ms step_avg:133.13ms
step:1137/6250 train_loss:2.5895 aux_loss:0.0435 train_time:150037ms step_avg:133.13ms
step:1138/6250 train_loss:2.5547 aux_loss:0.0437 train_time:150178ms step_avg:133.14ms
step:1139/6250 train_loss:2.5367 aux_loss:0.0438 train_time:150320ms step_avg:133.14ms
step:1140/6250 train_loss:2.4881 aux_loss:0.0438 train_time:150462ms step_avg:133.15ms
step:1141/6250 train_loss:2.5802 aux_loss:0.0439 train_time:150604ms step_avg:133.16ms
step:1142/6250 train_loss:2.5583 aux_loss:0.0438 train_time:150745ms step_avg:133.17ms
step:1143/6250 train_loss:2.4383 aux_loss:0.0441 train_time:150884ms step_avg:133.17ms
step:1144/6250 train_loss:2.6047 aux_loss:0.0438 train_time:151022ms step_avg:133.18ms
step:1145/6250 train_loss:2.5253 aux_loss:0.0437 train_time:151155ms step_avg:133.18ms
step:1146/6250 train_loss:2.5510 aux_loss:0.0436 train_time:151287ms step_avg:133.18ms
step:1147/6250 train_loss:2.4812 aux_loss:0.0433 train_time:151419ms step_avg:133.17ms
step:1148/6250 train_loss:2.5900 aux_loss:0.0434 train_time:151554ms step_avg:133.18ms
step:1149/6250 train_loss:2.4926 aux_loss:0.0435 train_time:151685ms step_avg:133.17ms
step:1150/6250 train_loss:2.4996 aux_loss:0.0434 train_time:151815ms step_avg:133.17ms
step:1151/6250 train_loss:2.5799 aux_loss:0.0434 train_time:151947ms step_avg:133.17ms
step:1152/6250 train_loss:2.5819 aux_loss:0.0433 train_time:152082ms step_avg:133.17ms
step:1153/6250 train_loss:2.5104 aux_loss:0.0436 train_time:152214ms step_avg:133.17ms
step:1154/6250 train_loss:2.4884 aux_loss:0.0436 train_time:152350ms step_avg:133.17ms
step:1155/6250 train_loss:2.6119 aux_loss:0.0434 train_time:152488ms step_avg:133.18ms
step:1156/6250 train_loss:2.5003 aux_loss:0.0434 train_time:152624ms step_avg:133.18ms
step:1157/6250 train_loss:2.5707 aux_loss:0.0436 train_time:152762ms step_avg:133.18ms
step:1158/6250 train_loss:2.6300 aux_loss:0.0433 train_time:152906ms step_avg:133.19ms
step:1159/6250 train_loss:2.5449 aux_loss:0.0434 train_time:153050ms step_avg:133.20ms
step:1160/6250 train_loss:2.4767 aux_loss:0.0433 train_time:153194ms step_avg:133.21ms
step:1161/6250 train_loss:2.5145 aux_loss:0.0435 train_time:153339ms step_avg:133.22ms
step:1162/6250 train_loss:2.4031 aux_loss:0.0434 train_time:153482ms step_avg:133.23ms
step:1163/6250 train_loss:2.5439 aux_loss:0.0437 train_time:153619ms step_avg:133.23ms
step:1164/6250 train_loss:2.5041 aux_loss:0.0437 train_time:153754ms step_avg:133.24ms
step:1165/6250 train_loss:2.5439 aux_loss:0.0437 train_time:153888ms step_avg:133.24ms
step:1166/6250 train_loss:2.5182 aux_loss:0.0437 train_time:154025ms step_avg:133.24ms
step:1167/6250 train_loss:2.5813 aux_loss:0.0435 train_time:154158ms step_avg:133.24ms
step:1168/6250 train_loss:2.4699 aux_loss:0.0438 train_time:154293ms step_avg:133.24ms
step:1169/6250 train_loss:2.5026 aux_loss:0.0436 train_time:154430ms step_avg:133.24ms
step:1170/6250 train_loss:2.5128 aux_loss:0.0434 train_time:154564ms step_avg:133.24ms
step:1171/6250 train_loss:2.4839 aux_loss:0.0433 train_time:154700ms step_avg:133.25ms
step:1172/6250 train_loss:2.5475 aux_loss:0.0430 train_time:154833ms step_avg:133.25ms
step:1173/6250 train_loss:2.5217 aux_loss:0.0432 train_time:154965ms step_avg:133.25ms
step:1174/6250 train_loss:2.5231 aux_loss:0.0432 train_time:155096ms step_avg:133.24ms
step:1175/6250 train_loss:2.6361 aux_loss:0.0432 train_time:155228ms step_avg:133.24ms
step:1176/6250 train_loss:2.4889 aux_loss:0.0434 train_time:155359ms step_avg:133.24ms
step:1177/6250 train_loss:2.5051 aux_loss:0.0434 train_time:155491ms step_avg:133.24ms
step:1178/6250 train_loss:2.5063 aux_loss:0.0434 train_time:155621ms step_avg:133.24ms
step:1179/6250 train_loss:2.5310 aux_loss:0.0433 train_time:155752ms step_avg:133.24ms
step:1180/6250 train_loss:2.5515 aux_loss:0.0433 train_time:155884ms step_avg:133.23ms
step:1181/6250 train_loss:2.5419 aux_loss:0.0433 train_time:156016ms step_avg:133.23ms
step:1182/6250 train_loss:2.4826 aux_loss:0.0433 train_time:156147ms step_avg:133.23ms
step:1183/6250 train_loss:2.5923 aux_loss:0.0433 train_time:156277ms step_avg:133.23ms
step:1184/6250 train_loss:2.5994 aux_loss:0.0431 train_time:156409ms step_avg:133.23ms
step:1185/6250 train_loss:2.5025 aux_loss:0.0430 train_time:156541ms step_avg:133.23ms
step:1186/6250 train_loss:2.3735 aux_loss:0.0433 train_time:156672ms step_avg:133.22ms
step:1187/6250 train_loss:2.4893 aux_loss:0.0431 train_time:156808ms step_avg:133.23ms
step:1188/6250 train_loss:2.5897 aux_loss:0.0430 train_time:156943ms step_avg:133.23ms
step:1189/6250 train_loss:2.5193 aux_loss:0.0431 train_time:157074ms step_avg:133.23ms
step:1190/6250 train_loss:2.5509 aux_loss:0.0433 train_time:157205ms step_avg:133.22ms
step:1191/6250 train_loss:2.5542 aux_loss:0.0433 train_time:157342ms step_avg:133.23ms
step:1192/6250 train_loss:2.4567 aux_loss:0.0434 train_time:157482ms step_avg:133.23ms
step:1193/6250 train_loss:2.6099 aux_loss:0.0432 train_time:157617ms step_avg:133.24ms
step:1194/6250 train_loss:2.6130 aux_loss:0.0431 train_time:157754ms step_avg:133.24ms
step:1195/6250 train_loss:2.4682 aux_loss:0.0431 train_time:157893ms step_avg:133.24ms
step:1196/6250 train_loss:2.5605 aux_loss:0.0431 train_time:158025ms step_avg:133.24ms
step:1197/6250 train_loss:2.5313 aux_loss:0.0431 train_time:158155ms step_avg:133.24ms
step:1198/6250 train_loss:2.4661 aux_loss:0.0432 train_time:158287ms step_avg:133.24ms
step:1199/6250 train_loss:2.5762 aux_loss:0.0432 train_time:158419ms step_avg:133.24ms
step:1200/6250 train_loss:2.4490 aux_loss:0.0434 train_time:158555ms step_avg:133.24ms
step:1201/6250 train_loss:2.6168 aux_loss:0.0433 train_time:158693ms step_avg:133.24ms
step:1202/6250 train_loss:2.4675 aux_loss:0.0434 train_time:158830ms step_avg:133.25ms
step:1203/6250 train_loss:2.5086 aux_loss:0.0432 train_time:158966ms step_avg:133.25ms
step:1204/6250 train_loss:2.4877 aux_loss:0.0433 train_time:159098ms step_avg:133.25ms
step:1205/6250 train_loss:2.5105 aux_loss:0.0431 train_time:159229ms step_avg:133.25ms
step:1206/6250 train_loss:2.5420 aux_loss:0.0430 train_time:159364ms step_avg:133.25ms
step:1207/6250 train_loss:2.4914 aux_loss:0.0430 train_time:159502ms step_avg:133.25ms
step:1208/6250 train_loss:2.5058 aux_loss:0.0427 train_time:159633ms step_avg:133.25ms
step:1209/6250 train_loss:2.4490 aux_loss:0.0433 train_time:159768ms step_avg:133.25ms
step:1210/6250 train_loss:2.4791 aux_loss:0.0435 train_time:159900ms step_avg:133.25ms
step:1211/6250 train_loss:2.5090 aux_loss:0.0434 train_time:160032ms step_avg:133.25ms
step:1212/6250 train_loss:2.5552 aux_loss:0.0433 train_time:160171ms step_avg:133.25ms
step:1213/6250 train_loss:2.4276 aux_loss:0.0432 train_time:160313ms step_avg:133.26ms
step:1214/6250 train_loss:2.5887 aux_loss:0.0429 train_time:160455ms step_avg:133.27ms
step:1215/6250 train_loss:2.5025 aux_loss:0.0429 train_time:160598ms step_avg:133.28ms
step:1216/6250 train_loss:2.5185 aux_loss:0.0431 train_time:160740ms step_avg:133.28ms
step:1217/6250 train_loss:2.5682 aux_loss:0.0433 train_time:160882ms step_avg:133.29ms
step:1218/6250 train_loss:2.5240 aux_loss:0.0432 train_time:161025ms step_avg:133.30ms
step:1219/6250 train_loss:2.5185 aux_loss:0.0432 train_time:161167ms step_avg:133.31ms
step:1220/6250 train_loss:2.5184 aux_loss:0.0430 train_time:161309ms step_avg:133.31ms
step:1221/6250 train_loss:2.5408 aux_loss:0.0429 train_time:161451ms step_avg:133.32ms
step:1222/6250 train_loss:2.4933 aux_loss:0.0432 train_time:161593ms step_avg:133.33ms
step:1223/6250 train_loss:2.4400 aux_loss:0.0431 train_time:161735ms step_avg:133.33ms
step:1224/6250 train_loss:2.4581 aux_loss:0.0430 train_time:161878ms step_avg:133.34ms
step:1225/6250 train_loss:2.5391 aux_loss:0.0430 train_time:162020ms step_avg:133.35ms
step:1226/6250 train_loss:2.5317 aux_loss:0.0432 train_time:162162ms step_avg:133.36ms
step:1227/6250 train_loss:2.5652 aux_loss:0.0430 train_time:162304ms step_avg:133.36ms
step:1228/6250 train_loss:2.4699 aux_loss:0.0431 train_time:162447ms step_avg:133.37ms
step:1229/6250 train_loss:2.5269 aux_loss:0.0429 train_time:162589ms step_avg:133.38ms
step:1230/6250 train_loss:2.5213 aux_loss:0.0430 train_time:162731ms step_avg:133.39ms
step:1231/6250 train_loss:2.4784 aux_loss:0.0429 train_time:162884ms step_avg:133.40ms
step:1232/6250 train_loss:2.4893 aux_loss:0.0427 train_time:163031ms step_avg:133.41ms
step:1233/6250 train_loss:2.6132 aux_loss:0.0429 train_time:163172ms step_avg:133.42ms
step:1234/6250 train_loss:2.5555 aux_loss:0.0430 train_time:163313ms step_avg:133.43ms
step:1235/6250 train_loss:2.5185 aux_loss:0.0428 train_time:163452ms step_avg:133.43ms
step:1236/6250 train_loss:2.4543 aux_loss:0.0429 train_time:163592ms step_avg:133.44ms
step:1237/6250 train_loss:2.5170 aux_loss:0.0428 train_time:163732ms step_avg:133.44ms
step:1238/6250 train_loss:2.4884 aux_loss:0.0431 train_time:163874ms step_avg:133.45ms
step:1239/6250 train_loss:2.5683 aux_loss:0.0429 train_time:164008ms step_avg:133.45ms
step:1240/6250 train_loss:2.5539 aux_loss:0.0428 train_time:164143ms step_avg:133.45ms
step:1241/6250 train_loss:2.5679 aux_loss:0.0425 train_time:164274ms step_avg:133.45ms
step:1242/6250 train_loss:2.5135 aux_loss:0.0424 train_time:164415ms step_avg:133.45ms
step:1243/6250 train_loss:2.5369 aux_loss:0.0425 train_time:164548ms step_avg:133.45ms
step:1244/6250 train_loss:2.5420 aux_loss:0.0427 train_time:164685ms step_avg:133.46ms
step:1245/6250 train_loss:2.5112 aux_loss:0.0426 train_time:164823ms step_avg:133.46ms
step:1246/6250 train_loss:2.4904 aux_loss:0.0428 train_time:164955ms step_avg:133.46ms
step:1247/6250 train_loss:2.4695 aux_loss:0.0426 train_time:165092ms step_avg:133.46ms
step:1248/6250 train_loss:2.5450 aux_loss:0.0426 train_time:165228ms step_avg:133.46ms
step:1249/6250 train_loss:2.4515 aux_loss:0.0428 train_time:165364ms step_avg:133.47ms
step:1250/6250 train_loss:2.4353 aux_loss:0.0428 train_time:165498ms step_avg:133.47ms
step:1251/6250 train_loss:2.4658 aux_loss:0.0429 train_time:165630ms step_avg:133.47ms
step:1252/6250 train_loss:2.5284 aux_loss:0.0429 train_time:165762ms step_avg:133.46ms
step:1253/6250 train_loss:2.5480 aux_loss:0.0426 train_time:165895ms step_avg:133.46ms
step:1254/6250 train_loss:2.4838 aux_loss:0.0427 train_time:166028ms step_avg:133.46ms
step:1255/6250 train_loss:2.4628 aux_loss:0.0428 train_time:166160ms step_avg:133.46ms
step:1256/6250 train_loss:2.5005 aux_loss:0.0428 train_time:166290ms step_avg:133.46ms
step:1257/6250 train_loss:2.4478 aux_loss:0.0427 train_time:166421ms step_avg:133.46ms
step:1258/6250 train_loss:2.5519 aux_loss:0.0426 train_time:166552ms step_avg:133.46ms
step:1259/6250 train_loss:2.4947 aux_loss:0.0430 train_time:166683ms step_avg:133.45ms
step:1260/6250 train_loss:2.5069 aux_loss:0.0429 train_time:166813ms step_avg:133.45ms
step:1261/6250 train_loss:2.4778 aux_loss:0.0427 train_time:166946ms step_avg:133.45ms
step:1262/6250 train_loss:2.5498 aux_loss:0.0425 train_time:167079ms step_avg:133.45ms
step:1263/6250 train_loss:2.5187 aux_loss:0.0424 train_time:167210ms step_avg:133.45ms
step:1264/6250 train_loss:2.4293 aux_loss:0.0424 train_time:167343ms step_avg:133.45ms
step:1265/6250 train_loss:2.5391 aux_loss:0.0422 train_time:167474ms step_avg:133.45ms
step:1266/6250 train_loss:2.5295 aux_loss:0.0425 train_time:167605ms step_avg:133.44ms
step:1267/6250 train_loss:2.5184 aux_loss:0.0425 train_time:167736ms step_avg:133.44ms
step:1268/6250 train_loss:2.4175 aux_loss:0.0424 train_time:167867ms step_avg:133.44ms
step:1269/6250 train_loss:2.4896 aux_loss:0.0428 train_time:168000ms step_avg:133.44ms
step:1270/6250 train_loss:2.5968 aux_loss:0.0428 train_time:168131ms step_avg:133.44ms
step:1271/6250 train_loss:2.4840 aux_loss:0.0429 train_time:168263ms step_avg:133.44ms
step:1272/6250 train_loss:2.4979 aux_loss:0.0425 train_time:168397ms step_avg:133.44ms
step:1273/6250 train_loss:2.4419 aux_loss:0.0424 train_time:168530ms step_avg:133.44ms
step:1274/6250 train_loss:2.5781 aux_loss:0.0422 train_time:168662ms step_avg:133.43ms
step:1275/6250 train_loss:2.5044 aux_loss:0.0426 train_time:168793ms step_avg:133.43ms
step:1276/6250 train_loss:2.5333 aux_loss:0.0426 train_time:168925ms step_avg:133.43ms
step:1277/6250 train_loss:2.5226 aux_loss:0.0425 train_time:169058ms step_avg:133.43ms
step:1278/6250 train_loss:2.5259 aux_loss:0.0428 train_time:169189ms step_avg:133.43ms
step:1279/6250 train_loss:2.5973 aux_loss:0.0427 train_time:169322ms step_avg:133.43ms
step:1280/6250 train_loss:2.4681 aux_loss:0.0427 train_time:169453ms step_avg:133.43ms
step:1281/6250 train_loss:2.4458 aux_loss:0.0424 train_time:169588ms step_avg:133.43ms
step:1282/6250 train_loss:2.4699 aux_loss:0.0424 train_time:169741ms step_avg:133.44ms
step:1283/6250 train_loss:2.5641 aux_loss:0.0427 train_time:169884ms step_avg:133.45ms
step:1284/6250 train_loss:2.5215 aux_loss:0.0427 train_time:170021ms step_avg:133.45ms
step:1285/6250 train_loss:2.4359 aux_loss:0.0424 train_time:170159ms step_avg:133.46ms
step:1286/6250 train_loss:2.5211 aux_loss:0.0423 train_time:170292ms step_avg:133.46ms
step:1287/6250 train_loss:2.5122 aux_loss:0.0424 train_time:170431ms step_avg:133.46ms
step:1288/6250 train_loss:2.4217 aux_loss:0.0425 train_time:170567ms step_avg:133.46ms
step:1289/6250 train_loss:2.4752 aux_loss:0.0424 train_time:170704ms step_avg:133.47ms
step:1290/6250 train_loss:2.4955 aux_loss:0.0426 train_time:170836ms step_avg:133.47ms
step:1291/6250 train_loss:2.3906 aux_loss:0.0426 train_time:170967ms step_avg:133.46ms
step:1292/6250 train_loss:2.4072 aux_loss:0.0425 train_time:171098ms step_avg:133.46ms
step:1293/6250 train_loss:2.4901 aux_loss:0.0425 train_time:171233ms step_avg:133.46ms
step:1294/6250 train_loss:2.4460 aux_loss:0.0423 train_time:171365ms step_avg:133.46ms
step:1295/6250 train_loss:2.5484 aux_loss:0.0423 train_time:171500ms step_avg:133.46ms
step:1296/6250 train_loss:2.5307 aux_loss:0.0424 train_time:171632ms step_avg:133.46ms
step:1297/6250 train_loss:2.4896 aux_loss:0.0424 train_time:171765ms step_avg:133.46ms
step:1298/6250 train_loss:2.5529 aux_loss:0.0422 train_time:171896ms step_avg:133.46ms
step:1299/6250 train_loss:2.4701 aux_loss:0.0421 train_time:172028ms step_avg:133.46ms
step:1300/6250 train_loss:2.4671 aux_loss:0.0424 train_time:172159ms step_avg:133.46ms
step:1301/6250 train_loss:2.4779 aux_loss:0.0423 train_time:172289ms step_avg:133.45ms
step:1302/6250 train_loss:2.3560 aux_loss:0.0420 train_time:172420ms step_avg:133.45ms
step:1303/6250 train_loss:2.4904 aux_loss:0.0420 train_time:172551ms step_avg:133.45ms
step:1304/6250 train_loss:2.4730 aux_loss:0.0422 train_time:172683ms step_avg:133.45ms
step:1305/6250 train_loss:2.4817 aux_loss:0.0422 train_time:172815ms step_avg:133.45ms
step:1306/6250 train_loss:2.5008 aux_loss:0.0420 train_time:172946ms step_avg:133.45ms
step:1307/6250 train_loss:2.5221 aux_loss:0.0421 train_time:173078ms step_avg:133.44ms
step:1308/6250 train_loss:2.5051 aux_loss:0.0420 train_time:173208ms step_avg:133.44ms
step:1309/6250 train_loss:2.4394 aux_loss:0.0420 train_time:173339ms step_avg:133.44ms
step:1310/6250 train_loss:2.5754 aux_loss:0.0421 train_time:173470ms step_avg:133.44ms
step:1311/6250 train_loss:2.5540 aux_loss:0.0422 train_time:173602ms step_avg:133.44ms
step:1312/6250 train_loss:2.5723 aux_loss:0.0420 train_time:173733ms step_avg:133.44ms
step:1313/6250 train_loss:2.5103 aux_loss:0.0419 train_time:173865ms step_avg:133.43ms
step:1314/6250 train_loss:2.5043 aux_loss:0.0421 train_time:173996ms step_avg:133.43ms
step:1315/6250 train_loss:2.5860 aux_loss:0.0419 train_time:174127ms step_avg:133.43ms
step:1316/6250 train_loss:2.5076 aux_loss:0.0421 train_time:174258ms step_avg:133.43ms
step:1317/6250 train_loss:2.5475 aux_loss:0.0423 train_time:174388ms step_avg:133.43ms
step:1318/6250 train_loss:2.5686 aux_loss:0.0422 train_time:174520ms step_avg:133.43ms
step:1319/6250 train_loss:2.5402 aux_loss:0.0421 train_time:174652ms step_avg:133.42ms
step:1320/6250 train_loss:2.3748 aux_loss:0.0421 train_time:174785ms step_avg:133.42ms
step:1321/6250 train_loss:2.6069 aux_loss:0.0422 train_time:174916ms step_avg:133.42ms
step:1322/6250 train_loss:2.4633 aux_loss:0.0421 train_time:175049ms step_avg:133.42ms
step:1323/6250 train_loss:2.4650 aux_loss:0.0420 train_time:175180ms step_avg:133.42ms
step:1324/6250 train_loss:2.5259 aux_loss:0.0421 train_time:175310ms step_avg:133.42ms
step:1325/6250 train_loss:2.5423 aux_loss:0.0419 train_time:175441ms step_avg:133.42ms
step:1326/6250 train_loss:2.6431 aux_loss:0.0418 train_time:175572ms step_avg:133.41ms
step:1327/6250 train_loss:2.4359 aux_loss:0.0421 train_time:175703ms step_avg:133.41ms
step:1328/6250 train_loss:2.5360 aux_loss:0.0421 train_time:175835ms step_avg:133.41ms
step:1329/6250 train_loss:2.5779 aux_loss:0.0422 train_time:175967ms step_avg:133.41ms
step:1330/6250 train_loss:2.4675 aux_loss:0.0420 train_time:176098ms step_avg:133.41ms
step:1331/6250 train_loss:2.5049 aux_loss:0.0416 train_time:176228ms step_avg:133.41ms
step:1332/6250 train_loss:2.5377 aux_loss:0.0416 train_time:176360ms step_avg:133.40ms
step:1333/6250 train_loss:2.3997 aux_loss:0.0422 train_time:176492ms step_avg:133.40ms
step:1334/6250 train_loss:2.5410 aux_loss:0.0421 train_time:176623ms step_avg:133.40ms
step:1335/6250 train_loss:2.4702 aux_loss:0.0421 train_time:176754ms step_avg:133.40ms
step:1336/6250 train_loss:2.4641 aux_loss:0.0418 train_time:176886ms step_avg:133.40ms
step:1337/6250 train_loss:2.4560 aux_loss:0.0418 train_time:177019ms step_avg:133.40ms
step:1338/6250 train_loss:2.4620 aux_loss:0.0418 train_time:177149ms step_avg:133.40ms
step:1339/6250 train_loss:2.5802 aux_loss:0.0419 train_time:177280ms step_avg:133.39ms
step:1340/6250 train_loss:2.5377 aux_loss:0.0418 train_time:177411ms step_avg:133.39ms
step:1341/6250 train_loss:2.5123 aux_loss:0.0416 train_time:177543ms step_avg:133.39ms
step:1342/6250 train_loss:2.5172 aux_loss:0.0416 train_time:177674ms step_avg:133.39ms
step:1343/6250 train_loss:2.5545 aux_loss:0.0418 train_time:177806ms step_avg:133.39ms
step:1344/6250 train_loss:2.4181 aux_loss:0.0421 train_time:177938ms step_avg:133.39ms
step:1345/6250 train_loss:2.5222 aux_loss:0.0419 train_time:178069ms step_avg:133.39ms
step:1346/6250 train_loss:2.4383 aux_loss:0.0418 train_time:178200ms step_avg:133.38ms
step:1347/6250 train_loss:2.4274 aux_loss:0.0417 train_time:178330ms step_avg:133.38ms
step:1348/6250 train_loss:2.4644 aux_loss:0.0419 train_time:178462ms step_avg:133.38ms
step:1349/6250 train_loss:2.5357 aux_loss:0.0420 train_time:178593ms step_avg:133.38ms
step:1350/6250 train_loss:2.4196 aux_loss:0.0421 train_time:178725ms step_avg:133.38ms
step:1351/6250 train_loss:2.5296 aux_loss:0.0420 train_time:178857ms step_avg:133.38ms
step:1352/6250 train_loss:2.5377 aux_loss:0.0417 train_time:178988ms step_avg:133.37ms
step:1353/6250 train_loss:2.5264 aux_loss:0.0418 train_time:179119ms step_avg:133.37ms
step:1354/6250 train_loss:2.4372 aux_loss:0.0418 train_time:179250ms step_avg:133.37ms
step:1355/6250 train_loss:2.5086 aux_loss:0.0418 train_time:179386ms step_avg:133.37ms
step:1356/6250 train_loss:2.4724 aux_loss:0.0420 train_time:179519ms step_avg:133.37ms
step:1357/6250 train_loss:2.4855 aux_loss:0.0420 train_time:179651ms step_avg:133.37ms
step:1358/6250 train_loss:2.3876 aux_loss:0.0420 train_time:179784ms step_avg:133.37ms
step:1359/6250 train_loss:2.4278 aux_loss:0.0417 train_time:179916ms step_avg:133.37ms
step:1360/6250 train_loss:2.4218 aux_loss:0.0418 train_time:180047ms step_avg:133.37ms
step:1361/6250 train_loss:2.5083 aux_loss:0.0418 train_time:180180ms step_avg:133.37ms
step:1362/6250 train_loss:2.3919 aux_loss:0.0419 train_time:180312ms step_avg:133.37ms
step:1363/6250 train_loss:2.4954 aux_loss:0.0421 train_time:180445ms step_avg:133.37ms
step:1364/6250 train_loss:2.5922 aux_loss:0.0419 train_time:180577ms step_avg:133.37ms
step:1365/6250 train_loss:2.4542 aux_loss:0.0418 train_time:180709ms step_avg:133.36ms
step:1366/6250 train_loss:2.5395 aux_loss:0.0417 train_time:180842ms step_avg:133.36ms
step:1367/6250 train_loss:2.5036 aux_loss:0.0417 train_time:180973ms step_avg:133.36ms
step:1368/6250 train_loss:2.5157 aux_loss:0.0419 train_time:181110ms step_avg:133.36ms
step:1369/6250 train_loss:2.4230 aux_loss:0.0418 train_time:181268ms step_avg:133.38ms
step:1370/6250 train_loss:2.4230 aux_loss:0.0418 train_time:181400ms step_avg:133.38ms
step:1371/6250 train_loss:2.5053 aux_loss:0.0417 train_time:181530ms step_avg:133.38ms
step:1372/6250 train_loss:2.5009 aux_loss:0.0419 train_time:181662ms step_avg:133.38ms
step:1373/6250 train_loss:2.4715 aux_loss:0.0416 train_time:181793ms step_avg:133.38ms
step:1374/6250 train_loss:2.4616 aux_loss:0.0415 train_time:181924ms step_avg:133.38ms
step:1375/6250 train_loss:2.4600 aux_loss:0.0417 train_time:182056ms step_avg:133.37ms
step:1376/6250 train_loss:2.5258 aux_loss:0.0415 train_time:182188ms step_avg:133.37ms
step:1377/6250 train_loss:2.4947 aux_loss:0.0414 train_time:182319ms step_avg:133.37ms
step:1378/6250 train_loss:2.4460 aux_loss:0.0415 train_time:182450ms step_avg:133.37ms
step:1379/6250 train_loss:2.4462 aux_loss:0.0416 train_time:182582ms step_avg:133.37ms
step:1380/6250 train_loss:2.3264 aux_loss:0.0416 train_time:182714ms step_avg:133.37ms
step:1381/6250 train_loss:2.4300 aux_loss:0.0414 train_time:182846ms step_avg:133.37ms
step:1382/6250 train_loss:2.4727 aux_loss:0.0415 train_time:182979ms step_avg:133.37ms
step:1383/6250 train_loss:2.4928 aux_loss:0.0415 train_time:183116ms step_avg:133.37ms
step:1384/6250 train_loss:2.5378 aux_loss:0.0416 train_time:183254ms step_avg:133.37ms
step:1385/6250 train_loss:2.4602 aux_loss:0.0415 train_time:183385ms step_avg:133.37ms
step:1386/6250 train_loss:2.5159 aux_loss:0.0414 train_time:183522ms step_avg:133.37ms
step:1387/6250 train_loss:2.4189 aux_loss:0.0415 train_time:183658ms step_avg:133.38ms
step:1388/6250 train_loss:2.5228 aux_loss:0.0416 train_time:183794ms step_avg:133.38ms
step:1389/6250 train_loss:2.4425 aux_loss:0.0416 train_time:183932ms step_avg:133.38ms
step:1390/6250 train_loss:2.4557 aux_loss:0.0415 train_time:184066ms step_avg:133.38ms
step:1391/6250 train_loss:2.5045 aux_loss:0.0413 train_time:184198ms step_avg:133.38ms
step:1392/6250 train_loss:2.4802 aux_loss:0.0415 train_time:184328ms step_avg:133.38ms
step:1393/6250 train_loss:2.4852 aux_loss:0.0415 train_time:184459ms step_avg:133.38ms
step:1394/6250 train_loss:2.4517 aux_loss:0.0415 train_time:184590ms step_avg:133.37ms
step:1395/6250 train_loss:2.4695 aux_loss:0.0413 train_time:184723ms step_avg:133.37ms
step:1396/6250 train_loss:2.5246 aux_loss:0.0412 train_time:184854ms step_avg:133.37ms
step:1397/6250 train_loss:2.3504 aux_loss:0.0414 train_time:184985ms step_avg:133.37ms
step:1398/6250 train_loss:2.4821 aux_loss:0.0415 train_time:185117ms step_avg:133.37ms
step:1399/6250 train_loss:2.4820 aux_loss:0.0416 train_time:185248ms step_avg:133.37ms
step:1400/6250 train_loss:2.4449 aux_loss:0.0416 train_time:185384ms step_avg:133.37ms
step:1401/6250 train_loss:2.4224 aux_loss:0.0418 train_time:185516ms step_avg:133.37ms
step:1402/6250 train_loss:2.5482 aux_loss:0.0412 train_time:185647ms step_avg:133.37ms
step:1403/6250 train_loss:2.4508 aux_loss:0.0412 train_time:185782ms step_avg:133.37ms
step:1404/6250 train_loss:2.4472 aux_loss:0.0413 train_time:185918ms step_avg:133.37ms
step:1405/6250 train_loss:2.4445 aux_loss:0.0416 train_time:186048ms step_avg:133.37ms
step:1406/6250 train_loss:2.4156 aux_loss:0.0413 train_time:186181ms step_avg:133.37ms
step:1407/6250 train_loss:2.4351 aux_loss:0.0413 train_time:186312ms step_avg:133.37ms
step:1408/6250 train_loss:2.4583 aux_loss:0.0415 train_time:186443ms step_avg:133.36ms
step:1409/6250 train_loss:2.5536 aux_loss:0.0414 train_time:186590ms step_avg:133.37ms
step:1410/6250 train_loss:2.4633 aux_loss:0.0412 train_time:186747ms step_avg:133.39ms
step:1411/6250 train_loss:2.4918 aux_loss:0.0413 train_time:186889ms step_avg:133.40ms
step:1412/6250 train_loss:2.3772 aux_loss:0.0413 train_time:187020ms step_avg:133.40ms
step:1413/6250 train_loss:2.4659 aux_loss:0.0413 train_time:187151ms step_avg:133.39ms
step:1414/6250 train_loss:2.4637 aux_loss:0.0412 train_time:187284ms step_avg:133.39ms
step:1415/6250 train_loss:2.4233 aux_loss:0.0411 train_time:187416ms step_avg:133.39ms
step:1416/6250 train_loss:2.5757 aux_loss:0.0407 train_time:187548ms step_avg:133.39ms
step:1417/6250 train_loss:2.5156 aux_loss:0.0409 train_time:187681ms step_avg:133.39ms
step:1418/6250 train_loss:2.4992 aux_loss:0.0412 train_time:187812ms step_avg:133.39ms
step:1419/6250 train_loss:2.4577 aux_loss:0.0414 train_time:187944ms step_avg:133.39ms
step:1420/6250 train_loss:2.5147 aux_loss:0.0413 train_time:188076ms step_avg:133.39ms
step:1421/6250 train_loss:2.5606 aux_loss:0.0410 train_time:188209ms step_avg:133.39ms
step:1422/6250 train_loss:2.4955 aux_loss:0.0413 train_time:188346ms step_avg:133.39ms
step:1423/6250 train_loss:2.4714 aux_loss:0.0416 train_time:188483ms step_avg:133.39ms
step:1424/6250 train_loss:2.4419 aux_loss:0.0415 train_time:188617ms step_avg:133.39ms
step:1425/6250 train_loss:2.5294 aux_loss:0.0413 train_time:188749ms step_avg:133.39ms
step:1426/6250 train_loss:2.3825 aux_loss:0.0411 train_time:188882ms step_avg:133.39ms
step:1427/6250 train_loss:2.4222 aux_loss:0.0411 train_time:189014ms step_avg:133.39ms
step:1428/6250 train_loss:2.5071 aux_loss:0.0409 train_time:189145ms step_avg:133.39ms
step:1429/6250 train_loss:2.4143 aux_loss:0.0410 train_time:189278ms step_avg:133.39ms
step:1430/6250 train_loss:2.5365 aux_loss:0.0411 train_time:189411ms step_avg:133.39ms
step:1431/6250 train_loss:2.4913 aux_loss:0.0413 train_time:189544ms step_avg:133.39ms
step:1432/6250 train_loss:2.4513 aux_loss:0.0411 train_time:189676ms step_avg:133.39ms
step:1433/6250 train_loss:2.4414 aux_loss:0.0411 train_time:189807ms step_avg:133.39ms
step:1434/6250 train_loss:2.4395 aux_loss:0.0412 train_time:189939ms step_avg:133.38ms
step:1435/6250 train_loss:2.4539 aux_loss:0.0414 train_time:190069ms step_avg:133.38ms
step:1436/6250 train_loss:2.4276 aux_loss:0.0411 train_time:190202ms step_avg:133.38ms
step:1437/6250 train_loss:2.4706 aux_loss:0.0410 train_time:190332ms step_avg:133.38ms
step:1438/6250 train_loss:2.4285 aux_loss:0.0413 train_time:190463ms step_avg:133.38ms
step:1439/6250 train_loss:2.3751 aux_loss:0.0411 train_time:190594ms step_avg:133.38ms
step:1440/6250 train_loss:2.4524 aux_loss:0.0408 train_time:190725ms step_avg:133.37ms
step:1441/6250 train_loss:2.5445 aux_loss:0.0410 train_time:190857ms step_avg:133.37ms
step:1442/6250 train_loss:2.5353 aux_loss:0.0411 train_time:190988ms step_avg:133.37ms
step:1443/6250 train_loss:2.3933 aux_loss:0.0412 train_time:191119ms step_avg:133.37ms
step:1444/6250 train_loss:2.4294 aux_loss:0.0411 train_time:191250ms step_avg:133.37ms
step:1445/6250 train_loss:2.5191 aux_loss:0.0413 train_time:191381ms step_avg:133.37ms
step:1446/6250 train_loss:2.4970 aux_loss:0.0412 train_time:191513ms step_avg:133.37ms
step:1447/6250 train_loss:2.3935 aux_loss:0.0412 train_time:191645ms step_avg:133.36ms
step:1448/6250 train_loss:2.5229 aux_loss:0.0413 train_time:191778ms step_avg:133.36ms
step:1449/6250 train_loss:2.4964 aux_loss:0.0410 train_time:191910ms step_avg:133.36ms
step:1450/6250 train_loss:2.5456 aux_loss:0.0410 train_time:192043ms step_avg:133.36ms
step:1451/6250 train_loss:2.4603 aux_loss:0.0411 train_time:192175ms step_avg:133.36ms
step:1452/6250 train_loss:2.4598 aux_loss:0.0410 train_time:192307ms step_avg:133.36ms
step:1453/6250 train_loss:2.4598 aux_loss:0.0409 train_time:192438ms step_avg:133.36ms
step:1454/6250 train_loss:2.4448 aux_loss:0.0412 train_time:192569ms step_avg:133.36ms
step:1455/6250 train_loss:2.5091 aux_loss:0.0411 train_time:192701ms step_avg:133.36ms
step:1456/6250 train_loss:2.4793 aux_loss:0.0412 train_time:192832ms step_avg:133.36ms
step:1457/6250 train_loss:2.4769 aux_loss:0.0412 train_time:192963ms step_avg:133.35ms
step:1458/6250 train_loss:2.5642 aux_loss:0.0412 train_time:193094ms step_avg:133.35ms
step:1459/6250 train_loss:2.4223 aux_loss:0.0412 train_time:193226ms step_avg:133.35ms
step:1460/6250 train_loss:2.4749 aux_loss:0.0409 train_time:193360ms step_avg:133.35ms
step:1461/6250 train_loss:2.4823 aux_loss:0.0411 train_time:193491ms step_avg:133.35ms
step:1462/6250 train_loss:2.4845 aux_loss:0.0408 train_time:193622ms step_avg:133.35ms
step:1463/6250 train_loss:2.4286 aux_loss:0.0412 train_time:193753ms step_avg:133.35ms
step:1464/6250 train_loss:2.5068 aux_loss:0.0408 train_time:193885ms step_avg:133.35ms
step:1465/6250 train_loss:2.4403 aux_loss:0.0408 train_time:194018ms step_avg:133.35ms
step:1466/6250 train_loss:2.4177 aux_loss:0.0406 train_time:194150ms step_avg:133.34ms
step:1467/6250 train_loss:2.4127 aux_loss:0.0408 train_time:194281ms step_avg:133.34ms
step:1468/6250 train_loss:2.4347 aux_loss:0.0409 train_time:194412ms step_avg:133.34ms
step:1469/6250 train_loss:2.5037 aux_loss:0.0407 train_time:194544ms step_avg:133.34ms
step:1470/6250 train_loss:2.4474 aux_loss:0.0408 train_time:194675ms step_avg:133.34ms
step:1471/6250 train_loss:2.4553 aux_loss:0.0410 train_time:194806ms step_avg:133.34ms
step:1472/6250 train_loss:2.4995 aux_loss:0.0411 train_time:194937ms step_avg:133.34ms
step:1473/6250 train_loss:2.4452 aux_loss:0.0413 train_time:195068ms step_avg:133.33ms
step:1474/6250 train_loss:2.3829 aux_loss:0.0412 train_time:195201ms step_avg:133.33ms
step:1475/6250 train_loss:2.3688 aux_loss:0.0408 train_time:195332ms step_avg:133.33ms
step:1476/6250 train_loss:2.4676 aux_loss:0.0409 train_time:195463ms step_avg:133.33ms
step:1477/6250 train_loss:2.4516 aux_loss:0.0406 train_time:195594ms step_avg:133.33ms
step:1478/6250 train_loss:2.4388 aux_loss:0.0407 train_time:195725ms step_avg:133.33ms
step:1479/6250 train_loss:2.3939 aux_loss:0.0405 train_time:195857ms step_avg:133.33ms
step:1480/6250 train_loss:2.3350 aux_loss:0.0406 train_time:195987ms step_avg:133.32ms
step:1481/6250 train_loss:2.5735 aux_loss:0.0407 train_time:196119ms step_avg:133.32ms
step:1482/6250 train_loss:2.3994 aux_loss:0.0407 train_time:196251ms step_avg:133.32ms
step:1483/6250 train_loss:2.5613 aux_loss:0.0406 train_time:196382ms step_avg:133.32ms
step:1484/6250 train_loss:2.3765 aux_loss:0.0407 train_time:196519ms step_avg:133.32ms
step:1485/6250 train_loss:2.4874 aux_loss:0.0408 train_time:196651ms step_avg:133.32ms
step:1486/6250 train_loss:2.5451 aux_loss:0.0407 train_time:196783ms step_avg:133.32ms
step:1487/6250 train_loss:2.3924 aux_loss:0.0406 train_time:196914ms step_avg:133.32ms
step:1488/6250 train_loss:2.4369 aux_loss:0.0403 train_time:197050ms step_avg:133.32ms
step:1489/6250 train_loss:2.4285 aux_loss:0.0403 train_time:197185ms step_avg:133.32ms
step:1490/6250 train_loss:2.4683 aux_loss:0.0406 train_time:197321ms step_avg:133.33ms
step:1491/6250 train_loss:2.4000 aux_loss:0.0407 train_time:197454ms step_avg:133.33ms
step:1492/6250 train_loss:2.4968 aux_loss:0.0410 train_time:197584ms step_avg:133.32ms
step:1493/6250 train_loss:2.4910 aux_loss:0.0409 train_time:197716ms step_avg:133.32ms
step:1494/6250 train_loss:2.4073 aux_loss:0.0406 train_time:197850ms step_avg:133.32ms
step:1495/6250 train_loss:2.4776 aux_loss:0.0404 train_time:198001ms step_avg:133.33ms
step:1496/6250 train_loss:2.4098 aux_loss:0.0405 train_time:198154ms step_avg:133.35ms
step:1497/6250 train_loss:2.4959 aux_loss:0.0408 train_time:198293ms step_avg:133.35ms
step:1498/6250 train_loss:2.4424 aux_loss:0.0407 train_time:198424ms step_avg:133.35ms
step:1499/6250 train_loss:2.4408 aux_loss:0.0405 train_time:198556ms step_avg:133.35ms
step:1500/6250 train_loss:2.5393 aux_loss:0.0405 train_time:198687ms step_avg:133.35ms
step:1501/6250 train_loss:2.4224 aux_loss:0.0405 train_time:198821ms step_avg:133.35ms
step:1502/6250 train_loss:2.4130 aux_loss:0.0408 train_time:198954ms step_avg:133.35ms
step:1503/6250 train_loss:2.4418 aux_loss:0.0406 train_time:199086ms step_avg:133.35ms
step:1504/6250 train_loss:2.4278 aux_loss:0.0406 train_time:199219ms step_avg:133.35ms
step:1505/6250 train_loss:2.4789 aux_loss:0.0407 train_time:199353ms step_avg:133.35ms
step:1506/6250 train_loss:2.5215 aux_loss:0.0407 train_time:199485ms step_avg:133.35ms
step:1507/6250 train_loss:2.5236 aux_loss:0.0409 train_time:199617ms step_avg:133.34ms
step:1508/6250 train_loss:2.4516 aux_loss:0.0407 train_time:199751ms step_avg:133.34ms
step:1509/6250 train_loss:2.5006 aux_loss:0.0405 train_time:199893ms step_avg:133.35ms
step:1510/6250 train_loss:2.5303 aux_loss:0.0405 train_time:200051ms step_avg:133.37ms
step:1511/6250 train_loss:2.5014 aux_loss:0.0406 train_time:200192ms step_avg:133.37ms
step:1512/6250 train_loss:2.5231 aux_loss:0.0406 train_time:200325ms step_avg:133.37ms
step:1513/6250 train_loss:2.4611 aux_loss:0.0406 train_time:200456ms step_avg:133.37ms
step:1514/6250 train_loss:2.5298 aux_loss:0.0408 train_time:200587ms step_avg:133.37ms
step:1515/6250 train_loss:2.3916 aux_loss:0.0404 train_time:200719ms step_avg:133.37ms
step:1516/6250 train_loss:2.5609 aux_loss:0.0405 train_time:200853ms step_avg:133.37ms
step:1517/6250 train_loss:2.4844 aux_loss:0.0403 train_time:200998ms step_avg:133.38ms
step:1518/6250 train_loss:2.4748 aux_loss:0.0404 train_time:201140ms step_avg:133.38ms
step:1519/6250 train_loss:2.3661 aux_loss:0.0406 train_time:201284ms step_avg:133.39ms
step:1520/6250 train_loss:2.4510 aux_loss:0.0403 train_time:201427ms step_avg:133.40ms
step:1521/6250 train_loss:2.3991 aux_loss:0.0405 train_time:201569ms step_avg:133.40ms
step:1522/6250 train_loss:2.4146 aux_loss:0.0403 train_time:201711ms step_avg:133.41ms
step:1523/6250 train_loss:2.3851 aux_loss:0.0402 train_time:201854ms step_avg:133.41ms
step:1524/6250 train_loss:2.3710 aux_loss:0.0403 train_time:201997ms step_avg:133.42ms
step:1525/6250 train_loss:2.5088 aux_loss:0.0404 train_time:202141ms step_avg:133.43ms
step:1526/6250 train_loss:2.4715 aux_loss:0.0403 train_time:202500ms step_avg:133.58ms
step:1527/6250 train_loss:2.3879 aux_loss:0.0404 train_time:202642ms step_avg:133.58ms
step:1528/6250 train_loss:2.4117 aux_loss:0.0403 train_time:202783ms step_avg:133.59ms
step:1529/6250 train_loss:2.3167 aux_loss:0.0403 train_time:202927ms step_avg:133.59ms
step:1530/6250 train_loss:2.4260 aux_loss:0.0401 train_time:203070ms step_avg:133.60ms
step:1531/6250 train_loss:2.4392 aux_loss:0.0403 train_time:203212ms step_avg:133.60ms
step:1532/6250 train_loss:2.4181 aux_loss:0.0400 train_time:203355ms step_avg:133.61ms
step:1533/6250 train_loss:2.4259 aux_loss:0.0402 train_time:203498ms step_avg:133.62ms
step:1534/6250 train_loss:2.4737 aux_loss:0.0404 train_time:203641ms step_avg:133.62ms
step:1535/6250 train_loss:2.4103 aux_loss:0.0405 train_time:203784ms step_avg:133.63ms
step:1536/6250 train_loss:2.4071 aux_loss:0.0402 train_time:203926ms step_avg:133.63ms
step:1537/6250 train_loss:2.4740 aux_loss:0.0401 train_time:204069ms step_avg:133.64ms
step:1538/6250 train_loss:2.3575 aux_loss:0.0404 train_time:204212ms step_avg:133.65ms
step:1539/6250 train_loss:2.5607 aux_loss:0.0399 train_time:204354ms step_avg:133.65ms
step:1540/6250 train_loss:2.4425 aux_loss:0.0401 train_time:204497ms step_avg:133.66ms
step:1541/6250 train_loss:2.4619 aux_loss:0.0404 train_time:204639ms step_avg:133.66ms
step:1542/6250 train_loss:2.4942 aux_loss:0.0404 train_time:204781ms step_avg:133.67ms
step:1543/6250 train_loss:2.4341 aux_loss:0.0402 train_time:204924ms step_avg:133.68ms
step:1544/6250 train_loss:2.4135 aux_loss:0.0401 train_time:205067ms step_avg:133.68ms
step:1545/6250 train_loss:2.4677 aux_loss:0.0402 train_time:205216ms step_avg:133.69ms
step:1546/6250 train_loss:2.4615 aux_loss:0.0404 train_time:205348ms step_avg:133.69ms
step:1547/6250 train_loss:2.4346 aux_loss:0.0407 train_time:205481ms step_avg:133.69ms
step:1548/6250 train_loss:2.4647 aux_loss:0.0403 train_time:205619ms step_avg:133.69ms
step:1549/6250 train_loss:2.5256 aux_loss:0.0400 train_time:205757ms step_avg:133.70ms
step:1550/6250 train_loss:2.4615 aux_loss:0.0401 train_time:205894ms step_avg:133.70ms
step:1551/6250 train_loss:2.4127 aux_loss:0.0401 train_time:206031ms step_avg:133.70ms
step:1552/6250 train_loss:2.5897 aux_loss:0.0399 train_time:206167ms step_avg:133.70ms
step:1553/6250 train_loss:2.4621 aux_loss:0.0400 train_time:206303ms step_avg:133.70ms
step:1554/6250 train_loss:2.4094 aux_loss:0.0402 train_time:206436ms step_avg:133.70ms
step:1555/6250 train_loss:2.4761 aux_loss:0.0403 train_time:206567ms step_avg:133.70ms
step:1556/6250 train_loss:2.5524 aux_loss:0.0399 train_time:206698ms step_avg:133.70ms
step:1557/6250 train_loss:2.4244 aux_loss:0.0402 train_time:206831ms step_avg:133.70ms
step:1558/6250 train_loss:2.4455 aux_loss:0.0399 train_time:206963ms step_avg:133.70ms
step:1559/6250 train_loss:2.3639 aux_loss:0.0400 train_time:207096ms step_avg:133.70ms
step:1560/6250 train_loss:2.5493 aux_loss:0.0401 train_time:207229ms step_avg:133.70ms
step:1561/6250 train_loss:2.3491 aux_loss:0.0403 train_time:207362ms step_avg:133.70ms
step:1562/6250 train_loss:2.3923 aux_loss:0.0402 train_time:207494ms step_avg:133.69ms
step:1563/6250 train_loss:2.3783 aux_loss:0.0401 train_time:207628ms step_avg:133.70ms
step:1564/6250 train_loss:2.3982 aux_loss:0.0401 train_time:207762ms step_avg:133.70ms
step:1565/6250 train_loss:2.4590 aux_loss:0.0401 train_time:207899ms step_avg:133.70ms
step:1566/6250 train_loss:2.4037 aux_loss:0.0400 train_time:208030ms step_avg:133.70ms
step:1567/6250 train_loss:2.4139 aux_loss:0.0399 train_time:208163ms step_avg:133.70ms
step:1568/6250 train_loss:2.3955 aux_loss:0.0399 train_time:208297ms step_avg:133.70ms
step:1569/6250 train_loss:2.4198 aux_loss:0.0400 train_time:208428ms step_avg:133.69ms
step:1570/6250 train_loss:2.3344 aux_loss:0.0400 train_time:208560ms step_avg:133.69ms
step:1571/6250 train_loss:2.4090 aux_loss:0.0397 train_time:208692ms step_avg:133.69ms
step:1572/6250 train_loss:2.4391 aux_loss:0.0397 train_time:208825ms step_avg:133.69ms
step:1573/6250 train_loss:2.4429 aux_loss:0.0399 train_time:208957ms step_avg:133.69ms
step:1574/6250 train_loss:2.4342 aux_loss:0.0399 train_time:209089ms step_avg:133.69ms
step:1575/6250 train_loss:2.3429 aux_loss:0.0397 train_time:209220ms step_avg:133.69ms
step:1576/6250 train_loss:2.3330 aux_loss:0.0396 train_time:209359ms step_avg:133.69ms
step:1577/6250 train_loss:2.5869 aux_loss:0.0398 train_time:209496ms step_avg:133.69ms
step:1578/6250 train_loss:2.3128 aux_loss:0.0398 train_time:209633ms step_avg:133.69ms
step:1579/6250 train_loss:2.4080 aux_loss:0.0398 train_time:209771ms step_avg:133.70ms
step:1580/6250 train_loss:2.4797 aux_loss:0.0399 train_time:209902ms step_avg:133.70ms
step:1581/6250 train_loss:2.4253 aux_loss:0.0399 train_time:210038ms step_avg:133.70ms
step:1582/6250 train_loss:2.5045 aux_loss:0.0400 train_time:210176ms step_avg:133.70ms
step:1583/6250 train_loss:2.4635 aux_loss:0.0398 train_time:210313ms step_avg:133.70ms
step:1584/6250 train_loss:2.4326 aux_loss:0.0395 train_time:210450ms step_avg:133.70ms
step:1585/6250 train_loss:2.3198 aux_loss:0.0395 train_time:210588ms step_avg:133.71ms
step:1586/6250 train_loss:2.6007 aux_loss:0.0395 train_time:210721ms step_avg:133.71ms
step:1587/6250 train_loss:2.4398 aux_loss:0.0400 train_time:210853ms step_avg:133.71ms
step:1588/6250 train_loss:2.3879 aux_loss:0.0399 train_time:210984ms step_avg:133.70ms
step:1589/6250 train_loss:2.3274 aux_loss:0.0400 train_time:211118ms step_avg:133.70ms
step:1590/6250 train_loss:2.4945 aux_loss:0.0395 train_time:211251ms step_avg:133.70ms
step:1591/6250 train_loss:2.4768 aux_loss:0.0395 train_time:211384ms step_avg:133.70ms
step:1592/6250 train_loss:2.4123 aux_loss:0.0399 train_time:211517ms step_avg:133.70ms
step:1593/6250 train_loss:2.4109 aux_loss:0.0398 train_time:211649ms step_avg:133.70ms
step:1594/6250 train_loss:2.4406 aux_loss:0.0395 train_time:211786ms step_avg:133.70ms
step:1595/6250 train_loss:2.4511 aux_loss:0.0395 train_time:211920ms step_avg:133.70ms
step:1596/6250 train_loss:2.4824 aux_loss:0.0395 train_time:212055ms step_avg:133.70ms
step:1597/6250 train_loss:2.5357 aux_loss:0.0394 train_time:212191ms step_avg:133.71ms
step:1598/6250 train_loss:2.3366 aux_loss:0.0394 train_time:212327ms step_avg:133.71ms
step:1599/6250 train_loss:2.3922 aux_loss:0.0398 train_time:212459ms step_avg:133.71ms
step:1600/6250 train_loss:2.4105 aux_loss:0.0396 train_time:212591ms step_avg:133.71ms
step:1601/6250 train_loss:2.5092 aux_loss:0.0397 train_time:212722ms step_avg:133.70ms
step:1602/6250 train_loss:2.4690 aux_loss:0.0397 train_time:212855ms step_avg:133.70ms
step:1603/6250 train_loss:2.5019 aux_loss:0.0399 train_time:212986ms step_avg:133.70ms
step:1604/6250 train_loss:2.4375 aux_loss:0.0397 train_time:213124ms step_avg:133.70ms
step:1605/6250 train_loss:2.4913 aux_loss:0.0397 train_time:213268ms step_avg:133.71ms
step:1606/6250 train_loss:2.3611 aux_loss:0.0397 train_time:213402ms step_avg:133.71ms
step:1607/6250 train_loss:2.4683 aux_loss:0.0398 train_time:213534ms step_avg:133.71ms
step:1608/6250 train_loss:2.4051 aux_loss:0.0398 train_time:213664ms step_avg:133.71ms
step:1609/6250 train_loss:2.4840 aux_loss:0.0396 train_time:213797ms step_avg:133.71ms
step:1610/6250 train_loss:2.4575 aux_loss:0.0398 train_time:213928ms step_avg:133.70ms
step:1611/6250 train_loss:2.4411 aux_loss:0.0397 train_time:214061ms step_avg:133.70ms
step:1612/6250 train_loss:2.4524 aux_loss:0.0397 train_time:214193ms step_avg:133.70ms
step:1613/6250 train_loss:2.4388 aux_loss:0.0395 train_time:214324ms step_avg:133.70ms
step:1614/6250 train_loss:2.4315 aux_loss:0.0394 train_time:214456ms step_avg:133.70ms
step:1615/6250 train_loss:2.3397 aux_loss:0.0396 train_time:214586ms step_avg:133.70ms
step:1616/6250 train_loss:2.3516 aux_loss:0.0398 train_time:214718ms step_avg:133.70ms
step:1617/6250 train_loss:2.4105 aux_loss:0.0394 train_time:214850ms step_avg:133.70ms
step:1618/6250 train_loss:2.4131 aux_loss:0.0393 train_time:214981ms step_avg:133.69ms
step:1619/6250 train_loss:2.4096 aux_loss:0.0392 train_time:215114ms step_avg:133.69ms
step:1620/6250 train_loss:2.4283 aux_loss:0.0395 train_time:215245ms step_avg:133.69ms
step:1621/6250 train_loss:2.2897 aux_loss:0.0397 train_time:215377ms step_avg:133.69ms
step:1622/6250 train_loss:2.5541 aux_loss:0.0395 train_time:215509ms step_avg:133.69ms
step:1623/6250 train_loss:2.4606 aux_loss:0.0395 train_time:215640ms step_avg:133.69ms
step:1624/6250 train_loss:2.4365 aux_loss:0.0397 train_time:215773ms step_avg:133.69ms
step:1625/6250 train_loss:2.5085 aux_loss:0.0398 train_time:215905ms step_avg:133.69ms
step:1626/6250 train_loss:2.4501 aux_loss:0.0396 train_time:216037ms step_avg:133.69ms
step:1627/6250 train_loss:2.4536 aux_loss:0.0395 train_time:216169ms step_avg:133.69ms
step:1628/6250 train_loss:2.5386 aux_loss:0.0398 train_time:216303ms step_avg:133.69ms
step:1629/6250 train_loss:2.4216 aux_loss:0.0395 train_time:216434ms step_avg:133.68ms
step:1630/6250 train_loss:2.4051 aux_loss:0.0395 train_time:216565ms step_avg:133.68ms
step:1631/6250 train_loss:2.3938 aux_loss:0.0394 train_time:216695ms step_avg:133.68ms
step:1632/6250 train_loss:2.4864 aux_loss:0.0394 train_time:216827ms step_avg:133.68ms
step:1633/6250 train_loss:2.3801 aux_loss:0.0395 train_time:216958ms step_avg:133.68ms
step:1634/6250 train_loss:2.4647 aux_loss:0.0394 train_time:217090ms step_avg:133.68ms
step:1635/6250 train_loss:2.3602 aux_loss:0.0392 train_time:217221ms step_avg:133.67ms
step:1636/6250 train_loss:2.3931 aux_loss:0.0390 train_time:217354ms step_avg:133.67ms
step:1637/6250 train_loss:2.4904 aux_loss:0.0394 train_time:217485ms step_avg:133.67ms
step:1638/6250 train_loss:2.4773 aux_loss:0.0394 train_time:217617ms step_avg:133.67ms
step:1639/6250 train_loss:2.3802 aux_loss:0.0394 train_time:217748ms step_avg:133.67ms
step:1640/6250 train_loss:2.3937 aux_loss:0.0396 train_time:217880ms step_avg:133.67ms
step:1641/6250 train_loss:2.4332 aux_loss:0.0396 train_time:218013ms step_avg:133.67ms
step:1642/6250 train_loss:2.4469 aux_loss:0.0399 train_time:218144ms step_avg:133.67ms
step:1643/6250 train_loss:2.4360 aux_loss:0.0395 train_time:218277ms step_avg:133.67ms
step:1644/6250 train_loss:2.4865 aux_loss:0.0392 train_time:218408ms step_avg:133.66ms
step:1645/6250 train_loss:2.4202 aux_loss:0.0393 train_time:218538ms step_avg:133.66ms
step:1646/6250 train_loss:2.4999 aux_loss:0.0391 train_time:218669ms step_avg:133.66ms
step:1647/6250 train_loss:2.3390 aux_loss:0.0390 train_time:218801ms step_avg:133.66ms
step:1648/6250 train_loss:2.4608 aux_loss:0.0390 train_time:218932ms step_avg:133.66ms
step:1649/6250 train_loss:2.4591 aux_loss:0.0390 train_time:219064ms step_avg:133.66ms
step:1650/6250 train_loss:2.4651 aux_loss:0.0392 train_time:219196ms step_avg:133.66ms
step:1651/6250 train_loss:2.4444 aux_loss:0.0392 train_time:219327ms step_avg:133.65ms
step:1652/6250 train_loss:2.4029 aux_loss:0.0391 train_time:219458ms step_avg:133.65ms
step:1653/6250 train_loss:2.3845 aux_loss:0.0389 train_time:219589ms step_avg:133.65ms
step:1654/6250 train_loss:2.3163 aux_loss:0.0392 train_time:219718ms step_avg:133.65ms
step:1655/6250 train_loss:2.4897 aux_loss:0.0391 train_time:219851ms step_avg:133.65ms
step:1656/6250 train_loss:2.5512 aux_loss:0.0390 train_time:219983ms step_avg:133.65ms
step:1657/6250 train_loss:2.4425 aux_loss:0.0392 train_time:220115ms step_avg:133.65ms
step:1658/6250 train_loss:2.3370 aux_loss:0.0394 train_time:220247ms step_avg:133.65ms
step:1659/6250 train_loss:2.5198 aux_loss:0.0392 train_time:220378ms step_avg:133.64ms
step:1660/6250 train_loss:2.3881 aux_loss:0.0392 train_time:220509ms step_avg:133.64ms
step:1661/6250 train_loss:2.4431 aux_loss:0.0391 train_time:220641ms step_avg:133.64ms
step:1662/6250 train_loss:2.3636 aux_loss:0.0392 train_time:220775ms step_avg:133.64ms
step:1663/6250 train_loss:2.3285 aux_loss:0.0389 train_time:220906ms step_avg:133.64ms
step:1664/6250 train_loss:2.4943 aux_loss:0.0387 train_time:221038ms step_avg:133.64ms
step:1665/6250 train_loss:2.5121 aux_loss:0.0387 train_time:221169ms step_avg:133.64ms
step:1666/6250 train_loss:2.4395 aux_loss:0.0386 train_time:221301ms step_avg:133.64ms
step:1667/6250 train_loss:2.3554 aux_loss:0.0388 train_time:221432ms step_avg:133.63ms
step:1668/6250 train_loss:2.4818 aux_loss:0.0389 train_time:221563ms step_avg:133.63ms
step:1669/6250 train_loss:2.5095 aux_loss:0.0389 train_time:221694ms step_avg:133.63ms
step:1670/6250 train_loss:2.4168 aux_loss:0.0390 train_time:221827ms step_avg:133.63ms
step:1671/6250 train_loss:2.4035 aux_loss:0.0392 train_time:221957ms step_avg:133.63ms
step:1672/6250 train_loss:2.4598 aux_loss:0.0390 train_time:222089ms step_avg:133.63ms
step:1673/6250 train_loss:2.3937 aux_loss:0.0390 train_time:222221ms step_avg:133.63ms
step:1674/6250 train_loss:2.3624 aux_loss:0.0392 train_time:222353ms step_avg:133.63ms
step:1675/6250 train_loss:2.3906 aux_loss:0.0389 train_time:222485ms step_avg:133.62ms
step:1676/6250 train_loss:2.5172 aux_loss:0.0391 train_time:222616ms step_avg:133.62ms
step:1677/6250 train_loss:2.4667 aux_loss:0.0392 train_time:222748ms step_avg:133.62ms
step:1678/6250 train_loss:2.4526 aux_loss:0.0392 train_time:222879ms step_avg:133.62ms
step:1679/6250 train_loss:2.2781 aux_loss:0.0387 train_time:223010ms step_avg:133.62ms
step:1680/6250 train_loss:2.4840 aux_loss:0.0390 train_time:223141ms step_avg:133.62ms
step:1681/6250 train_loss:2.3856 aux_loss:0.0389 train_time:223274ms step_avg:133.62ms
step:1682/6250 train_loss:2.3837 aux_loss:0.0389 train_time:223407ms step_avg:133.62ms
step:1683/6250 train_loss:2.4257 aux_loss:0.0388 train_time:223538ms step_avg:133.61ms
step:1684/6250 train_loss:2.3793 aux_loss:0.0393 train_time:223670ms step_avg:133.61ms
step:1685/6250 train_loss:2.5034 aux_loss:0.0395 train_time:223802ms step_avg:133.61ms
step:1686/6250 train_loss:2.3623 aux_loss:0.0390 train_time:223934ms step_avg:133.61ms
step:1687/6250 train_loss:2.4190 aux_loss:0.0393 train_time:224064ms step_avg:133.61ms
step:1688/6250 train_loss:2.3386 aux_loss:0.0392 train_time:224196ms step_avg:133.61ms
step:1689/6250 train_loss:2.3855 aux_loss:0.0390 train_time:224327ms step_avg:133.61ms
step:1690/6250 train_loss:2.4787 aux_loss:0.0389 train_time:224458ms step_avg:133.61ms
step:1691/6250 train_loss:2.4504 aux_loss:0.0390 train_time:224589ms step_avg:133.60ms
step:1692/6250 train_loss:2.4562 aux_loss:0.0388 train_time:224721ms step_avg:133.60ms
step:1693/6250 train_loss:2.4127 aux_loss:0.0388 train_time:224853ms step_avg:133.60ms
step:1694/6250 train_loss:2.4831 aux_loss:0.0388 train_time:224984ms step_avg:133.60ms
step:1695/6250 train_loss:2.3805 aux_loss:0.0387 train_time:225115ms step_avg:133.60ms
step:1696/6250 train_loss:2.4185 aux_loss:0.0385 train_time:225251ms step_avg:133.60ms
step:1697/6250 train_loss:2.3541 aux_loss:0.0386 train_time:225383ms step_avg:133.60ms
step:1698/6250 train_loss:2.5114 aux_loss:0.0387 train_time:225516ms step_avg:133.60ms
step:1699/6250 train_loss:2.4021 aux_loss:0.0386 train_time:225648ms step_avg:133.60ms
step:1700/6250 train_loss:2.3867 aux_loss:0.0386 train_time:225779ms step_avg:133.60ms
step:1701/6250 train_loss:2.3993 aux_loss:0.0386 train_time:225911ms step_avg:133.60ms
step:1702/6250 train_loss:2.3773 aux_loss:0.0387 train_time:226042ms step_avg:133.59ms
step:1703/6250 train_loss:2.3934 aux_loss:0.0385 train_time:226173ms step_avg:133.59ms
step:1704/6250 train_loss:2.4181 aux_loss:0.0384 train_time:226304ms step_avg:133.59ms
step:1705/6250 train_loss:2.3347 aux_loss:0.0386 train_time:226436ms step_avg:133.59ms
step:1706/6250 train_loss:2.3479 aux_loss:0.0387 train_time:226567ms step_avg:133.59ms
step:1707/6250 train_loss:2.4149 aux_loss:0.0387 train_time:226699ms step_avg:133.59ms
step:1708/6250 train_loss:2.3829 aux_loss:0.0387 train_time:226832ms step_avg:133.59ms
step:1709/6250 train_loss:2.4315 aux_loss:0.0384 train_time:226966ms step_avg:133.59ms
step:1710/6250 train_loss:2.4883 aux_loss:0.0386 train_time:227100ms step_avg:133.59ms
step:1711/6250 train_loss:2.4041 aux_loss:0.0387 train_time:227235ms step_avg:133.59ms
step:1712/6250 train_loss:2.3751 aux_loss:0.0388 train_time:227367ms step_avg:133.59ms
step:1713/6250 train_loss:2.3565 aux_loss:0.0386 train_time:227502ms step_avg:133.59ms
step:1714/6250 train_loss:2.3963 aux_loss:0.0385 train_time:227640ms step_avg:133.59ms
step:1715/6250 train_loss:2.4483 aux_loss:0.0386 train_time:227778ms step_avg:133.59ms
step:1716/6250 train_loss:2.4507 aux_loss:0.0384 train_time:227921ms step_avg:133.60ms
step:1717/6250 train_loss:2.4164 aux_loss:0.0389 train_time:228062ms step_avg:133.60ms
step:1718/6250 train_loss:2.4854 aux_loss:0.0389 train_time:228201ms step_avg:133.61ms
step:1719/6250 train_loss:2.4178 aux_loss:0.0385 train_time:228342ms step_avg:133.61ms
step:1720/6250 train_loss:2.3832 aux_loss:0.0384 train_time:228482ms step_avg:133.62ms
step:1721/6250 train_loss:2.4301 aux_loss:0.0386 train_time:228616ms step_avg:133.62ms
step:1722/6250 train_loss:2.3400 aux_loss:0.0385 train_time:228750ms step_avg:133.62ms
step:1723/6250 train_loss:2.4159 aux_loss:0.0383 train_time:228884ms step_avg:133.62ms
step:1724/6250 train_loss:2.3583 aux_loss:0.0386 train_time:229031ms step_avg:133.62ms
step:1725/6250 train_loss:2.3970 aux_loss:0.0386 train_time:229165ms step_avg:133.62ms
step:1726/6250 train_loss:2.4326 aux_loss:0.0385 train_time:229297ms step_avg:133.62ms
step:1727/6250 train_loss:2.4481 aux_loss:0.0384 train_time:229430ms step_avg:133.62ms
step:1728/6250 train_loss:2.4464 aux_loss:0.0384 train_time:229562ms step_avg:133.62ms
step:1729/6250 train_loss:2.4617 aux_loss:0.0384 train_time:229695ms step_avg:133.62ms
step:1730/6250 train_loss:2.3596 aux_loss:0.0381 train_time:229828ms step_avg:133.62ms
step:1731/6250 train_loss:2.4877 aux_loss:0.0386 train_time:229958ms step_avg:133.62ms
step:1732/6250 train_loss:2.3724 aux_loss:0.0385 train_time:230095ms step_avg:133.62ms
step:1733/6250 train_loss:2.3972 aux_loss:0.0384 train_time:230228ms step_avg:133.62ms
step:1734/6250 train_loss:2.3520 aux_loss:0.0385 train_time:230361ms step_avg:133.62ms
step:1735/6250 train_loss:2.3018 aux_loss:0.0385 train_time:230493ms step_avg:133.62ms
step:1736/6250 train_loss:2.3827 aux_loss:0.0386 train_time:230626ms step_avg:133.62ms
step:1737/6250 train_loss:2.4206 aux_loss:0.0387 train_time:230759ms step_avg:133.62ms
step:1738/6250 train_loss:2.4693 aux_loss:0.0385 train_time:230892ms step_avg:133.62ms
step:1739/6250 train_loss:2.4706 aux_loss:0.0383 train_time:231023ms step_avg:133.62ms
step:1740/6250 train_loss:2.3355 aux_loss:0.0385 train_time:231153ms step_avg:133.61ms
step:1741/6250 train_loss:2.3621 aux_loss:0.0387 train_time:231285ms step_avg:133.61ms
step:1742/6250 train_loss:2.4242 aux_loss:0.0390 train_time:231416ms step_avg:133.61ms
step:1743/6250 train_loss:2.4308 aux_loss:0.0386 train_time:231549ms step_avg:133.61ms
step:1744/6250 train_loss:2.4128 aux_loss:0.0387 train_time:231681ms step_avg:133.61ms
step:1745/6250 train_loss:2.3723 aux_loss:0.0386 train_time:231813ms step_avg:133.61ms
step:1746/6250 train_loss:2.4481 aux_loss:0.0386 train_time:231946ms step_avg:133.61ms
step:1747/6250 train_loss:2.4063 aux_loss:0.0384 train_time:232077ms step_avg:133.61ms
step:1748/6250 train_loss:2.3853 aux_loss:0.0383 train_time:232208ms step_avg:133.61ms
step:1749/6250 train_loss:2.3555 aux_loss:0.0383 train_time:232339ms step_avg:133.60ms
step:1750/6250 train_loss:2.4316 aux_loss:0.0383 train_time:232471ms step_avg:133.60ms
step:1751/6250 train_loss:2.4032 aux_loss:0.0383 train_time:232602ms step_avg:133.60ms
step:1752/6250 train_loss:2.3380 aux_loss:0.0384 train_time:232733ms step_avg:133.60ms
step:1753/6250 train_loss:2.3583 aux_loss:0.0384 train_time:232865ms step_avg:133.60ms
step:1754/6250 train_loss:2.4037 aux_loss:0.0384 train_time:232997ms step_avg:133.60ms
step:1755/6250 train_loss:2.4315 aux_loss:0.0382 train_time:233128ms step_avg:133.60ms
step:1756/6250 train_loss:2.3347 aux_loss:0.0380 train_time:233262ms step_avg:133.60ms
step:1757/6250 train_loss:2.4559 aux_loss:0.0379 train_time:233394ms step_avg:133.60ms
step:1758/6250 train_loss:2.4550 aux_loss:0.0381 train_time:233526ms step_avg:133.60ms
step:1759/6250 train_loss:2.3891 aux_loss:0.0381 train_time:233658ms step_avg:133.60ms
step:1760/6250 train_loss:2.4119 aux_loss:0.0380 train_time:233791ms step_avg:133.59ms
step:1761/6250 train_loss:2.3822 aux_loss:0.0380 train_time:233924ms step_avg:133.59ms
step:1762/6250 train_loss:2.3242 aux_loss:0.0380 train_time:234063ms step_avg:133.60ms
step:1763/6250 train_loss:2.3059 aux_loss:0.0382 train_time:234200ms step_avg:133.60ms
step:1764/6250 train_loss:2.5026 aux_loss:0.0384 train_time:234340ms step_avg:133.60ms
step:1765/6250 train_loss:2.3587 aux_loss:0.0382 train_time:234479ms step_avg:133.61ms
step:1766/6250 train_loss:2.4196 aux_loss:0.0382 train_time:234617ms step_avg:133.61ms
step:1767/6250 train_loss:2.4358 aux_loss:0.0382 train_time:234755ms step_avg:133.61ms
step:1768/6250 train_loss:2.4053 aux_loss:0.0379 train_time:234916ms step_avg:133.63ms
step:1769/6250 train_loss:2.4585 aux_loss:0.0378 train_time:235063ms step_avg:133.63ms
step:1770/6250 train_loss:2.4055 aux_loss:0.0382 train_time:235193ms step_avg:133.63ms
step:1771/6250 train_loss:2.3958 aux_loss:0.0383 train_time:235324ms step_avg:133.63ms
step:1772/6250 train_loss:2.4076 aux_loss:0.0382 train_time:235456ms step_avg:133.63ms
step:1773/6250 train_loss:2.4264 aux_loss:0.0383 train_time:235589ms step_avg:133.63ms
step:1774/6250 train_loss:2.3996 aux_loss:0.0384 train_time:235721ms step_avg:133.63ms
step:1775/6250 train_loss:2.4060 aux_loss:0.0381 train_time:235854ms step_avg:133.63ms
step:1776/6250 train_loss:2.4686 aux_loss:0.0381 train_time:235987ms step_avg:133.63ms
step:1777/6250 train_loss:2.3951 aux_loss:0.0381 train_time:236118ms step_avg:133.63ms
step:1778/6250 train_loss:2.3831 aux_loss:0.0379 train_time:236250ms step_avg:133.63ms
step:1779/6250 train_loss:2.3207 aux_loss:0.0378 train_time:236382ms step_avg:133.62ms
step:1780/6250 train_loss:2.4536 aux_loss:0.0380 train_time:236515ms step_avg:133.62ms
step:1781/6250 train_loss:2.4348 aux_loss:0.0376 train_time:236662ms step_avg:133.63ms
step:1782/6250 train_loss:2.4047 aux_loss:0.0377 train_time:236806ms step_avg:133.64ms
step:1783/6250 train_loss:2.3974 aux_loss:0.0378 train_time:236950ms step_avg:133.64ms
step:1784/6250 train_loss:2.3539 aux_loss:0.0378 train_time:237093ms step_avg:133.65ms
step:1785/6250 train_loss:2.3719 aux_loss:0.0375 train_time:237234ms step_avg:133.65ms
step:1786/6250 train_loss:2.3804 aux_loss:0.0377 train_time:237376ms step_avg:133.66ms
step:1787/6250 train_loss:2.4426 aux_loss:0.0380 train_time:237519ms step_avg:133.66ms
step:1788/6250 train_loss:2.4304 aux_loss:0.0378 train_time:237662ms step_avg:133.67ms
step:1789/6250 train_loss:2.2850 aux_loss:0.0375 train_time:237805ms step_avg:133.67ms
step:1790/6250 train_loss:2.3649 aux_loss:0.0377 train_time:237949ms step_avg:133.68ms
step:1791/6250 train_loss:2.4772 aux_loss:0.0379 train_time:238092ms step_avg:133.68ms
step:1792/6250 train_loss:2.4025 aux_loss:0.0377 train_time:238234ms step_avg:133.69ms
step:1793/6250 train_loss:2.3663 aux_loss:0.0375 train_time:238376ms step_avg:133.69ms
step:1794/6250 train_loss:2.4274 aux_loss:0.0377 train_time:238521ms step_avg:133.70ms
step:1795/6250 train_loss:2.3440 aux_loss:0.0381 train_time:238665ms step_avg:133.71ms
step:1796/6250 train_loss:2.3678 aux_loss:0.0382 train_time:238797ms step_avg:133.70ms
step:1797/6250 train_loss:2.3977 aux_loss:0.0381 train_time:238932ms step_avg:133.71ms
step:1798/6250 train_loss:2.3567 aux_loss:0.0379 train_time:239065ms step_avg:133.71ms
step:1799/6250 train_loss:2.4736 aux_loss:0.0379 train_time:239198ms step_avg:133.70ms
step:1800/6250 train_loss:2.4181 aux_loss:0.0379 train_time:239335ms step_avg:133.71ms
step:1801/6250 train_loss:2.4256 aux_loss:0.0377 train_time:239471ms step_avg:133.71ms
step:1802/6250 train_loss:2.3722 aux_loss:0.0378 train_time:239610ms step_avg:133.71ms
step:1803/6250 train_loss:2.4504 aux_loss:0.0377 train_time:239744ms step_avg:133.71ms
step:1804/6250 train_loss:2.4973 aux_loss:0.0379 train_time:239879ms step_avg:133.71ms
step:1805/6250 train_loss:2.4314 aux_loss:0.0380 train_time:240014ms step_avg:133.71ms
step:1806/6250 train_loss:2.3929 aux_loss:0.0378 train_time:240153ms step_avg:133.72ms
step:1807/6250 train_loss:2.4413 aux_loss:0.0379 train_time:240291ms step_avg:133.72ms
step:1808/6250 train_loss:2.3378 aux_loss:0.0378 train_time:240424ms step_avg:133.72ms
step:1809/6250 train_loss:2.3697 aux_loss:0.0375 train_time:240555ms step_avg:133.72ms
step:1810/6250 train_loss:2.4029 aux_loss:0.0380 train_time:240689ms step_avg:133.72ms
step:1811/6250 train_loss:2.4702 aux_loss:0.0375 train_time:240823ms step_avg:133.72ms
step:1812/6250 train_loss:2.3336 aux_loss:0.0374 train_time:240955ms step_avg:133.72ms
step:1813/6250 train_loss:2.4361 aux_loss:0.0376 train_time:241088ms step_avg:133.71ms
step:1814/6250 train_loss:2.3871 aux_loss:0.0375 train_time:241221ms step_avg:133.71ms
step:1815/6250 train_loss:2.3245 aux_loss:0.0376 train_time:241355ms step_avg:133.71ms
step:1816/6250 train_loss:2.3482 aux_loss:0.0377 train_time:241491ms step_avg:133.72ms
step:1817/6250 train_loss:2.4309 aux_loss:0.0376 train_time:241626ms step_avg:133.72ms
step:1818/6250 train_loss:2.3668 aux_loss:0.0375 train_time:241762ms step_avg:133.72ms
step:1819/6250 train_loss:2.3112 aux_loss:0.0378 train_time:241898ms step_avg:133.72ms
step:1820/6250 train_loss:2.3735 aux_loss:0.0377 train_time:242037ms step_avg:133.72ms
step:1821/6250 train_loss:2.3807 aux_loss:0.0380 train_time:242175ms step_avg:133.72ms
step:1822/6250 train_loss:2.4136 aux_loss:0.0379 train_time:242311ms step_avg:133.73ms
step:1823/6250 train_loss:2.3752 aux_loss:0.0378 train_time:242444ms step_avg:133.73ms
step:1824/6250 train_loss:2.4451 aux_loss:0.0379 train_time:242575ms step_avg:133.72ms
step:1825/6250 train_loss:2.3820 aux_loss:0.0380 train_time:242707ms step_avg:133.72ms
step:1826/6250 train_loss:2.4344 aux_loss:0.0380 train_time:242838ms step_avg:133.72ms
step:1827/6250 train_loss:2.3888 aux_loss:0.0379 train_time:242970ms step_avg:133.72ms
step:1828/6250 train_loss:2.3511 aux_loss:0.0377 train_time:243101ms step_avg:133.72ms
step:1829/6250 train_loss:2.5846 aux_loss:0.0376 train_time:243232ms step_avg:133.72ms
step:1830/6250 train_loss:2.3311 aux_loss:0.0376 train_time:243364ms step_avg:133.72ms
step:1831/6250 train_loss:2.4734 aux_loss:0.0374 train_time:243497ms step_avg:133.72ms
step:1832/6250 train_loss:2.3208 aux_loss:0.0371 train_time:243630ms step_avg:133.72ms
step:1833/6250 train_loss:2.4980 aux_loss:0.0375 train_time:243762ms step_avg:133.71ms
step:1834/6250 train_loss:2.4108 aux_loss:0.0374 train_time:243893ms step_avg:133.71ms
step:1835/6250 train_loss:2.3921 aux_loss:0.0373 train_time:244027ms step_avg:133.71ms
step:1836/6250 train_loss:2.4933 aux_loss:0.0372 train_time:244159ms step_avg:133.71ms
step:1837/6250 train_loss:2.3761 aux_loss:0.0374 train_time:244290ms step_avg:133.71ms
step:1838/6250 train_loss:2.4223 aux_loss:0.0374 train_time:244422ms step_avg:133.71ms
step:1839/6250 train_loss:2.3674 aux_loss:0.0373 train_time:244553ms step_avg:133.71ms
step:1840/6250 train_loss:2.3670 aux_loss:0.0374 train_time:244687ms step_avg:133.71ms
step:1841/6250 train_loss:2.4348 aux_loss:0.0374 train_time:244819ms step_avg:133.71ms
step:1842/6250 train_loss:2.3666 aux_loss:0.0371 train_time:244951ms step_avg:133.71ms
step:1843/6250 train_loss:2.4465 aux_loss:0.0375 train_time:245083ms step_avg:133.71ms
step:1844/6250 train_loss:2.4595 aux_loss:0.0378 train_time:245215ms step_avg:133.70ms
step:1845/6250 train_loss:2.3512 aux_loss:0.0375 train_time:245348ms step_avg:133.70ms
step:1846/6250 train_loss:2.4302 aux_loss:0.0374 train_time:245480ms step_avg:133.70ms
step:1847/6250 train_loss:2.4389 aux_loss:0.0376 train_time:245611ms step_avg:133.70ms
step:1848/6250 train_loss:2.2870 aux_loss:0.0373 train_time:245743ms step_avg:133.70ms
step:1849/6250 train_loss:2.4367 aux_loss:0.0372 train_time:245875ms step_avg:133.70ms
step:1850/6250 train_loss:2.4497 aux_loss:0.0371 train_time:246009ms step_avg:133.70ms
step:1851/6250 train_loss:2.3120 aux_loss:0.0372 train_time:246140ms step_avg:133.70ms
step:1852/6250 train_loss:2.3335 aux_loss:0.0375 train_time:246271ms step_avg:133.70ms
step:1853/6250 train_loss:2.4348 aux_loss:0.0378 train_time:246404ms step_avg:133.70ms
step:1854/6250 train_loss:2.3596 aux_loss:0.0377 train_time:246536ms step_avg:133.70ms
step:1855/6250 train_loss:2.3605 aux_loss:0.0379 train_time:246668ms step_avg:133.70ms
step:1856/6250 train_loss:2.4662 aux_loss:0.0377 train_time:246799ms step_avg:133.69ms
step:1857/6250 train_loss:2.4490 aux_loss:0.0376 train_time:246931ms step_avg:133.69ms
step:1858/6250 train_loss:2.3636 aux_loss:0.0376 train_time:247063ms step_avg:133.69ms
step:1859/6250 train_loss:2.4883 aux_loss:0.0380 train_time:247196ms step_avg:133.69ms
step:1860/6250 train_loss:2.2943 aux_loss:0.0376 train_time:247328ms step_avg:133.69ms
step:1861/6250 train_loss:2.5346 aux_loss:0.0376 train_time:247461ms step_avg:133.69ms
step:1862/6250 train_loss:2.4195 aux_loss:0.0377 train_time:247592ms step_avg:133.69ms
step:1863/6250 train_loss:2.3732 aux_loss:0.0375 train_time:247723ms step_avg:133.69ms
step:1864/6250 train_loss:2.4480 aux_loss:0.0376 train_time:247854ms step_avg:133.69ms
step:1865/6250 train_loss:2.3204 aux_loss:0.0373 train_time:247987ms step_avg:133.69ms
step:1866/6250 train_loss:2.3406 aux_loss:0.0373 train_time:248118ms step_avg:133.68ms
step:1867/6250 train_loss:2.4249 aux_loss:0.0375 train_time:248250ms step_avg:133.68ms
step:1868/6250 train_loss:2.3541 aux_loss:0.0373 train_time:248388ms step_avg:133.69ms
step:1869/6250 train_loss:2.3882 aux_loss:0.0372 train_time:248520ms step_avg:133.68ms
step:1870/6250 train_loss:2.4157 aux_loss:0.0375 train_time:248654ms step_avg:133.69ms
step:1871/6250 train_loss:2.3867 aux_loss:0.0372 train_time:248786ms step_avg:133.68ms
step:1872/6250 train_loss:2.4972 aux_loss:0.0368 train_time:248917ms step_avg:133.68ms
step:1873/6250 train_loss:2.3042 aux_loss:0.0374 train_time:249049ms step_avg:133.68ms
step:1874/6250 train_loss:2.3875 aux_loss:0.0378 train_time:249181ms step_avg:133.68ms
step:1875/6250 train_loss:2.2710 aux_loss:0.0374 train_time:249312ms step_avg:133.68ms
step:1876/6250 train_loss:2.4540 aux_loss:0.0371 train_time:249444ms step_avg:133.68ms
step:1877/6250 train_loss:2.4115 aux_loss:0.0371 train_time:249575ms step_avg:133.68ms
step:1878/6250 train_loss:2.4268 aux_loss:0.0369 train_time:249706ms step_avg:133.68ms
step:1879/6250 train_loss:2.4401 aux_loss:0.0372 train_time:249838ms step_avg:133.67ms
step:1880/6250 train_loss:2.4102 aux_loss:0.0373 train_time:249969ms step_avg:133.67ms
step:1881/6250 train_loss:2.3783 aux_loss:0.0371 train_time:250101ms step_avg:133.67ms
step:1882/6250 train_loss:2.3345 aux_loss:0.0371 train_time:250232ms step_avg:133.67ms
step:1883/6250 train_loss:2.4456 aux_loss:0.0368 train_time:250364ms step_avg:133.67ms
step:1884/6250 train_loss:2.3707 aux_loss:0.0366 train_time:250496ms step_avg:133.67ms
step:1885/6250 train_loss:2.3890 aux_loss:0.0366 train_time:250629ms step_avg:133.67ms
step:1886/6250 train_loss:2.3913 aux_loss:0.0366 train_time:250761ms step_avg:133.67ms
step:1887/6250 train_loss:2.3558 aux_loss:0.0367 train_time:250892ms step_avg:133.67ms
step:1888/6250 train_loss:2.3329 aux_loss:0.0369 train_time:251024ms step_avg:133.67ms
step:1889/6250 train_loss:2.3647 aux_loss:0.0370 train_time:251155ms step_avg:133.66ms
step:1890/6250 train_loss:2.3767 aux_loss:0.0365 train_time:251286ms step_avg:133.66ms
step:1891/6250 train_loss:2.3648 aux_loss:0.0366 train_time:251418ms step_avg:133.66ms
step:1892/6250 train_loss:2.3758 aux_loss:0.0371 train_time:251549ms step_avg:133.66ms
step:1893/6250 train_loss:2.4354 aux_loss:0.0373 train_time:251681ms step_avg:133.66ms
step:1894/6250 train_loss:2.4042 aux_loss:0.0374 train_time:251812ms step_avg:133.66ms
step:1895/6250 train_loss:2.3957 aux_loss:0.0372 train_time:251944ms step_avg:133.66ms
step:1896/6250 train_loss:2.4310 aux_loss:0.0372 train_time:252076ms step_avg:133.66ms
step:1897/6250 train_loss:2.3284 aux_loss:0.0371 train_time:252207ms step_avg:133.65ms
step:1898/6250 train_loss:2.3995 aux_loss:0.0370 train_time:252340ms step_avg:133.65ms
step:1899/6250 train_loss:2.4630 aux_loss:0.0370 train_time:252472ms step_avg:133.65ms
step:1900/6250 train_loss:2.3741 aux_loss:0.0369 train_time:252604ms step_avg:133.65ms
step:1901/6250 train_loss:2.3218 aux_loss:0.0367 train_time:252735ms step_avg:133.65ms
step:1902/6250 train_loss:2.3578 aux_loss:0.0369 train_time:252867ms step_avg:133.65ms
step:1903/6250 train_loss:2.3875 aux_loss:0.0368 train_time:252999ms step_avg:133.65ms
step:1904/6250 train_loss:2.3766 aux_loss:0.0370 train_time:253130ms step_avg:133.65ms
step:1905/6250 train_loss:2.3713 aux_loss:0.0371 train_time:253262ms step_avg:133.65ms
step:1906/6250 train_loss:2.3237 aux_loss:0.0370 train_time:253392ms step_avg:133.65ms
step:1907/6250 train_loss:2.4373 aux_loss:0.0370 train_time:253525ms step_avg:133.65ms
step:1908/6250 train_loss:2.4031 aux_loss:0.0371 train_time:253656ms step_avg:133.64ms
step:1909/6250 train_loss:2.3239 aux_loss:0.0369 train_time:253788ms step_avg:133.64ms
step:1910/6250 train_loss:2.3945 aux_loss:0.0369 train_time:253922ms step_avg:133.64ms
step:1911/6250 train_loss:2.3591 aux_loss:0.0370 train_time:254054ms step_avg:133.64ms
step:1912/6250 train_loss:2.5158 aux_loss:0.0371 train_time:254186ms step_avg:133.64ms
step:1913/6250 train_loss:2.3316 aux_loss:0.0366 train_time:254317ms step_avg:133.64ms
step:1914/6250 train_loss:2.3661 aux_loss:0.0371 train_time:254448ms step_avg:133.64ms
step:1915/6250 train_loss:2.3807 aux_loss:0.0372 train_time:254581ms step_avg:133.64ms
step:1916/6250 train_loss:2.4563 aux_loss:0.0370 train_time:254713ms step_avg:133.64ms
step:1917/6250 train_loss:2.2725 aux_loss:0.0366 train_time:254846ms step_avg:133.64ms
step:1918/6250 train_loss:2.4326 aux_loss:0.0369 train_time:254980ms step_avg:133.64ms
step:1919/6250 train_loss:2.3853 aux_loss:0.0370 train_time:255113ms step_avg:133.64ms
step:1920/6250 train_loss:2.4219 aux_loss:0.0369 train_time:255245ms step_avg:133.64ms
step:1921/6250 train_loss:2.3589 aux_loss:0.0368 train_time:255376ms step_avg:133.63ms
step:1922/6250 train_loss:2.3373 aux_loss:0.0366 train_time:255508ms step_avg:133.63ms
step:1923/6250 train_loss:2.4399 aux_loss:0.0369 train_time:255639ms step_avg:133.63ms
step:1924/6250 train_loss:2.4167 aux_loss:0.0370 train_time:255769ms step_avg:133.63ms
step:1925/6250 train_loss:2.4666 aux_loss:0.0368 train_time:255901ms step_avg:133.63ms
step:1926/6250 train_loss:2.3425 aux_loss:0.0366 train_time:256034ms step_avg:133.63ms
step:1927/6250 train_loss:2.3323 aux_loss:0.0365 train_time:256168ms step_avg:133.63ms
step:1928/6250 train_loss:2.3918 aux_loss:0.0366 train_time:256301ms step_avg:133.63ms
step:1929/6250 train_loss:2.4238 aux_loss:0.0367 train_time:256434ms step_avg:133.63ms
step:1930/6250 train_loss:2.3482 aux_loss:0.0368 train_time:256566ms step_avg:133.63ms
step:1931/6250 train_loss:2.3871 aux_loss:0.0370 train_time:256697ms step_avg:133.63ms
step:1932/6250 train_loss:2.3528 aux_loss:0.0369 train_time:256828ms step_avg:133.63ms
step:1933/6250 train_loss:2.3513 aux_loss:0.0366 train_time:256960ms step_avg:133.62ms
step:1934/6250 train_loss:2.4378 aux_loss:0.0369 train_time:257092ms step_avg:133.62ms
step:1935/6250 train_loss:2.4114 aux_loss:0.0367 train_time:257224ms step_avg:133.62ms
step:1936/6250 train_loss:2.4161 aux_loss:0.0367 train_time:257355ms step_avg:133.62ms
step:1937/6250 train_loss:2.4011 aux_loss:0.0367 train_time:257486ms step_avg:133.62ms
step:1938/6250 train_loss:2.3599 aux_loss:0.0366 train_time:257617ms step_avg:133.62ms
step:1939/6250 train_loss:2.3792 aux_loss:0.0369 train_time:257748ms step_avg:133.62ms
step:1940/6250 train_loss:2.3551 aux_loss:0.0370 train_time:257880ms step_avg:133.62ms
step:1941/6250 train_loss:2.3699 aux_loss:0.0376 train_time:258012ms step_avg:133.62ms
step:1942/6250 train_loss:2.4965 aux_loss:0.0369 train_time:258144ms step_avg:133.61ms
step:1943/6250 train_loss:2.3460 aux_loss:0.0367 train_time:258275ms step_avg:133.61ms
step:1944/6250 train_loss:2.4014 aux_loss:0.0363 train_time:258406ms step_avg:133.61ms
step:1945/6250 train_loss:2.3783 aux_loss:0.0365 train_time:258538ms step_avg:133.61ms
step:1946/6250 train_loss:2.4543 aux_loss:0.0367 train_time:258668ms step_avg:133.61ms
step:1947/6250 train_loss:2.4452 aux_loss:0.0367 train_time:258801ms step_avg:133.61ms
step:1948/6250 train_loss:2.3985 aux_loss:0.0363 train_time:258934ms step_avg:133.61ms
step:1949/6250 train_loss:2.3365 aux_loss:0.0366 train_time:259065ms step_avg:133.61ms
step:1950/6250 train_loss:2.2995 aux_loss:0.0367 train_time:259197ms step_avg:133.61ms
step:1951/6250 train_loss:2.3512 aux_loss:0.0368 train_time:259329ms step_avg:133.61ms
step:1952/6250 train_loss:2.3750 aux_loss:0.0366 train_time:259463ms step_avg:133.61ms
step:1953/6250 train_loss:2.2847 aux_loss:0.0369 train_time:259596ms step_avg:133.61ms
step:1954/6250 train_loss:2.3528 aux_loss:0.0369 train_time:259727ms step_avg:133.60ms
step:1955/6250 train_loss:2.3612 aux_loss:0.0368 train_time:259859ms step_avg:133.60ms
step:1956/6250 train_loss:2.2996 aux_loss:0.0364 train_time:259992ms step_avg:133.60ms
step:1957/6250 train_loss:2.3763 aux_loss:0.0367 train_time:260125ms step_avg:133.60ms
step:1958/6250 train_loss:2.3855 aux_loss:0.0369 train_time:260256ms step_avg:133.60ms
step:1959/6250 train_loss:2.3746 aux_loss:0.0365 train_time:260389ms step_avg:133.60ms
step:1960/6250 train_loss:2.4051 aux_loss:0.0366 train_time:260521ms step_avg:133.60ms
step:1961/6250 train_loss:2.4136 aux_loss:0.0366 train_time:260654ms step_avg:133.60ms
step:1962/6250 train_loss:2.3397 aux_loss:0.0366 train_time:260786ms step_avg:133.60ms
step:1963/6250 train_loss:2.4130 aux_loss:0.0367 train_time:260918ms step_avg:133.60ms
step:1964/6250 train_loss:2.3307 aux_loss:0.0368 train_time:261050ms step_avg:133.60ms
step:1965/6250 train_loss:2.5267 aux_loss:0.0365 train_time:261181ms step_avg:133.60ms
step:1966/6250 train_loss:2.2904 aux_loss:0.0364 train_time:261313ms step_avg:133.60ms
step:1967/6250 train_loss:2.4282 aux_loss:0.0366 train_time:261446ms step_avg:133.60ms
step:1968/6250 train_loss:2.3684 aux_loss:0.0365 train_time:261579ms step_avg:133.59ms
step:1969/6250 train_loss:2.3583 aux_loss:0.0363 train_time:261711ms step_avg:133.59ms
step:1970/6250 train_loss:2.3523 aux_loss:0.0365 train_time:261844ms step_avg:133.59ms
step:1971/6250 train_loss:2.3598 aux_loss:0.0366 train_time:261976ms step_avg:133.59ms
step:1972/6250 train_loss:2.3624 aux_loss:0.0363 train_time:262109ms step_avg:133.59ms
step:1973/6250 train_loss:2.3709 aux_loss:0.0362 train_time:262241ms step_avg:133.59ms
step:1974/6250 train_loss:2.3162 aux_loss:0.0363 train_time:262374ms step_avg:133.59ms
step:1975/6250 train_loss:2.2960 aux_loss:0.0363 train_time:262507ms step_avg:133.59ms
step:1976/6250 train_loss:2.3899 aux_loss:0.0361 train_time:262639ms step_avg:133.59ms
step:1977/6250 train_loss:2.4146 aux_loss:0.0361 train_time:262772ms step_avg:133.59ms
step:1978/6250 train_loss:2.3594 aux_loss:0.0361 train_time:262906ms step_avg:133.59ms
step:1979/6250 train_loss:2.2722 aux_loss:0.0360 train_time:263040ms step_avg:133.59ms
step:1980/6250 train_loss:2.4277 aux_loss:0.0362 train_time:263171ms step_avg:133.59ms
step:1981/6250 train_loss:2.3071 aux_loss:0.0359 train_time:263304ms step_avg:133.59ms
step:1982/6250 train_loss:2.3553 aux_loss:0.0361 train_time:263435ms step_avg:133.59ms
step:1983/6250 train_loss:2.3183 aux_loss:0.0363 train_time:263567ms step_avg:133.59ms
step:1984/6250 train_loss:2.3727 aux_loss:0.0362 train_time:263699ms step_avg:133.59ms
step:1985/6250 train_loss:2.4310 aux_loss:0.0361 train_time:263831ms step_avg:133.59ms
step:1986/6250 train_loss:2.4564 aux_loss:0.0360 train_time:263963ms step_avg:133.58ms
step:1987/6250 train_loss:2.3837 aux_loss:0.0362 train_time:264094ms step_avg:133.58ms
step:1988/6250 train_loss:2.3338 aux_loss:0.0359 train_time:264226ms step_avg:133.58ms
step:1989/6250 train_loss:2.4306 aux_loss:0.0363 train_time:264357ms step_avg:133.58ms
step:1990/6250 train_loss:2.4265 aux_loss:0.0364 train_time:264488ms step_avg:133.58ms
step:1991/6250 train_loss:2.3151 aux_loss:0.0363 train_time:264620ms step_avg:133.58ms
step:1992/6250 train_loss:2.3247 aux_loss:0.0362 train_time:264751ms step_avg:133.58ms
step:1993/6250 train_loss:2.4858 aux_loss:0.0363 train_time:264883ms step_avg:133.58ms
step:1994/6250 train_loss:2.4763 aux_loss:0.0364 train_time:265015ms step_avg:133.58ms
step:1995/6250 train_loss:2.2866 aux_loss:0.0363 train_time:265145ms step_avg:133.57ms
step:1996/6250 train_loss:2.3536 aux_loss:0.0361 train_time:265278ms step_avg:133.57ms
step:1997/6250 train_loss:2.3750 aux_loss:0.0364 train_time:265409ms step_avg:133.57ms
step:1998/6250 train_loss:2.4302 aux_loss:0.0365 train_time:265542ms step_avg:133.57ms
step:1999/6250 train_loss:2.4113 aux_loss:0.0364 train_time:265674ms step_avg:133.57ms
step:2000/6250 train_loss:2.3583 aux_loss:0.0365 train_time:265805ms step_avg:133.57ms
step:2000/6250 val_loss:2.3652 val_aux_loss:0.0363 train_time:265805ms step_avg:133.57ms
step:2001/6250 train_loss:2.4422 aux_loss:0.0362 train_time:265937ms step_avg:133.57ms
step:2002/6250 train_loss:2.3602 aux_loss:0.0363 train_time:266068ms step_avg:133.57ms
step:2003/6250 train_loss:2.3275 aux_loss:0.0364 train_time:266199ms step_avg:133.57ms
step:2004/6250 train_loss:2.3895 aux_loss:0.0366 train_time:266331ms step_avg:133.57ms
step:2005/6250 train_loss:2.2916 aux_loss:0.0364 train_time:266464ms step_avg:133.57ms
step:2006/6250 train_loss:2.3382 aux_loss:0.0362 train_time:266595ms step_avg:133.56ms
step:2007/6250 train_loss:2.4151 aux_loss:0.0363 train_time:266729ms step_avg:133.56ms
step:2008/6250 train_loss:2.3381 aux_loss:0.0360 train_time:266862ms step_avg:133.56ms
step:2009/6250 train_loss:2.3706 aux_loss:0.0360 train_time:266995ms step_avg:133.56ms
step:2010/6250 train_loss:2.3403 aux_loss:0.0362 train_time:267127ms step_avg:133.56ms
step:2011/6250 train_loss:2.3759 aux_loss:0.0367 train_time:267257ms step_avg:133.56ms
step:2012/6250 train_loss:2.4018 aux_loss:0.0364 train_time:267389ms step_avg:133.56ms
step:2013/6250 train_loss:2.3301 aux_loss:0.0361 train_time:267520ms step_avg:133.56ms
step:2014/6250 train_loss:2.2999 aux_loss:0.0363 train_time:267651ms step_avg:133.56ms
step:2015/6250 train_loss:2.3851 aux_loss:0.0362 train_time:267783ms step_avg:133.56ms
step:2016/6250 train_loss:2.4009 aux_loss:0.0361 train_time:267915ms step_avg:133.56ms
step:2017/6250 train_loss:2.4006 aux_loss:0.0360 train_time:268047ms step_avg:133.56ms
step:2018/6250 train_loss:2.3521 aux_loss:0.0359 train_time:268179ms step_avg:133.56ms
step:2019/6250 train_loss:2.3646 aux_loss:0.0361 train_time:268311ms step_avg:133.55ms
step:2020/6250 train_loss:2.3605 aux_loss:0.0360 train_time:268445ms step_avg:133.55ms
step:2021/6250 train_loss:2.3813 aux_loss:0.0364 train_time:268578ms step_avg:133.55ms
step:2022/6250 train_loss:2.3596 aux_loss:0.0367 train_time:268710ms step_avg:133.55ms
step:2023/6250 train_loss:2.3083 aux_loss:0.0365 train_time:268842ms step_avg:133.55ms
step:2024/6250 train_loss:2.3936 aux_loss:0.0365 train_time:268975ms step_avg:133.55ms
step:2025/6250 train_loss:2.3893 aux_loss:0.0365 train_time:269109ms step_avg:133.55ms
step:2026/6250 train_loss:2.3470 aux_loss:0.0365 train_time:269240ms step_avg:133.55ms
step:2027/6250 train_loss:2.3369 aux_loss:0.0363 train_time:269373ms step_avg:133.55ms
step:2028/6250 train_loss:2.3025 aux_loss:0.0362 train_time:269505ms step_avg:133.55ms
step:2029/6250 train_loss:2.3997 aux_loss:0.0363 train_time:269637ms step_avg:133.55ms
step:2030/6250 train_loss:2.3726 aux_loss:0.0362 train_time:269770ms step_avg:133.55ms
step:2031/6250 train_loss:2.4171 aux_loss:0.0363 train_time:269904ms step_avg:133.55ms
step:2032/6250 train_loss:2.3333 aux_loss:0.0364 train_time:270037ms step_avg:133.55ms
step:2033/6250 train_loss:2.3294 aux_loss:0.0363 train_time:270169ms step_avg:133.55ms
step:2034/6250 train_loss:2.4027 aux_loss:0.0360 train_time:270302ms step_avg:133.55ms
step:2035/6250 train_loss:2.3410 aux_loss:0.0362 train_time:270436ms step_avg:133.55ms
step:2036/6250 train_loss:2.3003 aux_loss:0.0362 train_time:270568ms step_avg:133.55ms
step:2037/6250 train_loss:2.4496 aux_loss:0.0360 train_time:270699ms step_avg:133.55ms
step:2038/6250 train_loss:2.3403 aux_loss:0.0359 train_time:270830ms step_avg:133.55ms
step:2039/6250 train_loss:2.3294 aux_loss:0.0359 train_time:270963ms step_avg:133.55ms
step:2040/6250 train_loss:2.2936 aux_loss:0.0358 train_time:271094ms step_avg:133.54ms
step:2041/6250 train_loss:2.3857 aux_loss:0.0359 train_time:271225ms step_avg:133.54ms
step:2042/6250 train_loss:2.3918 aux_loss:0.0359 train_time:271357ms step_avg:133.54ms
step:2043/6250 train_loss:2.3167 aux_loss:0.0360 train_time:271488ms step_avg:133.54ms
step:2044/6250 train_loss:2.2593 aux_loss:0.0359 train_time:271621ms step_avg:133.54ms
step:2045/6250 train_loss:2.3223 aux_loss:0.0360 train_time:271752ms step_avg:133.54ms
step:2046/6250 train_loss:2.4678 aux_loss:0.0360 train_time:271885ms step_avg:133.54ms
step:2047/6250 train_loss:2.4140 aux_loss:0.0357 train_time:272016ms step_avg:133.54ms
step:2048/6250 train_loss:2.3363 aux_loss:0.0358 train_time:272147ms step_avg:133.54ms
step:2049/6250 train_loss:2.3706 aux_loss:0.0359 train_time:272278ms step_avg:133.53ms
step:2050/6250 train_loss:2.2672 aux_loss:0.0362 train_time:272410ms step_avg:133.53ms
step:2051/6250 train_loss:2.3573 aux_loss:0.0360 train_time:272542ms step_avg:133.53ms
step:2052/6250 train_loss:2.3460 aux_loss:0.0358 train_time:272672ms step_avg:133.53ms
step:2053/6250 train_loss:2.2946 aux_loss:0.0358 train_time:272806ms step_avg:133.53ms
step:2054/6250 train_loss:2.3743 aux_loss:0.0361 train_time:272939ms step_avg:133.53ms
step:2055/6250 train_loss:2.2424 aux_loss:0.0357 train_time:273069ms step_avg:133.53ms
step:2056/6250 train_loss:2.4752 aux_loss:0.0355 train_time:273200ms step_avg:133.53ms
step:2057/6250 train_loss:2.3234 aux_loss:0.0353 train_time:273331ms step_avg:133.53ms
step:2058/6250 train_loss:2.3222 aux_loss:0.0355 train_time:273464ms step_avg:133.53ms
step:2059/6250 train_loss:2.4075 aux_loss:0.0357 train_time:273595ms step_avg:133.53ms
step:2060/6250 train_loss:2.3533 aux_loss:0.0358 train_time:273726ms step_avg:133.52ms
step:2061/6250 train_loss:2.2920 aux_loss:0.0358 train_time:273858ms step_avg:133.52ms
step:2062/6250 train_loss:2.4445 aux_loss:0.0359 train_time:273988ms step_avg:133.52ms
step:2063/6250 train_loss:2.3392 aux_loss:0.0362 train_time:274120ms step_avg:133.52ms
step:2064/6250 train_loss:2.3121 aux_loss:0.0362 train_time:274250ms step_avg:133.52ms
step:2065/6250 train_loss:2.4198 aux_loss:0.0356 train_time:274384ms step_avg:133.52ms
step:2066/6250 train_loss:2.3374 aux_loss:0.0357 train_time:274515ms step_avg:133.52ms
step:2067/6250 train_loss:2.3683 aux_loss:0.0360 train_time:274646ms step_avg:133.52ms
step:2068/6250 train_loss:2.2840 aux_loss:0.0360 train_time:274778ms step_avg:133.52ms
step:2069/6250 train_loss:2.3026 aux_loss:0.0359 train_time:274910ms step_avg:133.52ms
step:2070/6250 train_loss:2.4159 aux_loss:0.0359 train_time:275043ms step_avg:133.52ms
step:2071/6250 train_loss:2.3848 aux_loss:0.0363 train_time:275174ms step_avg:133.51ms
step:2072/6250 train_loss:2.3952 aux_loss:0.0363 train_time:275306ms step_avg:133.51ms
step:2073/6250 train_loss:2.3193 aux_loss:0.0359 train_time:275438ms step_avg:133.51ms
step:2074/6250 train_loss:2.3500 aux_loss:0.0357 train_time:275569ms step_avg:133.51ms
step:2075/6250 train_loss:2.4104 aux_loss:0.0359 train_time:275701ms step_avg:133.51ms
step:2076/6250 train_loss:2.3651 aux_loss:0.0358 train_time:275833ms step_avg:133.51ms
step:2077/6250 train_loss:2.3803 aux_loss:0.0357 train_time:275967ms step_avg:133.51ms
step:2078/6250 train_loss:2.4253 aux_loss:0.0355 train_time:276099ms step_avg:133.51ms
step:2079/6250 train_loss:2.4223 aux_loss:0.0358 train_time:276230ms step_avg:133.51ms
step:2080/6250 train_loss:2.2993 aux_loss:0.0357 train_time:276362ms step_avg:133.51ms
step:2081/6250 train_loss:2.3882 aux_loss:0.0355 train_time:276493ms step_avg:133.51ms
step:2082/6250 train_loss:2.3212 aux_loss:0.0358 train_time:276625ms step_avg:133.51ms
step:2083/6250 train_loss:2.4394 aux_loss:0.0359 train_time:276756ms step_avg:133.50ms
step:2084/6250 train_loss:2.3886 aux_loss:0.0355 train_time:276888ms step_avg:133.50ms
step:2085/6250 train_loss:2.2709 aux_loss:0.0355 train_time:277019ms step_avg:133.50ms
step:2086/6250 train_loss:2.2759 aux_loss:0.0351 train_time:277152ms step_avg:133.50ms
step:2087/6250 train_loss:2.4468 aux_loss:0.0355 train_time:277286ms step_avg:133.50ms
step:2088/6250 train_loss:2.3784 aux_loss:0.0354 train_time:277420ms step_avg:133.50ms
step:2089/6250 train_loss:2.4112 aux_loss:0.0353 train_time:277552ms step_avg:133.50ms
step:2090/6250 train_loss:2.3005 aux_loss:0.0354 train_time:277688ms step_avg:133.50ms
step:2091/6250 train_loss:2.4157 aux_loss:0.0354 train_time:277823ms step_avg:133.50ms
step:2092/6250 train_loss:2.3679 aux_loss:0.0357 train_time:277957ms step_avg:133.50ms
step:2093/6250 train_loss:2.3000 aux_loss:0.0353 train_time:278089ms step_avg:133.50ms
step:2094/6250 train_loss:2.2136 aux_loss:0.0354 train_time:278223ms step_avg:133.50ms
step:2095/6250 train_loss:2.3488 aux_loss:0.0355 train_time:278359ms step_avg:133.51ms
step:2096/6250 train_loss:2.3554 aux_loss:0.0354 train_time:278497ms step_avg:133.51ms
step:2097/6250 train_loss:2.3165 aux_loss:0.0356 train_time:278630ms step_avg:133.51ms
step:2098/6250 train_loss:2.3873 aux_loss:0.0357 train_time:278764ms step_avg:133.51ms
step:2099/6250 train_loss:2.2989 aux_loss:0.0353 train_time:278895ms step_avg:133.51ms
step:2100/6250 train_loss:2.3612 aux_loss:0.0356 train_time:279027ms step_avg:133.51ms
step:2101/6250 train_loss:2.2579 aux_loss:0.0356 train_time:279160ms step_avg:133.51ms
step:2102/6250 train_loss:2.4255 aux_loss:0.0352 train_time:279292ms step_avg:133.50ms
step:2103/6250 train_loss:2.2846 aux_loss:0.0349 train_time:279424ms step_avg:133.50ms
step:2104/6250 train_loss:2.3864 aux_loss:0.0353 train_time:279556ms step_avg:133.50ms
step:2105/6250 train_loss:2.3656 aux_loss:0.0354 train_time:279689ms step_avg:133.50ms
step:2106/6250 train_loss:2.2971 aux_loss:0.0350 train_time:279822ms step_avg:133.50ms
step:2107/6250 train_loss:2.2935 aux_loss:0.0351 train_time:279953ms step_avg:133.50ms
step:2108/6250 train_loss:2.3886 aux_loss:0.0350 train_time:280084ms step_avg:133.50ms
step:2109/6250 train_loss:2.3558 aux_loss:0.0353 train_time:280216ms step_avg:133.50ms
step:2110/6250 train_loss:2.4086 aux_loss:0.0352 train_time:280347ms step_avg:133.50ms
step:2111/6250 train_loss:2.3277 aux_loss:0.0349 train_time:280480ms step_avg:133.50ms
step:2112/6250 train_loss:2.2972 aux_loss:0.0356 train_time:280611ms step_avg:133.50ms
step:2113/6250 train_loss:2.3894 aux_loss:0.0354 train_time:280744ms step_avg:133.50ms
step:2114/6250 train_loss:2.2940 aux_loss:0.0354 train_time:280876ms step_avg:133.50ms
step:2115/6250 train_loss:2.3887 aux_loss:0.0356 train_time:281008ms step_avg:133.50ms
step:2116/6250 train_loss:2.2773 aux_loss:0.0354 train_time:281140ms step_avg:133.49ms
step:2117/6250 train_loss:2.4349 aux_loss:0.0352 train_time:281272ms step_avg:133.49ms
step:2118/6250 train_loss:2.3509 aux_loss:0.0352 train_time:281404ms step_avg:133.49ms
step:2119/6250 train_loss:2.3812 aux_loss:0.0354 train_time:281536ms step_avg:133.49ms
step:2120/6250 train_loss:2.3133 aux_loss:0.0351 train_time:281668ms step_avg:133.49ms
step:2121/6250 train_loss:2.4037 aux_loss:0.0352 train_time:281802ms step_avg:133.49ms
step:2122/6250 train_loss:2.3363 aux_loss:0.0354 train_time:281935ms step_avg:133.49ms
step:2123/6250 train_loss:2.4144 aux_loss:0.0357 train_time:282067ms step_avg:133.49ms
step:2124/6250 train_loss:2.4205 aux_loss:0.0352 train_time:282202ms step_avg:133.49ms
step:2125/6250 train_loss:2.3437 aux_loss:0.0351 train_time:282336ms step_avg:133.49ms
step:2126/6250 train_loss:2.3058 aux_loss:0.0352 train_time:282475ms step_avg:133.49ms
step:2127/6250 train_loss:2.3851 aux_loss:0.0356 train_time:282610ms step_avg:133.50ms
step:2128/6250 train_loss:2.3291 aux_loss:0.0353 train_time:282748ms step_avg:133.50ms
step:2129/6250 train_loss:2.3572 aux_loss:0.0353 train_time:282882ms step_avg:133.50ms
step:2130/6250 train_loss:2.3055 aux_loss:0.0354 train_time:283017ms step_avg:133.50ms
step:2131/6250 train_loss:2.3368 aux_loss:0.0354 train_time:283149ms step_avg:133.50ms
step:2132/6250 train_loss:2.3389 aux_loss:0.0354 train_time:283282ms step_avg:133.50ms
step:2133/6250 train_loss:2.4194 aux_loss:0.0353 train_time:283414ms step_avg:133.50ms
step:2134/6250 train_loss:2.2476 aux_loss:0.0350 train_time:283546ms step_avg:133.50ms
step:2135/6250 train_loss:2.2622 aux_loss:0.0349 train_time:283679ms step_avg:133.50ms
step:2136/6250 train_loss:2.1954 aux_loss:0.0351 train_time:283811ms step_avg:133.50ms
step:2137/6250 train_loss:2.3706 aux_loss:0.0354 train_time:283944ms step_avg:133.50ms
step:2138/6250 train_loss:2.2476 aux_loss:0.0354 train_time:284075ms step_avg:133.49ms
step:2139/6250 train_loss:2.3661 aux_loss:0.0354 train_time:284208ms step_avg:133.49ms
step:2140/6250 train_loss:2.3715 aux_loss:0.0353 train_time:284352ms step_avg:133.50ms
step:2141/6250 train_loss:2.3437 aux_loss:0.0354 train_time:284497ms step_avg:133.50ms
step:2142/6250 train_loss:2.3441 aux_loss:0.0351 train_time:284640ms step_avg:133.51ms
step:2143/6250 train_loss:2.2951 aux_loss:0.0350 train_time:284782ms step_avg:133.51ms
step:2144/6250 train_loss:2.3186 aux_loss:0.0350 train_time:284926ms step_avg:133.52ms
step:2145/6250 train_loss:2.3930 aux_loss:0.0350 train_time:285069ms step_avg:133.52ms
step:2146/6250 train_loss:2.4367 aux_loss:0.0346 train_time:285212ms step_avg:133.53ms
step:2147/6250 train_loss:2.3375 aux_loss:0.0349 train_time:285354ms step_avg:133.53ms
step:2148/6250 train_loss:2.3895 aux_loss:0.0345 train_time:285497ms step_avg:133.53ms
step:2149/6250 train_loss:2.3176 aux_loss:0.0348 train_time:285640ms step_avg:133.54ms
step:2150/6250 train_loss:2.3265 aux_loss:0.0351 train_time:285783ms step_avg:133.54ms
step:2151/6250 train_loss:2.4143 aux_loss:0.0350 train_time:285925ms step_avg:133.55ms
step:2152/6250 train_loss:2.3254 aux_loss:0.0353 train_time:286068ms step_avg:133.55ms
step:2153/6250 train_loss:2.3429 aux_loss:0.0350 train_time:286212ms step_avg:133.56ms
step:2154/6250 train_loss:2.3294 aux_loss:0.0353 train_time:286354ms step_avg:133.56ms
step:2155/6250 train_loss:2.2485 aux_loss:0.0350 train_time:286497ms step_avg:133.56ms
step:2156/6250 train_loss:2.3150 aux_loss:0.0349 train_time:286646ms step_avg:133.57ms
step:2157/6250 train_loss:2.2423 aux_loss:0.0347 train_time:286780ms step_avg:133.57ms
step:2158/6250 train_loss:2.3348 aux_loss:0.0349 train_time:286916ms step_avg:133.57ms
step:2159/6250 train_loss:2.3785 aux_loss:0.0348 train_time:287053ms step_avg:133.58ms
step:2160/6250 train_loss:2.3422 aux_loss:0.0346 train_time:287185ms step_avg:133.57ms
step:2161/6250 train_loss:2.2832 aux_loss:0.0346 train_time:287318ms step_avg:133.57ms
step:2162/6250 train_loss:2.3165 aux_loss:0.0348 train_time:287453ms step_avg:133.57ms
step:2163/6250 train_loss:2.4625 aux_loss:0.0348 train_time:287591ms step_avg:133.58ms
step:2164/6250 train_loss:2.2321 aux_loss:0.0347 train_time:287731ms step_avg:133.58ms
step:2165/6250 train_loss:2.3202 aux_loss:0.0346 train_time:287871ms step_avg:133.58ms
step:2166/6250 train_loss:2.3351 aux_loss:0.0343 train_time:288010ms step_avg:133.59ms
step:2167/6250 train_loss:2.3857 aux_loss:0.0349 train_time:288145ms step_avg:133.59ms
step:2168/6250 train_loss:2.3361 aux_loss:0.0349 train_time:288287ms step_avg:133.59ms
step:2169/6250 train_loss:2.3262 aux_loss:0.0355 train_time:288426ms step_avg:133.59ms
step:2170/6250 train_loss:2.3444 aux_loss:0.0349 train_time:288559ms step_avg:133.59ms
step:2171/6250 train_loss:2.4131 aux_loss:0.0348 train_time:288692ms step_avg:133.59ms
step:2172/6250 train_loss:2.4101 aux_loss:0.0346 train_time:288829ms step_avg:133.59ms
step:2173/6250 train_loss:2.3607 aux_loss:0.0345 train_time:288970ms step_avg:133.60ms
step:2174/6250 train_loss:2.3545 aux_loss:0.0344 train_time:289112ms step_avg:133.60ms
step:2175/6250 train_loss:2.2799 aux_loss:0.0345 train_time:289248ms step_avg:133.60ms
step:2176/6250 train_loss:2.3987 aux_loss:0.0345 train_time:289381ms step_avg:133.60ms
step:2177/6250 train_loss:2.3689 aux_loss:0.0346 train_time:289515ms step_avg:133.60ms
step:2178/6250 train_loss:2.3604 aux_loss:0.0348 train_time:289647ms step_avg:133.60ms
step:2179/6250 train_loss:2.3401 aux_loss:0.0348 train_time:289781ms step_avg:133.60ms
step:2180/6250 train_loss:2.4011 aux_loss:0.0349 train_time:289913ms step_avg:133.60ms
step:2181/6250 train_loss:2.2894 aux_loss:0.0348 train_time:290045ms step_avg:133.60ms
step:2182/6250 train_loss:2.3181 aux_loss:0.0351 train_time:290177ms step_avg:133.60ms
step:2183/6250 train_loss:2.2982 aux_loss:0.0350 train_time:290311ms step_avg:133.60ms
step:2184/6250 train_loss:2.3958 aux_loss:0.0350 train_time:290446ms step_avg:133.60ms
step:2185/6250 train_loss:2.3511 aux_loss:0.0350 train_time:290582ms step_avg:133.60ms
step:2186/6250 train_loss:2.3983 aux_loss:0.0350 train_time:290716ms step_avg:133.60ms
step:2187/6250 train_loss:2.3786 aux_loss:0.0350 train_time:290849ms step_avg:133.60ms
step:2188/6250 train_loss:2.3476 aux_loss:0.0347 train_time:290982ms step_avg:133.60ms
step:2189/6250 train_loss:2.4173 aux_loss:0.0349 train_time:291117ms step_avg:133.60ms
step:2190/6250 train_loss:2.3942 aux_loss:0.0349 train_time:291255ms step_avg:133.60ms
step:2191/6250 train_loss:2.2902 aux_loss:0.0346 train_time:291387ms step_avg:133.60ms
step:2192/6250 train_loss:2.2496 aux_loss:0.0348 train_time:291522ms step_avg:133.60ms
step:2193/6250 train_loss:2.3999 aux_loss:0.0348 train_time:291665ms step_avg:133.61ms
step:2194/6250 train_loss:2.4215 aux_loss:0.0346 train_time:291805ms step_avg:133.61ms
step:2195/6250 train_loss:2.2718 aux_loss:0.0345 train_time:291938ms step_avg:133.61ms
step:2196/6250 train_loss:2.4401 aux_loss:0.0348 train_time:292069ms step_avg:133.61ms
step:2197/6250 train_loss:2.3522 aux_loss:0.0345 train_time:292200ms step_avg:133.61ms
step:2198/6250 train_loss:2.4215 aux_loss:0.0347 train_time:292332ms step_avg:133.61ms
step:2199/6250 train_loss:2.3348 aux_loss:0.0347 train_time:292464ms step_avg:133.61ms
step:2200/6250 train_loss:2.3366 aux_loss:0.0346 train_time:292598ms step_avg:133.61ms
step:2201/6250 train_loss:2.3223 aux_loss:0.0343 train_time:292730ms step_avg:133.61ms
step:2202/6250 train_loss:2.3079 aux_loss:0.0346 train_time:292862ms step_avg:133.61ms
step:2203/6250 train_loss:2.3383 aux_loss:0.0349 train_time:292996ms step_avg:133.61ms
step:2204/6250 train_loss:2.3432 aux_loss:0.0345 train_time:293129ms step_avg:133.60ms
step:2205/6250 train_loss:2.4042 aux_loss:0.0344 train_time:293261ms step_avg:133.60ms
step:2206/6250 train_loss:2.3829 aux_loss:0.0346 train_time:293393ms step_avg:133.60ms
step:2207/6250 train_loss:2.4138 aux_loss:0.0351 train_time:293527ms step_avg:133.60ms
step:2208/6250 train_loss:2.2022 aux_loss:0.0346 train_time:293659ms step_avg:133.60ms
step:2209/6250 train_loss:2.3796 aux_loss:0.0346 train_time:293791ms step_avg:133.60ms
step:2210/6250 train_loss:2.3059 aux_loss:0.0346 train_time:293923ms step_avg:133.60ms
step:2211/6250 train_loss:2.3138 aux_loss:0.0348 train_time:294057ms step_avg:133.60ms
step:2212/6250 train_loss:2.2148 aux_loss:0.0343 train_time:294189ms step_avg:133.60ms
step:2213/6250 train_loss:2.2709 aux_loss:0.0344 train_time:294320ms step_avg:133.60ms
step:2214/6250 train_loss:2.2655 aux_loss:0.0345 train_time:294452ms step_avg:133.60ms
step:2215/6250 train_loss:2.3200 aux_loss:0.0346 train_time:294583ms step_avg:133.60ms
step:2216/6250 train_loss:2.3572 aux_loss:0.0341 train_time:294714ms step_avg:133.60ms
step:2217/6250 train_loss:2.3441 aux_loss:0.0345 train_time:294846ms step_avg:133.60ms
step:2218/6250 train_loss:2.3112 aux_loss:0.0347 train_time:294978ms step_avg:133.60ms
step:2219/6250 train_loss:2.3314 aux_loss:0.0347 train_time:295110ms step_avg:133.59ms
step:2220/6250 train_loss:2.2585 aux_loss:0.0344 train_time:295241ms step_avg:133.59ms
step:2221/6250 train_loss:2.4029 aux_loss:0.0348 train_time:295372ms step_avg:133.59ms
step:2222/6250 train_loss:2.4063 aux_loss:0.0350 train_time:295503ms step_avg:133.59ms
step:2223/6250 train_loss:2.2679 aux_loss:0.0346 train_time:295635ms step_avg:133.59ms
step:2224/6250 train_loss:2.3266 aux_loss:0.0344 train_time:295767ms step_avg:133.59ms
step:2225/6250 train_loss:2.3283 aux_loss:0.0344 train_time:295899ms step_avg:133.59ms
step:2226/6250 train_loss:2.3896 aux_loss:0.0346 train_time:296031ms step_avg:133.59ms
step:2227/6250 train_loss:2.3687 aux_loss:0.0346 train_time:296164ms step_avg:133.59ms
step:2228/6250 train_loss:2.3497 aux_loss:0.0345 train_time:296296ms step_avg:133.59ms
step:2229/6250 train_loss:2.3514 aux_loss:0.0346 train_time:296427ms step_avg:133.59ms
step:2230/6250 train_loss:2.2830 aux_loss:0.0342 train_time:296558ms step_avg:133.58ms
step:2231/6250 train_loss:2.5487 aux_loss:0.0342 train_time:296690ms step_avg:133.58ms
step:2232/6250 train_loss:2.3983 aux_loss:0.0345 train_time:296821ms step_avg:133.58ms
step:2233/6250 train_loss:2.3302 aux_loss:0.0344 train_time:296954ms step_avg:133.58ms
step:2234/6250 train_loss:2.3093 aux_loss:0.0340 train_time:297085ms step_avg:133.58ms
step:2235/6250 train_loss:2.3453 aux_loss:0.0340 train_time:297217ms step_avg:133.58ms
step:2236/6250 train_loss:2.3857 aux_loss:0.0346 train_time:297348ms step_avg:133.58ms
step:2237/6250 train_loss:2.3796 aux_loss:0.0346 train_time:297479ms step_avg:133.58ms
step:2238/6250 train_loss:2.3633 aux_loss:0.0346 train_time:297612ms step_avg:133.58ms
step:2239/6250 train_loss:2.2325 aux_loss:0.0344 train_time:297742ms step_avg:133.58ms
step:2240/6250 train_loss:2.2642 aux_loss:0.0340 train_time:297875ms step_avg:133.58ms
step:2241/6250 train_loss:2.2547 aux_loss:0.0343 train_time:298007ms step_avg:133.58ms
step:2242/6250 train_loss:2.3406 aux_loss:0.0343 train_time:298138ms step_avg:133.57ms
step:2243/6250 train_loss:2.3237 aux_loss:0.0343 train_time:298270ms step_avg:133.57ms
step:2244/6250 train_loss:2.3838 aux_loss:0.0340 train_time:298400ms step_avg:133.57ms
step:2245/6250 train_loss:2.2763 aux_loss:0.0341 train_time:298532ms step_avg:133.57ms
step:2246/6250 train_loss:2.3366 aux_loss:0.0344 train_time:298663ms step_avg:133.57ms
step:2247/6250 train_loss:2.3422 aux_loss:0.0342 train_time:298796ms step_avg:133.57ms
step:2248/6250 train_loss:2.2818 aux_loss:0.0346 train_time:298927ms step_avg:133.57ms
step:2249/6250 train_loss:2.4089 aux_loss:0.0346 train_time:299059ms step_avg:133.57ms
step:2250/6250 train_loss:2.2640 aux_loss:0.0343 train_time:299190ms step_avg:133.57ms
step:2251/6250 train_loss:2.3172 aux_loss:0.0344 train_time:299321ms step_avg:133.57ms
step:2252/6250 train_loss:2.3412 aux_loss:0.0344 train_time:299453ms step_avg:133.57ms
step:2253/6250 train_loss:2.3890 aux_loss:0.0345 train_time:299583ms step_avg:133.56ms
step:2254/6250 train_loss:2.3146 aux_loss:0.0344 train_time:299715ms step_avg:133.56ms
step:2255/6250 train_loss:2.3350 aux_loss:0.0345 train_time:299847ms step_avg:133.56ms
step:2256/6250 train_loss:2.3065 aux_loss:0.0340 train_time:299979ms step_avg:133.56ms
step:2257/6250 train_loss:2.3080 aux_loss:0.0342 train_time:300110ms step_avg:133.56ms
step:2258/6250 train_loss:2.2329 aux_loss:0.0339 train_time:300241ms step_avg:133.56ms
step:2259/6250 train_loss:2.3661 aux_loss:0.0342 train_time:300373ms step_avg:133.56ms
step:2260/6250 train_loss:2.2528 aux_loss:0.0345 train_time:300504ms step_avg:133.56ms
step:2261/6250 train_loss:2.2550 aux_loss:0.0342 train_time:300636ms step_avg:133.56ms
step:2262/6250 train_loss:2.2934 aux_loss:0.0343 train_time:300769ms step_avg:133.56ms
step:2263/6250 train_loss:2.2424 aux_loss:0.0340 train_time:300900ms step_avg:133.56ms
step:2264/6250 train_loss:2.3286 aux_loss:0.0339 train_time:301032ms step_avg:133.55ms
step:2265/6250 train_loss:2.3232 aux_loss:0.0340 train_time:301163ms step_avg:133.55ms
step:2266/6250 train_loss:2.2576 aux_loss:0.0337 train_time:301295ms step_avg:133.55ms
step:2267/6250 train_loss:2.3054 aux_loss:0.0338 train_time:301426ms step_avg:133.55ms
step:2268/6250 train_loss:2.3322 aux_loss:0.0341 train_time:301558ms step_avg:133.55ms
step:2269/6250 train_loss:2.5246 aux_loss:0.0342 train_time:301691ms step_avg:133.55ms
step:2270/6250 train_loss:2.3277 aux_loss:0.0342 train_time:301822ms step_avg:133.55ms
step:2271/6250 train_loss:2.3484 aux_loss:0.0344 train_time:301955ms step_avg:133.55ms
step:2272/6250 train_loss:2.2719 aux_loss:0.0343 train_time:302086ms step_avg:133.55ms
step:2273/6250 train_loss:2.3511 aux_loss:0.0340 train_time:302218ms step_avg:133.55ms
step:2274/6250 train_loss:2.3330 aux_loss:0.0347 train_time:302350ms step_avg:133.55ms
step:2275/6250 train_loss:2.3246 aux_loss:0.0345 train_time:302482ms step_avg:133.55ms
step:2276/6250 train_loss:2.4621 aux_loss:0.0346 train_time:302614ms step_avg:133.55ms
step:2277/6250 train_loss:2.3105 aux_loss:0.0340 train_time:302745ms step_avg:133.54ms
step:2278/6250 train_loss:2.3452 aux_loss:0.0337 train_time:302877ms step_avg:133.54ms
step:2279/6250 train_loss:2.2824 aux_loss:0.0341 train_time:303011ms step_avg:133.54ms
step:2280/6250 train_loss:2.3178 aux_loss:0.0342 train_time:303143ms step_avg:133.54ms
step:2281/6250 train_loss:2.3451 aux_loss:0.0338 train_time:303276ms step_avg:133.54ms
step:2282/6250 train_loss:2.3175 aux_loss:0.0336 train_time:303408ms step_avg:133.54ms
step:2283/6250 train_loss:2.2426 aux_loss:0.0336 train_time:303541ms step_avg:133.54ms
step:2284/6250 train_loss:2.5060 aux_loss:0.0337 train_time:303673ms step_avg:133.54ms
step:2285/6250 train_loss:2.3468 aux_loss:0.0338 train_time:303805ms step_avg:133.54ms
step:2286/6250 train_loss:2.3537 aux_loss:0.0337 train_time:303937ms step_avg:133.54ms
step:2287/6250 train_loss:2.3414 aux_loss:0.0335 train_time:304070ms step_avg:133.54ms
step:2288/6250 train_loss:2.3009 aux_loss:0.0339 train_time:304203ms step_avg:133.54ms
step:2289/6250 train_loss:2.2760 aux_loss:0.0337 train_time:304335ms step_avg:133.54ms
step:2290/6250 train_loss:2.4286 aux_loss:0.0337 train_time:304468ms step_avg:133.54ms
step:2291/6250 train_loss:2.3754 aux_loss:0.0339 train_time:304599ms step_avg:133.54ms
step:2292/6250 train_loss:2.3544 aux_loss:0.0338 train_time:304732ms step_avg:133.54ms
step:2293/6250 train_loss:2.3510 aux_loss:0.0338 train_time:304864ms step_avg:133.54ms
step:2294/6250 train_loss:2.2832 aux_loss:0.0339 train_time:304996ms step_avg:133.54ms
step:2295/6250 train_loss:2.3714 aux_loss:0.0339 train_time:305128ms step_avg:133.54ms
step:2296/6250 train_loss:2.4131 aux_loss:0.0339 train_time:305259ms step_avg:133.53ms
step:2297/6250 train_loss:2.3796 aux_loss:0.0340 train_time:305392ms step_avg:133.53ms
step:2298/6250 train_loss:2.3326 aux_loss:0.0340 train_time:305524ms step_avg:133.53ms
step:2299/6250 train_loss:2.3559 aux_loss:0.0341 train_time:305656ms step_avg:133.53ms
step:2300/6250 train_loss:2.3991 aux_loss:0.0340 train_time:305790ms step_avg:133.53ms
step:2301/6250 train_loss:2.3273 aux_loss:0.0337 train_time:305922ms step_avg:133.53ms
step:2302/6250 train_loss:2.2634 aux_loss:0.0341 train_time:306055ms step_avg:133.53ms
step:2303/6250 train_loss:2.2993 aux_loss:0.0340 train_time:306188ms step_avg:133.53ms
step:2304/6250 train_loss:2.2776 aux_loss:0.0339 train_time:306320ms step_avg:133.53ms
step:2305/6250 train_loss:2.3467 aux_loss:0.0340 train_time:306452ms step_avg:133.53ms
step:2306/6250 train_loss:2.2812 aux_loss:0.0341 train_time:306584ms step_avg:133.53ms
step:2307/6250 train_loss:2.2766 aux_loss:0.0341 train_time:306715ms step_avg:133.53ms
step:2308/6250 train_loss:2.3042 aux_loss:0.0338 train_time:306847ms step_avg:133.53ms
step:2309/6250 train_loss:2.2932 aux_loss:0.0340 train_time:306980ms step_avg:133.53ms
step:2310/6250 train_loss:2.2774 aux_loss:0.0341 train_time:307110ms step_avg:133.53ms
step:2311/6250 train_loss:2.3018 aux_loss:0.0341 train_time:307242ms step_avg:133.53ms
step:2312/6250 train_loss:2.2948 aux_loss:0.0339 train_time:307374ms step_avg:133.52ms
step:2313/6250 train_loss:2.3617 aux_loss:0.0337 train_time:307508ms step_avg:133.53ms
step:2314/6250 train_loss:2.2791 aux_loss:0.0335 train_time:307640ms step_avg:133.52ms
step:2315/6250 train_loss:2.3947 aux_loss:0.0339 train_time:307773ms step_avg:133.52ms
step:2316/6250 train_loss:2.3472 aux_loss:0.0341 train_time:307904ms step_avg:133.52ms
step:2317/6250 train_loss:2.3869 aux_loss:0.0339 train_time:308036ms step_avg:133.52ms
step:2318/6250 train_loss:2.3237 aux_loss:0.0333 train_time:308167ms step_avg:133.52ms
step:2319/6250 train_loss:2.3355 aux_loss:0.0336 train_time:308299ms step_avg:133.52ms
step:2320/6250 train_loss:2.2789 aux_loss:0.0337 train_time:308434ms step_avg:133.52ms
step:2321/6250 train_loss:2.3004 aux_loss:0.0332 train_time:308567ms step_avg:133.52ms
step:2322/6250 train_loss:2.3151 aux_loss:0.0332 train_time:308698ms step_avg:133.52ms
step:2323/6250 train_loss:2.2833 aux_loss:0.0338 train_time:308832ms step_avg:133.52ms
step:2324/6250 train_loss:2.3600 aux_loss:0.0337 train_time:308965ms step_avg:133.52ms
step:2325/6250 train_loss:2.3695 aux_loss:0.0338 train_time:309096ms step_avg:133.52ms
step:2326/6250 train_loss:2.3554 aux_loss:0.0337 train_time:309227ms step_avg:133.52ms
step:2327/6250 train_loss:2.3475 aux_loss:0.0338 train_time:309358ms step_avg:133.52ms
step:2328/6250 train_loss:2.3000 aux_loss:0.0338 train_time:309491ms step_avg:133.52ms
step:2329/6250 train_loss:2.3589 aux_loss:0.0339 train_time:309622ms step_avg:133.52ms
step:2330/6250 train_loss:2.3230 aux_loss:0.0338 train_time:309755ms step_avg:133.52ms
step:2331/6250 train_loss:2.2650 aux_loss:0.0338 train_time:309885ms step_avg:133.51ms
step:2332/6250 train_loss:2.4122 aux_loss:0.0339 train_time:310017ms step_avg:133.51ms
step:2333/6250 train_loss:2.2893 aux_loss:0.0341 train_time:310148ms step_avg:133.51ms
step:2334/6250 train_loss:2.2706 aux_loss:0.0334 train_time:310279ms step_avg:133.51ms
step:2335/6250 train_loss:2.3293 aux_loss:0.0334 train_time:310411ms step_avg:133.51ms
step:2336/6250 train_loss:2.3171 aux_loss:0.0336 train_time:310543ms step_avg:133.51ms
step:2337/6250 train_loss:2.3302 aux_loss:0.0336 train_time:310675ms step_avg:133.51ms
step:2338/6250 train_loss:2.3338 aux_loss:0.0335 train_time:310807ms step_avg:133.51ms
step:2339/6250 train_loss:2.3440 aux_loss:0.0337 train_time:310938ms step_avg:133.51ms
step:2340/6250 train_loss:2.3264 aux_loss:0.0336 train_time:311070ms step_avg:133.51ms
step:2341/6250 train_loss:2.3276 aux_loss:0.0335 train_time:311202ms step_avg:133.51ms
step:2342/6250 train_loss:2.2978 aux_loss:0.0338 train_time:311334ms step_avg:133.51ms
step:2343/6250 train_loss:2.3448 aux_loss:0.0331 train_time:311466ms step_avg:133.50ms
step:2344/6250 train_loss:2.3075 aux_loss:0.0331 train_time:311597ms step_avg:133.50ms
step:2345/6250 train_loss:2.2716 aux_loss:0.0334 train_time:311729ms step_avg:133.50ms
step:2346/6250 train_loss:2.4466 aux_loss:0.0336 train_time:311861ms step_avg:133.50ms
step:2347/6250 train_loss:2.2361 aux_loss:0.0334 train_time:311994ms step_avg:133.50ms
step:2348/6250 train_loss:2.4233 aux_loss:0.0334 train_time:312131ms step_avg:133.50ms
step:2349/6250 train_loss:2.3795 aux_loss:0.0337 train_time:312267ms step_avg:133.50ms
step:2350/6250 train_loss:2.2243 aux_loss:0.0336 train_time:312404ms step_avg:133.51ms
step:2351/6250 train_loss:2.2456 aux_loss:0.0336 train_time:312536ms step_avg:133.51ms
step:2352/6250 train_loss:2.3565 aux_loss:0.0333 train_time:312675ms step_avg:133.51ms
step:2353/6250 train_loss:2.3103 aux_loss:0.0331 train_time:312812ms step_avg:133.51ms
step:2354/6250 train_loss:2.3283 aux_loss:0.0335 train_time:312944ms step_avg:133.51ms
step:2355/6250 train_loss:2.3177 aux_loss:0.0336 train_time:313076ms step_avg:133.51ms
step:2356/6250 train_loss:2.3421 aux_loss:0.0334 train_time:313208ms step_avg:133.51ms
step:2357/6250 train_loss:2.3816 aux_loss:0.0338 train_time:313339ms step_avg:133.51ms
step:2358/6250 train_loss:2.2514 aux_loss:0.0335 train_time:313471ms step_avg:133.51ms
step:2359/6250 train_loss:2.3427 aux_loss:0.0333 train_time:313603ms step_avg:133.50ms
step:2360/6250 train_loss:2.2170 aux_loss:0.0331 train_time:313734ms step_avg:133.50ms
step:2361/6250 train_loss:2.4174 aux_loss:0.0335 train_time:313866ms step_avg:133.50ms
step:2362/6250 train_loss:2.2973 aux_loss:0.0336 train_time:313998ms step_avg:133.50ms
step:2363/6250 train_loss:2.4125 aux_loss:0.0336 train_time:314130ms step_avg:133.50ms
step:2364/6250 train_loss:2.2774 aux_loss:0.0335 train_time:314261ms step_avg:133.50ms
step:2365/6250 train_loss:2.2739 aux_loss:0.0334 train_time:314393ms step_avg:133.50ms
step:2366/6250 train_loss:2.3859 aux_loss:0.0337 train_time:314524ms step_avg:133.50ms
step:2367/6250 train_loss:2.3487 aux_loss:0.0338 train_time:314655ms step_avg:133.50ms
step:2368/6250 train_loss:2.2989 aux_loss:0.0337 train_time:314787ms step_avg:133.50ms
step:2369/6250 train_loss:2.3281 aux_loss:0.0337 train_time:314919ms step_avg:133.50ms
step:2370/6250 train_loss:2.2408 aux_loss:0.0336 train_time:315052ms step_avg:133.50ms
step:2371/6250 train_loss:2.2655 aux_loss:0.0336 train_time:315183ms step_avg:133.50ms
step:2372/6250 train_loss:2.3373 aux_loss:0.0335 train_time:315314ms step_avg:133.49ms
step:2373/6250 train_loss:2.3844 aux_loss:0.0334 train_time:315446ms step_avg:133.49ms
step:2374/6250 train_loss:2.3126 aux_loss:0.0335 train_time:315576ms step_avg:133.49ms
step:2375/6250 train_loss:2.3054 aux_loss:0.0334 train_time:315708ms step_avg:133.49ms
step:2376/6250 train_loss:2.3032 aux_loss:0.0336 train_time:315841ms step_avg:133.49ms
step:2377/6250 train_loss:2.3478 aux_loss:0.0336 train_time:315974ms step_avg:133.49ms
step:2378/6250 train_loss:2.2248 aux_loss:0.0333 train_time:316105ms step_avg:133.49ms
step:2379/6250 train_loss:2.3802 aux_loss:0.0334 train_time:316236ms step_avg:133.49ms
step:2380/6250 train_loss:2.3576 aux_loss:0.0335 train_time:316369ms step_avg:133.49ms
step:2381/6250 train_loss:2.3269 aux_loss:0.0332 train_time:316501ms step_avg:133.49ms
step:2382/6250 train_loss:2.3079 aux_loss:0.0329 train_time:316633ms step_avg:133.49ms
step:2383/6250 train_loss:2.4051 aux_loss:0.0335 train_time:316764ms step_avg:133.49ms
step:2384/6250 train_loss:2.3234 aux_loss:0.0332 train_time:316896ms step_avg:133.49ms
step:2385/6250 train_loss:2.3759 aux_loss:0.0335 train_time:317030ms step_avg:133.49ms
step:2386/6250 train_loss:2.4017 aux_loss:0.0335 train_time:317163ms step_avg:133.49ms
step:2387/6250 train_loss:2.3645 aux_loss:0.0340 train_time:317295ms step_avg:133.49ms
step:2388/6250 train_loss:2.3195 aux_loss:0.0334 train_time:317427ms step_avg:133.48ms
step:2389/6250 train_loss:2.3789 aux_loss:0.0331 train_time:317559ms step_avg:133.48ms
step:2390/6250 train_loss:2.3450 aux_loss:0.0330 train_time:317691ms step_avg:133.48ms
step:2391/6250 train_loss:2.2960 aux_loss:0.0333 train_time:317823ms step_avg:133.48ms
step:2392/6250 train_loss:2.3616 aux_loss:0.0334 train_time:317955ms step_avg:133.48ms
step:2393/6250 train_loss:2.2671 aux_loss:0.0332 train_time:318089ms step_avg:133.48ms
step:2394/6250 train_loss:2.3327 aux_loss:0.0333 train_time:318222ms step_avg:133.48ms
step:2395/6250 train_loss:2.3861 aux_loss:0.0333 train_time:318353ms step_avg:133.48ms
step:2396/6250 train_loss:2.4284 aux_loss:0.0332 train_time:318485ms step_avg:133.48ms
step:2397/6250 train_loss:2.2687 aux_loss:0.0330 train_time:318617ms step_avg:133.48ms
step:2398/6250 train_loss:2.3681 aux_loss:0.0333 train_time:318749ms step_avg:133.48ms
step:2399/6250 train_loss:2.4562 aux_loss:0.0333 train_time:318881ms step_avg:133.48ms
step:2400/6250 train_loss:2.3676 aux_loss:0.0334 train_time:319013ms step_avg:133.48ms
step:2401/6250 train_loss:2.3681 aux_loss:0.0335 train_time:319144ms step_avg:133.48ms
step:2402/6250 train_loss:2.3375 aux_loss:0.0330 train_time:319275ms step_avg:133.48ms
step:2403/6250 train_loss:2.3464 aux_loss:0.0331 train_time:319406ms step_avg:133.48ms
step:2404/6250 train_loss:2.3496 aux_loss:0.0332 train_time:319538ms step_avg:133.47ms
step:2405/6250 train_loss:2.3291 aux_loss:0.0332 train_time:319670ms step_avg:133.47ms
step:2406/6250 train_loss:2.3630 aux_loss:0.0331 train_time:319802ms step_avg:133.47ms
step:2407/6250 train_loss:2.3237 aux_loss:0.0329 train_time:319934ms step_avg:133.47ms
step:2408/6250 train_loss:2.3913 aux_loss:0.0332 train_time:320065ms step_avg:133.47ms
step:2409/6250 train_loss:2.2653 aux_loss:0.0330 train_time:320197ms step_avg:133.47ms
step:2410/6250 train_loss:2.4246 aux_loss:0.0331 train_time:320329ms step_avg:133.47ms
step:2411/6250 train_loss:2.3670 aux_loss:0.0329 train_time:320461ms step_avg:133.47ms
step:2412/6250 train_loss:2.3760 aux_loss:0.0331 train_time:320592ms step_avg:133.47ms
step:2413/6250 train_loss:2.3751 aux_loss:0.0332 train_time:320726ms step_avg:133.47ms
step:2414/6250 train_loss:2.2234 aux_loss:0.0331 train_time:320862ms step_avg:133.47ms
step:2415/6250 train_loss:2.3232 aux_loss:0.0331 train_time:321004ms step_avg:133.47ms
step:2416/6250 train_loss:2.3074 aux_loss:0.0329 train_time:321142ms step_avg:133.48ms
step:2417/6250 train_loss:2.4468 aux_loss:0.0334 train_time:321277ms step_avg:133.48ms
step:2418/6250 train_loss:2.3121 aux_loss:0.0332 train_time:321416ms step_avg:133.48ms
step:2419/6250 train_loss:2.3245 aux_loss:0.0333 train_time:321556ms step_avg:133.48ms
step:2420/6250 train_loss:2.1951 aux_loss:0.0331 train_time:321687ms step_avg:133.48ms
step:2421/6250 train_loss:2.2965 aux_loss:0.0332 train_time:321819ms step_avg:133.48ms
step:2422/6250 train_loss:2.3112 aux_loss:0.0331 train_time:321952ms step_avg:133.48ms
step:2423/6250 train_loss:2.2855 aux_loss:0.0331 train_time:322095ms step_avg:133.48ms
step:2424/6250 train_loss:2.3284 aux_loss:0.0327 train_time:322238ms step_avg:133.49ms
step:2425/6250 train_loss:2.3411 aux_loss:0.0329 train_time:322378ms step_avg:133.49ms
step:2426/6250 train_loss:2.3537 aux_loss:0.0327 train_time:322520ms step_avg:133.49ms
step:2427/6250 train_loss:2.3020 aux_loss:0.0329 train_time:322664ms step_avg:133.50ms
step:2428/6250 train_loss:2.3690 aux_loss:0.0331 train_time:322807ms step_avg:133.50ms
step:2429/6250 train_loss:2.2517 aux_loss:0.0331 train_time:322948ms step_avg:133.50ms
step:2430/6250 train_loss:2.2864 aux_loss:0.0329 train_time:323082ms step_avg:133.50ms
step:2431/6250 train_loss:2.3345 aux_loss:0.0325 train_time:323219ms step_avg:133.51ms
step:2432/6250 train_loss:2.2845 aux_loss:0.0327 train_time:323352ms step_avg:133.51ms
step:2433/6250 train_loss:2.3249 aux_loss:0.0334 train_time:323485ms step_avg:133.51ms
step:2434/6250 train_loss:2.2719 aux_loss:0.0328 train_time:323624ms step_avg:133.51ms
step:2435/6250 train_loss:2.4104 aux_loss:0.0330 train_time:323768ms step_avg:133.51ms
step:2436/6250 train_loss:2.3349 aux_loss:0.0332 train_time:323911ms step_avg:133.52ms
step:2437/6250 train_loss:2.2872 aux_loss:0.0333 train_time:324044ms step_avg:133.52ms
step:2438/6250 train_loss:2.3491 aux_loss:0.0330 train_time:324178ms step_avg:133.52ms
step:2439/6250 train_loss:2.3261 aux_loss:0.0328 train_time:324320ms step_avg:133.52ms
step:2440/6250 train_loss:2.2493 aux_loss:0.0328 train_time:324464ms step_avg:133.52ms
step:2441/6250 train_loss:2.2600 aux_loss:0.0326 train_time:324604ms step_avg:133.53ms
step:2442/6250 train_loss:2.2530 aux_loss:0.0325 train_time:324741ms step_avg:133.53ms
step:2443/6250 train_loss:2.3257 aux_loss:0.0328 train_time:324886ms step_avg:133.53ms
step:2444/6250 train_loss:2.4141 aux_loss:0.0330 train_time:325031ms step_avg:133.54ms
step:2445/6250 train_loss:2.2787 aux_loss:0.0329 train_time:325171ms step_avg:133.54ms
step:2446/6250 train_loss:2.4054 aux_loss:0.0329 train_time:325304ms step_avg:133.54ms
step:2447/6250 train_loss:2.3088 aux_loss:0.0330 train_time:325437ms step_avg:133.54ms
step:2448/6250 train_loss:2.3418 aux_loss:0.0329 train_time:325569ms step_avg:133.54ms
step:2449/6250 train_loss:2.3271 aux_loss:0.0325 train_time:325712ms step_avg:133.54ms
step:2450/6250 train_loss:2.2732 aux_loss:0.0329 train_time:325852ms step_avg:133.55ms
step:2451/6250 train_loss:2.2832 aux_loss:0.0328 train_time:325986ms step_avg:133.55ms
step:2452/6250 train_loss:2.2447 aux_loss:0.0328 train_time:326119ms step_avg:133.55ms
step:2453/6250 train_loss:2.2280 aux_loss:0.0328 train_time:326255ms step_avg:133.55ms
step:2454/6250 train_loss:2.3747 aux_loss:0.0328 train_time:326396ms step_avg:133.55ms
step:2455/6250 train_loss:2.3242 aux_loss:0.0328 train_time:326535ms step_avg:133.55ms
step:2456/6250 train_loss:2.2969 aux_loss:0.0327 train_time:326673ms step_avg:133.55ms
step:2457/6250 train_loss:2.2689 aux_loss:0.0329 train_time:326810ms step_avg:133.56ms
step:2458/6250 train_loss:2.3065 aux_loss:0.0327 train_time:326945ms step_avg:133.56ms
step:2459/6250 train_loss:2.2333 aux_loss:0.0328 train_time:327083ms step_avg:133.56ms
step:2460/6250 train_loss:2.3300 aux_loss:0.0329 train_time:327221ms step_avg:133.56ms
step:2461/6250 train_loss:2.2882 aux_loss:0.0327 train_time:327354ms step_avg:133.56ms
step:2462/6250 train_loss:2.2944 aux_loss:0.0327 train_time:327487ms step_avg:133.56ms
step:2463/6250 train_loss:2.3403 aux_loss:0.0325 train_time:327618ms step_avg:133.56ms
step:2464/6250 train_loss:2.3127 aux_loss:0.0325 train_time:327749ms step_avg:133.56ms
step:2465/6250 train_loss:2.2848 aux_loss:0.0324 train_time:327881ms step_avg:133.56ms
step:2466/6250 train_loss:2.3016 aux_loss:0.0325 train_time:328012ms step_avg:133.56ms
step:2467/6250 train_loss:2.2568 aux_loss:0.0325 train_time:328144ms step_avg:133.55ms
step:2468/6250 train_loss:2.3181 aux_loss:0.0323 train_time:328275ms step_avg:133.55ms
step:2469/6250 train_loss:2.2896 aux_loss:0.0323 train_time:328407ms step_avg:133.55ms
step:2470/6250 train_loss:2.2606 aux_loss:0.0324 train_time:328539ms step_avg:133.55ms
step:2471/6250 train_loss:2.3360 aux_loss:0.0324 train_time:328670ms step_avg:133.55ms
step:2472/6250 train_loss:2.3692 aux_loss:0.0324 train_time:328803ms step_avg:133.55ms
step:2473/6250 train_loss:2.3366 aux_loss:0.0327 train_time:328934ms step_avg:133.55ms
step:2474/6250 train_loss:2.3651 aux_loss:0.0324 train_time:329067ms step_avg:133.55ms
step:2475/6250 train_loss:2.2852 aux_loss:0.0326 train_time:329198ms step_avg:133.55ms
step:2476/6250 train_loss:2.3001 aux_loss:0.0328 train_time:329329ms step_avg:133.55ms
step:2477/6250 train_loss:2.2969 aux_loss:0.0329 train_time:329462ms step_avg:133.55ms
step:2478/6250 train_loss:2.3580 aux_loss:0.0324 train_time:329592ms step_avg:133.55ms
step:2479/6250 train_loss:2.3035 aux_loss:0.0326 train_time:329724ms step_avg:133.55ms
step:2480/6250 train_loss:2.3481 aux_loss:0.0326 train_time:329856ms step_avg:133.54ms
step:2481/6250 train_loss:2.3563 aux_loss:0.0325 train_time:329989ms step_avg:133.54ms
step:2482/6250 train_loss:2.3629 aux_loss:0.0324 train_time:330120ms step_avg:133.54ms
step:2483/6250 train_loss:2.2500 aux_loss:0.0324 train_time:330251ms step_avg:133.54ms
step:2484/6250 train_loss:2.3586 aux_loss:0.0326 train_time:330383ms step_avg:133.54ms
step:2485/6250 train_loss:2.2941 aux_loss:0.0325 train_time:330516ms step_avg:133.54ms
step:2486/6250 train_loss:2.2707 aux_loss:0.0325 train_time:330647ms step_avg:133.54ms
step:2487/6250 train_loss:2.2277 aux_loss:0.0324 train_time:330779ms step_avg:133.54ms
step:2488/6250 train_loss:2.2990 aux_loss:0.0328 train_time:330910ms step_avg:133.54ms
step:2489/6250 train_loss:2.2586 aux_loss:0.0329 train_time:331042ms step_avg:133.54ms
step:2490/6250 train_loss:2.3092 aux_loss:0.0326 train_time:331173ms step_avg:133.54ms
step:2491/6250 train_loss:2.3350 aux_loss:0.0325 train_time:331305ms step_avg:133.54ms
step:2492/6250 train_loss:2.2851 aux_loss:0.0326 train_time:331437ms step_avg:133.54ms
step:2493/6250 train_loss:2.4170 aux_loss:0.0326 train_time:331568ms step_avg:133.54ms
step:2494/6250 train_loss:2.2073 aux_loss:0.0325 train_time:331700ms step_avg:133.53ms
step:2495/6250 train_loss:2.2916 aux_loss:0.0328 train_time:331831ms step_avg:133.53ms
step:2496/6250 train_loss:2.2659 aux_loss:0.0325 train_time:331964ms step_avg:133.53ms
step:2497/6250 train_loss:2.2360 aux_loss:0.0322 train_time:332096ms step_avg:133.53ms
step:2498/6250 train_loss:2.2931 aux_loss:0.0326 train_time:332228ms step_avg:133.53ms
step:2499/6250 train_loss:2.2894 aux_loss:0.0323 train_time:332359ms step_avg:133.53ms
step:2500/6250 train_loss:2.3434 aux_loss:0.0321 train_time:332491ms step_avg:133.53ms
step:2501/6250 train_loss:2.3747 aux_loss:0.0325 train_time:332622ms step_avg:133.53ms
step:2502/6250 train_loss:2.2212 aux_loss:0.0323 train_time:332753ms step_avg:133.53ms
step:2503/6250 train_loss:2.2660 aux_loss:0.0324 train_time:332887ms step_avg:133.53ms
step:2504/6250 train_loss:2.3315 aux_loss:0.0324 train_time:333020ms step_avg:133.53ms
step:2505/6250 train_loss:2.3027 aux_loss:0.0321 train_time:333159ms step_avg:133.53ms
step:2506/6250 train_loss:2.2858 aux_loss:0.0323 train_time:333297ms step_avg:133.53ms
step:2507/6250 train_loss:2.2211 aux_loss:0.0321 train_time:333435ms step_avg:133.53ms
step:2508/6250 train_loss:2.2190 aux_loss:0.0320 train_time:333566ms step_avg:133.53ms
step:2509/6250 train_loss:2.3005 aux_loss:0.0320 train_time:333698ms step_avg:133.53ms
step:2510/6250 train_loss:2.3228 aux_loss:0.0323 train_time:333837ms step_avg:133.53ms
step:2511/6250 train_loss:2.2945 aux_loss:0.0323 train_time:333972ms step_avg:133.54ms
step:2512/6250 train_loss:2.4038 aux_loss:0.0321 train_time:334105ms step_avg:133.54ms
step:2513/6250 train_loss:2.2803 aux_loss:0.0323 train_time:334237ms step_avg:133.53ms
step:2514/6250 train_loss:2.3591 aux_loss:0.0324 train_time:334370ms step_avg:133.53ms
step:2515/6250 train_loss:2.2965 aux_loss:0.0323 train_time:334503ms step_avg:133.53ms
step:2516/6250 train_loss:2.2946 aux_loss:0.0319 train_time:334646ms step_avg:133.54ms
step:2517/6250 train_loss:2.2855 aux_loss:0.0325 train_time:334794ms step_avg:133.54ms
step:2518/6250 train_loss:2.2674 aux_loss:0.0323 train_time:334940ms step_avg:133.55ms
step:2519/6250 train_loss:2.2816 aux_loss:0.0322 train_time:335083ms step_avg:133.55ms
step:2520/6250 train_loss:2.2795 aux_loss:0.0326 train_time:335226ms step_avg:133.56ms
step:2521/6250 train_loss:2.2289 aux_loss:0.0329 train_time:335367ms step_avg:133.56ms
step:2522/6250 train_loss:2.3386 aux_loss:0.0323 train_time:335500ms step_avg:133.56ms
step:2523/6250 train_loss:2.3713 aux_loss:0.0323 train_time:335632ms step_avg:133.56ms
step:2524/6250 train_loss:2.3105 aux_loss:0.0325 train_time:335765ms step_avg:133.56ms
step:2525/6250 train_loss:2.3006 aux_loss:0.0324 train_time:335897ms step_avg:133.56ms
step:2526/6250 train_loss:2.2880 aux_loss:0.0318 train_time:336029ms step_avg:133.56ms
step:2527/6250 train_loss:2.3291 aux_loss:0.0318 train_time:336160ms step_avg:133.56ms
step:2528/6250 train_loss:2.3353 aux_loss:0.0322 train_time:336290ms step_avg:133.55ms
step:2529/6250 train_loss:2.2486 aux_loss:0.0324 train_time:336422ms step_avg:133.55ms
step:2530/6250 train_loss:2.4317 aux_loss:0.0321 train_time:336554ms step_avg:133.55ms
step:2531/6250 train_loss:2.2892 aux_loss:0.0321 train_time:336686ms step_avg:133.55ms
step:2532/6250 train_loss:2.3447 aux_loss:0.0325 train_time:336818ms step_avg:133.55ms
step:2533/6250 train_loss:2.3879 aux_loss:0.0325 train_time:336953ms step_avg:133.55ms
step:2534/6250 train_loss:2.2447 aux_loss:0.0319 train_time:337092ms step_avg:133.55ms
step:2535/6250 train_loss:2.4333 aux_loss:0.0321 train_time:337231ms step_avg:133.56ms
step:2536/6250 train_loss:2.3723 aux_loss:0.0324 train_time:337366ms step_avg:133.56ms
step:2537/6250 train_loss:2.2508 aux_loss:0.0321 train_time:337498ms step_avg:133.56ms
step:2538/6250 train_loss:2.2823 aux_loss:0.0321 train_time:337629ms step_avg:133.56ms
step:2539/6250 train_loss:2.2777 aux_loss:0.0320 train_time:337763ms step_avg:133.56ms
step:2540/6250 train_loss:2.3151 aux_loss:0.0319 train_time:337894ms step_avg:133.55ms
step:2541/6250 train_loss:2.2398 aux_loss:0.0324 train_time:338027ms step_avg:133.55ms
step:2542/6250 train_loss:2.2518 aux_loss:0.0326 train_time:338158ms step_avg:133.55ms
step:2543/6250 train_loss:2.2579 aux_loss:0.0324 train_time:338295ms step_avg:133.55ms
step:2544/6250 train_loss:2.2463 aux_loss:0.0323 train_time:338435ms step_avg:133.56ms
step:2545/6250 train_loss:2.2633 aux_loss:0.0325 train_time:338575ms step_avg:133.56ms
step:2546/6250 train_loss:2.3582 aux_loss:0.0324 train_time:338712ms step_avg:133.56ms
step:2547/6250 train_loss:2.2092 aux_loss:0.0320 train_time:338849ms step_avg:133.56ms
step:2548/6250 train_loss:2.2955 aux_loss:0.0321 train_time:338986ms step_avg:133.56ms
step:2549/6250 train_loss:2.3075 aux_loss:0.0322 train_time:339120ms step_avg:133.56ms
step:2550/6250 train_loss:2.3161 aux_loss:0.0320 train_time:339256ms step_avg:133.57ms
step:2551/6250 train_loss:2.3768 aux_loss:0.0319 train_time:339393ms step_avg:133.57ms
step:2552/6250 train_loss:2.3306 aux_loss:0.0318 train_time:339529ms step_avg:133.57ms
step:2553/6250 train_loss:2.2758 aux_loss:0.0323 train_time:339665ms step_avg:133.57ms
step:2554/6250 train_loss:2.2845 aux_loss:0.0316 train_time:339804ms step_avg:133.57ms
step:2555/6250 train_loss:2.2829 aux_loss:0.0321 train_time:339942ms step_avg:133.57ms
step:2556/6250 train_loss:2.3090 aux_loss:0.0321 train_time:340079ms step_avg:133.57ms
step:2557/6250 train_loss:2.3142 aux_loss:0.0318 train_time:340215ms step_avg:133.57ms
step:2558/6250 train_loss:2.3271 aux_loss:0.0319 train_time:340346ms step_avg:133.57ms
step:2559/6250 train_loss:2.2471 aux_loss:0.0322 train_time:340479ms step_avg:133.57ms
step:2560/6250 train_loss:2.2720 aux_loss:0.0323 train_time:340616ms step_avg:133.57ms
step:2561/6250 train_loss:2.2875 aux_loss:0.0320 train_time:340752ms step_avg:133.58ms
step:2562/6250 train_loss:2.2662 aux_loss:0.0320 train_time:340884ms step_avg:133.58ms
step:2563/6250 train_loss:2.2311 aux_loss:0.0319 train_time:341016ms step_avg:133.57ms
step:2564/6250 train_loss:2.2224 aux_loss:0.0314 train_time:341154ms step_avg:133.58ms
step:2565/6250 train_loss:2.3651 aux_loss:0.0321 train_time:341291ms step_avg:133.58ms
step:2566/6250 train_loss:2.3273 aux_loss:0.0320 train_time:341423ms step_avg:133.58ms
step:2567/6250 train_loss:2.3061 aux_loss:0.0317 train_time:341558ms step_avg:133.58ms
step:2568/6250 train_loss:2.2142 aux_loss:0.0316 train_time:341690ms step_avg:133.58ms
step:2569/6250 train_loss:2.2994 aux_loss:0.0321 train_time:341823ms step_avg:133.58ms
step:2570/6250 train_loss:2.3660 aux_loss:0.0323 train_time:341955ms step_avg:133.58ms
step:2571/6250 train_loss:2.3547 aux_loss:0.0322 train_time:342088ms step_avg:133.58ms
step:2572/6250 train_loss:2.2086 aux_loss:0.0321 train_time:342220ms step_avg:133.58ms
step:2573/6250 train_loss:2.3333 aux_loss:0.0323 train_time:342351ms step_avg:133.57ms
step:2574/6250 train_loss:2.2549 aux_loss:0.0323 train_time:342484ms step_avg:133.57ms
step:2575/6250 train_loss:2.2865 aux_loss:0.0324 train_time:342615ms step_avg:133.57ms
step:2576/6250 train_loss:2.3543 aux_loss:0.0322 train_time:342747ms step_avg:133.57ms
step:2577/6250 train_loss:2.3043 aux_loss:0.0325 train_time:342886ms step_avg:133.57ms
step:2578/6250 train_loss:2.3715 aux_loss:0.0318 train_time:343023ms step_avg:133.58ms
step:2579/6250 train_loss:2.2892 aux_loss:0.0319 train_time:343161ms step_avg:133.58ms
step:2580/6250 train_loss:2.2932 aux_loss:0.0317 train_time:343299ms step_avg:133.58ms
step:2581/6250 train_loss:2.2977 aux_loss:0.0318 train_time:343433ms step_avg:133.58ms
step:2582/6250 train_loss:2.2514 aux_loss:0.0316 train_time:343565ms step_avg:133.58ms
step:2583/6250 train_loss:2.3127 aux_loss:0.0318 train_time:343700ms step_avg:133.58ms
step:2584/6250 train_loss:2.3073 aux_loss:0.0317 train_time:343837ms step_avg:133.58ms
step:2585/6250 train_loss:2.3140 aux_loss:0.0315 train_time:343975ms step_avg:133.58ms
step:2586/6250 train_loss:2.2646 aux_loss:0.0317 train_time:344111ms step_avg:133.58ms
step:2587/6250 train_loss:2.3115 aux_loss:0.0317 train_time:344249ms step_avg:133.59ms
step:2588/6250 train_loss:2.2690 aux_loss:0.0315 train_time:344387ms step_avg:133.59ms
step:2589/6250 train_loss:2.3868 aux_loss:0.0318 train_time:344522ms step_avg:133.59ms
step:2590/6250 train_loss:2.2362 aux_loss:0.0318 train_time:344659ms step_avg:133.59ms
step:2591/6250 train_loss:2.2943 aux_loss:0.0319 train_time:344797ms step_avg:133.59ms
step:2592/6250 train_loss:2.3303 aux_loss:0.0318 train_time:344929ms step_avg:133.59ms
step:2593/6250 train_loss:2.2806 aux_loss:0.0316 train_time:345062ms step_avg:133.59ms
step:2594/6250 train_loss:2.2971 aux_loss:0.0319 train_time:345195ms step_avg:133.59ms
step:2595/6250 train_loss:2.2957 aux_loss:0.0316 train_time:345333ms step_avg:133.59ms
step:2596/6250 train_loss:2.3092 aux_loss:0.0316 train_time:345470ms step_avg:133.59ms
step:2597/6250 train_loss:2.4148 aux_loss:0.0320 train_time:345609ms step_avg:133.59ms
step:2598/6250 train_loss:2.2559 aux_loss:0.0318 train_time:345746ms step_avg:133.60ms
step:2599/6250 train_loss:2.2407 aux_loss:0.0312 train_time:345884ms step_avg:133.60ms
step:2600/6250 train_loss:2.2061 aux_loss:0.0314 train_time:346016ms step_avg:133.60ms
step:2601/6250 train_loss:2.2739 aux_loss:0.0317 train_time:346153ms step_avg:133.60ms
step:2602/6250 train_loss:2.2433 aux_loss:0.0319 train_time:346289ms step_avg:133.60ms
step:2603/6250 train_loss:2.3229 aux_loss:0.0315 train_time:346421ms step_avg:133.60ms
step:2604/6250 train_loss:2.2772 aux_loss:0.0316 train_time:346553ms step_avg:133.60ms
step:2605/6250 train_loss:2.3599 aux_loss:0.0317 train_time:346686ms step_avg:133.60ms
step:2606/6250 train_loss:2.3115 aux_loss:0.0315 train_time:346818ms step_avg:133.60ms
step:2607/6250 train_loss:2.4304 aux_loss:0.0320 train_time:346952ms step_avg:133.60ms
step:2608/6250 train_loss:2.2482 aux_loss:0.0316 train_time:347085ms step_avg:133.60ms
step:2609/6250 train_loss:2.3002 aux_loss:0.0314 train_time:347217ms step_avg:133.60ms
step:2610/6250 train_loss:2.3433 aux_loss:0.0317 train_time:347351ms step_avg:133.60ms
step:2611/6250 train_loss:2.3048 aux_loss:0.0316 train_time:347484ms step_avg:133.60ms
step:2612/6250 train_loss:2.2931 aux_loss:0.0319 train_time:347618ms step_avg:133.60ms
step:2613/6250 train_loss:2.2866 aux_loss:0.0316 train_time:347749ms step_avg:133.60ms
step:2614/6250 train_loss:2.3819 aux_loss:0.0319 train_time:347883ms step_avg:133.60ms
step:2615/6250 train_loss:2.2374 aux_loss:0.0313 train_time:348015ms step_avg:133.59ms
step:2616/6250 train_loss:2.3109 aux_loss:0.0313 train_time:348146ms step_avg:133.59ms
step:2617/6250 train_loss:2.3464 aux_loss:0.0317 train_time:348279ms step_avg:133.59ms
step:2618/6250 train_loss:2.3268 aux_loss:0.0320 train_time:348412ms step_avg:133.59ms
step:2619/6250 train_loss:2.2468 aux_loss:0.0315 train_time:348544ms step_avg:133.59ms
step:2620/6250 train_loss:2.2641 aux_loss:0.0315 train_time:348676ms step_avg:133.59ms
step:2621/6250 train_loss:2.2824 aux_loss:0.0316 train_time:348808ms step_avg:133.59ms
step:2622/6250 train_loss:2.3079 aux_loss:0.0317 train_time:348941ms step_avg:133.59ms
step:2623/6250 train_loss:2.1794 aux_loss:0.0312 train_time:349074ms step_avg:133.59ms
step:2624/6250 train_loss:2.2727 aux_loss:0.0315 train_time:349206ms step_avg:133.59ms
step:2625/6250 train_loss:2.3495 aux_loss:0.0316 train_time:349339ms step_avg:133.59ms
step:2626/6250 train_loss:2.2631 aux_loss:0.0317 train_time:349471ms step_avg:133.59ms
step:2627/6250 train_loss:2.2691 aux_loss:0.0312 train_time:349604ms step_avg:133.59ms
step:2628/6250 train_loss:2.2730 aux_loss:0.0313 train_time:349735ms step_avg:133.59ms
step:2629/6250 train_loss:2.3307 aux_loss:0.0316 train_time:349869ms step_avg:133.59ms
step:2630/6250 train_loss:2.3044 aux_loss:0.0317 train_time:350001ms step_avg:133.59ms
step:2631/6250 train_loss:2.2284 aux_loss:0.0315 train_time:350134ms step_avg:133.59ms
step:2632/6250 train_loss:2.2798 aux_loss:0.0315 train_time:350265ms step_avg:133.59ms
step:2633/6250 train_loss:2.2422 aux_loss:0.0316 train_time:350398ms step_avg:133.59ms
step:2634/6250 train_loss:2.3485 aux_loss:0.0314 train_time:350529ms step_avg:133.59ms
step:2635/6250 train_loss:2.3539 aux_loss:0.0315 train_time:350661ms step_avg:133.59ms
step:2636/6250 train_loss:2.2569 aux_loss:0.0312 train_time:350793ms step_avg:133.58ms
step:2637/6250 train_loss:2.3034 aux_loss:0.0313 train_time:350926ms step_avg:133.58ms
step:2638/6250 train_loss:2.2618 aux_loss:0.0317 train_time:351058ms step_avg:133.58ms
step:2639/6250 train_loss:2.2947 aux_loss:0.0316 train_time:351189ms step_avg:133.58ms
step:2640/6250 train_loss:2.3818 aux_loss:0.0312 train_time:351322ms step_avg:133.58ms
step:2641/6250 train_loss:2.2710 aux_loss:0.0316 train_time:351455ms step_avg:133.58ms
step:2642/6250 train_loss:2.2723 aux_loss:0.0317 train_time:351586ms step_avg:133.58ms
step:2643/6250 train_loss:2.3545 aux_loss:0.0316 train_time:351719ms step_avg:133.58ms
step:2644/6250 train_loss:2.2034 aux_loss:0.0316 train_time:351852ms step_avg:133.58ms
step:2645/6250 train_loss:2.2405 aux_loss:0.0318 train_time:351983ms step_avg:133.58ms
step:2646/6250 train_loss:2.2068 aux_loss:0.0312 train_time:352116ms step_avg:133.58ms
step:2647/6250 train_loss:2.3973 aux_loss:0.0315 train_time:352249ms step_avg:133.58ms
step:2648/6250 train_loss:2.3563 aux_loss:0.0319 train_time:352382ms step_avg:133.58ms
step:2649/6250 train_loss:2.3125 aux_loss:0.0314 train_time:352514ms step_avg:133.58ms
step:2650/6250 train_loss:2.3355 aux_loss:0.0314 train_time:352645ms step_avg:133.58ms
step:2651/6250 train_loss:2.3764 aux_loss:0.0314 train_time:352777ms step_avg:133.58ms
step:2652/6250 train_loss:2.4035 aux_loss:0.0316 train_time:352910ms step_avg:133.58ms
step:2653/6250 train_loss:2.3028 aux_loss:0.0313 train_time:353041ms step_avg:133.58ms
step:2654/6250 train_loss:2.2862 aux_loss:0.0311 train_time:353173ms step_avg:133.58ms
step:2655/6250 train_loss:2.3005 aux_loss:0.0314 train_time:353304ms step_avg:133.57ms
step:2656/6250 train_loss:2.3623 aux_loss:0.0315 train_time:353438ms step_avg:133.57ms
step:2657/6250 train_loss:2.3411 aux_loss:0.0314 train_time:353569ms step_avg:133.57ms
step:2658/6250 train_loss:2.3254 aux_loss:0.0311 train_time:353700ms step_avg:133.57ms
step:2659/6250 train_loss:2.1757 aux_loss:0.0312 train_time:353832ms step_avg:133.57ms
step:2660/6250 train_loss:2.3310 aux_loss:0.0315 train_time:353965ms step_avg:133.57ms
step:2661/6250 train_loss:2.2326 aux_loss:0.0313 train_time:354097ms step_avg:133.57ms
step:2662/6250 train_loss:2.2796 aux_loss:0.0310 train_time:354227ms step_avg:133.57ms
step:2663/6250 train_loss:2.3327 aux_loss:0.0309 train_time:354361ms step_avg:133.57ms
step:2664/6250 train_loss:2.2506 aux_loss:0.0312 train_time:354493ms step_avg:133.57ms
step:2665/6250 train_loss:2.2209 aux_loss:0.0313 train_time:354624ms step_avg:133.57ms
step:2666/6250 train_loss:2.3185 aux_loss:0.0312 train_time:354755ms step_avg:133.57ms
step:2667/6250 train_loss:2.3767 aux_loss:0.0316 train_time:354887ms step_avg:133.57ms
step:2668/6250 train_loss:2.3058 aux_loss:0.0311 train_time:355020ms step_avg:133.57ms
step:2669/6250 train_loss:2.2540 aux_loss:0.0309 train_time:355151ms step_avg:133.57ms
step:2670/6250 train_loss:2.3238 aux_loss:0.0315 train_time:355283ms step_avg:133.56ms
step:2671/6250 train_loss:2.2642 aux_loss:0.0314 train_time:355415ms step_avg:133.56ms
step:2672/6250 train_loss:2.2722 aux_loss:0.0314 train_time:355546ms step_avg:133.56ms
step:2673/6250 train_loss:2.2577 aux_loss:0.0313 train_time:355678ms step_avg:133.56ms
step:2674/6250 train_loss:2.4483 aux_loss:0.0316 train_time:355810ms step_avg:133.56ms
step:2675/6250 train_loss:2.2720 aux_loss:0.0314 train_time:355943ms step_avg:133.56ms
step:2676/6250 train_loss:2.1873 aux_loss:0.0314 train_time:356075ms step_avg:133.56ms
step:2677/6250 train_loss:2.3537 aux_loss:0.0314 train_time:356207ms step_avg:133.56ms
step:2678/6250 train_loss:2.3035 aux_loss:0.0312 train_time:356340ms step_avg:133.56ms
step:2679/6250 train_loss:2.2778 aux_loss:0.0313 train_time:356473ms step_avg:133.56ms
step:2680/6250 train_loss:2.3344 aux_loss:0.0314 train_time:356604ms step_avg:133.56ms
step:2681/6250 train_loss:2.3405 aux_loss:0.0313 train_time:356737ms step_avg:133.56ms
step:2682/6250 train_loss:2.2446 aux_loss:0.0311 train_time:356871ms step_avg:133.56ms
step:2683/6250 train_loss:2.2542 aux_loss:0.0311 train_time:357003ms step_avg:133.56ms
step:2684/6250 train_loss:2.2156 aux_loss:0.0311 train_time:357135ms step_avg:133.56ms
step:2685/6250 train_loss:2.2819 aux_loss:0.0308 train_time:357268ms step_avg:133.56ms
step:2686/6250 train_loss:2.3096 aux_loss:0.0315 train_time:357402ms step_avg:133.56ms
step:2687/6250 train_loss:2.3457 aux_loss:0.0313 train_time:357534ms step_avg:133.56ms
step:2688/6250 train_loss:2.2312 aux_loss:0.0313 train_time:357665ms step_avg:133.56ms
step:2689/6250 train_loss:2.2748 aux_loss:0.0310 train_time:357798ms step_avg:133.56ms
step:2690/6250 train_loss:2.2510 aux_loss:0.0311 train_time:357930ms step_avg:133.56ms
step:2691/6250 train_loss:2.2838 aux_loss:0.0314 train_time:358062ms step_avg:133.56ms
step:2692/6250 train_loss:2.3437 aux_loss:0.0310 train_time:358194ms step_avg:133.55ms
step:2693/6250 train_loss:2.3232 aux_loss:0.0311 train_time:358325ms step_avg:133.55ms
step:2694/6250 train_loss:2.2486 aux_loss:0.0308 train_time:358459ms step_avg:133.55ms
step:2695/6250 train_loss:2.3176 aux_loss:0.0310 train_time:358591ms step_avg:133.55ms
step:2696/6250 train_loss:2.3487 aux_loss:0.0312 train_time:358723ms step_avg:133.55ms
step:2697/6250 train_loss:2.2703 aux_loss:0.0311 train_time:358857ms step_avg:133.55ms
step:2698/6250 train_loss:2.2402 aux_loss:0.0309 train_time:358989ms step_avg:133.55ms
step:2699/6250 train_loss:2.2685 aux_loss:0.0306 train_time:359120ms step_avg:133.55ms
step:2700/6250 train_loss:2.3400 aux_loss:0.0313 train_time:359252ms step_avg:133.55ms
step:2701/6250 train_loss:2.3165 aux_loss:0.0313 train_time:359386ms step_avg:133.55ms
step:2702/6250 train_loss:2.2567 aux_loss:0.0310 train_time:359520ms step_avg:133.55ms
step:2703/6250 train_loss:2.2694 aux_loss:0.0311 train_time:359651ms step_avg:133.55ms
step:2704/6250 train_loss:2.1756 aux_loss:0.0312 train_time:359783ms step_avg:133.55ms
step:2705/6250 train_loss:2.2418 aux_loss:0.0314 train_time:359916ms step_avg:133.55ms
step:2706/6250 train_loss:2.2468 aux_loss:0.0310 train_time:360048ms step_avg:133.55ms
step:2707/6250 train_loss:2.2448 aux_loss:0.0314 train_time:360180ms step_avg:133.55ms
step:2708/6250 train_loss:2.2880 aux_loss:0.0310 train_time:360313ms step_avg:133.55ms
step:2709/6250 train_loss:2.2040 aux_loss:0.0310 train_time:360447ms step_avg:133.55ms
step:2710/6250 train_loss:2.3138 aux_loss:0.0311 train_time:360580ms step_avg:133.55ms
step:2711/6250 train_loss:2.2799 aux_loss:0.0311 train_time:360712ms step_avg:133.55ms
step:2712/6250 train_loss:2.3417 aux_loss:0.0307 train_time:360844ms step_avg:133.55ms
step:2713/6250 train_loss:2.3776 aux_loss:0.0310 train_time:360977ms step_avg:133.55ms
step:2714/6250 train_loss:2.3572 aux_loss:0.0310 train_time:361110ms step_avg:133.55ms
step:2715/6250 train_loss:2.2592 aux_loss:0.0312 train_time:361241ms step_avg:133.55ms
step:2716/6250 train_loss:2.1919 aux_loss:0.0306 train_time:361374ms step_avg:133.55ms
step:2717/6250 train_loss:2.4395 aux_loss:0.0307 train_time:361508ms step_avg:133.55ms
step:2718/6250 train_loss:2.2054 aux_loss:0.0309 train_time:361640ms step_avg:133.55ms
step:2719/6250 train_loss:2.3595 aux_loss:0.0311 train_time:361771ms step_avg:133.54ms
step:2720/6250 train_loss:2.3409 aux_loss:0.0309 train_time:361905ms step_avg:133.54ms
step:2721/6250 train_loss:2.3150 aux_loss:0.0312 train_time:362037ms step_avg:133.54ms
step:2722/6250 train_loss:2.2974 aux_loss:0.0310 train_time:362169ms step_avg:133.54ms
step:2723/6250 train_loss:2.2135 aux_loss:0.0310 train_time:362300ms step_avg:133.54ms
step:2724/6250 train_loss:2.2930 aux_loss:0.0311 train_time:362432ms step_avg:133.54ms
step:2725/6250 train_loss:2.2059 aux_loss:0.0308 train_time:362563ms step_avg:133.54ms
step:2726/6250 train_loss:2.2921 aux_loss:0.0309 train_time:362695ms step_avg:133.54ms
step:2727/6250 train_loss:2.2971 aux_loss:0.0309 train_time:362827ms step_avg:133.54ms
step:2728/6250 train_loss:2.2261 aux_loss:0.0311 train_time:362959ms step_avg:133.54ms
step:2729/6250 train_loss:2.4334 aux_loss:0.0314 train_time:363091ms step_avg:133.54ms
step:2730/6250 train_loss:2.2721 aux_loss:0.0312 train_time:363222ms step_avg:133.54ms
step:2731/6250 train_loss:2.3120 aux_loss:0.0308 train_time:363354ms step_avg:133.54ms
step:2732/6250 train_loss:2.2708 aux_loss:0.0310 train_time:363487ms step_avg:133.54ms
step:2733/6250 train_loss:2.3551 aux_loss:0.0309 train_time:363620ms step_avg:133.54ms
step:2734/6250 train_loss:2.3133 aux_loss:0.0312 train_time:363751ms step_avg:133.54ms
step:2735/6250 train_loss:2.2612 aux_loss:0.0311 train_time:363883ms step_avg:133.54ms
step:2736/6250 train_loss:2.3201 aux_loss:0.0309 train_time:364016ms step_avg:133.53ms
step:2737/6250 train_loss:2.2575 aux_loss:0.0312 train_time:364147ms step_avg:133.53ms
step:2738/6250 train_loss:2.2634 aux_loss:0.0308 train_time:364280ms step_avg:133.53ms
step:2739/6250 train_loss:2.2666 aux_loss:0.0306 train_time:364412ms step_avg:133.53ms
step:2740/6250 train_loss:2.3260 aux_loss:0.0308 train_time:364543ms step_avg:133.53ms
step:2741/6250 train_loss:2.2840 aux_loss:0.0311 train_time:364675ms step_avg:133.53ms
step:2742/6250 train_loss:2.2897 aux_loss:0.0313 train_time:364807ms step_avg:133.53ms
step:2743/6250 train_loss:2.2681 aux_loss:0.0308 train_time:364939ms step_avg:133.53ms
step:2744/6250 train_loss:2.3301 aux_loss:0.0310 train_time:365071ms step_avg:133.53ms
step:2745/6250 train_loss:2.2825 aux_loss:0.0309 train_time:365202ms step_avg:133.53ms
step:2746/6250 train_loss:2.2665 aux_loss:0.0308 train_time:365334ms step_avg:133.53ms
step:2747/6250 train_loss:2.3481 aux_loss:0.0304 train_time:365466ms step_avg:133.53ms
step:2748/6250 train_loss:2.2337 aux_loss:0.0308 train_time:365598ms step_avg:133.53ms
step:2749/6250 train_loss:2.2971 aux_loss:0.0311 train_time:365729ms step_avg:133.53ms
step:2750/6250 train_loss:2.3077 aux_loss:0.0307 train_time:365862ms step_avg:133.53ms
step:2751/6250 train_loss:2.3141 aux_loss:0.0307 train_time:365995ms step_avg:133.53ms
step:2752/6250 train_loss:2.2850 aux_loss:0.0304 train_time:366126ms step_avg:133.53ms
step:2753/6250 train_loss:2.2861 aux_loss:0.0308 train_time:366257ms step_avg:133.52ms
step:2754/6250 train_loss:2.2597 aux_loss:0.0306 train_time:366389ms step_avg:133.52ms
step:2755/6250 train_loss:2.2938 aux_loss:0.0303 train_time:366520ms step_avg:133.52ms
step:2756/6250 train_loss:2.2861 aux_loss:0.0303 train_time:366653ms step_avg:133.52ms
step:2757/6250 train_loss:2.2201 aux_loss:0.0306 train_time:366785ms step_avg:133.52ms
step:2758/6250 train_loss:2.1903 aux_loss:0.0306 train_time:366918ms step_avg:133.52ms
step:2759/6250 train_loss:2.1971 aux_loss:0.0306 train_time:367049ms step_avg:133.52ms
step:2760/6250 train_loss:2.2010 aux_loss:0.0307 train_time:367180ms step_avg:133.52ms
step:2761/6250 train_loss:2.2793 aux_loss:0.0305 train_time:367312ms step_avg:133.52ms
step:2762/6250 train_loss:2.3580 aux_loss:0.0306 train_time:367444ms step_avg:133.52ms
step:2763/6250 train_loss:2.2105 aux_loss:0.0311 train_time:367576ms step_avg:133.52ms
step:2764/6250 train_loss:2.2623 aux_loss:0.0310 train_time:367707ms step_avg:133.52ms
step:2765/6250 train_loss:2.2282 aux_loss:0.0307 train_time:367839ms step_avg:133.52ms
step:2766/6250 train_loss:2.2481 aux_loss:0.0307 train_time:367972ms step_avg:133.52ms
step:2767/6250 train_loss:2.2935 aux_loss:0.0309 train_time:368103ms step_avg:133.52ms
step:2768/6250 train_loss:2.2924 aux_loss:0.0306 train_time:368235ms step_avg:133.52ms
step:2769/6250 train_loss:2.2982 aux_loss:0.0307 train_time:368367ms step_avg:133.51ms
step:2770/6250 train_loss:2.4404 aux_loss:0.0312 train_time:368499ms step_avg:133.51ms
step:2771/6250 train_loss:2.3124 aux_loss:0.0310 train_time:368631ms step_avg:133.51ms
step:2772/6250 train_loss:2.3296 aux_loss:0.0311 train_time:368762ms step_avg:133.51ms
step:2773/6250 train_loss:2.2804 aux_loss:0.0310 train_time:368896ms step_avg:133.51ms
step:2774/6250 train_loss:2.2858 aux_loss:0.0309 train_time:369028ms step_avg:133.51ms
step:2775/6250 train_loss:2.3410 aux_loss:0.0309 train_time:369160ms step_avg:133.51ms
step:2776/6250 train_loss:2.2290 aux_loss:0.0310 train_time:369293ms step_avg:133.51ms
step:2777/6250 train_loss:2.4103 aux_loss:0.0309 train_time:369425ms step_avg:133.51ms
step:2778/6250 train_loss:2.2655 aux_loss:0.0306 train_time:369557ms step_avg:133.51ms
step:2779/6250 train_loss:2.2673 aux_loss:0.0309 train_time:369689ms step_avg:133.51ms
step:2780/6250 train_loss:2.3955 aux_loss:0.0307 train_time:369823ms step_avg:133.51ms
step:2781/6250 train_loss:2.2676 aux_loss:0.0303 train_time:369955ms step_avg:133.51ms
step:2782/6250 train_loss:2.3434 aux_loss:0.0303 train_time:370088ms step_avg:133.51ms
step:2783/6250 train_loss:2.2712 aux_loss:0.0304 train_time:370220ms step_avg:133.51ms
step:2784/6250 train_loss:2.1245 aux_loss:0.0303 train_time:370353ms step_avg:133.51ms
step:2785/6250 train_loss:2.1774 aux_loss:0.0304 train_time:370486ms step_avg:133.51ms
step:2786/6250 train_loss:2.3081 aux_loss:0.0308 train_time:370618ms step_avg:133.51ms
step:2787/6250 train_loss:2.3390 aux_loss:0.0306 train_time:370751ms step_avg:133.51ms
step:2788/6250 train_loss:2.3606 aux_loss:0.0308 train_time:370883ms step_avg:133.51ms
step:2789/6250 train_loss:2.3052 aux_loss:0.0304 train_time:371015ms step_avg:133.51ms
step:2790/6250 train_loss:2.2983 aux_loss:0.0302 train_time:371146ms step_avg:133.51ms
step:2791/6250 train_loss:2.3421 aux_loss:0.0306 train_time:371278ms step_avg:133.51ms
step:2792/6250 train_loss:2.2789 aux_loss:0.0305 train_time:371410ms step_avg:133.50ms
step:2793/6250 train_loss:2.2509 aux_loss:0.0303 train_time:371541ms step_avg:133.50ms
step:2794/6250 train_loss:2.1543 aux_loss:0.0303 train_time:371674ms step_avg:133.50ms
step:2795/6250 train_loss:2.3031 aux_loss:0.0309 train_time:371805ms step_avg:133.50ms
step:2796/6250 train_loss:2.2478 aux_loss:0.0304 train_time:371938ms step_avg:133.50ms
step:2797/6250 train_loss:2.3418 aux_loss:0.0301 train_time:372070ms step_avg:133.50ms
step:2798/6250 train_loss:2.2025 aux_loss:0.0302 train_time:372202ms step_avg:133.50ms
step:2799/6250 train_loss:2.2816 aux_loss:0.0306 train_time:372334ms step_avg:133.50ms
step:2800/6250 train_loss:2.3664 aux_loss:0.0303 train_time:372466ms step_avg:133.50ms
step:2801/6250 train_loss:2.2549 aux_loss:0.0304 train_time:372597ms step_avg:133.50ms
step:2802/6250 train_loss:2.2742 aux_loss:0.0305 train_time:372729ms step_avg:133.50ms
step:2803/6250 train_loss:2.3160 aux_loss:0.0307 train_time:372861ms step_avg:133.50ms
step:2804/6250 train_loss:2.1959 aux_loss:0.0304 train_time:372996ms step_avg:133.50ms
step:2805/6250 train_loss:2.2315 aux_loss:0.0302 train_time:373127ms step_avg:133.50ms
step:2806/6250 train_loss:2.3773 aux_loss:0.0304 train_time:373259ms step_avg:133.50ms
step:2807/6250 train_loss:2.3176 aux_loss:0.0303 train_time:373390ms step_avg:133.50ms
step:2808/6250 train_loss:2.2340 aux_loss:0.0298 train_time:373522ms step_avg:133.50ms
step:2809/6250 train_loss:2.1675 aux_loss:0.0305 train_time:373655ms step_avg:133.50ms
step:2810/6250 train_loss:2.2481 aux_loss:0.0310 train_time:373789ms step_avg:133.50ms
step:2811/6250 train_loss:2.2232 aux_loss:0.0308 train_time:373922ms step_avg:133.50ms
step:2812/6250 train_loss:2.2982 aux_loss:0.0305 train_time:374055ms step_avg:133.50ms
step:2813/6250 train_loss:2.4459 aux_loss:0.0306 train_time:374187ms step_avg:133.50ms
step:2814/6250 train_loss:2.3966 aux_loss:0.0304 train_time:374318ms step_avg:133.49ms
step:2815/6250 train_loss:2.3653 aux_loss:0.0302 train_time:374451ms step_avg:133.49ms
step:2816/6250 train_loss:2.2493 aux_loss:0.0304 train_time:374583ms step_avg:133.49ms
step:2817/6250 train_loss:2.2319 aux_loss:0.0301 train_time:374715ms step_avg:133.49ms
step:2818/6250 train_loss:2.3126 aux_loss:0.0301 train_time:374848ms step_avg:133.49ms
step:2819/6250 train_loss:2.3180 aux_loss:0.0301 train_time:374981ms step_avg:133.49ms
step:2820/6250 train_loss:2.2254 aux_loss:0.0300 train_time:375114ms step_avg:133.49ms
step:2821/6250 train_loss:2.2471 aux_loss:0.0299 train_time:375247ms step_avg:133.49ms
step:2822/6250 train_loss:2.3179 aux_loss:0.0304 train_time:375378ms step_avg:133.49ms
step:2823/6250 train_loss:2.4032 aux_loss:0.0305 train_time:375511ms step_avg:133.49ms
step:2824/6250 train_loss:2.2655 aux_loss:0.0304 train_time:375643ms step_avg:133.49ms
step:2825/6250 train_loss:2.3192 aux_loss:0.0303 train_time:375777ms step_avg:133.49ms
step:2826/6250 train_loss:2.3609 aux_loss:0.0301 train_time:375910ms step_avg:133.49ms
step:2827/6250 train_loss:2.2692 aux_loss:0.0304 train_time:376042ms step_avg:133.49ms
step:2828/6250 train_loss:2.2250 aux_loss:0.0295 train_time:376174ms step_avg:133.49ms
step:2829/6250 train_loss:2.3366 aux_loss:0.0308 train_time:376305ms step_avg:133.49ms
step:2830/6250 train_loss:2.2190 aux_loss:0.0304 train_time:376438ms step_avg:133.49ms
step:2831/6250 train_loss:2.3131 aux_loss:0.0305 train_time:376570ms step_avg:133.49ms
step:2832/6250 train_loss:2.2609 aux_loss:0.0304 train_time:376702ms step_avg:133.49ms
step:2833/6250 train_loss:2.2237 aux_loss:0.0303 train_time:376836ms step_avg:133.49ms
step:2834/6250 train_loss:2.2675 aux_loss:0.0303 train_time:376969ms step_avg:133.49ms
step:2835/6250 train_loss:2.3563 aux_loss:0.0302 train_time:377101ms step_avg:133.49ms
step:2836/6250 train_loss:2.3170 aux_loss:0.0302 train_time:377233ms step_avg:133.49ms
step:2837/6250 train_loss:2.3225 aux_loss:0.0304 train_time:377367ms step_avg:133.49ms
step:2838/6250 train_loss:2.2753 aux_loss:0.0301 train_time:377500ms step_avg:133.49ms
step:2839/6250 train_loss:2.1560 aux_loss:0.0296 train_time:377632ms step_avg:133.49ms
step:2840/6250 train_loss:2.3118 aux_loss:0.0300 train_time:377765ms step_avg:133.49ms
step:2841/6250 train_loss:2.2446 aux_loss:0.0300 train_time:377897ms step_avg:133.49ms
step:2842/6250 train_loss:2.3125 aux_loss:0.0297 train_time:378029ms step_avg:133.48ms
step:2843/6250 train_loss:2.3160 aux_loss:0.0303 train_time:378161ms step_avg:133.48ms
step:2844/6250 train_loss:2.2106 aux_loss:0.0301 train_time:378294ms step_avg:133.48ms
step:2845/6250 train_loss:2.2593 aux_loss:0.0299 train_time:378427ms step_avg:133.48ms
step:2846/6250 train_loss:2.2468 aux_loss:0.0301 train_time:378559ms step_avg:133.48ms
step:2847/6250 train_loss:2.2314 aux_loss:0.0302 train_time:378693ms step_avg:133.48ms
step:2848/6250 train_loss:2.3042 aux_loss:0.0302 train_time:378827ms step_avg:133.48ms
step:2849/6250 train_loss:2.2656 aux_loss:0.0301 train_time:378959ms step_avg:133.48ms
step:2850/6250 train_loss:2.2646 aux_loss:0.0301 train_time:379091ms step_avg:133.48ms
step:2851/6250 train_loss:2.3148 aux_loss:0.0305 train_time:379224ms step_avg:133.48ms
step:2852/6250 train_loss:2.3679 aux_loss:0.0303 train_time:379356ms step_avg:133.48ms
step:2853/6250 train_loss:2.2568 aux_loss:0.0298 train_time:379488ms step_avg:133.48ms
step:2854/6250 train_loss:2.3053 aux_loss:0.0301 train_time:379621ms step_avg:133.48ms
step:2855/6250 train_loss:2.3144 aux_loss:0.0304 train_time:379754ms step_avg:133.48ms
step:2856/6250 train_loss:2.2441 aux_loss:0.0300 train_time:379887ms step_avg:133.48ms
step:2857/6250 train_loss:2.2277 aux_loss:0.0298 train_time:380019ms step_avg:133.48ms
step:2858/6250 train_loss:2.2330 aux_loss:0.0303 train_time:380151ms step_avg:133.48ms
step:2859/6250 train_loss:2.2526 aux_loss:0.0300 train_time:380283ms step_avg:133.48ms
step:2860/6250 train_loss:2.2880 aux_loss:0.0297 train_time:380415ms step_avg:133.48ms
step:2861/6250 train_loss:2.2665 aux_loss:0.0297 train_time:380549ms step_avg:133.48ms
step:2862/6250 train_loss:2.2377 aux_loss:0.0298 train_time:380682ms step_avg:133.48ms
step:2863/6250 train_loss:2.2377 aux_loss:0.0297 train_time:380815ms step_avg:133.48ms
step:2864/6250 train_loss:2.3998 aux_loss:0.0303 train_time:380948ms step_avg:133.48ms
step:2865/6250 train_loss:2.2609 aux_loss:0.0306 train_time:381081ms step_avg:133.48ms
step:2866/6250 train_loss:2.3322 aux_loss:0.0304 train_time:381214ms step_avg:133.48ms
step:2867/6250 train_loss:2.2685 aux_loss:0.0299 train_time:381347ms step_avg:133.48ms
step:2868/6250 train_loss:2.2725 aux_loss:0.0300 train_time:381480ms step_avg:133.48ms
step:2869/6250 train_loss:2.2906 aux_loss:0.0295 train_time:381614ms step_avg:133.48ms
step:2870/6250 train_loss:2.2883 aux_loss:0.0300 train_time:381746ms step_avg:133.48ms
step:2871/6250 train_loss:2.3127 aux_loss:0.0301 train_time:381879ms step_avg:133.48ms
step:2872/6250 train_loss:2.2483 aux_loss:0.0304 train_time:382013ms step_avg:133.48ms
step:2873/6250 train_loss:2.3005 aux_loss:0.0302 train_time:382145ms step_avg:133.48ms
step:2874/6250 train_loss:2.2348 aux_loss:0.0301 train_time:382276ms step_avg:133.48ms
step:2875/6250 train_loss:2.2915 aux_loss:0.0302 train_time:382408ms step_avg:133.48ms
step:2876/6250 train_loss:2.2611 aux_loss:0.0298 train_time:382540ms step_avg:133.48ms
step:2877/6250 train_loss:2.2709 aux_loss:0.0299 train_time:382672ms step_avg:133.47ms
step:2878/6250 train_loss:2.3284 aux_loss:0.0301 train_time:382804ms step_avg:133.47ms
step:2879/6250 train_loss:2.3539 aux_loss:0.0299 train_time:382936ms step_avg:133.47ms
step:2880/6250 train_loss:2.2042 aux_loss:0.0300 train_time:383068ms step_avg:133.47ms
step:2881/6250 train_loss:2.2576 aux_loss:0.0298 train_time:383199ms step_avg:133.47ms
step:2882/6250 train_loss:2.2372 aux_loss:0.0296 train_time:383331ms step_avg:133.47ms
step:2883/6250 train_loss:2.2396 aux_loss:0.0300 train_time:383463ms step_avg:133.47ms
step:2884/6250 train_loss:2.1967 aux_loss:0.0299 train_time:383595ms step_avg:133.47ms
step:2885/6250 train_loss:2.3681 aux_loss:0.0297 train_time:383726ms step_avg:133.47ms
step:2886/6250 train_loss:2.2336 aux_loss:0.0295 train_time:383858ms step_avg:133.47ms
step:2887/6250 train_loss:2.3105 aux_loss:0.0302 train_time:383992ms step_avg:133.47ms
step:2888/6250 train_loss:2.2833 aux_loss:0.0302 train_time:384123ms step_avg:133.47ms
step:2889/6250 train_loss:2.3651 aux_loss:0.0300 train_time:384255ms step_avg:133.47ms
step:2890/6250 train_loss:2.2304 aux_loss:0.0301 train_time:384386ms step_avg:133.47ms
step:2891/6250 train_loss:2.2345 aux_loss:0.0297 train_time:384518ms step_avg:133.47ms
step:2892/6250 train_loss:2.2293 aux_loss:0.0297 train_time:384650ms step_avg:133.47ms
step:2893/6250 train_loss:2.2790 aux_loss:0.0298 train_time:384782ms step_avg:133.47ms
step:2894/6250 train_loss:2.3173 aux_loss:0.0296 train_time:384915ms step_avg:133.47ms
step:2895/6250 train_loss:2.3392 aux_loss:0.0298 train_time:385046ms step_avg:133.46ms
step:2896/6250 train_loss:2.1914 aux_loss:0.0297 train_time:385178ms step_avg:133.46ms
step:2897/6250 train_loss:2.3437 aux_loss:0.0301 train_time:385311ms step_avg:133.46ms
step:2898/6250 train_loss:2.3078 aux_loss:0.0305 train_time:385444ms step_avg:133.46ms
step:2899/6250 train_loss:2.2466 aux_loss:0.0299 train_time:385576ms step_avg:133.46ms
step:2900/6250 train_loss:2.3776 aux_loss:0.0296 train_time:385708ms step_avg:133.46ms
step:2901/6250 train_loss:2.2800 aux_loss:0.0297 train_time:385840ms step_avg:133.46ms
step:2902/6250 train_loss:2.3206 aux_loss:0.0297 train_time:385974ms step_avg:133.46ms
step:2903/6250 train_loss:2.2255 aux_loss:0.0292 train_time:386105ms step_avg:133.46ms
step:2904/6250 train_loss:2.2862 aux_loss:0.0297 train_time:386237ms step_avg:133.46ms
step:2905/6250 train_loss:2.2110 aux_loss:0.0299 train_time:386369ms step_avg:133.46ms
step:2906/6250 train_loss:2.2543 aux_loss:0.0298 train_time:386503ms step_avg:133.46ms
step:2907/6250 train_loss:2.2841 aux_loss:0.0295 train_time:386635ms step_avg:133.46ms
step:2908/6250 train_loss:2.2630 aux_loss:0.0295 train_time:386767ms step_avg:133.46ms
step:2909/6250 train_loss:2.3004 aux_loss:0.0299 train_time:386900ms step_avg:133.46ms
step:2910/6250 train_loss:2.2550 aux_loss:0.0299 train_time:387033ms step_avg:133.46ms
step:2911/6250 train_loss:2.2764 aux_loss:0.0297 train_time:387165ms step_avg:133.46ms
step:2912/6250 train_loss:2.3080 aux_loss:0.0297 train_time:387297ms step_avg:133.46ms
step:2913/6250 train_loss:2.2907 aux_loss:0.0297 train_time:387431ms step_avg:133.46ms
step:2914/6250 train_loss:2.2824 aux_loss:0.0298 train_time:387563ms step_avg:133.46ms
step:2915/6250 train_loss:2.3802 aux_loss:0.0298 train_time:387695ms step_avg:133.46ms
step:2916/6250 train_loss:2.2347 aux_loss:0.0297 train_time:387827ms step_avg:133.46ms
step:2917/6250 train_loss:2.1506 aux_loss:0.0299 train_time:387961ms step_avg:133.46ms
step:2918/6250 train_loss:2.2129 aux_loss:0.0296 train_time:388093ms step_avg:133.46ms
step:2919/6250 train_loss:2.1624 aux_loss:0.0298 train_time:388224ms step_avg:133.46ms
step:2920/6250 train_loss:2.2489 aux_loss:0.0300 train_time:388356ms step_avg:133.46ms
step:2921/6250 train_loss:2.2868 aux_loss:0.0297 train_time:388490ms step_avg:133.46ms
step:2922/6250 train_loss:2.2925 aux_loss:0.0297 train_time:388623ms step_avg:133.46ms
step:2923/6250 train_loss:2.1742 aux_loss:0.0293 train_time:388754ms step_avg:133.45ms
step:2924/6250 train_loss:2.3094 aux_loss:0.0294 train_time:388887ms step_avg:133.45ms
step:2925/6250 train_loss:2.2317 aux_loss:0.0297 train_time:389019ms step_avg:133.45ms
step:2926/6250 train_loss:2.2978 aux_loss:0.0300 train_time:389152ms step_avg:133.45ms
step:2927/6250 train_loss:2.2044 aux_loss:0.0294 train_time:389284ms step_avg:133.45ms
step:2928/6250 train_loss:2.2447 aux_loss:0.0292 train_time:389417ms step_avg:133.45ms
step:2929/6250 train_loss:2.1986 aux_loss:0.0293 train_time:389550ms step_avg:133.45ms
step:2930/6250 train_loss:2.3607 aux_loss:0.0293 train_time:389682ms step_avg:133.45ms
step:2931/6250 train_loss:2.2335 aux_loss:0.0293 train_time:389815ms step_avg:133.45ms
step:2932/6250 train_loss:2.2911 aux_loss:0.0296 train_time:389947ms step_avg:133.45ms
step:2933/6250 train_loss:2.2518 aux_loss:0.0295 train_time:390079ms step_avg:133.45ms
step:2934/6250 train_loss:2.4383 aux_loss:0.0298 train_time:390211ms step_avg:133.45ms
step:2935/6250 train_loss:2.3530 aux_loss:0.0297 train_time:390343ms step_avg:133.45ms
step:2936/6250 train_loss:2.1671 aux_loss:0.0294 train_time:390475ms step_avg:133.45ms
step:2937/6250 train_loss:2.3407 aux_loss:0.0296 train_time:390607ms step_avg:133.45ms
step:2938/6250 train_loss:2.3396 aux_loss:0.0294 train_time:390740ms step_avg:133.45ms
step:2939/6250 train_loss:2.3240 aux_loss:0.0296 train_time:390872ms step_avg:133.45ms
step:2940/6250 train_loss:2.1566 aux_loss:0.0295 train_time:391005ms step_avg:133.45ms
step:2941/6250 train_loss:2.1895 aux_loss:0.0296 train_time:391138ms step_avg:133.45ms
step:2942/6250 train_loss:2.2721 aux_loss:0.0296 train_time:391271ms step_avg:133.45ms
step:2943/6250 train_loss:2.2456 aux_loss:0.0294 train_time:391403ms step_avg:133.45ms
step:2944/6250 train_loss:2.2644 aux_loss:0.0293 train_time:391535ms step_avg:133.45ms
step:2945/6250 train_loss:2.3747 aux_loss:0.0295 train_time:391668ms step_avg:133.45ms
step:2946/6250 train_loss:2.1271 aux_loss:0.0297 train_time:391802ms step_avg:133.45ms
step:2947/6250 train_loss:2.3342 aux_loss:0.0299 train_time:391935ms step_avg:133.45ms
step:2948/6250 train_loss:2.1691 aux_loss:0.0293 train_time:392067ms step_avg:133.45ms
step:2949/6250 train_loss:2.3438 aux_loss:0.0298 train_time:392200ms step_avg:133.45ms
step:2950/6250 train_loss:2.2559 aux_loss:0.0299 train_time:392332ms step_avg:133.45ms
step:2951/6250 train_loss:2.2929 aux_loss:0.0295 train_time:392465ms step_avg:133.45ms
step:2952/6250 train_loss:2.2753 aux_loss:0.0298 train_time:392596ms step_avg:133.45ms
step:2953/6250 train_loss:2.3723 aux_loss:0.0296 train_time:392729ms step_avg:133.45ms
step:2954/6250 train_loss:2.2346 aux_loss:0.0292 train_time:392861ms step_avg:133.44ms
step:2955/6250 train_loss:2.2768 aux_loss:0.0294 train_time:392994ms step_avg:133.44ms
step:2956/6250 train_loss:2.2624 aux_loss:0.0298 train_time:393126ms step_avg:133.44ms
step:2957/6250 train_loss:2.2081 aux_loss:0.0296 train_time:393258ms step_avg:133.44ms
step:2958/6250 train_loss:2.1422 aux_loss:0.0295 train_time:393390ms step_avg:133.44ms
step:2959/6250 train_loss:2.2515 aux_loss:0.0299 train_time:393522ms step_avg:133.44ms
step:2960/6250 train_loss:2.3039 aux_loss:0.0297 train_time:393653ms step_avg:133.44ms
step:2961/6250 train_loss:2.2561 aux_loss:0.0294 train_time:393786ms step_avg:133.44ms
step:2962/6250 train_loss:2.2289 aux_loss:0.0291 train_time:393918ms step_avg:133.44ms
step:2963/6250 train_loss:2.2674 aux_loss:0.0297 train_time:394050ms step_avg:133.44ms
step:2964/6250 train_loss:2.2730 aux_loss:0.0297 train_time:394182ms step_avg:133.44ms
step:2965/6250 train_loss:2.2929 aux_loss:0.0291 train_time:394313ms step_avg:133.44ms
step:2966/6250 train_loss:2.2145 aux_loss:0.0289 train_time:394446ms step_avg:133.44ms
step:2967/6250 train_loss:2.2532 aux_loss:0.0290 train_time:394577ms step_avg:133.44ms
step:2968/6250 train_loss:2.2774 aux_loss:0.0294 train_time:394709ms step_avg:133.44ms
step:2969/6250 train_loss:2.2254 aux_loss:0.0297 train_time:394842ms step_avg:133.44ms
step:2970/6250 train_loss:2.2550 aux_loss:0.0298 train_time:394974ms step_avg:133.44ms
step:2971/6250 train_loss:2.1844 aux_loss:0.0295 train_time:395105ms step_avg:133.44ms
step:2972/6250 train_loss:2.1784 aux_loss:0.0294 train_time:395236ms step_avg:133.44ms
step:2973/6250 train_loss:2.3154 aux_loss:0.0294 train_time:395369ms step_avg:133.44ms
step:2974/6250 train_loss:2.3434 aux_loss:0.0294 train_time:395502ms step_avg:133.44ms
step:2975/6250 train_loss:2.2825 aux_loss:0.0290 train_time:395634ms step_avg:133.43ms
step:2976/6250 train_loss:2.2055 aux_loss:0.0294 train_time:395766ms step_avg:133.43ms
step:2977/6250 train_loss:2.2139 aux_loss:0.0290 train_time:395898ms step_avg:133.43ms
step:2978/6250 train_loss:2.2108 aux_loss:0.0288 train_time:396030ms step_avg:133.43ms
step:2979/6250 train_loss:2.2852 aux_loss:0.0294 train_time:396162ms step_avg:133.43ms
step:2980/6250 train_loss:2.3343 aux_loss:0.0293 train_time:396293ms step_avg:133.43ms
step:2981/6250 train_loss:2.2746 aux_loss:0.0293 train_time:396426ms step_avg:133.43ms
step:2982/6250 train_loss:2.3113 aux_loss:0.0290 train_time:396559ms step_avg:133.43ms
step:2983/6250 train_loss:2.3621 aux_loss:0.0296 train_time:396690ms step_avg:133.43ms
step:2984/6250 train_loss:2.2629 aux_loss:0.0296 train_time:396823ms step_avg:133.43ms
step:2985/6250 train_loss:2.2155 aux_loss:0.0295 train_time:396956ms step_avg:133.43ms
step:2986/6250 train_loss:2.4130 aux_loss:0.0295 train_time:397090ms step_avg:133.43ms
step:2987/6250 train_loss:2.3203 aux_loss:0.0297 train_time:397221ms step_avg:133.43ms
step:2988/6250 train_loss:2.2492 aux_loss:0.0295 train_time:397353ms step_avg:133.43ms
step:2989/6250 train_loss:2.2776 aux_loss:0.0291 train_time:397486ms step_avg:133.43ms
step:2990/6250 train_loss:2.2588 aux_loss:0.0290 train_time:397618ms step_avg:133.43ms
step:2991/6250 train_loss:2.3016 aux_loss:0.0294 train_time:397750ms step_avg:133.43ms
step:2992/6250 train_loss:2.3680 aux_loss:0.0292 train_time:397883ms step_avg:133.43ms
step:2993/6250 train_loss:2.2173 aux_loss:0.0289 train_time:398016ms step_avg:133.43ms
step:2994/6250 train_loss:2.3504 aux_loss:0.0293 train_time:398148ms step_avg:133.43ms
step:2995/6250 train_loss:2.3309 aux_loss:0.0297 train_time:398281ms step_avg:133.43ms
step:2996/6250 train_loss:2.3600 aux_loss:0.0293 train_time:398413ms step_avg:133.43ms
step:2997/6250 train_loss:2.2898 aux_loss:0.0295 train_time:398546ms step_avg:133.43ms
step:2998/6250 train_loss:2.2768 aux_loss:0.0294 train_time:398679ms step_avg:133.43ms
step:2999/6250 train_loss:2.3025 aux_loss:0.0294 train_time:398812ms step_avg:133.43ms
step:3000/6250 train_loss:2.2606 aux_loss:0.0293 train_time:398945ms step_avg:133.43ms
step:3001/6250 train_loss:2.3278 aux_loss:0.0293 train_time:399077ms step_avg:133.43ms
step:3002/6250 train_loss:2.2513 aux_loss:0.0292 train_time:399210ms step_avg:133.43ms
step:3003/6250 train_loss:2.1982 aux_loss:0.0292 train_time:399342ms step_avg:133.43ms
step:3004/6250 train_loss:2.2452 aux_loss:0.0293 train_time:399475ms step_avg:133.43ms
step:3005/6250 train_loss:2.1732 aux_loss:0.0292 train_time:399608ms step_avg:133.43ms
step:3006/6250 train_loss:2.1610 aux_loss:0.0291 train_time:399740ms step_avg:133.42ms
step:3007/6250 train_loss:2.3259 aux_loss:0.0293 train_time:399872ms step_avg:133.42ms
step:3008/6250 train_loss:2.2701 aux_loss:0.0288 train_time:400005ms step_avg:133.42ms
step:3009/6250 train_loss:2.2203 aux_loss:0.0289 train_time:400137ms step_avg:133.42ms
step:3010/6250 train_loss:2.2533 aux_loss:0.0292 train_time:400269ms step_avg:133.42ms
step:3011/6250 train_loss:2.2735 aux_loss:0.0287 train_time:400401ms step_avg:133.42ms
step:3012/6250 train_loss:2.3149 aux_loss:0.0292 train_time:400533ms step_avg:133.42ms
step:3013/6250 train_loss:2.1857 aux_loss:0.0290 train_time:400666ms step_avg:133.42ms
step:3014/6250 train_loss:2.3145 aux_loss:0.0297 train_time:400800ms step_avg:133.42ms
step:3015/6250 train_loss:2.3187 aux_loss:0.0291 train_time:400932ms step_avg:133.42ms
step:3016/6250 train_loss:2.3327 aux_loss:0.0291 train_time:401065ms step_avg:133.42ms
step:3017/6250 train_loss:2.2599 aux_loss:0.0294 train_time:401197ms step_avg:133.42ms
step:3018/6250 train_loss:2.3620 aux_loss:0.0289 train_time:401329ms step_avg:133.42ms
step:3019/6250 train_loss:2.2306 aux_loss:0.0289 train_time:401462ms step_avg:133.42ms
step:3020/6250 train_loss:2.2657 aux_loss:0.0294 train_time:401595ms step_avg:133.42ms
step:3021/6250 train_loss:2.3045 aux_loss:0.0295 train_time:401728ms step_avg:133.42ms
step:3022/6250 train_loss:2.2511 aux_loss:0.0290 train_time:401860ms step_avg:133.42ms
step:3023/6250 train_loss:2.2915 aux_loss:0.0294 train_time:401992ms step_avg:133.42ms
step:3024/6250 train_loss:2.2070 aux_loss:0.0290 train_time:402126ms step_avg:133.42ms
step:3025/6250 train_loss:2.2747 aux_loss:0.0287 train_time:402257ms step_avg:133.42ms
step:3026/6250 train_loss:2.2523 aux_loss:0.0289 train_time:402389ms step_avg:133.42ms
step:3027/6250 train_loss:2.3118 aux_loss:0.0290 train_time:402521ms step_avg:133.42ms
step:3028/6250 train_loss:2.1906 aux_loss:0.0287 train_time:402653ms step_avg:133.42ms
step:3029/6250 train_loss:2.2240 aux_loss:0.0287 train_time:402787ms step_avg:133.42ms
step:3030/6250 train_loss:2.2114 aux_loss:0.0288 train_time:402918ms step_avg:133.42ms
step:3031/6250 train_loss:2.3361 aux_loss:0.0290 train_time:403050ms step_avg:133.42ms
step:3032/6250 train_loss:2.2761 aux_loss:0.0292 train_time:403183ms step_avg:133.42ms
step:3033/6250 train_loss:2.1684 aux_loss:0.0289 train_time:403314ms step_avg:133.42ms
step:3034/6250 train_loss:2.2509 aux_loss:0.0292 train_time:403447ms step_avg:133.41ms
step:3035/6250 train_loss:2.2550 aux_loss:0.0293 train_time:403578ms step_avg:133.41ms
step:3036/6250 train_loss:2.2557 aux_loss:0.0290 train_time:403710ms step_avg:133.41ms
step:3037/6250 train_loss:2.2021 aux_loss:0.0289 train_time:403842ms step_avg:133.41ms
step:3038/6250 train_loss:2.4151 aux_loss:0.0291 train_time:403973ms step_avg:133.41ms
step:3039/6250 train_loss:2.2740 aux_loss:0.0290 train_time:404106ms step_avg:133.41ms
step:3040/6250 train_loss:2.2629 aux_loss:0.0287 train_time:404238ms step_avg:133.41ms
step:3041/6250 train_loss:2.2473 aux_loss:0.0289 train_time:404370ms step_avg:133.41ms
step:3042/6250 train_loss:2.3384 aux_loss:0.0290 train_time:404503ms step_avg:133.41ms
step:3043/6250 train_loss:2.2510 aux_loss:0.0293 train_time:404636ms step_avg:133.41ms
step:3044/6250 train_loss:2.3133 aux_loss:0.0293 train_time:404768ms step_avg:133.41ms
step:3045/6250 train_loss:2.2765 aux_loss:0.0290 train_time:404901ms step_avg:133.41ms
step:3046/6250 train_loss:2.2387 aux_loss:0.0292 train_time:405034ms step_avg:133.41ms
step:3047/6250 train_loss:2.3566 aux_loss:0.0295 train_time:405166ms step_avg:133.41ms
step:3048/6250 train_loss:2.2677 aux_loss:0.0294 train_time:405299ms step_avg:133.41ms
step:3049/6250 train_loss:2.3195 aux_loss:0.0291 train_time:405431ms step_avg:133.41ms
step:3050/6250 train_loss:2.2239 aux_loss:0.0288 train_time:405563ms step_avg:133.41ms
step:3051/6250 train_loss:2.3338 aux_loss:0.0288 train_time:405694ms step_avg:133.41ms
step:3052/6250 train_loss:2.2849 aux_loss:0.0288 train_time:405953ms step_avg:133.45ms
step:3053/6250 train_loss:2.2176 aux_loss:0.0287 train_time:406085ms step_avg:133.45ms
step:3054/6250 train_loss:2.1838 aux_loss:0.0289 train_time:406217ms step_avg:133.45ms
step:3055/6250 train_loss:2.2209 aux_loss:0.0286 train_time:406349ms step_avg:133.45ms
step:3056/6250 train_loss:2.1352 aux_loss:0.0285 train_time:406481ms step_avg:133.45ms
step:3057/6250 train_loss:2.3094 aux_loss:0.0286 train_time:406612ms step_avg:133.45ms
step:3058/6250 train_loss:2.1868 aux_loss:0.0288 train_time:406745ms step_avg:133.45ms
step:3059/6250 train_loss:2.2786 aux_loss:0.0285 train_time:406880ms step_avg:133.45ms
step:3060/6250 train_loss:2.2627 aux_loss:0.0287 train_time:407014ms step_avg:133.45ms
step:3061/6250 train_loss:2.2428 aux_loss:0.0287 train_time:407146ms step_avg:133.45ms
step:3062/6250 train_loss:2.2221 aux_loss:0.0287 train_time:407277ms step_avg:133.45ms
step:3063/6250 train_loss:2.2660 aux_loss:0.0289 train_time:407408ms step_avg:133.45ms
step:3064/6250 train_loss:2.2483 aux_loss:0.0289 train_time:407541ms step_avg:133.45ms
step:3065/6250 train_loss:2.2472 aux_loss:0.0290 train_time:407671ms step_avg:133.44ms
step:3066/6250 train_loss:2.2589 aux_loss:0.0291 train_time:407804ms step_avg:133.44ms
step:3067/6250 train_loss:2.2768 aux_loss:0.0286 train_time:407938ms step_avg:133.44ms
step:3068/6250 train_loss:2.2547 aux_loss:0.0289 train_time:408071ms step_avg:133.44ms
step:3069/6250 train_loss:2.1930 aux_loss:0.0289 train_time:408204ms step_avg:133.44ms
step:3070/6250 train_loss:2.3029 aux_loss:0.0285 train_time:408336ms step_avg:133.44ms
step:3071/6250 train_loss:2.2413 aux_loss:0.0287 train_time:408468ms step_avg:133.44ms
step:3072/6250 train_loss:2.2064 aux_loss:0.0288 train_time:408600ms step_avg:133.44ms
step:3073/6250 train_loss:2.2959 aux_loss:0.0287 train_time:408732ms step_avg:133.44ms
step:3074/6250 train_loss:2.2720 aux_loss:0.0288 train_time:408867ms step_avg:133.44ms
step:3075/6250 train_loss:2.2291 aux_loss:0.0286 train_time:409000ms step_avg:133.44ms
step:3076/6250 train_loss:2.2311 aux_loss:0.0283 train_time:409131ms step_avg:133.44ms
step:3077/6250 train_loss:2.2528 aux_loss:0.0282 train_time:409264ms step_avg:133.44ms
step:3078/6250 train_loss:2.1682 aux_loss:0.0285 train_time:409396ms step_avg:133.44ms
step:3079/6250 train_loss:2.2520 aux_loss:0.0283 train_time:409530ms step_avg:133.44ms
step:3080/6250 train_loss:2.2575 aux_loss:0.0282 train_time:409661ms step_avg:133.44ms
step:3081/6250 train_loss:2.1569 aux_loss:0.0286 train_time:409794ms step_avg:133.44ms
step:3082/6250 train_loss:2.3549 aux_loss:0.0288 train_time:409927ms step_avg:133.44ms
step:3083/6250 train_loss:2.2455 aux_loss:0.0283 train_time:410060ms step_avg:133.44ms
step:3084/6250 train_loss:2.2921 aux_loss:0.0288 train_time:410192ms step_avg:133.44ms
step:3085/6250 train_loss:2.2257 aux_loss:0.0289 train_time:410325ms step_avg:133.44ms
step:3086/6250 train_loss:2.2209 aux_loss:0.0287 train_time:410457ms step_avg:133.44ms
step:3087/6250 train_loss:2.2048 aux_loss:0.0288 train_time:410589ms step_avg:133.44ms
step:3088/6250 train_loss:2.2688 aux_loss:0.0289 train_time:410720ms step_avg:133.44ms
step:3089/6250 train_loss:2.2843 aux_loss:0.0283 train_time:410854ms step_avg:133.44ms
step:3090/6250 train_loss:2.2755 aux_loss:0.0282 train_time:410988ms step_avg:133.44ms
step:3091/6250 train_loss:2.1928 aux_loss:0.0288 train_time:411121ms step_avg:133.44ms
step:3092/6250 train_loss:2.2913 aux_loss:0.0288 train_time:411253ms step_avg:133.44ms
step:3093/6250 train_loss:2.2793 aux_loss:0.0285 train_time:411386ms step_avg:133.44ms
step:3094/6250 train_loss:2.2765 aux_loss:0.0282 train_time:411518ms step_avg:133.44ms
step:3095/6250 train_loss:2.2505 aux_loss:0.0284 train_time:411649ms step_avg:133.44ms
step:3096/6250 train_loss:2.2586 aux_loss:0.0291 train_time:411782ms step_avg:133.44ms
step:3097/6250 train_loss:2.2641 aux_loss:0.0282 train_time:411915ms step_avg:133.44ms
step:3098/6250 train_loss:2.1472 aux_loss:0.0283 train_time:412046ms step_avg:133.43ms
step:3099/6250 train_loss:2.2760 aux_loss:0.0282 train_time:412177ms step_avg:133.43ms
step:3100/6250 train_loss:2.1980 aux_loss:0.0279 train_time:412309ms step_avg:133.43ms
step:3101/6250 train_loss:2.2800 aux_loss:0.0281 train_time:412442ms step_avg:133.43ms
step:3102/6250 train_loss:2.2407 aux_loss:0.0281 train_time:412573ms step_avg:133.43ms
step:3103/6250 train_loss:2.1915 aux_loss:0.0279 train_time:412705ms step_avg:133.43ms
step:3104/6250 train_loss:2.2826 aux_loss:0.0283 train_time:412837ms step_avg:133.43ms
step:3105/6250 train_loss:2.2434 aux_loss:0.0285 train_time:412969ms step_avg:133.43ms
step:3106/6250 train_loss:2.2730 aux_loss:0.0283 train_time:413101ms step_avg:133.43ms
step:3107/6250 train_loss:2.3151 aux_loss:0.0285 train_time:413233ms step_avg:133.43ms
step:3108/6250 train_loss:2.2654 aux_loss:0.0288 train_time:413365ms step_avg:133.43ms
step:3109/6250 train_loss:2.2353 aux_loss:0.0284 train_time:413497ms step_avg:133.43ms
step:3110/6250 train_loss:2.2601 aux_loss:0.0284 train_time:413628ms step_avg:133.43ms
step:3111/6250 train_loss:2.2578 aux_loss:0.0282 train_time:413760ms step_avg:133.43ms
step:3112/6250 train_loss:2.2479 aux_loss:0.0281 train_time:413892ms step_avg:133.43ms
step:3113/6250 train_loss:2.2855 aux_loss:0.0280 train_time:414025ms step_avg:133.43ms
step:3114/6250 train_loss:2.1714 aux_loss:0.0280 train_time:414156ms step_avg:133.43ms
step:3115/6250 train_loss:2.1218 aux_loss:0.0281 train_time:414287ms step_avg:133.43ms
step:3116/6250 train_loss:2.2808 aux_loss:0.0278 train_time:414420ms step_avg:133.43ms
step:3117/6250 train_loss:2.3560 aux_loss:0.0283 train_time:414551ms step_avg:133.42ms
step:3118/6250 train_loss:2.1430 aux_loss:0.0283 train_time:414683ms step_avg:133.42ms
step:3119/6250 train_loss:2.2927 aux_loss:0.0284 train_time:414816ms step_avg:133.42ms
step:3120/6250 train_loss:2.2888 aux_loss:0.0285 train_time:414950ms step_avg:133.42ms
step:3121/6250 train_loss:2.2822 aux_loss:0.0282 train_time:415083ms step_avg:133.42ms
step:3122/6250 train_loss:2.2487 aux_loss:0.0283 train_time:415215ms step_avg:133.42ms
step:3123/6250 train_loss:2.2493 aux_loss:0.0283 train_time:415346ms step_avg:133.42ms
step:3124/6250 train_loss:2.2532 aux_loss:0.0284 train_time:415478ms step_avg:133.42ms
step:3125/6250 train_loss:2.2972 aux_loss:0.0287 train_time:415610ms step_avg:133.42ms
step:3126/6250 train_loss:2.2626 aux_loss:0.0286 train_time:415742ms step_avg:133.42ms
step:3127/6250 train_loss:2.2635 aux_loss:0.0285 train_time:415874ms step_avg:133.42ms
step:3128/6250 train_loss:2.2252 aux_loss:0.0286 train_time:416006ms step_avg:133.42ms
step:3129/6250 train_loss:2.2118 aux_loss:0.0285 train_time:416139ms step_avg:133.42ms
step:3130/6250 train_loss:2.2078 aux_loss:0.0286 train_time:416271ms step_avg:133.42ms
step:3131/6250 train_loss:2.2625 aux_loss:0.0287 train_time:416402ms step_avg:133.42ms
step:3132/6250 train_loss:2.3741 aux_loss:0.0287 train_time:416534ms step_avg:133.42ms
step:3133/6250 train_loss:2.3172 aux_loss:0.0285 train_time:416665ms step_avg:133.42ms
step:3134/6250 train_loss:2.1414 aux_loss:0.0283 train_time:416798ms step_avg:133.42ms
step:3135/6250 train_loss:2.2290 aux_loss:0.0280 train_time:416931ms step_avg:133.42ms
step:3136/6250 train_loss:2.2258 aux_loss:0.0283 train_time:417065ms step_avg:133.42ms
step:3137/6250 train_loss:2.2671 aux_loss:0.0281 train_time:417196ms step_avg:133.42ms
step:3138/6250 train_loss:2.2685 aux_loss:0.0282 train_time:417328ms step_avg:133.42ms
step:3139/6250 train_loss:2.3002 aux_loss:0.0284 train_time:417460ms step_avg:133.42ms
step:3140/6250 train_loss:2.2395 aux_loss:0.0283 train_time:417592ms step_avg:133.42ms
step:3141/6250 train_loss:2.2685 aux_loss:0.0286 train_time:417724ms step_avg:133.42ms
step:3142/6250 train_loss:2.1924 aux_loss:0.0286 train_time:417857ms step_avg:133.42ms
step:3143/6250 train_loss:2.2504 aux_loss:0.0286 train_time:417990ms step_avg:133.42ms
step:3144/6250 train_loss:2.2388 aux_loss:0.0283 train_time:418123ms step_avg:133.42ms
step:3145/6250 train_loss:2.2508 aux_loss:0.0281 train_time:418255ms step_avg:133.41ms
step:3146/6250 train_loss:2.3407 aux_loss:0.0285 train_time:418387ms step_avg:133.41ms
step:3147/6250 train_loss:2.2399 aux_loss:0.0281 train_time:418519ms step_avg:133.41ms
step:3148/6250 train_loss:2.2201 aux_loss:0.0283 train_time:418651ms step_avg:133.41ms
step:3149/6250 train_loss:2.3058 aux_loss:0.0281 train_time:418784ms step_avg:133.41ms
step:3150/6250 train_loss:2.2607 aux_loss:0.0285 train_time:418919ms step_avg:133.41ms
step:3151/6250 train_loss:2.3129 aux_loss:0.0284 train_time:419051ms step_avg:133.41ms
step:3152/6250 train_loss:2.2265 aux_loss:0.0281 train_time:419185ms step_avg:133.41ms
step:3153/6250 train_loss:2.2518 aux_loss:0.0283 train_time:419322ms step_avg:133.41ms
step:3154/6250 train_loss:2.3497 aux_loss:0.0284 train_time:419456ms step_avg:133.41ms
step:3155/6250 train_loss:2.2484 aux_loss:0.0277 train_time:419588ms step_avg:133.41ms
step:3156/6250 train_loss:2.2025 aux_loss:0.0283 train_time:419723ms step_avg:133.41ms
step:3157/6250 train_loss:2.2310 aux_loss:0.0287 train_time:419857ms step_avg:133.41ms
step:3158/6250 train_loss:2.3631 aux_loss:0.0287 train_time:419990ms step_avg:133.41ms
step:3159/6250 train_loss:2.2547 aux_loss:0.0280 train_time:420123ms step_avg:133.41ms
step:3160/6250 train_loss:2.2654 aux_loss:0.0278 train_time:420256ms step_avg:133.41ms
step:3161/6250 train_loss:2.2800 aux_loss:0.0308 train_time:420388ms step_avg:133.41ms
step:3162/6250 train_loss:2.2904 aux_loss:0.0285 train_time:420521ms step_avg:133.41ms
step:3163/6250 train_loss:2.2165 aux_loss:0.0282 train_time:420654ms step_avg:133.41ms
step:3164/6250 train_loss:2.2709 aux_loss:0.0281 train_time:420786ms step_avg:133.41ms
step:3165/6250 train_loss:2.2689 aux_loss:0.0279 train_time:420922ms step_avg:133.41ms
step:3166/6250 train_loss:2.2272 aux_loss:0.0282 train_time:421055ms step_avg:133.41ms
step:3167/6250 train_loss:2.2516 aux_loss:0.0283 train_time:421187ms step_avg:133.41ms
step:3168/6250 train_loss:2.2789 aux_loss:0.0283 train_time:421321ms step_avg:133.41ms
step:3169/6250 train_loss:2.3462 aux_loss:0.0283 train_time:421456ms step_avg:133.41ms
step:3170/6250 train_loss:2.3137 aux_loss:0.0283 train_time:421608ms step_avg:133.42ms
step:3171/6250 train_loss:2.2046 aux_loss:0.0279 train_time:421746ms step_avg:133.42ms
step:3172/6250 train_loss:2.2678 aux_loss:0.0283 train_time:421883ms step_avg:133.42ms
step:3173/6250 train_loss:2.2164 aux_loss:0.0286 train_time:422018ms step_avg:133.42ms
step:3174/6250 train_loss:2.2630 aux_loss:0.0285 train_time:422153ms step_avg:133.42ms
step:3175/6250 train_loss:2.2746 aux_loss:0.0282 train_time:422285ms step_avg:133.42ms
step:3176/6250 train_loss:2.2777 aux_loss:0.0286 train_time:422417ms step_avg:133.42ms
step:3177/6250 train_loss:2.1723 aux_loss:0.0282 train_time:422551ms step_avg:133.42ms
step:3178/6250 train_loss:2.2514 aux_loss:0.0283 train_time:422684ms step_avg:133.42ms
step:3179/6250 train_loss:2.2584 aux_loss:0.0281 train_time:422816ms step_avg:133.42ms
step:3180/6250 train_loss:2.2278 aux_loss:0.0282 train_time:422948ms step_avg:133.42ms
step:3181/6250 train_loss:2.2581 aux_loss:0.0281 train_time:423082ms step_avg:133.42ms
step:3182/6250 train_loss:2.1876 aux_loss:0.0280 train_time:423216ms step_avg:133.42ms
step:3183/6250 train_loss:2.1751 aux_loss:0.0279 train_time:423349ms step_avg:133.42ms
step:3184/6250 train_loss:2.2731 aux_loss:0.0276 train_time:423486ms step_avg:133.42ms
step:3185/6250 train_loss:2.2568 aux_loss:0.0276 train_time:423621ms step_avg:133.42ms
step:3186/6250 train_loss:2.2696 aux_loss:0.0278 train_time:423754ms step_avg:133.42ms
step:3187/6250 train_loss:2.3052 aux_loss:0.0283 train_time:423887ms step_avg:133.42ms
step:3188/6250 train_loss:2.2028 aux_loss:0.0279 train_time:424023ms step_avg:133.42ms
step:3189/6250 train_loss:2.2960 aux_loss:0.0280 train_time:424159ms step_avg:133.43ms
step:3190/6250 train_loss:2.3303 aux_loss:0.0283 train_time:424291ms step_avg:133.42ms
step:3191/6250 train_loss:2.1956 aux_loss:0.0281 train_time:424432ms step_avg:133.43ms
step:3192/6250 train_loss:2.2603 aux_loss:0.0280 train_time:424568ms step_avg:133.43ms
step:3193/6250 train_loss:2.2326 aux_loss:0.0279 train_time:424701ms step_avg:133.43ms
step:3194/6250 train_loss:2.1834 aux_loss:0.0278 train_time:424845ms step_avg:133.43ms
step:3195/6250 train_loss:2.2780 aux_loss:0.0279 train_time:424982ms step_avg:133.43ms
step:3196/6250 train_loss:2.2185 aux_loss:0.0278 train_time:425114ms step_avg:133.43ms
step:3197/6250 train_loss:2.3694 aux_loss:0.0283 train_time:425255ms step_avg:133.43ms
step:3198/6250 train_loss:2.1803 aux_loss:0.0284 train_time:425398ms step_avg:133.44ms
step:3199/6250 train_loss:2.2287 aux_loss:0.0282 train_time:425533ms step_avg:133.44ms
step:3200/6250 train_loss:2.1739 aux_loss:0.0279 train_time:425665ms step_avg:133.44ms
step:3201/6250 train_loss:2.2710 aux_loss:0.0279 train_time:425798ms step_avg:133.44ms
step:3202/6250 train_loss:2.1829 aux_loss:0.0276 train_time:425934ms step_avg:133.44ms
step:3203/6250 train_loss:2.2152 aux_loss:0.0283 train_time:426065ms step_avg:133.44ms
step:3204/6250 train_loss:2.2716 aux_loss:0.0283 train_time:426197ms step_avg:133.44ms
step:3205/6250 train_loss:2.3997 aux_loss:0.0282 train_time:426329ms step_avg:133.44ms
step:3206/6250 train_loss:2.1634 aux_loss:0.0280 train_time:426463ms step_avg:133.44ms
step:3207/6250 train_loss:2.2610 aux_loss:0.0280 train_time:426594ms step_avg:133.44ms
step:3208/6250 train_loss:2.2034 aux_loss:0.0280 train_time:426726ms step_avg:133.44ms
step:3209/6250 train_loss:2.2156 aux_loss:0.0276 train_time:426859ms step_avg:133.44ms
step:3210/6250 train_loss:2.3070 aux_loss:0.0280 train_time:426992ms step_avg:133.44ms
step:3211/6250 train_loss:2.2115 aux_loss:0.0279 train_time:427123ms step_avg:133.43ms
step:3212/6250 train_loss:2.2851 aux_loss:0.0281 train_time:427256ms step_avg:133.43ms
step:3213/6250 train_loss:2.2048 aux_loss:0.0281 train_time:427388ms step_avg:133.43ms
step:3214/6250 train_loss:2.2722 aux_loss:0.0283 train_time:427521ms step_avg:133.43ms
step:3215/6250 train_loss:2.3289 aux_loss:0.0280 train_time:427653ms step_avg:133.43ms
step:3216/6250 train_loss:2.2709 aux_loss:0.0281 train_time:427786ms step_avg:133.43ms
step:3217/6250 train_loss:2.2099 aux_loss:0.0278 train_time:427920ms step_avg:133.43ms
step:3218/6250 train_loss:2.4216 aux_loss:0.0279 train_time:428052ms step_avg:133.43ms
step:3219/6250 train_loss:2.2264 aux_loss:0.0278 train_time:428185ms step_avg:133.43ms
step:3220/6250 train_loss:2.2809 aux_loss:0.0273 train_time:428319ms step_avg:133.43ms
step:3221/6250 train_loss:2.2599 aux_loss:0.0277 train_time:428453ms step_avg:133.43ms
step:3222/6250 train_loss:2.2443 aux_loss:0.0276 train_time:428584ms step_avg:133.43ms
step:3223/6250 train_loss:2.2340 aux_loss:0.0274 train_time:428717ms step_avg:133.43ms
step:3224/6250 train_loss:2.2945 aux_loss:0.0278 train_time:428850ms step_avg:133.43ms
step:3225/6250 train_loss:2.3005 aux_loss:0.0277 train_time:428982ms step_avg:133.43ms
step:3226/6250 train_loss:2.2991 aux_loss:0.0277 train_time:429114ms step_avg:133.43ms
step:3227/6250 train_loss:2.1579 aux_loss:0.0276 train_time:429246ms step_avg:133.43ms
step:3228/6250 train_loss:2.2555 aux_loss:0.0276 train_time:429379ms step_avg:133.43ms
step:3229/6250 train_loss:2.2043 aux_loss:0.0276 train_time:429512ms step_avg:133.43ms
step:3230/6250 train_loss:2.2856 aux_loss:0.0279 train_time:429644ms step_avg:133.43ms
step:3231/6250 train_loss:2.1412 aux_loss:0.0276 train_time:429777ms step_avg:133.43ms
step:3232/6250 train_loss:2.1642 aux_loss:0.0273 train_time:429910ms step_avg:133.43ms
step:3233/6250 train_loss:2.2216 aux_loss:0.0276 train_time:430043ms step_avg:133.43ms
step:3234/6250 train_loss:2.2026 aux_loss:0.0277 train_time:430176ms step_avg:133.43ms
step:3235/6250 train_loss:2.2297 aux_loss:0.0274 train_time:430308ms step_avg:133.43ms
step:3236/6250 train_loss:2.2832 aux_loss:0.0276 train_time:430441ms step_avg:133.43ms
step:3237/6250 train_loss:2.2647 aux_loss:0.0278 train_time:430573ms step_avg:133.43ms
step:3238/6250 train_loss:2.2123 aux_loss:0.0276 train_time:430705ms step_avg:133.43ms
step:3239/6250 train_loss:2.2450 aux_loss:0.0275 train_time:430839ms step_avg:133.43ms
step:3240/6250 train_loss:2.1990 aux_loss:0.0275 train_time:430971ms step_avg:133.43ms
step:3241/6250 train_loss:2.2260 aux_loss:0.0279 train_time:431103ms step_avg:133.43ms
step:3242/6250 train_loss:2.2799 aux_loss:0.0276 train_time:431234ms step_avg:133.43ms
step:3243/6250 train_loss:2.2624 aux_loss:0.0277 train_time:431366ms step_avg:133.43ms
step:3244/6250 train_loss:2.2760 aux_loss:0.0277 train_time:431499ms step_avg:133.43ms
step:3245/6250 train_loss:2.2070 aux_loss:0.0273 train_time:431631ms step_avg:133.43ms
step:3246/6250 train_loss:2.3444 aux_loss:0.0280 train_time:431762ms step_avg:133.42ms
step:3247/6250 train_loss:2.2844 aux_loss:0.0280 train_time:431895ms step_avg:133.42ms
step:3248/6250 train_loss:2.1826 aux_loss:0.0274 train_time:432027ms step_avg:133.42ms
step:3249/6250 train_loss:2.1519 aux_loss:0.0273 train_time:432159ms step_avg:133.42ms
step:3250/6250 train_loss:2.2303 aux_loss:0.0276 train_time:432291ms step_avg:133.42ms
step:3251/6250 train_loss:2.3173 aux_loss:0.0277 train_time:432424ms step_avg:133.42ms
step:3252/6250 train_loss:2.2358 aux_loss:0.0277 train_time:432557ms step_avg:133.42ms
step:3253/6250 train_loss:2.2269 aux_loss:0.0279 train_time:432688ms step_avg:133.42ms
step:3254/6250 train_loss:2.1625 aux_loss:0.0279 train_time:432820ms step_avg:133.42ms
step:3255/6250 train_loss:2.2353 aux_loss:0.0278 train_time:432953ms step_avg:133.42ms
step:3256/6250 train_loss:2.2304 aux_loss:0.0274 train_time:433085ms step_avg:133.42ms
step:3257/6250 train_loss:2.1943 aux_loss:0.0279 train_time:433218ms step_avg:133.42ms
step:3258/6250 train_loss:2.2922 aux_loss:0.0278 train_time:433351ms step_avg:133.42ms
step:3259/6250 train_loss:2.2062 aux_loss:0.0277 train_time:433484ms step_avg:133.42ms
step:3260/6250 train_loss:2.3104 aux_loss:0.0277 train_time:433617ms step_avg:133.42ms
step:3261/6250 train_loss:2.3606 aux_loss:0.0279 train_time:433748ms step_avg:133.42ms
step:3262/6250 train_loss:2.2892 aux_loss:0.0278 train_time:433880ms step_avg:133.42ms
step:3263/6250 train_loss:2.2606 aux_loss:0.0281 train_time:434013ms step_avg:133.42ms
step:3264/6250 train_loss:2.2258 aux_loss:0.0279 train_time:434145ms step_avg:133.42ms
step:3265/6250 train_loss:2.2923 aux_loss:0.0279 train_time:434277ms step_avg:133.42ms
step:3266/6250 train_loss:2.2399 aux_loss:0.0275 train_time:434409ms step_avg:133.42ms
step:3267/6250 train_loss:2.2297 aux_loss:0.0279 train_time:434542ms step_avg:133.42ms
step:3268/6250 train_loss:2.2603 aux_loss:0.0277 train_time:434675ms step_avg:133.42ms
step:3269/6250 train_loss:2.2007 aux_loss:0.0275 train_time:434808ms step_avg:133.42ms
step:3270/6250 train_loss:2.2457 aux_loss:0.0277 train_time:434940ms step_avg:133.42ms
step:3271/6250 train_loss:2.2733 aux_loss:0.0277 train_time:435072ms step_avg:133.42ms
step:3272/6250 train_loss:2.2206 aux_loss:0.0276 train_time:435204ms step_avg:133.42ms
step:3273/6250 train_loss:2.2232 aux_loss:0.0273 train_time:435336ms step_avg:133.42ms
step:3274/6250 train_loss:2.2445 aux_loss:0.0274 train_time:435469ms step_avg:133.42ms
step:3275/6250 train_loss:2.2074 aux_loss:0.0275 train_time:435601ms step_avg:133.42ms
step:3276/6250 train_loss:2.3170 aux_loss:0.0278 train_time:435733ms step_avg:133.41ms
step:3277/6250 train_loss:2.1798 aux_loss:0.0272 train_time:435866ms step_avg:133.41ms
step:3278/6250 train_loss:2.2447 aux_loss:0.0273 train_time:435999ms step_avg:133.41ms
step:3279/6250 train_loss:2.2399 aux_loss:0.0273 train_time:436131ms step_avg:133.41ms
step:3280/6250 train_loss:2.4097 aux_loss:0.0274 train_time:436263ms step_avg:133.41ms
step:3281/6250 train_loss:2.3107 aux_loss:0.0272 train_time:436395ms step_avg:133.41ms
step:3282/6250 train_loss:2.1557 aux_loss:0.0273 train_time:436529ms step_avg:133.41ms
step:3283/6250 train_loss:2.2400 aux_loss:0.0275 train_time:436660ms step_avg:133.41ms
step:3284/6250 train_loss:2.1816 aux_loss:0.0274 train_time:436794ms step_avg:133.41ms
step:3285/6250 train_loss:2.3122 aux_loss:0.0274 train_time:436926ms step_avg:133.41ms
step:3286/6250 train_loss:2.3019 aux_loss:0.0275 train_time:437059ms step_avg:133.41ms
step:3287/6250 train_loss:2.2110 aux_loss:0.0274 train_time:437192ms step_avg:133.41ms
step:3288/6250 train_loss:2.2579 aux_loss:0.0275 train_time:437323ms step_avg:133.41ms
step:3289/6250 train_loss:2.1677 aux_loss:0.0276 train_time:437456ms step_avg:133.41ms
step:3290/6250 train_loss:2.2173 aux_loss:0.0276 train_time:437588ms step_avg:133.41ms
step:3291/6250 train_loss:2.2683 aux_loss:0.0276 train_time:437720ms step_avg:133.41ms
step:3292/6250 train_loss:2.2937 aux_loss:0.0273 train_time:437852ms step_avg:133.41ms
step:3293/6250 train_loss:2.2387 aux_loss:0.0279 train_time:437984ms step_avg:133.41ms
step:3294/6250 train_loss:2.1672 aux_loss:0.0277 train_time:438116ms step_avg:133.41ms
step:3295/6250 train_loss:2.1835 aux_loss:0.0273 train_time:438248ms step_avg:133.41ms
step:3296/6250 train_loss:2.1531 aux_loss:0.0274 train_time:438379ms step_avg:133.41ms
step:3297/6250 train_loss:2.2122 aux_loss:0.0275 train_time:438512ms step_avg:133.41ms
step:3298/6250 train_loss:2.2201 aux_loss:0.0272 train_time:438644ms step_avg:133.41ms
step:3299/6250 train_loss:2.2097 aux_loss:0.0273 train_time:438776ms step_avg:133.41ms
step:3300/6250 train_loss:2.1654 aux_loss:0.0275 train_time:438908ms step_avg:133.41ms
step:3301/6250 train_loss:2.2254 aux_loss:0.0275 train_time:439040ms step_avg:133.41ms
step:3302/6250 train_loss:2.1974 aux_loss:0.0271 train_time:439172ms step_avg:133.41ms
step:3303/6250 train_loss:2.3613 aux_loss:0.0275 train_time:439305ms step_avg:133.41ms
step:3304/6250 train_loss:2.3109 aux_loss:0.0276 train_time:439438ms step_avg:133.41ms
step:3305/6250 train_loss:2.1683 aux_loss:0.0278 train_time:439570ms step_avg:133.41ms
step:3306/6250 train_loss:2.2718 aux_loss:0.0278 train_time:439701ms step_avg:133.40ms
step:3307/6250 train_loss:2.1995 aux_loss:0.0275 train_time:439834ms step_avg:133.40ms
step:3308/6250 train_loss:2.3325 aux_loss:0.0277 train_time:439966ms step_avg:133.40ms
step:3309/6250 train_loss:2.2311 aux_loss:0.0275 train_time:440098ms step_avg:133.40ms
step:3310/6250 train_loss:2.2795 aux_loss:0.0273 train_time:440229ms step_avg:133.40ms
step:3311/6250 train_loss:2.2269 aux_loss:0.0278 train_time:440361ms step_avg:133.40ms
step:3312/6250 train_loss:2.2970 aux_loss:0.0276 train_time:440495ms step_avg:133.40ms
step:3313/6250 train_loss:2.2481 aux_loss:0.0273 train_time:440625ms step_avg:133.40ms
step:3314/6250 train_loss:2.2757 aux_loss:0.0271 train_time:440758ms step_avg:133.40ms
step:3315/6250 train_loss:2.2511 aux_loss:0.0275 train_time:440890ms step_avg:133.40ms
step:3316/6250 train_loss:2.3776 aux_loss:0.0273 train_time:441022ms step_avg:133.40ms
step:3317/6250 train_loss:2.2943 aux_loss:0.0275 train_time:441156ms step_avg:133.40ms
step:3318/6250 train_loss:2.1805 aux_loss:0.0275 train_time:441288ms step_avg:133.40ms
step:3319/6250 train_loss:2.2566 aux_loss:0.0277 train_time:441422ms step_avg:133.40ms
step:3320/6250 train_loss:2.2156 aux_loss:0.0274 train_time:441554ms step_avg:133.40ms
step:3321/6250 train_loss:2.2280 aux_loss:0.0271 train_time:441687ms step_avg:133.40ms
step:3322/6250 train_loss:2.2303 aux_loss:0.0275 train_time:441819ms step_avg:133.40ms
step:3323/6250 train_loss:2.1853 aux_loss:0.0273 train_time:441951ms step_avg:133.40ms
step:3324/6250 train_loss:2.2578 aux_loss:0.0273 train_time:442083ms step_avg:133.40ms
step:3325/6250 train_loss:2.1721 aux_loss:0.0270 train_time:442216ms step_avg:133.40ms
step:3326/6250 train_loss:2.2169 aux_loss:0.0267 train_time:442347ms step_avg:133.40ms
step:3327/6250 train_loss:2.2087 aux_loss:0.0267 train_time:442479ms step_avg:133.40ms
step:3328/6250 train_loss:2.2717 aux_loss:0.0269 train_time:442611ms step_avg:133.40ms
step:3329/6250 train_loss:2.2168 aux_loss:0.0269 train_time:442742ms step_avg:133.40ms
step:3330/6250 train_loss:2.2513 aux_loss:0.0266 train_time:442875ms step_avg:133.40ms
step:3331/6250 train_loss:2.2069 aux_loss:0.0267 train_time:443007ms step_avg:133.40ms
step:3332/6250 train_loss:2.2484 aux_loss:0.0271 train_time:443138ms step_avg:133.39ms
step:3333/6250 train_loss:2.2549 aux_loss:0.0267 train_time:443270ms step_avg:133.39ms
step:3334/6250 train_loss:2.2871 aux_loss:0.0268 train_time:443401ms step_avg:133.39ms
step:3335/6250 train_loss:2.1944 aux_loss:0.0272 train_time:443534ms step_avg:133.39ms
step:3336/6250 train_loss:2.2070 aux_loss:0.0273 train_time:443666ms step_avg:133.39ms
step:3337/6250 train_loss:2.2157 aux_loss:0.0271 train_time:443799ms step_avg:133.39ms
step:3338/6250 train_loss:2.1648 aux_loss:0.0271 train_time:443930ms step_avg:133.39ms
step:3339/6250 train_loss:2.2530 aux_loss:0.0271 train_time:444062ms step_avg:133.39ms
step:3340/6250 train_loss:2.1802 aux_loss:0.0270 train_time:444195ms step_avg:133.39ms
step:3341/6250 train_loss:2.2853 aux_loss:0.0268 train_time:444327ms step_avg:133.39ms
step:3342/6250 train_loss:2.2944 aux_loss:0.0271 train_time:444459ms step_avg:133.39ms
step:3343/6250 train_loss:2.2504 aux_loss:0.0268 train_time:444592ms step_avg:133.39ms
step:3344/6250 train_loss:2.3089 aux_loss:0.0270 train_time:444723ms step_avg:133.39ms
step:3345/6250 train_loss:2.2453 aux_loss:0.0270 train_time:444856ms step_avg:133.39ms
step:3346/6250 train_loss:2.1699 aux_loss:0.0270 train_time:444988ms step_avg:133.39ms
step:3347/6250 train_loss:2.2209 aux_loss:0.0271 train_time:445119ms step_avg:133.39ms
step:3348/6250 train_loss:2.3458 aux_loss:0.0272 train_time:445251ms step_avg:133.39ms
step:3349/6250 train_loss:2.2574 aux_loss:0.0272 train_time:445383ms step_avg:133.39ms
step:3350/6250 train_loss:2.2152 aux_loss:0.0270 train_time:445516ms step_avg:133.39ms
step:3351/6250 train_loss:2.2803 aux_loss:0.0271 train_time:445648ms step_avg:133.39ms
step:3352/6250 train_loss:2.3377 aux_loss:0.0268 train_time:445780ms step_avg:133.39ms
step:3353/6250 train_loss:2.1891 aux_loss:0.0268 train_time:445912ms step_avg:133.39ms
step:3354/6250 train_loss:2.1654 aux_loss:0.0267 train_time:446044ms step_avg:133.39ms
step:3355/6250 train_loss:2.2331 aux_loss:0.0270 train_time:446176ms step_avg:133.39ms
step:3356/6250 train_loss:2.2600 aux_loss:0.0268 train_time:446307ms step_avg:133.39ms
step:3357/6250 train_loss:2.2368 aux_loss:0.0269 train_time:446440ms step_avg:133.38ms
step:3358/6250 train_loss:2.2264 aux_loss:0.0270 train_time:446572ms step_avg:133.38ms
step:3359/6250 train_loss:2.1810 aux_loss:0.0269 train_time:446704ms step_avg:133.38ms
step:3360/6250 train_loss:2.1915 aux_loss:0.0269 train_time:446837ms step_avg:133.38ms
step:3361/6250 train_loss:2.2881 aux_loss:0.0268 train_time:446969ms step_avg:133.38ms
step:3362/6250 train_loss:2.2238 aux_loss:0.0267 train_time:447100ms step_avg:133.38ms
step:3363/6250 train_loss:2.2626 aux_loss:0.0270 train_time:447232ms step_avg:133.38ms
step:3364/6250 train_loss:2.2777 aux_loss:0.0271 train_time:447364ms step_avg:133.38ms
step:3365/6250 train_loss:2.2692 aux_loss:0.0271 train_time:447497ms step_avg:133.38ms
step:3366/6250 train_loss:2.1456 aux_loss:0.0266 train_time:447628ms step_avg:133.38ms
step:3367/6250 train_loss:2.2624 aux_loss:0.0269 train_time:447759ms step_avg:133.38ms
step:3368/6250 train_loss:2.3078 aux_loss:0.0272 train_time:447891ms step_avg:133.38ms
step:3369/6250 train_loss:2.2477 aux_loss:0.0273 train_time:448024ms step_avg:133.38ms
step:3370/6250 train_loss:2.2237 aux_loss:0.0271 train_time:448156ms step_avg:133.38ms
step:3371/6250 train_loss:2.2355 aux_loss:0.0270 train_time:448288ms step_avg:133.38ms
step:3372/6250 train_loss:2.3308 aux_loss:0.0268 train_time:448419ms step_avg:133.38ms
step:3373/6250 train_loss:2.2545 aux_loss:0.0271 train_time:448552ms step_avg:133.38ms
step:3374/6250 train_loss:2.2264 aux_loss:0.0268 train_time:448684ms step_avg:133.38ms
step:3375/6250 train_loss:2.2698 aux_loss:0.0267 train_time:448816ms step_avg:133.38ms
step:3376/6250 train_loss:2.2139 aux_loss:0.0269 train_time:448947ms step_avg:133.38ms
step:3377/6250 train_loss:2.2170 aux_loss:0.0275 train_time:449079ms step_avg:133.38ms
step:3378/6250 train_loss:2.2844 aux_loss:0.0269 train_time:449214ms step_avg:133.38ms
step:3379/6250 train_loss:2.3025 aux_loss:0.0271 train_time:449352ms step_avg:133.38ms
step:3380/6250 train_loss:2.3267 aux_loss:0.0272 train_time:449485ms step_avg:133.38ms
step:3381/6250 train_loss:2.3252 aux_loss:0.0272 train_time:449617ms step_avg:133.38ms
step:3382/6250 train_loss:2.2204 aux_loss:0.0268 train_time:449749ms step_avg:133.38ms
step:3383/6250 train_loss:2.2025 aux_loss:0.0272 train_time:449881ms step_avg:133.38ms
step:3384/6250 train_loss:2.1546 aux_loss:0.0269 train_time:450014ms step_avg:133.38ms
step:3385/6250 train_loss:2.1831 aux_loss:0.0269 train_time:450153ms step_avg:133.38ms
step:3386/6250 train_loss:2.2689 aux_loss:0.0270 train_time:450288ms step_avg:133.38ms
step:3387/6250 train_loss:2.2754 aux_loss:0.0269 train_time:450419ms step_avg:133.38ms
step:3388/6250 train_loss:2.2212 aux_loss:0.0270 train_time:450552ms step_avg:133.38ms
step:3389/6250 train_loss:2.3245 aux_loss:0.0271 train_time:450684ms step_avg:133.38ms
step:3390/6250 train_loss:2.2657 aux_loss:0.0267 train_time:450817ms step_avg:133.38ms
step:3391/6250 train_loss:2.3357 aux_loss:0.0272 train_time:450949ms step_avg:133.38ms
step:3392/6250 train_loss:2.1856 aux_loss:0.0271 train_time:451080ms step_avg:133.38ms
step:3393/6250 train_loss:2.1989 aux_loss:0.0266 train_time:451212ms step_avg:133.38ms
step:3394/6250 train_loss:2.2674 aux_loss:0.0266 train_time:451343ms step_avg:133.38ms
step:3395/6250 train_loss:2.2296 aux_loss:0.0267 train_time:451475ms step_avg:133.38ms
step:3396/6250 train_loss:2.2682 aux_loss:0.0267 train_time:451606ms step_avg:133.37ms
step:3397/6250 train_loss:2.2185 aux_loss:0.0267 train_time:451738ms step_avg:133.37ms
step:3398/6250 train_loss:2.2610 aux_loss:0.0268 train_time:451873ms step_avg:133.37ms
step:3399/6250 train_loss:2.2570 aux_loss:0.0269 train_time:452007ms step_avg:133.37ms
step:3400/6250 train_loss:2.2578 aux_loss:0.0267 train_time:452138ms step_avg:133.37ms
step:3401/6250 train_loss:2.2075 aux_loss:0.0270 train_time:452272ms step_avg:133.37ms
step:3402/6250 train_loss:2.2825 aux_loss:0.0273 train_time:452404ms step_avg:133.37ms
step:3403/6250 train_loss:2.2602 aux_loss:0.0269 train_time:452536ms step_avg:133.37ms
step:3404/6250 train_loss:2.1562 aux_loss:0.0266 train_time:452668ms step_avg:133.37ms
step:3405/6250 train_loss:2.2079 aux_loss:0.0268 train_time:452802ms step_avg:133.37ms
step:3406/6250 train_loss:2.2804 aux_loss:0.0272 train_time:452935ms step_avg:133.37ms
step:3407/6250 train_loss:2.2548 aux_loss:0.0268 train_time:453066ms step_avg:133.37ms
step:3408/6250 train_loss:2.1335 aux_loss:0.0268 train_time:453198ms step_avg:133.37ms
step:3409/6250 train_loss:2.1377 aux_loss:0.0268 train_time:453330ms step_avg:133.37ms
step:3410/6250 train_loss:2.2088 aux_loss:0.0265 train_time:453464ms step_avg:133.37ms
step:3411/6250 train_loss:2.2225 aux_loss:0.0270 train_time:453595ms step_avg:133.37ms
step:3412/6250 train_loss:2.2711 aux_loss:0.0270 train_time:453729ms step_avg:133.37ms
step:3413/6250 train_loss:2.3740 aux_loss:0.0268 train_time:453862ms step_avg:133.37ms
step:3414/6250 train_loss:2.2590 aux_loss:0.0270 train_time:453995ms step_avg:133.37ms
step:3415/6250 train_loss:2.1817 aux_loss:0.0270 train_time:454127ms step_avg:133.37ms
step:3416/6250 train_loss:2.1836 aux_loss:0.0270 train_time:454260ms step_avg:133.37ms
step:3417/6250 train_loss:2.2736 aux_loss:0.0267 train_time:454393ms step_avg:133.37ms
step:3418/6250 train_loss:2.2541 aux_loss:0.0269 train_time:454526ms step_avg:133.37ms
step:3419/6250 train_loss:2.2241 aux_loss:0.0271 train_time:454658ms step_avg:133.37ms
step:3420/6250 train_loss:2.2405 aux_loss:0.0271 train_time:454792ms step_avg:133.37ms
step:3421/6250 train_loss:2.2953 aux_loss:0.0264 train_time:454924ms step_avg:133.37ms
step:3422/6250 train_loss:2.2468 aux_loss:0.0270 train_time:455056ms step_avg:133.37ms
step:3423/6250 train_loss:2.2270 aux_loss:0.0270 train_time:455188ms step_avg:133.37ms
step:3424/6250 train_loss:2.2654 aux_loss:0.0270 train_time:455320ms step_avg:133.37ms
step:3425/6250 train_loss:2.2367 aux_loss:0.0266 train_time:455452ms step_avg:133.37ms
step:3426/6250 train_loss:2.3303 aux_loss:0.0266 train_time:455584ms step_avg:133.37ms
step:3427/6250 train_loss:2.1870 aux_loss:0.0264 train_time:455716ms step_avg:133.37ms
step:3428/6250 train_loss:2.2195 aux_loss:0.0265 train_time:455852ms step_avg:133.37ms
step:3429/6250 train_loss:2.1778 aux_loss:0.0267 train_time:455998ms step_avg:133.37ms
step:3430/6250 train_loss:2.2650 aux_loss:0.0264 train_time:456139ms step_avg:133.37ms
step:3431/6250 train_loss:2.2712 aux_loss:0.0263 train_time:456276ms step_avg:133.37ms
step:3432/6250 train_loss:2.2651 aux_loss:0.0269 train_time:456409ms step_avg:133.37ms
step:3433/6250 train_loss:2.2855 aux_loss:0.0268 train_time:456545ms step_avg:133.38ms
step:3434/6250 train_loss:2.2438 aux_loss:0.0263 train_time:456678ms step_avg:133.38ms
step:3435/6250 train_loss:2.2535 aux_loss:0.0264 train_time:456814ms step_avg:133.38ms
step:3436/6250 train_loss:2.2225 aux_loss:0.0268 train_time:456953ms step_avg:133.38ms
step:3437/6250 train_loss:2.2117 aux_loss:0.0264 train_time:457087ms step_avg:133.38ms
step:3438/6250 train_loss:2.1738 aux_loss:0.0262 train_time:457235ms step_avg:133.38ms
step:3439/6250 train_loss:2.1512 aux_loss:0.0262 train_time:457381ms step_avg:133.39ms
step:3440/6250 train_loss:2.3088 aux_loss:0.0265 train_time:457521ms step_avg:133.39ms
step:3441/6250 train_loss:2.2532 aux_loss:0.0267 train_time:457656ms step_avg:133.39ms
step:3442/6250 train_loss:2.2353 aux_loss:0.0264 train_time:457796ms step_avg:133.39ms
step:3443/6250 train_loss:2.2188 aux_loss:0.0265 train_time:457936ms step_avg:133.39ms
step:3444/6250 train_loss:2.3043 aux_loss:0.0268 train_time:458075ms step_avg:133.39ms
step:3445/6250 train_loss:2.1164 aux_loss:0.0265 train_time:458210ms step_avg:133.39ms
step:3446/6250 train_loss:2.2746 aux_loss:0.0266 train_time:458343ms step_avg:133.39ms
step:3447/6250 train_loss:2.2466 aux_loss:0.0264 train_time:458474ms step_avg:133.39ms
step:3448/6250 train_loss:2.0953 aux_loss:0.0264 train_time:458606ms step_avg:133.39ms
step:3449/6250 train_loss:2.1747 aux_loss:0.0265 train_time:458739ms step_avg:133.39ms
step:3450/6250 train_loss:2.2329 aux_loss:0.0263 train_time:458874ms step_avg:133.39ms
step:3451/6250 train_loss:2.1124 aux_loss:0.0265 train_time:459007ms step_avg:133.39ms
step:3452/6250 train_loss:2.2083 aux_loss:0.0269 train_time:459140ms step_avg:133.39ms
step:3453/6250 train_loss:2.2160 aux_loss:0.0265 train_time:459276ms step_avg:133.39ms
step:3454/6250 train_loss:2.2053 aux_loss:0.0264 train_time:459408ms step_avg:133.39ms
step:3455/6250 train_loss:2.3569 aux_loss:0.0265 train_time:459541ms step_avg:133.39ms
step:3456/6250 train_loss:2.2554 aux_loss:0.0262 train_time:459675ms step_avg:133.39ms
step:3457/6250 train_loss:2.2860 aux_loss:0.0268 train_time:459810ms step_avg:133.39ms
step:3458/6250 train_loss:2.3582 aux_loss:0.0265 train_time:459941ms step_avg:133.39ms
step:3459/6250 train_loss:2.2131 aux_loss:0.0260 train_time:460073ms step_avg:133.39ms
step:3460/6250 train_loss:2.2762 aux_loss:0.0265 train_time:460205ms step_avg:133.39ms
step:3461/6250 train_loss:2.1664 aux_loss:0.0266 train_time:460337ms step_avg:133.39ms
step:3462/6250 train_loss:2.1532 aux_loss:0.0259 train_time:460469ms step_avg:133.39ms
step:3463/6250 train_loss:2.2618 aux_loss:0.0266 train_time:460601ms step_avg:133.39ms
step:3464/6250 train_loss:2.2182 aux_loss:0.0270 train_time:460733ms step_avg:133.39ms
step:3465/6250 train_loss:2.2279 aux_loss:0.0267 train_time:460865ms step_avg:133.39ms
step:3466/6250 train_loss:2.2149 aux_loss:0.0267 train_time:460996ms step_avg:133.39ms
step:3467/6250 train_loss:2.2168 aux_loss:0.0265 train_time:461128ms step_avg:133.39ms
step:3468/6250 train_loss:2.2330 aux_loss:0.0265 train_time:461261ms step_avg:133.39ms
step:3469/6250 train_loss:2.3093 aux_loss:0.0265 train_time:461392ms step_avg:133.39ms
step:3470/6250 train_loss:2.1827 aux_loss:0.0266 train_time:461524ms step_avg:133.39ms
step:3471/6250 train_loss:2.1565 aux_loss:0.0263 train_time:461655ms step_avg:133.39ms
step:3472/6250 train_loss:2.1600 aux_loss:0.0263 train_time:461789ms step_avg:133.39ms
step:3473/6250 train_loss:2.2090 aux_loss:0.0265 train_time:461923ms step_avg:133.39ms
step:3474/6250 train_loss:2.2156 aux_loss:0.0265 train_time:462054ms step_avg:133.39ms
step:3475/6250 train_loss:2.1233 aux_loss:0.0263 train_time:462186ms step_avg:133.39ms
step:3476/6250 train_loss:2.1979 aux_loss:0.0263 train_time:462318ms step_avg:133.39ms
step:3477/6250 train_loss:2.3317 aux_loss:0.0264 train_time:462450ms step_avg:133.39ms
step:3478/6250 train_loss:2.1581 aux_loss:0.0263 train_time:462583ms step_avg:133.39ms
step:3479/6250 train_loss:2.1041 aux_loss:0.0263 train_time:462715ms step_avg:133.39ms
step:3480/6250 train_loss:2.1324 aux_loss:0.0263 train_time:462848ms step_avg:133.39ms
step:3481/6250 train_loss:2.3045 aux_loss:0.0266 train_time:462979ms step_avg:133.38ms
step:3482/6250 train_loss:2.1708 aux_loss:0.0262 train_time:463112ms step_avg:133.38ms
step:3483/6250 train_loss:2.2619 aux_loss:0.0263 train_time:463243ms step_avg:133.38ms
step:3484/6250 train_loss:2.3395 aux_loss:0.0264 train_time:463376ms step_avg:133.38ms
step:3485/6250 train_loss:2.2669 aux_loss:0.0260 train_time:463508ms step_avg:133.38ms
step:3486/6250 train_loss:2.1558 aux_loss:0.0263 train_time:463639ms step_avg:133.38ms
step:3487/6250 train_loss:2.4043 aux_loss:0.0278 train_time:463773ms step_avg:133.38ms
step:3488/6250 train_loss:2.2200 aux_loss:0.0264 train_time:463905ms step_avg:133.38ms
step:3489/6250 train_loss:2.2371 aux_loss:0.0265 train_time:464039ms step_avg:133.38ms
step:3490/6250 train_loss:2.1961 aux_loss:0.0269 train_time:464170ms step_avg:133.38ms
step:3491/6250 train_loss:2.3151 aux_loss:0.0267 train_time:464302ms step_avg:133.38ms
step:3492/6250 train_loss:2.2418 aux_loss:0.0264 train_time:464434ms step_avg:133.38ms
step:3493/6250 train_loss:2.2939 aux_loss:0.0268 train_time:464565ms step_avg:133.38ms
step:3494/6250 train_loss:2.1433 aux_loss:0.0266 train_time:464696ms step_avg:133.38ms
step:3495/6250 train_loss:2.2061 aux_loss:0.0263 train_time:464830ms step_avg:133.38ms
step:3496/6250 train_loss:2.1630 aux_loss:0.0263 train_time:464961ms step_avg:133.38ms
step:3497/6250 train_loss:2.2362 aux_loss:0.0264 train_time:465093ms step_avg:133.38ms
step:3498/6250 train_loss:2.2717 aux_loss:0.0265 train_time:465225ms step_avg:133.38ms
step:3499/6250 train_loss:2.1592 aux_loss:0.0263 train_time:465357ms step_avg:133.38ms
step:3500/6250 train_loss:2.2114 aux_loss:0.0264 train_time:465489ms step_avg:133.38ms
step:3501/6250 train_loss:2.2073 aux_loss:0.0267 train_time:465620ms step_avg:133.38ms
step:3502/6250 train_loss:2.2093 aux_loss:0.0268 train_time:465753ms step_avg:133.38ms
step:3503/6250 train_loss:2.2007 aux_loss:0.0267 train_time:465886ms step_avg:133.38ms
step:3504/6250 train_loss:2.2568 aux_loss:0.0268 train_time:466019ms step_avg:133.38ms
step:3505/6250 train_loss:2.2617 aux_loss:0.0266 train_time:466151ms step_avg:133.38ms
step:3506/6250 train_loss:2.3565 aux_loss:0.0265 train_time:466284ms step_avg:133.38ms
step:3507/6250 train_loss:2.2130 aux_loss:0.0266 train_time:466417ms step_avg:133.38ms
step:3508/6250 train_loss:2.1972 aux_loss:0.0265 train_time:466550ms step_avg:133.38ms
step:3509/6250 train_loss:2.2631 aux_loss:0.0262 train_time:466682ms step_avg:133.38ms
step:3510/6250 train_loss:2.2941 aux_loss:0.0265 train_time:466815ms step_avg:133.38ms
step:3511/6250 train_loss:2.2507 aux_loss:0.0265 train_time:466948ms step_avg:133.38ms
step:3512/6250 train_loss:2.2499 aux_loss:0.0264 train_time:467082ms step_avg:133.38ms
step:3513/6250 train_loss:2.2886 aux_loss:0.0265 train_time:467215ms step_avg:133.38ms
step:3514/6250 train_loss:2.2816 aux_loss:0.0265 train_time:467347ms step_avg:133.38ms
step:3515/6250 train_loss:2.2099 aux_loss:0.0264 train_time:467478ms step_avg:133.37ms
step:3516/6250 train_loss:2.1958 aux_loss:0.0263 train_time:467610ms step_avg:133.37ms
step:3517/6250 train_loss:2.2924 aux_loss:0.0264 train_time:467745ms step_avg:133.37ms
step:3518/6250 train_loss:2.2107 aux_loss:0.0261 train_time:467878ms step_avg:133.37ms
step:3519/6250 train_loss:2.1829 aux_loss:0.0261 train_time:468012ms step_avg:133.37ms
step:3520/6250 train_loss:2.1734 aux_loss:0.0257 train_time:468147ms step_avg:133.38ms
step:3521/6250 train_loss:2.1951 aux_loss:0.0260 train_time:468278ms step_avg:133.37ms
step:3522/6250 train_loss:2.2808 aux_loss:0.0260 train_time:468410ms step_avg:133.37ms
step:3523/6250 train_loss:2.2222 aux_loss:0.0260 train_time:468542ms step_avg:133.37ms
step:3524/6250 train_loss:2.2068 aux_loss:0.0263 train_time:468673ms step_avg:133.37ms
step:3525/6250 train_loss:2.2510 aux_loss:0.0263 train_time:468809ms step_avg:133.37ms
step:3526/6250 train_loss:2.2691 aux_loss:0.0257 train_time:468940ms step_avg:133.37ms
step:3527/6250 train_loss:2.2210 aux_loss:0.0261 train_time:469072ms step_avg:133.37ms
step:3528/6250 train_loss:2.2164 aux_loss:0.0261 train_time:469205ms step_avg:133.37ms
step:3529/6250 train_loss:2.3345 aux_loss:0.0258 train_time:469337ms step_avg:133.37ms
step:3530/6250 train_loss:2.1541 aux_loss:0.0254 train_time:469469ms step_avg:133.37ms
step:3531/6250 train_loss:2.2841 aux_loss:0.0259 train_time:469601ms step_avg:133.37ms
step:3532/6250 train_loss:2.2147 aux_loss:0.0259 train_time:469733ms step_avg:133.37ms
step:3533/6250 train_loss:2.1577 aux_loss:0.0254 train_time:469866ms step_avg:133.37ms
step:3534/6250 train_loss:2.2900 aux_loss:0.0258 train_time:469998ms step_avg:133.37ms
step:3535/6250 train_loss:2.2049 aux_loss:0.0259 train_time:470130ms step_avg:133.37ms
step:3536/6250 train_loss:2.2760 aux_loss:0.0267 train_time:470263ms step_avg:133.37ms
step:3537/6250 train_loss:2.1170 aux_loss:0.0260 train_time:470396ms step_avg:133.37ms
step:3538/6250 train_loss:2.2359 aux_loss:0.0258 train_time:470529ms step_avg:133.37ms
step:3539/6250 train_loss:2.2209 aux_loss:0.0260 train_time:470661ms step_avg:133.37ms
step:3540/6250 train_loss:2.2277 aux_loss:0.0264 train_time:470793ms step_avg:133.37ms
step:3541/6250 train_loss:2.2999 aux_loss:0.0270 train_time:470926ms step_avg:133.37ms
step:3542/6250 train_loss:2.2234 aux_loss:0.0264 train_time:471058ms step_avg:133.37ms
step:3543/6250 train_loss:2.3035 aux_loss:0.0260 train_time:471191ms step_avg:133.37ms
step:3544/6250 train_loss:2.1734 aux_loss:0.0262 train_time:471322ms step_avg:133.37ms
step:3545/6250 train_loss:2.2105 aux_loss:0.0262 train_time:471454ms step_avg:133.37ms
step:3546/6250 train_loss:2.2350 aux_loss:0.0259 train_time:471586ms step_avg:133.37ms
step:3547/6250 train_loss:2.2374 aux_loss:0.0260 train_time:471719ms step_avg:133.37ms
step:3548/6250 train_loss:2.2676 aux_loss:0.0261 train_time:471852ms step_avg:133.37ms
step:3549/6250 train_loss:2.3451 aux_loss:0.0265 train_time:471984ms step_avg:133.37ms
step:3550/6250 train_loss:2.2442 aux_loss:0.0258 train_time:472116ms step_avg:133.37ms
step:3551/6250 train_loss:2.2392 aux_loss:0.0257 train_time:472248ms step_avg:133.37ms
step:3552/6250 train_loss:2.2551 aux_loss:0.0259 train_time:472380ms step_avg:133.37ms
step:3553/6250 train_loss:2.2698 aux_loss:0.0259 train_time:472511ms step_avg:133.36ms
step:3554/6250 train_loss:2.1714 aux_loss:0.0258 train_time:472644ms step_avg:133.36ms
step:3555/6250 train_loss:2.2037 aux_loss:0.0258 train_time:472776ms step_avg:133.36ms
step:3556/6250 train_loss:2.2508 aux_loss:0.0260 train_time:472909ms step_avg:133.36ms
step:3557/6250 train_loss:2.1735 aux_loss:0.0256 train_time:473041ms step_avg:133.36ms
step:3558/6250 train_loss:2.2780 aux_loss:0.0261 train_time:473173ms step_avg:133.36ms
step:3559/6250 train_loss:2.1436 aux_loss:0.0259 train_time:473306ms step_avg:133.36ms
step:3560/6250 train_loss:2.2124 aux_loss:0.0260 train_time:473438ms step_avg:133.36ms
step:3561/6250 train_loss:2.2530 aux_loss:0.0261 train_time:473570ms step_avg:133.36ms
step:3562/6250 train_loss:2.1716 aux_loss:0.0256 train_time:473703ms step_avg:133.36ms
step:3563/6250 train_loss:2.2533 aux_loss:0.0255 train_time:473835ms step_avg:133.36ms
step:3564/6250 train_loss:2.1912 aux_loss:0.0256 train_time:473967ms step_avg:133.36ms
step:3565/6250 train_loss:2.2137 aux_loss:0.0254 train_time:474100ms step_avg:133.36ms
step:3566/6250 train_loss:2.2824 aux_loss:0.0256 train_time:474232ms step_avg:133.36ms
step:3567/6250 train_loss:2.2025 aux_loss:0.0257 train_time:474364ms step_avg:133.36ms
step:3568/6250 train_loss:2.1365 aux_loss:0.0285 train_time:474495ms step_avg:133.36ms
step:3569/6250 train_loss:2.1751 aux_loss:0.0256 train_time:474628ms step_avg:133.36ms
step:3570/6250 train_loss:2.1574 aux_loss:0.0260 train_time:474760ms step_avg:133.36ms
step:3571/6250 train_loss:2.2264 aux_loss:0.0258 train_time:474892ms step_avg:133.36ms
step:3572/6250 train_loss:2.3628 aux_loss:0.0259 train_time:475024ms step_avg:133.36ms
step:3573/6250 train_loss:2.2244 aux_loss:0.0260 train_time:475157ms step_avg:133.36ms
step:3574/6250 train_loss:2.2831 aux_loss:0.0265 train_time:475289ms step_avg:133.36ms
step:3575/6250 train_loss:2.1609 aux_loss:0.0264 train_time:475421ms step_avg:133.36ms
step:3576/6250 train_loss:2.2568 aux_loss:0.0261 train_time:475553ms step_avg:133.36ms
step:3577/6250 train_loss:2.2848 aux_loss:0.0266 train_time:475687ms step_avg:133.36ms
step:3578/6250 train_loss:2.2050 aux_loss:0.0262 train_time:475820ms step_avg:133.36ms
step:3579/6250 train_loss:2.2501 aux_loss:0.0261 train_time:475951ms step_avg:133.36ms
step:3580/6250 train_loss:2.2289 aux_loss:0.0262 train_time:476083ms step_avg:133.36ms
step:3581/6250 train_loss:2.2804 aux_loss:0.0261 train_time:476215ms step_avg:133.36ms
step:3582/6250 train_loss:2.1621 aux_loss:0.0260 train_time:476348ms step_avg:133.36ms
step:3583/6250 train_loss:2.2535 aux_loss:0.0261 train_time:476481ms step_avg:133.36ms
step:3584/6250 train_loss:2.1580 aux_loss:0.0257 train_time:476616ms step_avg:133.36ms
step:3585/6250 train_loss:2.0860 aux_loss:0.0257 train_time:476748ms step_avg:133.36ms
step:3586/6250 train_loss:2.2591 aux_loss:0.0259 train_time:476880ms step_avg:133.36ms
step:3587/6250 train_loss:2.1625 aux_loss:0.0256 train_time:477012ms step_avg:133.36ms
step:3588/6250 train_loss:2.1151 aux_loss:0.0254 train_time:477146ms step_avg:133.36ms
step:3589/6250 train_loss:2.2265 aux_loss:0.0256 train_time:477278ms step_avg:133.36ms
step:3590/6250 train_loss:2.2538 aux_loss:0.0255 train_time:477408ms step_avg:133.35ms
step:3591/6250 train_loss:2.2029 aux_loss:0.0252 train_time:477540ms step_avg:133.35ms
step:3592/6250 train_loss:2.3065 aux_loss:0.0254 train_time:477672ms step_avg:133.35ms
step:3593/6250 train_loss:2.1810 aux_loss:0.0258 train_time:477805ms step_avg:133.35ms
step:3594/6250 train_loss:2.1798 aux_loss:0.0257 train_time:477937ms step_avg:133.35ms
step:3595/6250 train_loss:2.2124 aux_loss:0.0262 train_time:478069ms step_avg:133.35ms
step:3596/6250 train_loss:2.1869 aux_loss:0.0260 train_time:478201ms step_avg:133.35ms
step:3597/6250 train_loss:2.1651 aux_loss:0.0259 train_time:478333ms step_avg:133.35ms
step:3598/6250 train_loss:2.1568 aux_loss:0.0259 train_time:478465ms step_avg:133.35ms
step:3599/6250 train_loss:2.1190 aux_loss:0.0257 train_time:478597ms step_avg:133.35ms
step:3600/6250 train_loss:2.1189 aux_loss:0.0260 train_time:478729ms step_avg:133.35ms
step:3601/6250 train_loss:2.2435 aux_loss:0.0260 train_time:478862ms step_avg:133.35ms
step:3602/6250 train_loss:2.1863 aux_loss:0.0259 train_time:478994ms step_avg:133.35ms
step:3603/6250 train_loss:2.1498 aux_loss:0.0257 train_time:479126ms step_avg:133.35ms
step:3604/6250 train_loss:2.1483 aux_loss:0.0254 train_time:479258ms step_avg:133.35ms
step:3605/6250 train_loss:2.2416 aux_loss:0.0259 train_time:479389ms step_avg:133.35ms
step:3606/6250 train_loss:2.1773 aux_loss:0.0258 train_time:479521ms step_avg:133.35ms
step:3607/6250 train_loss:2.1630 aux_loss:0.0258 train_time:479654ms step_avg:133.35ms
step:3608/6250 train_loss:2.3257 aux_loss:0.0256 train_time:479786ms step_avg:133.35ms
step:3609/6250 train_loss:2.2045 aux_loss:0.0254 train_time:479918ms step_avg:133.35ms
step:3610/6250 train_loss:2.1979 aux_loss:0.0254 train_time:480049ms step_avg:133.35ms
step:3611/6250 train_loss:2.2849 aux_loss:0.0254 train_time:480182ms step_avg:133.35ms
step:3612/6250 train_loss:2.3015 aux_loss:0.0255 train_time:480314ms step_avg:133.35ms
step:3613/6250 train_loss:2.1535 aux_loss:0.0257 train_time:480446ms step_avg:133.35ms
step:3614/6250 train_loss:2.1317 aux_loss:0.0255 train_time:480577ms step_avg:133.35ms
step:3615/6250 train_loss:2.1536 aux_loss:0.0252 train_time:480709ms step_avg:133.35ms
step:3616/6250 train_loss:2.2376 aux_loss:0.0254 train_time:480842ms step_avg:133.35ms
step:3617/6250 train_loss:2.2108 aux_loss:0.0255 train_time:480973ms step_avg:133.34ms
step:3618/6250 train_loss:2.2235 aux_loss:0.0257 train_time:481106ms step_avg:133.34ms
step:3619/6250 train_loss:2.2816 aux_loss:0.0259 train_time:481240ms step_avg:133.34ms
step:3620/6250 train_loss:2.2131 aux_loss:0.0262 train_time:481380ms step_avg:133.35ms
step:3621/6250 train_loss:2.2680 aux_loss:0.0256 train_time:481518ms step_avg:133.35ms
step:3622/6250 train_loss:2.2082 aux_loss:0.0255 train_time:481652ms step_avg:133.35ms
step:3623/6250 train_loss:2.2676 aux_loss:0.0257 train_time:481788ms step_avg:133.35ms
step:3624/6250 train_loss:2.1910 aux_loss:0.0256 train_time:481927ms step_avg:133.35ms
step:3625/6250 train_loss:2.2885 aux_loss:0.0260 train_time:482064ms step_avg:133.35ms
step:3626/6250 train_loss:2.3548 aux_loss:0.0260 train_time:482200ms step_avg:133.35ms
step:3627/6250 train_loss:2.2473 aux_loss:0.0255 train_time:482337ms step_avg:133.35ms
step:3628/6250 train_loss:2.1974 aux_loss:0.0257 train_time:482469ms step_avg:133.35ms
step:3629/6250 train_loss:2.1317 aux_loss:0.0258 train_time:482608ms step_avg:133.35ms
step:3630/6250 train_loss:2.2542 aux_loss:0.0257 train_time:482743ms step_avg:133.35ms
step:3631/6250 train_loss:2.2657 aux_loss:0.0262 train_time:482877ms step_avg:133.35ms
step:3632/6250 train_loss:2.1986 aux_loss:0.0257 train_time:483009ms step_avg:133.35ms
step:3633/6250 train_loss:2.1500 aux_loss:0.0256 train_time:483147ms step_avg:133.36ms
step:3634/6250 train_loss:2.2206 aux_loss:0.0260 train_time:483285ms step_avg:133.36ms
step:3635/6250 train_loss:2.2016 aux_loss:0.0259 train_time:483422ms step_avg:133.36ms
step:3636/6250 train_loss:2.2478 aux_loss:0.0265 train_time:483560ms step_avg:133.36ms
step:3637/6250 train_loss:2.2819 aux_loss:0.0264 train_time:483701ms step_avg:133.36ms
step:3638/6250 train_loss:2.2502 aux_loss:0.0261 train_time:483836ms step_avg:133.36ms
step:3639/6250 train_loss:2.2961 aux_loss:0.0254 train_time:483982ms step_avg:133.37ms
step:3640/6250 train_loss:2.1982 aux_loss:0.0253 train_time:484118ms step_avg:133.37ms
step:3641/6250 train_loss:2.2913 aux_loss:0.0257 train_time:484256ms step_avg:133.37ms
step:3642/6250 train_loss:2.1944 aux_loss:0.0260 train_time:484390ms step_avg:133.37ms
step:3643/6250 train_loss:2.3628 aux_loss:0.0257 train_time:484522ms step_avg:133.37ms
step:3644/6250 train_loss:2.2329 aux_loss:0.0257 train_time:484654ms step_avg:133.37ms
step:3645/6250 train_loss:2.2379 aux_loss:0.0260 train_time:484786ms step_avg:133.37ms
step:3646/6250 train_loss:2.2750 aux_loss:0.0255 train_time:484918ms step_avg:133.37ms
step:3647/6250 train_loss:2.2546 aux_loss:0.0251 train_time:485050ms step_avg:133.37ms
step:3648/6250 train_loss:2.2622 aux_loss:0.0253 train_time:485183ms step_avg:133.37ms
step:3649/6250 train_loss:2.2912 aux_loss:0.0254 train_time:485314ms step_avg:133.36ms
step:3650/6250 train_loss:2.1837 aux_loss:0.0251 train_time:485457ms step_avg:133.37ms
step:3651/6250 train_loss:2.3266 aux_loss:0.0252 train_time:485598ms step_avg:133.37ms
step:3652/6250 train_loss:2.1963 aux_loss:0.0250 train_time:485742ms step_avg:133.37ms
step:3653/6250 train_loss:2.1935 aux_loss:0.0253 train_time:485877ms step_avg:133.37ms
step:3654/6250 train_loss:2.2368 aux_loss:0.0250 train_time:486008ms step_avg:133.37ms
step:3655/6250 train_loss:2.2960 aux_loss:0.0249 train_time:486141ms step_avg:133.37ms
step:3656/6250 train_loss:2.3160 aux_loss:0.0254 train_time:486278ms step_avg:133.37ms
step:3657/6250 train_loss:2.2702 aux_loss:0.0252 train_time:486417ms step_avg:133.37ms
step:3658/6250 train_loss:2.2970 aux_loss:0.0252 train_time:486556ms step_avg:133.38ms
step:3659/6250 train_loss:2.3763 aux_loss:0.0251 train_time:486695ms step_avg:133.38ms
step:3660/6250 train_loss:2.1596 aux_loss:0.0249 train_time:486828ms step_avg:133.38ms
step:3661/6250 train_loss:2.2506 aux_loss:0.0251 train_time:486960ms step_avg:133.38ms
step:3662/6250 train_loss:2.2076 aux_loss:0.0251 train_time:487092ms step_avg:133.38ms
step:3663/6250 train_loss:2.2010 aux_loss:0.0254 train_time:487225ms step_avg:133.38ms
step:3664/6250 train_loss:2.2354 aux_loss:0.0254 train_time:487356ms step_avg:133.38ms
step:3665/6250 train_loss:2.2368 aux_loss:0.0254 train_time:487488ms step_avg:133.38ms
step:3666/6250 train_loss:2.2622 aux_loss:0.0258 train_time:487622ms step_avg:133.38ms
step:3667/6250 train_loss:2.2597 aux_loss:0.0257 train_time:487760ms step_avg:133.38ms
step:3668/6250 train_loss:2.2598 aux_loss:0.0254 train_time:487898ms step_avg:133.38ms
step:3669/6250 train_loss:2.2163 aux_loss:0.0257 train_time:488029ms step_avg:133.38ms
step:3670/6250 train_loss:2.2873 aux_loss:0.0257 train_time:488163ms step_avg:133.38ms
step:3671/6250 train_loss:2.2315 aux_loss:0.0256 train_time:488295ms step_avg:133.38ms
step:3672/6250 train_loss:2.2195 aux_loss:0.0258 train_time:488426ms step_avg:133.38ms
step:3673/6250 train_loss:2.1458 aux_loss:0.0254 train_time:488563ms step_avg:133.38ms
step:3674/6250 train_loss:2.2847 aux_loss:0.0255 train_time:488701ms step_avg:133.38ms
step:3675/6250 train_loss:2.2956 aux_loss:0.0256 train_time:488846ms step_avg:133.38ms
step:3676/6250 train_loss:2.2745 aux_loss:0.0256 train_time:488984ms step_avg:133.38ms
step:3677/6250 train_loss:2.2683 aux_loss:0.0257 train_time:489123ms step_avg:133.39ms
step:3678/6250 train_loss:2.2346 aux_loss:0.0260 train_time:489262ms step_avg:133.39ms
step:3679/6250 train_loss:2.2150 aux_loss:0.0258 train_time:489400ms step_avg:133.39ms
step:3680/6250 train_loss:2.2745 aux_loss:0.0257 train_time:489538ms step_avg:133.39ms
step:3681/6250 train_loss:2.2903 aux_loss:0.0258 train_time:489676ms step_avg:133.39ms
step:3682/6250 train_loss:2.2441 aux_loss:0.0259 train_time:489814ms step_avg:133.39ms
step:3683/6250 train_loss:2.2515 aux_loss:0.0262 train_time:489951ms step_avg:133.39ms
step:3684/6250 train_loss:2.1952 aux_loss:0.0257 train_time:490090ms step_avg:133.39ms
step:3685/6250 train_loss:2.2680 aux_loss:0.0257 train_time:490229ms step_avg:133.40ms
step:3686/6250 train_loss:2.2016 aux_loss:0.0254 train_time:490367ms step_avg:133.40ms
step:3687/6250 train_loss:2.2202 aux_loss:0.0255 train_time:490505ms step_avg:133.40ms
step:3688/6250 train_loss:2.2561 aux_loss:0.0257 train_time:490644ms step_avg:133.40ms
step:3689/6250 train_loss:2.1938 aux_loss:0.0252 train_time:490780ms step_avg:133.40ms
step:3690/6250 train_loss:2.1228 aux_loss:0.0254 train_time:490918ms step_avg:133.40ms
step:3691/6250 train_loss:2.1387 aux_loss:0.0256 train_time:491056ms step_avg:133.40ms
step:3692/6250 train_loss:2.2508 aux_loss:0.0255 train_time:491200ms step_avg:133.41ms
step:3693/6250 train_loss:2.2375 aux_loss:0.0255 train_time:491340ms step_avg:133.41ms
step:3694/6250 train_loss:2.2045 aux_loss:0.0248 train_time:491473ms step_avg:133.41ms
step:3695/6250 train_loss:2.1413 aux_loss:0.0251 train_time:491620ms step_avg:133.41ms
step:3696/6250 train_loss:2.2086 aux_loss:0.0252 train_time:491767ms step_avg:133.41ms
step:3697/6250 train_loss:2.3361 aux_loss:0.0254 train_time:491901ms step_avg:133.42ms
step:3698/6250 train_loss:2.2302 aux_loss:0.0248 train_time:492034ms step_avg:133.41ms
step:3699/6250 train_loss:2.2948 aux_loss:0.0250 train_time:492168ms step_avg:133.42ms
step:3700/6250 train_loss:2.2058 aux_loss:0.0250 train_time:492311ms step_avg:133.42ms
step:3701/6250 train_loss:2.1714 aux_loss:0.0249 train_time:492457ms step_avg:133.42ms
step:3702/6250 train_loss:2.2009 aux_loss:0.0248 train_time:492596ms step_avg:133.42ms
step:3703/6250 train_loss:2.1315 aux_loss:0.0247 train_time:492731ms step_avg:133.42ms
step:3704/6250 train_loss:2.1660 aux_loss:0.0250 train_time:492865ms step_avg:133.42ms
step:3705/6250 train_loss:2.2812 aux_loss:0.0253 train_time:492999ms step_avg:133.42ms
step:3706/6250 train_loss:2.1962 aux_loss:0.0250 train_time:493133ms step_avg:133.42ms
step:3707/6250 train_loss:2.1639 aux_loss:0.0251 train_time:493268ms step_avg:133.42ms
step:3708/6250 train_loss:2.2016 aux_loss:0.0253 train_time:493403ms step_avg:133.42ms
step:3709/6250 train_loss:2.1932 aux_loss:0.0253 train_time:493536ms step_avg:133.42ms
step:3710/6250 train_loss:2.2576 aux_loss:0.0253 train_time:493670ms step_avg:133.42ms
step:3711/6250 train_loss:2.2363 aux_loss:0.0251 train_time:493806ms step_avg:133.42ms
step:3712/6250 train_loss:2.1936 aux_loss:0.0250 train_time:493940ms step_avg:133.43ms
step:3713/6250 train_loss:2.2448 aux_loss:0.0252 train_time:494074ms step_avg:133.43ms
step:3714/6250 train_loss:2.2286 aux_loss:0.0251 train_time:494218ms step_avg:133.43ms
step:3715/6250 train_loss:2.2203 aux_loss:0.0249 train_time:494357ms step_avg:133.43ms
step:3716/6250 train_loss:2.2895 aux_loss:0.0251 train_time:494497ms step_avg:133.43ms
step:3717/6250 train_loss:2.2473 aux_loss:0.0257 train_time:494638ms step_avg:133.43ms
step:3718/6250 train_loss:2.2183 aux_loss:0.0255 train_time:494772ms step_avg:133.43ms
step:3719/6250 train_loss:2.2042 aux_loss:0.0249 train_time:494908ms step_avg:133.43ms
step:3720/6250 train_loss:2.2764 aux_loss:0.0252 train_time:495045ms step_avg:133.44ms
step:3721/6250 train_loss:2.1838 aux_loss:0.0252 train_time:495182ms step_avg:133.44ms
step:3722/6250 train_loss:2.2472 aux_loss:0.0251 train_time:495315ms step_avg:133.44ms
step:3723/6250 train_loss:2.1885 aux_loss:0.0251 train_time:495452ms step_avg:133.44ms
step:3724/6250 train_loss:2.1819 aux_loss:0.0252 train_time:495588ms step_avg:133.44ms
step:3725/6250 train_loss:2.2839 aux_loss:0.0249 train_time:495722ms step_avg:133.44ms
step:3726/6250 train_loss:2.1641 aux_loss:0.0251 train_time:495856ms step_avg:133.44ms
step:3727/6250 train_loss:2.2716 aux_loss:0.0251 train_time:495989ms step_avg:133.44ms
step:3728/6250 train_loss:2.1714 aux_loss:0.0249 train_time:496123ms step_avg:133.44ms
step:3729/6250 train_loss:2.1855 aux_loss:0.0247 train_time:496258ms step_avg:133.44ms
step:3730/6250 train_loss:2.2106 aux_loss:0.0249 train_time:496392ms step_avg:133.44ms
step:3731/6250 train_loss:2.1961 aux_loss:0.0253 train_time:496525ms step_avg:133.44ms
step:3732/6250 train_loss:2.3158 aux_loss:0.0252 train_time:496659ms step_avg:133.44ms
step:3733/6250 train_loss:2.1662 aux_loss:0.0251 train_time:496792ms step_avg:133.44ms
step:3734/6250 train_loss:2.2566 aux_loss:0.0251 train_time:496926ms step_avg:133.44ms
step:3735/6250 train_loss:2.2552 aux_loss:0.0251 train_time:497060ms step_avg:133.44ms
step:3736/6250 train_loss:2.2161 aux_loss:0.0248 train_time:497194ms step_avg:133.44ms
step:3737/6250 train_loss:2.2328 aux_loss:0.0251 train_time:497330ms step_avg:133.44ms
step:3738/6250 train_loss:2.2639 aux_loss:0.0253 train_time:497467ms step_avg:133.44ms
step:3739/6250 train_loss:2.1518 aux_loss:0.0251 train_time:497602ms step_avg:133.44ms
step:3740/6250 train_loss:2.2134 aux_loss:0.0250 train_time:497738ms step_avg:133.44ms
step:3741/6250 train_loss:2.2412 aux_loss:0.0249 train_time:497875ms step_avg:133.44ms
step:3742/6250 train_loss:2.1978 aux_loss:0.0251 train_time:498011ms step_avg:133.44ms
step:3743/6250 train_loss:2.1389 aux_loss:0.0251 train_time:498147ms step_avg:133.44ms
step:3744/6250 train_loss:2.2404 aux_loss:0.0252 train_time:498280ms step_avg:133.44ms
step:3745/6250 train_loss:2.2104 aux_loss:0.0250 train_time:498413ms step_avg:133.44ms
step:3746/6250 train_loss:2.2355 aux_loss:0.0254 train_time:498547ms step_avg:133.44ms
step:3747/6250 train_loss:2.2337 aux_loss:0.0252 train_time:498683ms step_avg:133.44ms
step:3748/6250 train_loss:2.1980 aux_loss:0.0252 train_time:498819ms step_avg:133.45ms
step:3749/6250 train_loss:2.1193 aux_loss:0.0253 train_time:498955ms step_avg:133.45ms
step:3750/6250 train_loss:2.2531 aux_loss:0.0250 train_time:499088ms step_avg:133.45ms
step:3751/6250 train_loss:2.2400 aux_loss:0.0248 train_time:499223ms step_avg:133.45ms
step:3752/6250 train_loss:2.1379 aux_loss:0.0248 train_time:499357ms step_avg:133.45ms
step:3753/6250 train_loss:2.1940 aux_loss:0.0248 train_time:499492ms step_avg:133.45ms
step:3754/6250 train_loss:2.1472 aux_loss:0.0244 train_time:499626ms step_avg:133.45ms
step:3755/6250 train_loss:2.2336 aux_loss:0.0247 train_time:499764ms step_avg:133.45ms
step:3756/6250 train_loss:2.3302 aux_loss:0.0250 train_time:499898ms step_avg:133.45ms
step:3757/6250 train_loss:2.2366 aux_loss:0.0241 train_time:500034ms step_avg:133.45ms
step:3758/6250 train_loss:2.1998 aux_loss:0.0248 train_time:500170ms step_avg:133.45ms
step:3759/6250 train_loss:2.1356 aux_loss:0.0249 train_time:500307ms step_avg:133.45ms
step:3760/6250 train_loss:2.2232 aux_loss:0.0247 train_time:500443ms step_avg:133.45ms
step:3761/6250 train_loss:2.1662 aux_loss:0.0248 train_time:500578ms step_avg:133.45ms
step:3762/6250 train_loss:2.2528 aux_loss:0.0250 train_time:500714ms step_avg:133.45ms
step:3763/6250 train_loss:2.3321 aux_loss:0.0252 train_time:500851ms step_avg:133.45ms
step:3764/6250 train_loss:2.2470 aux_loss:0.0252 train_time:500987ms step_avg:133.45ms
step:3765/6250 train_loss:2.1682 aux_loss:0.0255 train_time:501125ms step_avg:133.46ms
step:3766/6250 train_loss:2.1758 aux_loss:0.0250 train_time:501260ms step_avg:133.46ms
step:3767/6250 train_loss:2.1779 aux_loss:0.0252 train_time:501394ms step_avg:133.46ms
step:3768/6250 train_loss:2.2396 aux_loss:0.0254 train_time:501528ms step_avg:133.46ms
step:3769/6250 train_loss:2.2628 aux_loss:0.0253 train_time:501665ms step_avg:133.46ms
step:3770/6250 train_loss:2.3288 aux_loss:0.0250 train_time:501800ms step_avg:133.46ms
step:3771/6250 train_loss:2.1660 aux_loss:0.0249 train_time:501936ms step_avg:133.46ms
step:3772/6250 train_loss:2.1866 aux_loss:0.0251 train_time:502071ms step_avg:133.46ms
step:3773/6250 train_loss:2.2680 aux_loss:0.0248 train_time:502204ms step_avg:133.46ms
step:3774/6250 train_loss:2.2049 aux_loss:0.0250 train_time:502338ms step_avg:133.46ms
step:3775/6250 train_loss:2.1773 aux_loss:0.0250 train_time:502475ms step_avg:133.46ms
step:3776/6250 train_loss:2.1863 aux_loss:0.0249 train_time:502612ms step_avg:133.46ms
step:3777/6250 train_loss:2.1678 aux_loss:0.0252 train_time:502748ms step_avg:133.46ms
step:3778/6250 train_loss:2.2273 aux_loss:0.0251 train_time:502882ms step_avg:133.46ms
step:3779/6250 train_loss:2.2668 aux_loss:0.0250 train_time:503015ms step_avg:133.46ms
step:3780/6250 train_loss:2.1679 aux_loss:0.0248 train_time:503150ms step_avg:133.46ms
step:3781/6250 train_loss:2.1660 aux_loss:0.0249 train_time:503284ms step_avg:133.46ms
step:3782/6250 train_loss:2.2945 aux_loss:0.0249 train_time:503419ms step_avg:133.46ms
step:3783/6250 train_loss:2.1127 aux_loss:0.0248 train_time:503557ms step_avg:133.46ms
step:3784/6250 train_loss:2.1430 aux_loss:0.0249 train_time:503691ms step_avg:133.46ms
step:3785/6250 train_loss:2.2639 aux_loss:0.0247 train_time:503825ms step_avg:133.46ms
step:3786/6250 train_loss:2.2197 aux_loss:0.0247 train_time:503960ms step_avg:133.46ms
step:3787/6250 train_loss:2.1655 aux_loss:0.0250 train_time:504096ms step_avg:133.46ms
step:3788/6250 train_loss:2.2202 aux_loss:0.0246 train_time:504231ms step_avg:133.46ms
step:3789/6250 train_loss:2.1951 aux_loss:0.0243 train_time:504367ms step_avg:133.47ms
step:3790/6250 train_loss:2.1736 aux_loss:0.0248 train_time:504503ms step_avg:133.47ms
step:3791/6250 train_loss:2.2296 aux_loss:0.0250 train_time:504641ms step_avg:133.47ms
step:3792/6250 train_loss:2.2507 aux_loss:0.0245 train_time:504775ms step_avg:133.47ms
step:3793/6250 train_loss:2.2015 aux_loss:0.0245 train_time:504912ms step_avg:133.47ms
step:3794/6250 train_loss:2.1990 aux_loss:0.0244 train_time:505046ms step_avg:133.47ms
step:3795/6250 train_loss:2.2081 aux_loss:0.0247 train_time:505183ms step_avg:133.47ms
step:3796/6250 train_loss:2.2488 aux_loss:0.0246 train_time:505319ms step_avg:133.47ms
step:3797/6250 train_loss:2.2806 aux_loss:0.0244 train_time:505454ms step_avg:133.47ms
step:3798/6250 train_loss:2.2111 aux_loss:0.0245 train_time:505592ms step_avg:133.47ms
step:3799/6250 train_loss:2.2487 aux_loss:0.0246 train_time:505726ms step_avg:133.47ms
step:3800/6250 train_loss:2.1714 aux_loss:0.0244 train_time:505860ms step_avg:133.47ms
step:3801/6250 train_loss:2.2579 aux_loss:0.0244 train_time:505993ms step_avg:133.47ms
step:3802/6250 train_loss:2.2520 aux_loss:0.0245 train_time:506127ms step_avg:133.47ms
step:3803/6250 train_loss:2.1996 aux_loss:0.0245 train_time:506261ms step_avg:133.47ms
step:3804/6250 train_loss:2.1271 aux_loss:0.0241 train_time:506395ms step_avg:133.47ms
step:3805/6250 train_loss:2.1579 aux_loss:0.0243 train_time:506529ms step_avg:133.47ms
step:3806/6250 train_loss:2.2635 aux_loss:0.0245 train_time:506665ms step_avg:133.47ms
step:3807/6250 train_loss:2.2207 aux_loss:0.0248 train_time:506804ms step_avg:133.47ms
step:3808/6250 train_loss:2.2502 aux_loss:0.0245 train_time:506938ms step_avg:133.48ms
step:3809/6250 train_loss:2.2090 aux_loss:0.0243 train_time:507072ms step_avg:133.48ms
step:3810/6250 train_loss:2.2373 aux_loss:0.0244 train_time:507207ms step_avg:133.48ms
step:3811/6250 train_loss:2.2849 aux_loss:0.0241 train_time:507342ms step_avg:133.48ms
step:3812/6250 train_loss:2.2398 aux_loss:0.0246 train_time:507477ms step_avg:133.48ms
step:3813/6250 train_loss:2.1953 aux_loss:0.0246 train_time:507613ms step_avg:133.48ms
step:3814/6250 train_loss:2.1739 aux_loss:0.0246 train_time:507746ms step_avg:133.48ms
step:3815/6250 train_loss:2.1920 aux_loss:0.0246 train_time:507882ms step_avg:133.48ms
step:3816/6250 train_loss:2.1509 aux_loss:0.0245 train_time:508016ms step_avg:133.48ms
step:3817/6250 train_loss:2.2547 aux_loss:0.0246 train_time:508152ms step_avg:133.48ms
step:3818/6250 train_loss:2.1458 aux_loss:0.0248 train_time:508289ms step_avg:133.48ms
step:3819/6250 train_loss:2.2079 aux_loss:0.0247 train_time:508424ms step_avg:133.48ms
step:3820/6250 train_loss:2.1620 aux_loss:0.0247 train_time:508562ms step_avg:133.48ms
step:3821/6250 train_loss:2.2137 aux_loss:0.0246 train_time:508699ms step_avg:133.48ms
step:3822/6250 train_loss:2.2611 aux_loss:0.0243 train_time:508837ms step_avg:133.48ms
step:3823/6250 train_loss:2.1470 aux_loss:0.0249 train_time:508970ms step_avg:133.48ms
step:3824/6250 train_loss:2.2424 aux_loss:0.0248 train_time:509104ms step_avg:133.48ms
step:3825/6250 train_loss:2.2141 aux_loss:0.0243 train_time:509240ms step_avg:133.48ms
step:3826/6250 train_loss:2.2038 aux_loss:0.0249 train_time:509378ms step_avg:133.48ms
step:3827/6250 train_loss:2.2704 aux_loss:0.0249 train_time:509512ms step_avg:133.48ms
step:3828/6250 train_loss:2.1434 aux_loss:0.0243 train_time:509650ms step_avg:133.49ms
step:3829/6250 train_loss:2.1486 aux_loss:0.0244 train_time:509786ms step_avg:133.49ms
step:3830/6250 train_loss:2.2376 aux_loss:0.0249 train_time:509921ms step_avg:133.49ms
step:3831/6250 train_loss:2.2109 aux_loss:0.0245 train_time:510056ms step_avg:133.49ms
step:3832/6250 train_loss:2.2175 aux_loss:0.0246 train_time:510191ms step_avg:133.49ms
step:3833/6250 train_loss:2.0881 aux_loss:0.0249 train_time:510326ms step_avg:133.49ms
step:3834/6250 train_loss:2.2686 aux_loss:0.0247 train_time:510463ms step_avg:133.49ms
step:3835/6250 train_loss:2.2166 aux_loss:0.0247 train_time:510597ms step_avg:133.49ms
step:3836/6250 train_loss:2.1666 aux_loss:0.0243 train_time:510733ms step_avg:133.49ms
step:3837/6250 train_loss:2.2276 aux_loss:0.0243 train_time:510871ms step_avg:133.49ms
step:3838/6250 train_loss:2.2038 aux_loss:0.0248 train_time:511006ms step_avg:133.49ms
step:3839/6250 train_loss:2.2527 aux_loss:0.0248 train_time:511140ms step_avg:133.49ms
step:3840/6250 train_loss:2.1335 aux_loss:0.0241 train_time:511275ms step_avg:133.49ms
step:3841/6250 train_loss:2.1743 aux_loss:0.0243 train_time:511410ms step_avg:133.49ms
step:3842/6250 train_loss:2.3039 aux_loss:0.0248 train_time:511545ms step_avg:133.49ms
step:3843/6250 train_loss:2.2153 aux_loss:0.0248 train_time:511682ms step_avg:133.49ms
step:3844/6250 train_loss:2.2690 aux_loss:0.0245 train_time:511816ms step_avg:133.49ms
step:3845/6250 train_loss:2.2474 aux_loss:0.0246 train_time:511949ms step_avg:133.49ms
step:3846/6250 train_loss:2.2233 aux_loss:0.0250 train_time:512083ms step_avg:133.49ms
step:3847/6250 train_loss:2.1961 aux_loss:0.0249 train_time:512218ms step_avg:133.49ms
step:3848/6250 train_loss:2.1540 aux_loss:0.0247 train_time:512354ms step_avg:133.50ms
step:3849/6250 train_loss:2.2325 aux_loss:0.0245 train_time:512492ms step_avg:133.50ms
step:3850/6250 train_loss:2.3055 aux_loss:0.0244 train_time:512627ms step_avg:133.50ms
step:3851/6250 train_loss:2.2813 aux_loss:0.0244 train_time:512761ms step_avg:133.50ms
step:3852/6250 train_loss:2.2698 aux_loss:0.0247 train_time:512894ms step_avg:133.50ms
step:3853/6250 train_loss:2.3033 aux_loss:0.0245 train_time:513029ms step_avg:133.50ms
step:3854/6250 train_loss:2.2271 aux_loss:0.0247 train_time:513163ms step_avg:133.50ms
step:3855/6250 train_loss:2.2240 aux_loss:0.0246 train_time:513297ms step_avg:133.50ms
step:3856/6250 train_loss:2.1909 aux_loss:0.0246 train_time:513431ms step_avg:133.50ms
step:3857/6250 train_loss:2.1297 aux_loss:0.0246 train_time:513565ms step_avg:133.50ms
step:3858/6250 train_loss:2.2164 aux_loss:0.0245 train_time:513700ms step_avg:133.50ms
step:3859/6250 train_loss:2.1849 aux_loss:0.0246 train_time:513835ms step_avg:133.50ms
step:3860/6250 train_loss:2.2195 aux_loss:0.0248 train_time:513970ms step_avg:133.50ms
step:3861/6250 train_loss:2.2521 aux_loss:0.0246 train_time:514103ms step_avg:133.50ms
step:3862/6250 train_loss:2.1544 aux_loss:0.0247 train_time:514241ms step_avg:133.50ms
step:3863/6250 train_loss:2.2220 aux_loss:0.0248 train_time:514375ms step_avg:133.50ms
step:3864/6250 train_loss:2.1058 aux_loss:0.0243 train_time:514510ms step_avg:133.50ms
step:3865/6250 train_loss:2.1607 aux_loss:0.0245 train_time:514645ms step_avg:133.50ms
step:3866/6250 train_loss:2.2346 aux_loss:0.0245 train_time:514782ms step_avg:133.50ms
step:3867/6250 train_loss:2.1566 aux_loss:0.0243 train_time:514917ms step_avg:133.50ms
step:3868/6250 train_loss:2.1668 aux_loss:0.0245 train_time:515052ms step_avg:133.50ms
step:3869/6250 train_loss:2.2527 aux_loss:0.0245 train_time:515187ms step_avg:133.50ms
step:3870/6250 train_loss:2.1194 aux_loss:0.0245 train_time:515321ms step_avg:133.50ms
step:3871/6250 train_loss:2.2689 aux_loss:0.0249 train_time:515458ms step_avg:133.50ms
step:3872/6250 train_loss:2.1686 aux_loss:0.0245 train_time:515595ms step_avg:133.50ms
step:3873/6250 train_loss:2.2311 aux_loss:0.0248 train_time:515730ms step_avg:133.50ms
step:3874/6250 train_loss:2.1505 aux_loss:0.0244 train_time:515864ms step_avg:133.51ms
step:3875/6250 train_loss:2.1941 aux_loss:0.0243 train_time:515997ms step_avg:133.50ms
step:3876/6250 train_loss:2.1655 aux_loss:0.0245 train_time:516131ms step_avg:133.51ms
step:3877/6250 train_loss:2.2117 aux_loss:0.0242 train_time:516265ms step_avg:133.51ms
step:3878/6250 train_loss:2.2480 aux_loss:0.0247 train_time:516401ms step_avg:133.51ms
step:3879/6250 train_loss:2.1898 aux_loss:0.0243 train_time:516539ms step_avg:133.51ms
step:3880/6250 train_loss:2.1485 aux_loss:0.0241 train_time:516675ms step_avg:133.51ms
step:3881/6250 train_loss:2.2675 aux_loss:0.0245 train_time:516809ms step_avg:133.51ms
step:3882/6250 train_loss:2.2952 aux_loss:0.0242 train_time:516948ms step_avg:133.51ms
step:3883/6250 train_loss:2.1762 aux_loss:0.0237 train_time:517087ms step_avg:133.51ms
step:3884/6250 train_loss:2.1625 aux_loss:0.0241 train_time:517226ms step_avg:133.51ms
step:3885/6250 train_loss:2.1683 aux_loss:0.0242 train_time:517364ms step_avg:133.51ms
step:3886/6250 train_loss:2.2605 aux_loss:0.0245 train_time:517504ms step_avg:133.51ms
step:3887/6250 train_loss:2.2479 aux_loss:0.0244 train_time:517643ms step_avg:133.52ms
step:3888/6250 train_loss:2.2151 aux_loss:0.0245 train_time:517782ms step_avg:133.52ms
step:3889/6250 train_loss:2.2757 aux_loss:0.0243 train_time:517923ms step_avg:133.52ms
step:3890/6250 train_loss:2.2381 aux_loss:0.0245 train_time:518059ms step_avg:133.52ms
step:3891/6250 train_loss:2.1776 aux_loss:0.0243 train_time:518199ms step_avg:133.52ms
step:3892/6250 train_loss:2.1766 aux_loss:0.0243 train_time:518338ms step_avg:133.52ms
step:3893/6250 train_loss:2.1816 aux_loss:0.0242 train_time:518478ms step_avg:133.53ms
step:3894/6250 train_loss:2.1677 aux_loss:0.0241 train_time:518615ms step_avg:133.53ms
step:3895/6250 train_loss:2.1515 aux_loss:0.0245 train_time:518755ms step_avg:133.53ms
step:3896/6250 train_loss:2.1461 aux_loss:0.0240 train_time:518891ms step_avg:133.53ms
step:3897/6250 train_loss:2.1697 aux_loss:0.0246 train_time:519030ms step_avg:133.53ms
step:3898/6250 train_loss:2.1712 aux_loss:0.0245 train_time:519168ms step_avg:133.53ms
step:3899/6250 train_loss:2.3213 aux_loss:0.0247 train_time:519308ms step_avg:133.53ms
step:3900/6250 train_loss:2.2566 aux_loss:0.0243 train_time:519451ms step_avg:133.53ms
step:3901/6250 train_loss:2.2404 aux_loss:0.0244 train_time:519589ms step_avg:133.54ms
step:3902/6250 train_loss:2.2433 aux_loss:0.0246 train_time:519730ms step_avg:133.54ms
step:3903/6250 train_loss:2.1855 aux_loss:0.0241 train_time:519871ms step_avg:133.54ms
step:3904/6250 train_loss:2.1931 aux_loss:0.0242 train_time:520011ms step_avg:133.54ms
step:3905/6250 train_loss:2.2516 aux_loss:0.0240 train_time:520148ms step_avg:133.54ms
step:3906/6250 train_loss:2.2104 aux_loss:0.0241 train_time:520285ms step_avg:133.54ms
step:3907/6250 train_loss:2.1759 aux_loss:0.0242 train_time:520421ms step_avg:133.54ms
step:3908/6250 train_loss:2.2575 aux_loss:0.0243 train_time:520558ms step_avg:133.54ms
step:3909/6250 train_loss:2.1621 aux_loss:0.0239 train_time:520696ms step_avg:133.55ms
step:3910/6250 train_loss:2.2213 aux_loss:0.0240 train_time:520834ms step_avg:133.55ms
step:3911/6250 train_loss:2.1510 aux_loss:0.0245 train_time:520971ms step_avg:133.55ms
step:3912/6250 train_loss:2.1365 aux_loss:0.0249 train_time:521108ms step_avg:133.55ms
step:3913/6250 train_loss:2.1298 aux_loss:0.0246 train_time:521245ms step_avg:133.55ms
step:3914/6250 train_loss:2.1893 aux_loss:0.0243 train_time:521386ms step_avg:133.55ms
step:3915/6250 train_loss:2.2288 aux_loss:0.0238 train_time:521525ms step_avg:133.55ms
step:3916/6250 train_loss:2.2297 aux_loss:0.0245 train_time:521660ms step_avg:133.55ms
step:3917/6250 train_loss:2.2380 aux_loss:0.0246 train_time:521795ms step_avg:133.55ms
step:3918/6250 train_loss:2.2318 aux_loss:0.0243 train_time:521930ms step_avg:133.55ms
step:3919/6250 train_loss:2.0863 aux_loss:0.0242 train_time:522063ms step_avg:133.55ms
step:3920/6250 train_loss:2.2676 aux_loss:0.0245 train_time:522198ms step_avg:133.55ms
step:3921/6250 train_loss:2.1987 aux_loss:0.0243 train_time:522333ms step_avg:133.55ms
step:3922/6250 train_loss:2.2359 aux_loss:0.0245 train_time:522466ms step_avg:133.55ms
step:3923/6250 train_loss:2.2355 aux_loss:0.0244 train_time:522603ms step_avg:133.56ms
step:3924/6250 train_loss:2.2376 aux_loss:0.0243 train_time:522738ms step_avg:133.56ms
step:3925/6250 train_loss:2.1145 aux_loss:0.0244 train_time:522876ms step_avg:133.56ms
step:3926/6250 train_loss:2.2380 aux_loss:0.0246 train_time:523019ms step_avg:133.56ms
step:3927/6250 train_loss:2.1567 aux_loss:0.0246 train_time:523167ms step_avg:133.56ms
step:3928/6250 train_loss:2.2281 aux_loss:0.0243 train_time:523309ms step_avg:133.57ms
step:3929/6250 train_loss:2.3014 aux_loss:0.0247 train_time:523450ms step_avg:133.57ms
step:3930/6250 train_loss:2.2411 aux_loss:0.0242 train_time:523589ms step_avg:133.57ms
step:3931/6250 train_loss:2.2416 aux_loss:0.0240 train_time:523728ms step_avg:133.57ms
step:3932/6250 train_loss:2.2910 aux_loss:0.0243 train_time:523864ms step_avg:133.57ms
step:3933/6250 train_loss:2.3004 aux_loss:0.0245 train_time:523998ms step_avg:133.57ms
step:3934/6250 train_loss:2.1519 aux_loss:0.0241 train_time:524132ms step_avg:133.57ms
step:3935/6250 train_loss:2.1937 aux_loss:0.0245 train_time:524266ms step_avg:133.57ms
step:3936/6250 train_loss:2.2642 aux_loss:0.0246 train_time:524399ms step_avg:133.57ms
step:3937/6250 train_loss:2.2003 aux_loss:0.0244 train_time:524533ms step_avg:133.57ms
step:3938/6250 train_loss:2.2469 aux_loss:0.0242 train_time:524667ms step_avg:133.57ms
step:3939/6250 train_loss:2.1811 aux_loss:0.0243 train_time:524801ms step_avg:133.57ms
step:3940/6250 train_loss:2.2031 aux_loss:0.0244 train_time:524934ms step_avg:133.57ms
step:3941/6250 train_loss:2.2948 aux_loss:0.0240 train_time:525068ms step_avg:133.57ms
step:3942/6250 train_loss:2.2122 aux_loss:0.0239 train_time:525201ms step_avg:133.57ms
step:3943/6250 train_loss:2.1887 aux_loss:0.0239 train_time:525334ms step_avg:133.57ms
step:3944/6250 train_loss:2.2401 aux_loss:0.0237 train_time:525467ms step_avg:133.57ms
step:3945/6250 train_loss:2.2378 aux_loss:0.0239 train_time:525601ms step_avg:133.57ms
step:3946/6250 train_loss:2.1963 aux_loss:0.0239 train_time:525735ms step_avg:133.57ms
step:3947/6250 train_loss:2.2096 aux_loss:0.0242 train_time:525869ms step_avg:133.57ms
step:3948/6250 train_loss:2.2425 aux_loss:0.0238 train_time:526003ms step_avg:133.57ms
step:3949/6250 train_loss:2.1608 aux_loss:0.0238 train_time:526137ms step_avg:133.57ms
step:3950/6250 train_loss:2.2512 aux_loss:0.0239 train_time:526271ms step_avg:133.57ms
step:3951/6250 train_loss:2.1721 aux_loss:0.0241 train_time:526404ms step_avg:133.57ms
step:3952/6250 train_loss:2.1937 aux_loss:0.0239 train_time:526537ms step_avg:133.57ms
step:3953/6250 train_loss:2.2189 aux_loss:0.0239 train_time:526672ms step_avg:133.57ms
step:3954/6250 train_loss:2.2508 aux_loss:0.0238 train_time:526806ms step_avg:133.57ms
step:3955/6250 train_loss:2.2158 aux_loss:0.0239 train_time:526941ms step_avg:133.57ms
step:3956/6250 train_loss:2.2252 aux_loss:0.0235 train_time:527075ms step_avg:133.57ms
step:3957/6250 train_loss:2.2083 aux_loss:0.0238 train_time:527209ms step_avg:133.57ms
step:3958/6250 train_loss:2.1114 aux_loss:0.0238 train_time:527342ms step_avg:133.57ms
step:3959/6250 train_loss:2.1721 aux_loss:0.0239 train_time:527475ms step_avg:133.57ms
step:3960/6250 train_loss:2.2620 aux_loss:0.0233 train_time:527609ms step_avg:133.57ms
step:3961/6250 train_loss:2.1904 aux_loss:0.0234 train_time:527743ms step_avg:133.57ms
step:3962/6250 train_loss:2.1404 aux_loss:0.0238 train_time:527876ms step_avg:133.57ms
step:3963/6250 train_loss:2.1938 aux_loss:0.0236 train_time:528010ms step_avg:133.57ms
step:3964/6250 train_loss:2.2222 aux_loss:0.0239 train_time:528143ms step_avg:133.57ms
step:3965/6250 train_loss:2.1796 aux_loss:0.0241 train_time:528276ms step_avg:133.57ms
step:3966/6250 train_loss:2.1669 aux_loss:0.0240 train_time:528410ms step_avg:133.57ms
step:3967/6250 train_loss:2.1790 aux_loss:0.0239 train_time:528543ms step_avg:133.57ms
step:3968/6250 train_loss:2.2373 aux_loss:0.0237 train_time:528677ms step_avg:133.57ms
step:3969/6250 train_loss:2.1382 aux_loss:0.0235 train_time:528811ms step_avg:133.57ms
step:3970/6250 train_loss:2.2382 aux_loss:0.0237 train_time:528944ms step_avg:133.57ms
step:3971/6250 train_loss:2.2172 aux_loss:0.0238 train_time:529076ms step_avg:133.57ms
step:3972/6250 train_loss:2.2899 aux_loss:0.0236 train_time:529211ms step_avg:133.57ms
step:3973/6250 train_loss:2.2352 aux_loss:0.0242 train_time:529344ms step_avg:133.57ms
step:3974/6250 train_loss:2.2339 aux_loss:0.0241 train_time:529477ms step_avg:133.57ms
step:3975/6250 train_loss:2.2623 aux_loss:0.0240 train_time:529612ms step_avg:133.57ms
step:3976/6250 train_loss:2.2338 aux_loss:0.0241 train_time:529745ms step_avg:133.57ms
step:3977/6250 train_loss:2.2473 aux_loss:0.0238 train_time:529879ms step_avg:133.57ms
step:3978/6250 train_loss:2.2227 aux_loss:0.0234 train_time:530013ms step_avg:133.57ms
step:3979/6250 train_loss:2.1676 aux_loss:0.0234 train_time:530146ms step_avg:133.57ms
step:3980/6250 train_loss:2.2440 aux_loss:0.0241 train_time:530279ms step_avg:133.57ms
step:3981/6250 train_loss:2.1896 aux_loss:0.0241 train_time:530413ms step_avg:133.57ms
step:3982/6250 train_loss:2.1748 aux_loss:0.0239 train_time:530547ms step_avg:133.57ms
step:3983/6250 train_loss:2.1567 aux_loss:0.0238 train_time:530681ms step_avg:133.57ms
step:3984/6250 train_loss:2.2655 aux_loss:0.0238 train_time:530815ms step_avg:133.57ms
step:3985/6250 train_loss:2.1732 aux_loss:0.0240 train_time:530947ms step_avg:133.57ms
step:3986/6250 train_loss:2.2415 aux_loss:0.0242 train_time:531081ms step_avg:133.57ms
step:3987/6250 train_loss:2.2227 aux_loss:0.0238 train_time:531215ms step_avg:133.57ms
step:3988/6250 train_loss:2.2641 aux_loss:0.0237 train_time:531348ms step_avg:133.57ms
step:3989/6250 train_loss:2.2211 aux_loss:0.0237 train_time:531482ms step_avg:133.57ms
step:3990/6250 train_loss:2.2808 aux_loss:0.0238 train_time:531615ms step_avg:133.57ms
step:3991/6250 train_loss:2.3056 aux_loss:0.0239 train_time:531749ms step_avg:133.57ms
step:3992/6250 train_loss:2.2444 aux_loss:0.0238 train_time:531884ms step_avg:133.57ms
step:3993/6250 train_loss:2.2007 aux_loss:0.0235 train_time:532017ms step_avg:133.57ms
step:3994/6250 train_loss:2.1869 aux_loss:0.0237 train_time:532151ms step_avg:133.57ms
step:3995/6250 train_loss:2.1976 aux_loss:0.0237 train_time:532284ms step_avg:133.57ms
step:3996/6250 train_loss:2.1127 aux_loss:0.0236 train_time:532417ms step_avg:133.57ms
step:3997/6250 train_loss:2.1802 aux_loss:0.0234 train_time:532551ms step_avg:133.57ms
step:3998/6250 train_loss:2.1784 aux_loss:0.0235 train_time:532685ms step_avg:133.57ms
step:3999/6250 train_loss:2.2637 aux_loss:0.0238 train_time:532819ms step_avg:133.57ms
step:4000/6250 train_loss:2.2284 aux_loss:0.0238 train_time:532952ms step_avg:133.57ms
step:4000/6250 val_loss:2.1914 val_aux_loss:0.0238 train_time:532952ms step_avg:133.57ms
step:4001/6250 train_loss:2.1692 aux_loss:0.0238 train_time:533085ms step_avg:133.57ms
step:4002/6250 train_loss:2.1580 aux_loss:0.0237 train_time:533220ms step_avg:133.57ms
step:4003/6250 train_loss:2.1870 aux_loss:0.0238 train_time:533355ms step_avg:133.57ms
step:4004/6250 train_loss:2.1884 aux_loss:0.0238 train_time:533493ms step_avg:133.57ms
step:4005/6250 train_loss:2.1945 aux_loss:0.0240 train_time:533627ms step_avg:133.57ms
step:4006/6250 train_loss:2.0873 aux_loss:0.0238 train_time:533764ms step_avg:133.57ms
step:4007/6250 train_loss:2.2795 aux_loss:0.0242 train_time:533901ms step_avg:133.58ms
step:4008/6250 train_loss:2.1860 aux_loss:0.0241 train_time:534033ms step_avg:133.57ms
step:4009/6250 train_loss:2.2752 aux_loss:0.0242 train_time:534165ms step_avg:133.57ms
step:4010/6250 train_loss:2.1832 aux_loss:0.0241 train_time:534303ms step_avg:133.58ms
step:4011/6250 train_loss:2.2223 aux_loss:0.0241 train_time:534441ms step_avg:133.58ms
step:4012/6250 train_loss:2.1808 aux_loss:0.0241 train_time:534573ms step_avg:133.58ms
step:4013/6250 train_loss:2.3065 aux_loss:0.0240 train_time:534706ms step_avg:133.58ms
step:4014/6250 train_loss:2.1557 aux_loss:0.0239 train_time:534838ms step_avg:133.58ms
step:4015/6250 train_loss:2.1699 aux_loss:0.0240 train_time:534969ms step_avg:133.58ms
step:4016/6250 train_loss:2.2002 aux_loss:0.0240 train_time:535101ms step_avg:133.57ms
step:4017/6250 train_loss:2.2115 aux_loss:0.0241 train_time:535234ms step_avg:133.57ms
step:4018/6250 train_loss:2.1724 aux_loss:0.0240 train_time:535366ms step_avg:133.57ms
step:4019/6250 train_loss:2.1360 aux_loss:0.0236 train_time:535498ms step_avg:133.57ms
step:4020/6250 train_loss:2.2448 aux_loss:0.0238 train_time:535629ms step_avg:133.57ms
step:4021/6250 train_loss:2.2915 aux_loss:0.0241 train_time:535762ms step_avg:133.57ms
step:4022/6250 train_loss:2.1981 aux_loss:0.0237 train_time:535895ms step_avg:133.57ms
step:4023/6250 train_loss:2.1302 aux_loss:0.0237 train_time:536027ms step_avg:133.57ms
step:4024/6250 train_loss:2.1854 aux_loss:0.0239 train_time:536158ms step_avg:133.57ms
step:4025/6250 train_loss:2.2130 aux_loss:0.0241 train_time:536291ms step_avg:133.57ms
step:4026/6250 train_loss:2.2494 aux_loss:0.0239 train_time:536423ms step_avg:133.57ms
step:4027/6250 train_loss:2.1296 aux_loss:0.0238 train_time:536555ms step_avg:133.57ms
step:4028/6250 train_loss:2.0978 aux_loss:0.0238 train_time:536687ms step_avg:133.57ms
step:4029/6250 train_loss:2.1997 aux_loss:0.0239 train_time:536819ms step_avg:133.57ms
step:4030/6250 train_loss:2.1140 aux_loss:0.0237 train_time:536952ms step_avg:133.57ms
step:4031/6250 train_loss:2.2699 aux_loss:0.0240 train_time:537084ms step_avg:133.57ms
step:4032/6250 train_loss:2.2567 aux_loss:0.0238 train_time:537217ms step_avg:133.57ms
step:4033/6250 train_loss:2.0850 aux_loss:0.0232 train_time:537350ms step_avg:133.57ms
step:4034/6250 train_loss:2.1430 aux_loss:0.0228 train_time:537482ms step_avg:133.57ms
step:4035/6250 train_loss:2.2005 aux_loss:0.0231 train_time:537614ms step_avg:133.57ms
step:4036/6250 train_loss:2.2258 aux_loss:0.0230 train_time:537747ms step_avg:133.57ms
step:4037/6250 train_loss:2.2713 aux_loss:0.0231 train_time:537878ms step_avg:133.57ms
step:4038/6250 train_loss:2.2442 aux_loss:0.0229 train_time:538010ms step_avg:133.57ms
step:4039/6250 train_loss:2.1535 aux_loss:0.0234 train_time:538142ms step_avg:133.57ms
step:4040/6250 train_loss:2.2082 aux_loss:0.0233 train_time:538275ms step_avg:133.57ms
step:4041/6250 train_loss:2.2458 aux_loss:0.0229 train_time:538406ms step_avg:133.57ms
step:4042/6250 train_loss:2.2116 aux_loss:0.0230 train_time:538537ms step_avg:133.57ms
step:4043/6250 train_loss:2.1853 aux_loss:0.0231 train_time:538670ms step_avg:133.57ms
step:4044/6250 train_loss:2.2611 aux_loss:0.0236 train_time:538803ms step_avg:133.57ms
step:4045/6250 train_loss:2.1553 aux_loss:0.0232 train_time:538935ms step_avg:133.56ms
step:4046/6250 train_loss:2.1859 aux_loss:0.0235 train_time:539067ms step_avg:133.56ms
step:4047/6250 train_loss:2.1400 aux_loss:0.0237 train_time:539199ms step_avg:133.56ms
step:4048/6250 train_loss:2.2031 aux_loss:0.0236 train_time:539332ms step_avg:133.56ms
step:4049/6250 train_loss:2.2129 aux_loss:0.0235 train_time:539463ms step_avg:133.56ms
step:4050/6250 train_loss:2.3439 aux_loss:0.0238 train_time:539595ms step_avg:133.56ms
step:4051/6250 train_loss:2.2567 aux_loss:0.0239 train_time:539728ms step_avg:133.56ms
step:4052/6250 train_loss:2.2711 aux_loss:0.0240 train_time:539859ms step_avg:133.56ms
step:4053/6250 train_loss:2.1377 aux_loss:0.0236 train_time:539991ms step_avg:133.56ms
step:4054/6250 train_loss:2.2363 aux_loss:0.0239 train_time:540123ms step_avg:133.56ms
step:4055/6250 train_loss:2.1898 aux_loss:0.0238 train_time:540255ms step_avg:133.56ms
step:4056/6250 train_loss:2.2217 aux_loss:0.0234 train_time:540387ms step_avg:133.56ms
step:4057/6250 train_loss:2.2334 aux_loss:0.0240 train_time:540518ms step_avg:133.56ms
step:4058/6250 train_loss:2.1473 aux_loss:0.0235 train_time:540650ms step_avg:133.56ms
step:4059/6250 train_loss:2.3069 aux_loss:0.0239 train_time:540783ms step_avg:133.56ms
step:4060/6250 train_loss:2.2233 aux_loss:0.0238 train_time:540915ms step_avg:133.56ms
step:4061/6250 train_loss:2.2242 aux_loss:0.0237 train_time:541047ms step_avg:133.56ms
step:4062/6250 train_loss:2.1770 aux_loss:0.0234 train_time:541178ms step_avg:133.56ms
step:4063/6250 train_loss:2.1678 aux_loss:0.0239 train_time:541312ms step_avg:133.56ms
step:4064/6250 train_loss:2.1274 aux_loss:0.0240 train_time:541444ms step_avg:133.56ms
step:4065/6250 train_loss:2.2147 aux_loss:0.0235 train_time:541576ms step_avg:133.56ms
step:4066/6250 train_loss:2.1357 aux_loss:0.0230 train_time:541709ms step_avg:133.56ms
step:4067/6250 train_loss:2.1145 aux_loss:0.0240 train_time:541841ms step_avg:133.56ms
step:4068/6250 train_loss:2.2560 aux_loss:0.0236 train_time:541974ms step_avg:133.56ms
step:4069/6250 train_loss:2.2344 aux_loss:0.0231 train_time:542106ms step_avg:133.56ms
step:4070/6250 train_loss:2.2101 aux_loss:0.0236 train_time:542238ms step_avg:133.56ms
step:4071/6250 train_loss:2.1600 aux_loss:0.0237 train_time:542370ms step_avg:133.56ms
step:4072/6250 train_loss:2.1276 aux_loss:0.0234 train_time:542502ms step_avg:133.56ms
step:4073/6250 train_loss:2.1484 aux_loss:0.0231 train_time:542634ms step_avg:133.55ms
step:4074/6250 train_loss:2.1564 aux_loss:0.0231 train_time:542767ms step_avg:133.55ms
step:4075/6250 train_loss:2.1790 aux_loss:0.0233 train_time:542898ms step_avg:133.55ms
step:4076/6250 train_loss:2.1991 aux_loss:0.0233 train_time:543030ms step_avg:133.55ms
step:4077/6250 train_loss:2.2636 aux_loss:0.0231 train_time:543162ms step_avg:133.55ms
step:4078/6250 train_loss:2.1539 aux_loss:0.0229 train_time:543294ms step_avg:133.55ms
step:4079/6250 train_loss:2.2697 aux_loss:0.0234 train_time:543426ms step_avg:133.55ms
step:4080/6250 train_loss:2.1493 aux_loss:0.0233 train_time:543557ms step_avg:133.55ms
step:4081/6250 train_loss:2.2094 aux_loss:0.0236 train_time:543690ms step_avg:133.55ms
step:4082/6250 train_loss:2.1959 aux_loss:0.0230 train_time:543822ms step_avg:133.55ms
step:4083/6250 train_loss:2.1605 aux_loss:0.0232 train_time:543954ms step_avg:133.55ms
step:4084/6250 train_loss:2.1791 aux_loss:0.0237 train_time:544086ms step_avg:133.55ms
step:4085/6250 train_loss:2.3462 aux_loss:0.0236 train_time:544217ms step_avg:133.55ms
step:4086/6250 train_loss:2.3272 aux_loss:0.0236 train_time:544350ms step_avg:133.55ms
step:4087/6250 train_loss:2.2019 aux_loss:0.0235 train_time:544481ms step_avg:133.55ms
step:4088/6250 train_loss:2.1986 aux_loss:0.0232 train_time:544613ms step_avg:133.55ms
step:4089/6250 train_loss:2.1762 aux_loss:0.0232 train_time:544746ms step_avg:133.55ms
step:4090/6250 train_loss:2.2462 aux_loss:0.0236 train_time:544877ms step_avg:133.55ms
step:4091/6250 train_loss:2.2042 aux_loss:0.0233 train_time:545009ms step_avg:133.55ms
step:4092/6250 train_loss:2.1488 aux_loss:0.0235 train_time:545141ms step_avg:133.55ms
step:4093/6250 train_loss:2.2351 aux_loss:0.0236 train_time:545274ms step_avg:133.55ms
step:4094/6250 train_loss:2.1546 aux_loss:0.0232 train_time:545405ms step_avg:133.55ms
step:4095/6250 train_loss:2.1867 aux_loss:0.0235 train_time:545536ms step_avg:133.55ms
step:4096/6250 train_loss:2.1599 aux_loss:0.0231 train_time:545669ms step_avg:133.55ms
step:4097/6250 train_loss:2.2470 aux_loss:0.0230 train_time:545802ms step_avg:133.55ms
step:4098/6250 train_loss:2.2268 aux_loss:0.0231 train_time:545933ms step_avg:133.55ms
step:4099/6250 train_loss:2.2157 aux_loss:0.0232 train_time:546065ms step_avg:133.54ms
step:4100/6250 train_loss:2.2844 aux_loss:0.0238 train_time:546197ms step_avg:133.54ms
step:4101/6250 train_loss:2.1456 aux_loss:0.0237 train_time:546330ms step_avg:133.54ms
step:4102/6250 train_loss:2.1759 aux_loss:0.0235 train_time:546461ms step_avg:133.54ms
step:4103/6250 train_loss:2.2251 aux_loss:0.0236 train_time:546593ms step_avg:133.54ms
step:4104/6250 train_loss:2.2184 aux_loss:0.0234 train_time:546726ms step_avg:133.54ms
step:4105/6250 train_loss:2.2181 aux_loss:0.0233 train_time:546859ms step_avg:133.54ms
step:4106/6250 train_loss:2.0962 aux_loss:0.0239 train_time:546991ms step_avg:133.54ms
step:4107/6250 train_loss:2.1482 aux_loss:0.0236 train_time:547123ms step_avg:133.54ms
step:4108/6250 train_loss:2.2811 aux_loss:0.0239 train_time:547255ms step_avg:133.54ms
step:4109/6250 train_loss:2.1234 aux_loss:0.0234 train_time:547386ms step_avg:133.54ms
step:4110/6250 train_loss:2.1276 aux_loss:0.0235 train_time:547518ms step_avg:133.54ms
step:4111/6250 train_loss:2.2231 aux_loss:0.0236 train_time:547650ms step_avg:133.54ms
step:4112/6250 train_loss:2.2045 aux_loss:0.0235 train_time:547783ms step_avg:133.54ms
step:4113/6250 train_loss:2.1947 aux_loss:0.0236 train_time:547915ms step_avg:133.54ms
step:4114/6250 train_loss:2.1798 aux_loss:0.0235 train_time:548047ms step_avg:133.54ms
step:4115/6250 train_loss:2.1427 aux_loss:0.0232 train_time:548178ms step_avg:133.54ms
step:4116/6250 train_loss:2.2087 aux_loss:0.0236 train_time:548310ms step_avg:133.54ms
step:4117/6250 train_loss:2.2089 aux_loss:0.0235 train_time:548442ms step_avg:133.54ms
step:4118/6250 train_loss:2.2207 aux_loss:0.0234 train_time:548573ms step_avg:133.54ms
step:4119/6250 train_loss:2.1895 aux_loss:0.0233 train_time:548707ms step_avg:133.54ms
step:4120/6250 train_loss:2.2034 aux_loss:0.0232 train_time:548838ms step_avg:133.54ms
step:4121/6250 train_loss:2.1123 aux_loss:0.0230 train_time:548971ms step_avg:133.54ms
step:4122/6250 train_loss:2.2839 aux_loss:0.0231 train_time:549102ms step_avg:133.54ms
step:4123/6250 train_loss:2.2411 aux_loss:0.0228 train_time:549235ms step_avg:133.54ms
step:4124/6250 train_loss:2.1675 aux_loss:0.0231 train_time:549367ms step_avg:133.54ms
step:4125/6250 train_loss:2.2119 aux_loss:0.0232 train_time:549498ms step_avg:133.54ms
step:4126/6250 train_loss:2.1946 aux_loss:0.0230 train_time:549631ms step_avg:133.54ms
step:4127/6250 train_loss:2.1899 aux_loss:0.0233 train_time:549764ms step_avg:133.54ms
step:4128/6250 train_loss:2.1948 aux_loss:0.0230 train_time:549895ms step_avg:133.53ms
step:4129/6250 train_loss:2.1831 aux_loss:0.0229 train_time:550027ms step_avg:133.53ms
step:4130/6250 train_loss:2.1828 aux_loss:0.0233 train_time:550159ms step_avg:133.53ms
step:4131/6250 train_loss:2.1782 aux_loss:0.0233 train_time:550292ms step_avg:133.53ms
step:4132/6250 train_loss:2.1630 aux_loss:0.0231 train_time:550423ms step_avg:133.53ms
step:4133/6250 train_loss:2.1929 aux_loss:0.0233 train_time:550554ms step_avg:133.53ms
step:4134/6250 train_loss:2.2528 aux_loss:0.0235 train_time:550687ms step_avg:133.53ms
step:4135/6250 train_loss:2.1523 aux_loss:0.0228 train_time:550818ms step_avg:133.53ms
step:4136/6250 train_loss:2.2316 aux_loss:0.0232 train_time:550951ms step_avg:133.53ms
step:4137/6250 train_loss:2.2588 aux_loss:0.0236 train_time:551082ms step_avg:133.53ms
step:4138/6250 train_loss:2.1865 aux_loss:0.0234 train_time:551214ms step_avg:133.53ms
step:4139/6250 train_loss:2.1971 aux_loss:0.0231 train_time:551347ms step_avg:133.53ms
step:4140/6250 train_loss:2.1756 aux_loss:0.0234 train_time:551479ms step_avg:133.53ms
step:4141/6250 train_loss:2.1457 aux_loss:0.0231 train_time:551610ms step_avg:133.53ms
step:4142/6250 train_loss:2.0340 aux_loss:0.0230 train_time:551743ms step_avg:133.53ms
step:4143/6250 train_loss:2.1517 aux_loss:0.0236 train_time:551874ms step_avg:133.53ms
step:4144/6250 train_loss:2.1821 aux_loss:0.0238 train_time:552006ms step_avg:133.53ms
step:4145/6250 train_loss:2.2062 aux_loss:0.0231 train_time:552139ms step_avg:133.53ms
step:4146/6250 train_loss:2.1749 aux_loss:0.0228 train_time:552272ms step_avg:133.53ms
step:4147/6250 train_loss:2.2309 aux_loss:0.0231 train_time:552403ms step_avg:133.53ms
step:4148/6250 train_loss:2.1674 aux_loss:0.0233 train_time:552534ms step_avg:133.53ms
step:4149/6250 train_loss:2.1664 aux_loss:0.0230 train_time:552668ms step_avg:133.53ms
step:4150/6250 train_loss:2.2176 aux_loss:0.0232 train_time:552800ms step_avg:133.53ms
step:4151/6250 train_loss:2.3224 aux_loss:0.0233 train_time:552932ms step_avg:133.53ms
step:4152/6250 train_loss:2.1910 aux_loss:0.0230 train_time:553063ms step_avg:133.53ms
step:4153/6250 train_loss:2.2278 aux_loss:0.0231 train_time:553194ms step_avg:133.52ms
step:4154/6250 train_loss:2.1138 aux_loss:0.0230 train_time:553327ms step_avg:133.52ms
step:4155/6250 train_loss:2.2322 aux_loss:0.0233 train_time:553459ms step_avg:133.52ms
step:4156/6250 train_loss:2.1063 aux_loss:0.0233 train_time:553591ms step_avg:133.52ms
step:4157/6250 train_loss:2.2055 aux_loss:0.0232 train_time:553723ms step_avg:133.52ms
step:4158/6250 train_loss:2.1882 aux_loss:0.0233 train_time:553854ms step_avg:133.52ms
step:4159/6250 train_loss:2.1070 aux_loss:0.0233 train_time:553987ms step_avg:133.52ms
step:4160/6250 train_loss:2.1463 aux_loss:0.0231 train_time:554118ms step_avg:133.52ms
step:4161/6250 train_loss:2.3088 aux_loss:0.0230 train_time:554251ms step_avg:133.52ms
step:4162/6250 train_loss:2.1714 aux_loss:0.0232 train_time:554382ms step_avg:133.52ms
step:4163/6250 train_loss:2.2036 aux_loss:0.0231 train_time:554514ms step_avg:133.52ms
step:4164/6250 train_loss:2.1951 aux_loss:0.0230 train_time:554646ms step_avg:133.52ms
step:4165/6250 train_loss:2.1429 aux_loss:0.0233 train_time:554778ms step_avg:133.52ms
step:4166/6250 train_loss:2.1836 aux_loss:0.0234 train_time:554910ms step_avg:133.52ms
step:4167/6250 train_loss:2.2511 aux_loss:0.0230 train_time:555042ms step_avg:133.52ms
step:4168/6250 train_loss:2.1563 aux_loss:0.0229 train_time:555173ms step_avg:133.52ms
step:4169/6250 train_loss:2.1787 aux_loss:0.0231 train_time:555306ms step_avg:133.52ms
step:4170/6250 train_loss:2.2289 aux_loss:0.0229 train_time:555437ms step_avg:133.52ms
step:4171/6250 train_loss:2.2380 aux_loss:0.0231 train_time:555577ms step_avg:133.52ms
step:4172/6250 train_loss:2.0620 aux_loss:0.0227 train_time:555723ms step_avg:133.52ms
step:4173/6250 train_loss:2.2310 aux_loss:0.0231 train_time:555861ms step_avg:133.52ms
step:4174/6250 train_loss:2.1723 aux_loss:0.0229 train_time:555992ms step_avg:133.52ms
step:4175/6250 train_loss:2.1496 aux_loss:0.0229 train_time:556127ms step_avg:133.52ms
step:4176/6250 train_loss:2.1772 aux_loss:0.0231 train_time:556260ms step_avg:133.52ms
step:4177/6250 train_loss:2.1794 aux_loss:0.0232 train_time:556392ms step_avg:133.52ms
step:4178/6250 train_loss:2.1745 aux_loss:0.0230 train_time:556524ms step_avg:133.52ms
step:4179/6250 train_loss:2.2126 aux_loss:0.0230 train_time:556656ms step_avg:133.52ms
step:4180/6250 train_loss:2.2370 aux_loss:0.0230 train_time:556789ms step_avg:133.52ms
step:4181/6250 train_loss:2.2475 aux_loss:0.0229 train_time:556920ms step_avg:133.52ms
step:4182/6250 train_loss:2.1917 aux_loss:0.0228 train_time:557052ms step_avg:133.52ms
step:4183/6250 train_loss:2.1615 aux_loss:0.0227 train_time:557183ms step_avg:133.52ms
step:4184/6250 train_loss:2.2017 aux_loss:0.0227 train_time:557316ms step_avg:133.52ms
step:4185/6250 train_loss:2.1015 aux_loss:0.0224 train_time:557448ms step_avg:133.52ms
step:4186/6250 train_loss:2.1756 aux_loss:0.0223 train_time:557580ms step_avg:133.52ms
step:4187/6250 train_loss:2.1867 aux_loss:0.0223 train_time:557713ms step_avg:133.52ms
step:4188/6250 train_loss:2.1347 aux_loss:0.0225 train_time:557845ms step_avg:133.52ms
step:4189/6250 train_loss:2.1656 aux_loss:0.0227 train_time:557976ms step_avg:133.52ms
step:4190/6250 train_loss:2.1535 aux_loss:0.0225 train_time:558108ms step_avg:133.52ms
step:4191/6250 train_loss:2.2479 aux_loss:0.0229 train_time:558240ms step_avg:133.52ms
step:4192/6250 train_loss:2.1782 aux_loss:0.0224 train_time:558372ms step_avg:133.52ms
step:4193/6250 train_loss:2.1433 aux_loss:0.0222 train_time:558504ms step_avg:133.52ms
step:4194/6250 train_loss:2.2166 aux_loss:0.0225 train_time:558636ms step_avg:133.52ms
step:4195/6250 train_loss:2.1890 aux_loss:0.0227 train_time:558769ms step_avg:133.52ms
step:4196/6250 train_loss:2.2013 aux_loss:0.0227 train_time:558900ms step_avg:133.52ms
step:4197/6250 train_loss:2.2263 aux_loss:0.0225 train_time:559032ms step_avg:133.52ms
step:4198/6250 train_loss:2.1791 aux_loss:0.0230 train_time:559164ms step_avg:133.52ms
step:4199/6250 train_loss:2.1812 aux_loss:0.0226 train_time:559296ms step_avg:133.52ms
step:4200/6250 train_loss:2.2359 aux_loss:0.0227 train_time:559429ms step_avg:133.52ms
step:4201/6250 train_loss:2.1287 aux_loss:0.0231 train_time:559560ms step_avg:133.51ms
step:4202/6250 train_loss:2.2741 aux_loss:0.0234 train_time:559692ms step_avg:133.51ms
step:4203/6250 train_loss:2.1626 aux_loss:0.0230 train_time:559824ms step_avg:133.51ms
step:4204/6250 train_loss:2.1129 aux_loss:0.0228 train_time:559956ms step_avg:133.51ms
step:4205/6250 train_loss:2.1848 aux_loss:0.0226 train_time:560088ms step_avg:133.51ms
step:4206/6250 train_loss:2.1933 aux_loss:0.0230 train_time:560220ms step_avg:133.51ms
step:4207/6250 train_loss:2.2463 aux_loss:0.0229 train_time:560352ms step_avg:133.51ms
step:4208/6250 train_loss:2.0691 aux_loss:0.0227 train_time:560485ms step_avg:133.51ms
step:4209/6250 train_loss:2.1502 aux_loss:0.0227 train_time:560616ms step_avg:133.51ms
step:4210/6250 train_loss:2.2074 aux_loss:0.0229 train_time:560750ms step_avg:133.51ms
step:4211/6250 train_loss:2.1848 aux_loss:0.0228 train_time:560881ms step_avg:133.51ms
step:4212/6250 train_loss:2.0941 aux_loss:0.0224 train_time:561013ms step_avg:133.51ms
step:4213/6250 train_loss:2.2684 aux_loss:0.0227 train_time:561145ms step_avg:133.51ms
step:4214/6250 train_loss:2.2114 aux_loss:0.0233 train_time:561277ms step_avg:133.51ms
step:4215/6250 train_loss:2.2180 aux_loss:0.0233 train_time:561409ms step_avg:133.51ms
step:4216/6250 train_loss:2.2107 aux_loss:0.0227 train_time:561540ms step_avg:133.51ms
step:4217/6250 train_loss:2.1650 aux_loss:0.0226 train_time:561673ms step_avg:133.51ms
step:4218/6250 train_loss:2.2371 aux_loss:0.0227 train_time:561805ms step_avg:133.51ms
step:4219/6250 train_loss:2.1206 aux_loss:0.0223 train_time:561936ms step_avg:133.51ms
step:4220/6250 train_loss:2.2065 aux_loss:0.0226 train_time:562069ms step_avg:133.51ms
step:4221/6250 train_loss:2.1614 aux_loss:0.0225 train_time:562201ms step_avg:133.51ms
step:4222/6250 train_loss:2.3154 aux_loss:0.0225 train_time:562333ms step_avg:133.51ms
step:4223/6250 train_loss:2.2113 aux_loss:0.0227 train_time:562465ms step_avg:133.51ms
step:4224/6250 train_loss:2.1787 aux_loss:0.0228 train_time:562597ms step_avg:133.51ms
step:4225/6250 train_loss:2.2431 aux_loss:0.0225 train_time:562730ms step_avg:133.51ms
step:4226/6250 train_loss:2.1410 aux_loss:0.0226 train_time:562862ms step_avg:133.51ms
step:4227/6250 train_loss:2.2221 aux_loss:0.0225 train_time:562994ms step_avg:133.51ms
step:4228/6250 train_loss:2.2456 aux_loss:0.0223 train_time:563126ms step_avg:133.51ms
step:4229/6250 train_loss:2.2285 aux_loss:0.0223 train_time:563259ms step_avg:133.51ms
step:4230/6250 train_loss:2.2727 aux_loss:0.0224 train_time:563390ms step_avg:133.50ms
step:4231/6250 train_loss:2.1786 aux_loss:0.0228 train_time:563522ms step_avg:133.50ms
step:4232/6250 train_loss:2.1657 aux_loss:0.0223 train_time:563653ms step_avg:133.50ms
step:4233/6250 train_loss:2.2215 aux_loss:0.0223 train_time:563794ms step_avg:133.51ms
step:4234/6250 train_loss:2.2032 aux_loss:0.0224 train_time:563939ms step_avg:133.51ms
step:4235/6250 train_loss:2.2032 aux_loss:0.0230 train_time:564083ms step_avg:133.51ms
step:4236/6250 train_loss:2.1673 aux_loss:0.0231 train_time:564216ms step_avg:133.51ms
step:4237/6250 train_loss:2.1553 aux_loss:0.0227 train_time:564349ms step_avg:133.51ms
step:4238/6250 train_loss:2.2046 aux_loss:0.0226 train_time:564481ms step_avg:133.51ms
step:4239/6250 train_loss:2.2349 aux_loss:0.0224 train_time:564613ms step_avg:133.51ms
step:4240/6250 train_loss:2.1628 aux_loss:0.0222 train_time:564747ms step_avg:133.51ms
step:4241/6250 train_loss:2.2692 aux_loss:0.0225 train_time:564879ms step_avg:133.51ms
step:4242/6250 train_loss:2.1709 aux_loss:0.0225 train_time:565010ms step_avg:133.51ms
step:4243/6250 train_loss:2.1928 aux_loss:0.0227 train_time:565142ms step_avg:133.51ms
step:4244/6250 train_loss:2.1424 aux_loss:0.0224 train_time:565274ms step_avg:133.51ms
step:4245/6250 train_loss:2.2637 aux_loss:0.0226 train_time:565406ms step_avg:133.51ms
step:4246/6250 train_loss:2.2142 aux_loss:0.0226 train_time:565537ms step_avg:133.51ms
step:4247/6250 train_loss:2.2841 aux_loss:0.0225 train_time:565670ms step_avg:133.51ms
step:4248/6250 train_loss:2.2580 aux_loss:0.0225 train_time:565815ms step_avg:133.51ms
step:4249/6250 train_loss:2.1945 aux_loss:0.0228 train_time:565959ms step_avg:133.51ms
step:4250/6250 train_loss:2.1438 aux_loss:0.0228 train_time:566094ms step_avg:133.51ms
step:4251/6250 train_loss:2.2290 aux_loss:0.0226 train_time:566227ms step_avg:133.51ms
step:4252/6250 train_loss:2.2267 aux_loss:0.0225 train_time:566359ms step_avg:133.51ms
step:4253/6250 train_loss:2.2191 aux_loss:0.0226 train_time:566490ms step_avg:133.51ms
step:4254/6250 train_loss:2.2019 aux_loss:0.0229 train_time:566623ms step_avg:133.51ms
step:4255/6250 train_loss:2.1777 aux_loss:0.0231 train_time:566755ms step_avg:133.51ms
step:4256/6250 train_loss:2.2056 aux_loss:0.0229 train_time:566888ms step_avg:133.51ms
step:4257/6250 train_loss:2.1338 aux_loss:0.0228 train_time:567018ms step_avg:133.51ms
step:4258/6250 train_loss:2.2007 aux_loss:0.0229 train_time:567150ms step_avg:133.51ms
step:4259/6250 train_loss:2.2201 aux_loss:0.0224 train_time:567283ms step_avg:133.51ms
step:4260/6250 train_loss:2.2384 aux_loss:0.0225 train_time:567415ms step_avg:133.51ms
step:4261/6250 train_loss:2.0954 aux_loss:0.0224 train_time:567548ms step_avg:133.51ms
step:4262/6250 train_loss:2.1621 aux_loss:0.0220 train_time:567687ms step_avg:133.51ms
step:4263/6250 train_loss:2.0632 aux_loss:0.0223 train_time:567822ms step_avg:133.51ms
step:4264/6250 train_loss:2.2443 aux_loss:0.0224 train_time:567953ms step_avg:133.51ms
step:4265/6250 train_loss:2.1736 aux_loss:0.0225 train_time:568086ms step_avg:133.51ms
step:4266/6250 train_loss:2.1768 aux_loss:0.0225 train_time:568219ms step_avg:133.51ms
step:4267/6250 train_loss:2.2516 aux_loss:0.0224 train_time:568350ms step_avg:133.51ms
step:4268/6250 train_loss:2.2344 aux_loss:0.0226 train_time:568482ms step_avg:133.51ms
step:4269/6250 train_loss:2.2127 aux_loss:0.0226 train_time:568614ms step_avg:133.51ms
step:4270/6250 train_loss:2.2879 aux_loss:0.0226 train_time:568751ms step_avg:133.51ms
step:4271/6250 train_loss:2.1769 aux_loss:0.0226 train_time:568889ms step_avg:133.51ms
step:4272/6250 train_loss:2.2707 aux_loss:0.0232 train_time:569025ms step_avg:133.51ms
step:4273/6250 train_loss:2.1295 aux_loss:0.0228 train_time:569162ms step_avg:133.51ms
step:4274/6250 train_loss:2.2553 aux_loss:0.0227 train_time:569300ms step_avg:133.51ms
step:4275/6250 train_loss:2.1027 aux_loss:0.0224 train_time:569432ms step_avg:133.51ms
step:4276/6250 train_loss:2.2185 aux_loss:0.0224 train_time:569572ms step_avg:133.51ms
step:4277/6250 train_loss:2.1249 aux_loss:0.0225 train_time:569714ms step_avg:133.52ms
step:4278/6250 train_loss:2.2782 aux_loss:0.0224 train_time:569855ms step_avg:133.52ms
step:4279/6250 train_loss:2.2112 aux_loss:0.0223 train_time:570000ms step_avg:133.52ms
step:4280/6250 train_loss:2.1711 aux_loss:0.0223 train_time:570137ms step_avg:133.52ms
step:4281/6250 train_loss:2.1575 aux_loss:0.0226 train_time:570271ms step_avg:133.52ms
step:4282/6250 train_loss:2.1774 aux_loss:0.0226 train_time:570412ms step_avg:133.52ms
step:4283/6250 train_loss:2.2318 aux_loss:0.0230 train_time:570548ms step_avg:133.52ms
step:4284/6250 train_loss:2.1971 aux_loss:0.0229 train_time:570683ms step_avg:133.52ms
step:4285/6250 train_loss:2.1230 aux_loss:0.0224 train_time:570818ms step_avg:133.52ms
step:4286/6250 train_loss:2.1821 aux_loss:0.0225 train_time:570951ms step_avg:133.52ms
step:4287/6250 train_loss:2.1727 aux_loss:0.0227 train_time:571084ms step_avg:133.52ms
step:4288/6250 train_loss:2.2181 aux_loss:0.0227 train_time:571217ms step_avg:133.52ms
step:4289/6250 train_loss:2.1057 aux_loss:0.0226 train_time:571353ms step_avg:133.52ms
step:4290/6250 train_loss:2.1791 aux_loss:0.0224 train_time:571499ms step_avg:133.53ms
step:4291/6250 train_loss:2.2006 aux_loss:0.0224 train_time:571644ms step_avg:133.53ms
step:4292/6250 train_loss:2.1504 aux_loss:0.0226 train_time:571787ms step_avg:133.53ms
step:4293/6250 train_loss:2.1563 aux_loss:0.0225 train_time:571925ms step_avg:133.53ms
step:4294/6250 train_loss:2.1493 aux_loss:0.0224 train_time:572056ms step_avg:133.53ms
step:4295/6250 train_loss:2.1794 aux_loss:0.0222 train_time:572187ms step_avg:133.53ms
step:4296/6250 train_loss:2.2352 aux_loss:0.0223 train_time:572332ms step_avg:133.54ms
step:4297/6250 train_loss:2.1447 aux_loss:0.0222 train_time:572469ms step_avg:133.54ms
step:4298/6250 train_loss:2.2099 aux_loss:0.0222 train_time:572602ms step_avg:133.54ms
step:4299/6250 train_loss:2.1300 aux_loss:0.0227 train_time:572749ms step_avg:133.54ms
step:4300/6250 train_loss:2.2566 aux_loss:0.0223 train_time:572891ms step_avg:133.54ms
step:4301/6250 train_loss:2.1614 aux_loss:0.0223 train_time:573025ms step_avg:133.54ms
step:4302/6250 train_loss:2.1972 aux_loss:0.0225 train_time:573168ms step_avg:133.54ms
step:4303/6250 train_loss:2.1944 aux_loss:0.0226 train_time:573313ms step_avg:133.55ms
step:4304/6250 train_loss:2.1382 aux_loss:0.0226 train_time:573455ms step_avg:133.55ms
step:4305/6250 train_loss:2.1871 aux_loss:0.0229 train_time:573594ms step_avg:133.55ms
step:4306/6250 train_loss:2.1081 aux_loss:0.0225 train_time:573731ms step_avg:133.55ms
step:4307/6250 train_loss:2.1957 aux_loss:0.0225 train_time:573866ms step_avg:133.55ms
step:4308/6250 train_loss:2.2147 aux_loss:0.0230 train_time:573998ms step_avg:133.55ms
step:4309/6250 train_loss:2.1805 aux_loss:0.0227 train_time:574135ms step_avg:133.55ms
step:4310/6250 train_loss:2.2029 aux_loss:0.0224 train_time:574275ms step_avg:133.55ms
step:4311/6250 train_loss:2.0940 aux_loss:0.0225 train_time:574420ms step_avg:133.55ms
step:4312/6250 train_loss:2.1716 aux_loss:0.0226 train_time:574565ms step_avg:133.56ms
step:4313/6250 train_loss:2.2411 aux_loss:0.0224 train_time:574704ms step_avg:133.56ms
step:4314/6250 train_loss:2.1982 aux_loss:0.0225 train_time:574845ms step_avg:133.56ms
step:4315/6250 train_loss:2.3150 aux_loss:0.0227 train_time:574990ms step_avg:133.56ms
step:4316/6250 train_loss:2.1528 aux_loss:0.0225 train_time:575122ms step_avg:133.56ms
step:4317/6250 train_loss:2.2340 aux_loss:0.0222 train_time:575256ms step_avg:133.56ms
step:4318/6250 train_loss:2.0659 aux_loss:0.0221 train_time:575395ms step_avg:133.56ms
step:4319/6250 train_loss:2.1989 aux_loss:0.0223 train_time:575527ms step_avg:133.56ms
step:4320/6250 train_loss:2.2023 aux_loss:0.0222 train_time:575660ms step_avg:133.56ms
step:4321/6250 train_loss:2.1021 aux_loss:0.0221 train_time:575793ms step_avg:133.56ms
step:4322/6250 train_loss:2.1962 aux_loss:0.0224 train_time:575925ms step_avg:133.56ms
step:4323/6250 train_loss:2.1188 aux_loss:0.0223 train_time:576059ms step_avg:133.56ms
step:4324/6250 train_loss:2.2564 aux_loss:0.0226 train_time:576198ms step_avg:133.56ms
step:4325/6250 train_loss:2.2404 aux_loss:0.0228 train_time:576335ms step_avg:133.57ms
step:4326/6250 train_loss:2.1976 aux_loss:0.0224 train_time:576467ms step_avg:133.57ms
step:4327/6250 train_loss:2.2605 aux_loss:0.0224 train_time:576599ms step_avg:133.56ms
step:4328/6250 train_loss:2.2002 aux_loss:0.0224 train_time:576733ms step_avg:133.56ms
step:4329/6250 train_loss:2.1005 aux_loss:0.0224 train_time:576866ms step_avg:133.56ms
step:4330/6250 train_loss:2.1858 aux_loss:0.0229 train_time:576998ms step_avg:133.56ms
step:4331/6250 train_loss:2.2091 aux_loss:0.0227 train_time:577129ms step_avg:133.56ms
step:4332/6250 train_loss:2.1641 aux_loss:0.0224 train_time:577262ms step_avg:133.56ms
step:4333/6250 train_loss:2.1649 aux_loss:0.0220 train_time:577393ms step_avg:133.56ms
step:4334/6250 train_loss:2.0806 aux_loss:0.0220 train_time:577525ms step_avg:133.56ms
step:4335/6250 train_loss:2.2774 aux_loss:0.0220 train_time:577657ms step_avg:133.56ms
step:4336/6250 train_loss:2.1506 aux_loss:0.0219 train_time:577790ms step_avg:133.56ms
step:4337/6250 train_loss:2.1374 aux_loss:0.0222 train_time:577922ms step_avg:133.56ms
step:4338/6250 train_loss:2.2458 aux_loss:0.0225 train_time:578054ms step_avg:133.56ms
step:4339/6250 train_loss:2.2662 aux_loss:0.0222 train_time:578185ms step_avg:133.56ms
step:4340/6250 train_loss:2.1822 aux_loss:0.0222 train_time:578317ms step_avg:133.56ms
step:4341/6250 train_loss:2.1607 aux_loss:0.0224 train_time:578448ms step_avg:133.56ms
step:4342/6250 train_loss:2.1773 aux_loss:0.0220 train_time:578580ms step_avg:133.56ms
step:4343/6250 train_loss:2.2343 aux_loss:0.0219 train_time:578713ms step_avg:133.56ms
step:4344/6250 train_loss:2.1780 aux_loss:0.0221 train_time:578845ms step_avg:133.56ms
step:4345/6250 train_loss:2.1793 aux_loss:0.0219 train_time:578978ms step_avg:133.56ms
step:4346/6250 train_loss:2.1907 aux_loss:0.0221 train_time:579109ms step_avg:133.56ms
step:4347/6250 train_loss:2.2357 aux_loss:0.0221 train_time:579242ms step_avg:133.56ms
step:4348/6250 train_loss:2.1269 aux_loss:0.0219 train_time:579374ms step_avg:133.56ms
step:4349/6250 train_loss:2.1794 aux_loss:0.0219 train_time:579505ms step_avg:133.56ms
step:4350/6250 train_loss:2.2914 aux_loss:0.0224 train_time:579637ms step_avg:133.56ms
step:4351/6250 train_loss:2.1725 aux_loss:0.0223 train_time:579770ms step_avg:133.56ms
step:4352/6250 train_loss:2.1687 aux_loss:0.0220 train_time:579902ms step_avg:133.56ms
step:4353/6250 train_loss:2.1467 aux_loss:0.0221 train_time:580034ms step_avg:133.56ms
step:4354/6250 train_loss:2.1122 aux_loss:0.0221 train_time:580165ms step_avg:133.56ms
step:4355/6250 train_loss:2.1993 aux_loss:0.0220 train_time:580298ms step_avg:133.56ms
step:4356/6250 train_loss:2.0656 aux_loss:0.0217 train_time:580430ms step_avg:133.55ms
step:4357/6250 train_loss:2.1657 aux_loss:0.0218 train_time:580561ms step_avg:133.55ms
step:4358/6250 train_loss:2.2517 aux_loss:0.0221 train_time:580694ms step_avg:133.55ms
step:4359/6250 train_loss:2.2511 aux_loss:0.0219 train_time:580826ms step_avg:133.55ms
step:4360/6250 train_loss:2.1668 aux_loss:0.0225 train_time:580957ms step_avg:133.55ms
step:4361/6250 train_loss:2.2302 aux_loss:0.0223 train_time:581088ms step_avg:133.55ms
step:4362/6250 train_loss:2.1451 aux_loss:0.0223 train_time:581221ms step_avg:133.55ms
step:4363/6250 train_loss:2.1880 aux_loss:0.0220 train_time:581353ms step_avg:133.55ms
step:4364/6250 train_loss:2.2402 aux_loss:0.0221 train_time:581484ms step_avg:133.55ms
step:4365/6250 train_loss:2.1832 aux_loss:0.0221 train_time:581616ms step_avg:133.55ms
step:4366/6250 train_loss:2.2283 aux_loss:0.0218 train_time:581748ms step_avg:133.55ms
step:4367/6250 train_loss:2.0908 aux_loss:0.0218 train_time:581881ms step_avg:133.55ms
step:4368/6250 train_loss:2.1909 aux_loss:0.0223 train_time:582013ms step_avg:133.55ms
step:4369/6250 train_loss:2.0616 aux_loss:0.0221 train_time:582145ms step_avg:133.55ms
step:4370/6250 train_loss:2.1371 aux_loss:0.0222 train_time:582277ms step_avg:133.55ms
step:4371/6250 train_loss:2.2380 aux_loss:0.0221 train_time:582409ms step_avg:133.55ms
step:4372/6250 train_loss:2.1786 aux_loss:0.0219 train_time:582541ms step_avg:133.55ms
step:4373/6250 train_loss:2.2259 aux_loss:0.0221 train_time:582673ms step_avg:133.55ms
step:4374/6250 train_loss:2.1349 aux_loss:0.0222 train_time:582806ms step_avg:133.55ms
step:4375/6250 train_loss:2.1103 aux_loss:0.0220 train_time:582947ms step_avg:133.55ms
step:4376/6250 train_loss:2.1271 aux_loss:0.0220 train_time:583087ms step_avg:133.55ms
step:4377/6250 train_loss:2.1765 aux_loss:0.0222 train_time:583223ms step_avg:133.55ms
step:4378/6250 train_loss:2.1335 aux_loss:0.0220 train_time:583357ms step_avg:133.55ms
step:4379/6250 train_loss:2.1571 aux_loss:0.0218 train_time:583495ms step_avg:133.55ms
step:4380/6250 train_loss:2.1322 aux_loss:0.0217 train_time:583634ms step_avg:133.55ms
step:4381/6250 train_loss:2.1427 aux_loss:0.0217 train_time:583772ms step_avg:133.56ms
step:4382/6250 train_loss:2.1400 aux_loss:0.0218 train_time:583910ms step_avg:133.56ms
step:4383/6250 train_loss:2.0463 aux_loss:0.0221 train_time:584048ms step_avg:133.56ms
step:4384/6250 train_loss:2.1623 aux_loss:0.0223 train_time:584183ms step_avg:133.56ms
step:4385/6250 train_loss:2.1546 aux_loss:0.0223 train_time:584315ms step_avg:133.56ms
step:4386/6250 train_loss:2.1496 aux_loss:0.0219 train_time:584452ms step_avg:133.56ms
step:4387/6250 train_loss:2.0364 aux_loss:0.0221 train_time:584588ms step_avg:133.56ms
step:4388/6250 train_loss:2.2347 aux_loss:0.0225 train_time:584722ms step_avg:133.56ms
step:4389/6250 train_loss:2.1717 aux_loss:0.0219 train_time:584855ms step_avg:133.56ms
step:4390/6250 train_loss:2.1648 aux_loss:0.0223 train_time:584990ms step_avg:133.56ms
step:4391/6250 train_loss:2.2110 aux_loss:0.0230 train_time:585128ms step_avg:133.56ms
step:4392/6250 train_loss:2.2157 aux_loss:0.0222 train_time:585261ms step_avg:133.56ms
step:4393/6250 train_loss:2.1686 aux_loss:0.0223 train_time:585394ms step_avg:133.56ms
step:4394/6250 train_loss:2.2646 aux_loss:0.0222 train_time:585526ms step_avg:133.56ms
step:4395/6250 train_loss:2.1922 aux_loss:0.0221 train_time:585658ms step_avg:133.56ms
step:4396/6250 train_loss:2.1830 aux_loss:0.0221 train_time:585792ms step_avg:133.56ms
step:4397/6250 train_loss:2.1626 aux_loss:0.0219 train_time:585924ms step_avg:133.56ms
step:4398/6250 train_loss:2.2024 aux_loss:0.0221 train_time:586056ms step_avg:133.56ms
step:4399/6250 train_loss:2.1330 aux_loss:0.0218 train_time:586188ms step_avg:133.56ms
step:4400/6250 train_loss:2.1853 aux_loss:0.0218 train_time:586322ms step_avg:133.56ms
step:4401/6250 train_loss:2.2561 aux_loss:0.0222 train_time:586454ms step_avg:133.56ms
step:4402/6250 train_loss:2.1925 aux_loss:0.0219 train_time:586586ms step_avg:133.56ms
step:4403/6250 train_loss:2.1284 aux_loss:0.0217 train_time:586719ms step_avg:133.56ms
step:4404/6250 train_loss:2.1249 aux_loss:0.0221 train_time:586852ms step_avg:133.56ms
step:4405/6250 train_loss:2.2391 aux_loss:0.0222 train_time:586984ms step_avg:133.56ms
step:4406/6250 train_loss:2.2827 aux_loss:0.0221 train_time:587116ms step_avg:133.56ms
step:4407/6250 train_loss:2.1863 aux_loss:0.0221 train_time:587250ms step_avg:133.56ms
step:4408/6250 train_loss:2.1556 aux_loss:0.0221 train_time:587382ms step_avg:133.56ms
step:4409/6250 train_loss:2.1458 aux_loss:0.0219 train_time:587514ms step_avg:133.56ms
step:4410/6250 train_loss:2.2057 aux_loss:0.0218 train_time:587646ms step_avg:133.56ms
step:4411/6250 train_loss:2.2017 aux_loss:0.0219 train_time:587780ms step_avg:133.56ms
step:4412/6250 train_loss:2.2443 aux_loss:0.0217 train_time:587912ms step_avg:133.56ms
step:4413/6250 train_loss:2.0746 aux_loss:0.0216 train_time:588045ms step_avg:133.56ms
step:4414/6250 train_loss:2.1462 aux_loss:0.0215 train_time:588179ms step_avg:133.56ms
step:4415/6250 train_loss:2.2183 aux_loss:0.0215 train_time:588312ms step_avg:133.56ms
step:4416/6250 train_loss:2.1308 aux_loss:0.0216 train_time:588444ms step_avg:133.56ms
step:4417/6250 train_loss:2.1966 aux_loss:0.0218 train_time:588577ms step_avg:133.56ms
step:4418/6250 train_loss:2.0932 aux_loss:0.0218 train_time:588711ms step_avg:133.56ms
step:4419/6250 train_loss:2.1772 aux_loss:0.0218 train_time:588844ms step_avg:133.56ms
step:4420/6250 train_loss:2.2732 aux_loss:0.0220 train_time:588977ms step_avg:133.55ms
step:4421/6250 train_loss:2.2935 aux_loss:0.0221 train_time:589110ms step_avg:133.55ms
step:4422/6250 train_loss:2.2485 aux_loss:0.0218 train_time:589243ms step_avg:133.55ms
step:4423/6250 train_loss:2.2647 aux_loss:0.0221 train_time:589377ms step_avg:133.55ms
step:4424/6250 train_loss:2.1539 aux_loss:0.0220 train_time:589509ms step_avg:133.55ms
step:4425/6250 train_loss:2.1367 aux_loss:0.0223 train_time:589642ms step_avg:133.55ms
step:4426/6250 train_loss:2.0511 aux_loss:0.0219 train_time:589775ms step_avg:133.55ms
step:4427/6250 train_loss:2.1956 aux_loss:0.0221 train_time:589909ms step_avg:133.55ms
step:4428/6250 train_loss:2.1402 aux_loss:0.0219 train_time:590041ms step_avg:133.55ms
step:4429/6250 train_loss:2.2779 aux_loss:0.0220 train_time:590173ms step_avg:133.55ms
step:4430/6250 train_loss:2.1567 aux_loss:0.0223 train_time:590306ms step_avg:133.55ms
step:4431/6250 train_loss:2.1224 aux_loss:0.0219 train_time:590439ms step_avg:133.55ms
step:4432/6250 train_loss:2.2982 aux_loss:0.0224 train_time:590571ms step_avg:133.55ms
step:4433/6250 train_loss:2.1263 aux_loss:0.0217 train_time:590705ms step_avg:133.55ms
step:4434/6250 train_loss:2.2174 aux_loss:0.0219 train_time:590839ms step_avg:133.55ms
step:4435/6250 train_loss:2.2085 aux_loss:0.0217 train_time:590971ms step_avg:133.55ms
step:4436/6250 train_loss:2.1636 aux_loss:0.0214 train_time:591104ms step_avg:133.55ms
step:4437/6250 train_loss:2.1175 aux_loss:0.0216 train_time:591237ms step_avg:133.55ms
step:4438/6250 train_loss:2.1358 aux_loss:0.0219 train_time:591370ms step_avg:133.55ms
step:4439/6250 train_loss:2.0875 aux_loss:0.0222 train_time:591504ms step_avg:133.55ms
step:4440/6250 train_loss:2.2714 aux_loss:0.0219 train_time:591645ms step_avg:133.55ms
step:4441/6250 train_loss:2.1138 aux_loss:0.0217 train_time:591785ms step_avg:133.56ms
step:4442/6250 train_loss:2.0872 aux_loss:0.0216 train_time:591923ms step_avg:133.56ms
step:4443/6250 train_loss:2.2196 aux_loss:0.0218 train_time:592062ms step_avg:133.56ms
step:4444/6250 train_loss:2.2392 aux_loss:0.0219 train_time:592200ms step_avg:133.56ms
step:4445/6250 train_loss:2.1508 aux_loss:0.0217 train_time:592337ms step_avg:133.56ms
step:4446/6250 train_loss:2.1071 aux_loss:0.0216 train_time:592475ms step_avg:133.56ms
step:4447/6250 train_loss:2.2611 aux_loss:0.0222 train_time:592614ms step_avg:133.56ms
step:4448/6250 train_loss:2.0794 aux_loss:0.0221 train_time:592752ms step_avg:133.56ms
step:4449/6250 train_loss:2.0671 aux_loss:0.0218 train_time:592890ms step_avg:133.56ms
step:4450/6250 train_loss:2.0819 aux_loss:0.0216 train_time:593028ms step_avg:133.56ms
step:4451/6250 train_loss:2.1167 aux_loss:0.0218 train_time:593165ms step_avg:133.57ms
step:4452/6250 train_loss:2.3646 aux_loss:0.0218 train_time:593303ms step_avg:133.57ms
step:4453/6250 train_loss:2.3010 aux_loss:0.0216 train_time:593442ms step_avg:133.57ms
step:4454/6250 train_loss:2.1990 aux_loss:0.0218 train_time:593582ms step_avg:133.57ms
step:4455/6250 train_loss:2.1737 aux_loss:0.0218 train_time:593721ms step_avg:133.57ms
step:4456/6250 train_loss:2.2149 aux_loss:0.0220 train_time:593859ms step_avg:133.57ms
step:4457/6250 train_loss:2.1561 aux_loss:0.0218 train_time:593997ms step_avg:133.57ms
step:4458/6250 train_loss:2.2541 aux_loss:0.0223 train_time:594136ms step_avg:133.57ms
step:4459/6250 train_loss:2.1636 aux_loss:0.0222 train_time:594275ms step_avg:133.57ms
step:4460/6250 train_loss:2.1534 aux_loss:0.0220 train_time:594413ms step_avg:133.58ms
step:4461/6250 train_loss:2.0750 aux_loss:0.0217 train_time:594552ms step_avg:133.58ms
step:4462/6250 train_loss:2.2386 aux_loss:0.0218 train_time:594691ms step_avg:133.58ms
step:4463/6250 train_loss:2.1894 aux_loss:0.0219 train_time:594828ms step_avg:133.58ms
step:4464/6250 train_loss:2.1247 aux_loss:0.0217 train_time:594967ms step_avg:133.58ms
step:4465/6250 train_loss:2.3002 aux_loss:0.0216 train_time:595105ms step_avg:133.58ms
step:4466/6250 train_loss:2.2365 aux_loss:0.0220 train_time:595243ms step_avg:133.58ms
step:4467/6250 train_loss:2.1306 aux_loss:0.0218 train_time:595381ms step_avg:133.58ms
step:4468/6250 train_loss:2.1706 aux_loss:0.0214 train_time:595519ms step_avg:133.58ms
step:4469/6250 train_loss:2.1673 aux_loss:0.0214 train_time:595657ms step_avg:133.59ms
step:4470/6250 train_loss:2.1640 aux_loss:0.0217 train_time:595796ms step_avg:133.59ms
step:4471/6250 train_loss:2.1340 aux_loss:0.0215 train_time:595938ms step_avg:133.59ms
step:4472/6250 train_loss:2.1899 aux_loss:0.0214 train_time:596077ms step_avg:133.59ms
step:4473/6250 train_loss:2.1926 aux_loss:0.0218 train_time:596216ms step_avg:133.59ms
step:4474/6250 train_loss:2.2075 aux_loss:0.0218 train_time:596354ms step_avg:133.59ms
step:4475/6250 train_loss:2.1443 aux_loss:0.0216 train_time:596491ms step_avg:133.59ms
step:4476/6250 train_loss:2.2367 aux_loss:0.0215 train_time:596630ms step_avg:133.59ms
step:4477/6250 train_loss:2.1954 aux_loss:0.0216 train_time:596769ms step_avg:133.60ms
step:4478/6250 train_loss:2.2740 aux_loss:0.0216 train_time:596908ms step_avg:133.60ms
step:4479/6250 train_loss:2.1499 aux_loss:0.0216 train_time:597047ms step_avg:133.60ms
step:4480/6250 train_loss:2.1579 aux_loss:0.0217 train_time:597185ms step_avg:133.60ms
step:4481/6250 train_loss:2.1660 aux_loss:0.0218 train_time:597324ms step_avg:133.60ms
step:4482/6250 train_loss:2.1942 aux_loss:0.0216 train_time:597463ms step_avg:133.60ms
step:4483/6250 train_loss:2.2173 aux_loss:0.0215 train_time:597601ms step_avg:133.60ms
step:4484/6250 train_loss:2.2334 aux_loss:0.0213 train_time:597740ms step_avg:133.60ms
step:4485/6250 train_loss:2.1761 aux_loss:0.0212 train_time:597879ms step_avg:133.60ms
step:4486/6250 train_loss:2.1219 aux_loss:0.0213 train_time:598018ms step_avg:133.61ms
step:4487/6250 train_loss:2.3101 aux_loss:0.0216 train_time:598156ms step_avg:133.61ms
step:4488/6250 train_loss:2.2821 aux_loss:0.0214 train_time:598295ms step_avg:133.61ms
step:4489/6250 train_loss:2.2450 aux_loss:0.0217 train_time:598433ms step_avg:133.61ms
step:4490/6250 train_loss:2.1764 aux_loss:0.0218 train_time:598571ms step_avg:133.61ms
step:4491/6250 train_loss:2.1278 aux_loss:0.0214 train_time:598709ms step_avg:133.61ms
step:4492/6250 train_loss:2.2367 aux_loss:0.0217 train_time:598847ms step_avg:133.61ms
step:4493/6250 train_loss:2.2214 aux_loss:0.0216 train_time:598985ms step_avg:133.61ms
step:4494/6250 train_loss:2.2501 aux_loss:0.0216 train_time:599124ms step_avg:133.61ms
step:4495/6250 train_loss:2.2446 aux_loss:0.0215 train_time:599261ms step_avg:133.61ms
step:4496/6250 train_loss:2.2547 aux_loss:0.0217 train_time:599399ms step_avg:133.62ms
step:4497/6250 train_loss:2.2495 aux_loss:0.0215 train_time:599538ms step_avg:133.62ms
step:4498/6250 train_loss:2.2267 aux_loss:0.0216 train_time:599676ms step_avg:133.62ms
step:4499/6250 train_loss:2.2515 aux_loss:0.0216 train_time:599814ms step_avg:133.62ms
step:4500/6250 train_loss:2.1934 aux_loss:0.0217 train_time:599953ms step_avg:133.62ms
step:4501/6250 train_loss:2.1962 aux_loss:0.0219 train_time:600092ms step_avg:133.62ms
step:4502/6250 train_loss:2.1178 aux_loss:0.0214 train_time:600230ms step_avg:133.62ms
step:4503/6250 train_loss:2.1123 aux_loss:0.0214 train_time:600368ms step_avg:133.62ms
step:4504/6250 train_loss:2.2015 aux_loss:0.0218 train_time:600506ms step_avg:133.62ms
step:4505/6250 train_loss:2.2281 aux_loss:0.0221 train_time:600645ms step_avg:133.63ms
step:4506/6250 train_loss:2.1318 aux_loss:0.0215 train_time:600783ms step_avg:133.63ms
step:4507/6250 train_loss:2.2073 aux_loss:0.0219 train_time:600922ms step_avg:133.63ms
step:4508/6250 train_loss:2.1506 aux_loss:0.0213 train_time:601059ms step_avg:133.63ms
step:4509/6250 train_loss:2.1457 aux_loss:0.0221 train_time:601199ms step_avg:133.63ms
step:4510/6250 train_loss:2.1763 aux_loss:0.0218 train_time:601337ms step_avg:133.63ms
step:4511/6250 train_loss:2.2092 aux_loss:0.0216 train_time:601474ms step_avg:133.63ms
step:4512/6250 train_loss:2.1526 aux_loss:0.0217 train_time:601612ms step_avg:133.63ms
step:4513/6250 train_loss:2.2740 aux_loss:0.0216 train_time:601751ms step_avg:133.63ms
step:4514/6250 train_loss:2.1339 aux_loss:0.0215 train_time:601889ms step_avg:133.63ms
step:4515/6250 train_loss:2.2760 aux_loss:0.0218 train_time:602026ms step_avg:133.64ms
step:4516/6250 train_loss:2.2159 aux_loss:0.0217 train_time:602166ms step_avg:133.64ms
step:4517/6250 train_loss:2.2534 aux_loss:0.0218 train_time:602304ms step_avg:133.64ms
step:4518/6250 train_loss:2.1303 aux_loss:0.0217 train_time:602442ms step_avg:133.64ms
step:4519/6250 train_loss:2.1157 aux_loss:0.0215 train_time:602580ms step_avg:133.64ms
step:4520/6250 train_loss:2.2039 aux_loss:0.0215 train_time:602718ms step_avg:133.64ms
step:4521/6250 train_loss:2.2297 aux_loss:0.0215 train_time:602856ms step_avg:133.64ms
step:4522/6250 train_loss:2.2282 aux_loss:0.0216 train_time:602995ms step_avg:133.64ms
step:4523/6250 train_loss:2.2711 aux_loss:0.0215 train_time:603133ms step_avg:133.64ms
step:4524/6250 train_loss:2.2860 aux_loss:0.0218 train_time:603271ms step_avg:133.64ms
step:4525/6250 train_loss:2.1582 aux_loss:0.0215 train_time:603409ms step_avg:133.65ms
step:4526/6250 train_loss:2.1802 aux_loss:0.0217 train_time:603547ms step_avg:133.65ms
step:4527/6250 train_loss:2.2145 aux_loss:0.0217 train_time:603685ms step_avg:133.65ms
step:4528/6250 train_loss:2.1540 aux_loss:0.0215 train_time:603823ms step_avg:133.65ms
step:4529/6250 train_loss:2.1604 aux_loss:0.0215 train_time:603961ms step_avg:133.65ms
step:4530/6250 train_loss:2.1007 aux_loss:0.0215 train_time:604098ms step_avg:133.65ms
step:4531/6250 train_loss:2.1590 aux_loss:0.0214 train_time:604237ms step_avg:133.65ms
step:4532/6250 train_loss:2.1526 aux_loss:0.0215 train_time:604375ms step_avg:133.65ms
step:4533/6250 train_loss:2.1248 aux_loss:0.0214 train_time:604512ms step_avg:133.65ms
step:4534/6250 train_loss:2.1363 aux_loss:0.0216 train_time:604650ms step_avg:133.65ms
step:4535/6250 train_loss:2.1243 aux_loss:0.0214 train_time:604789ms step_avg:133.66ms
step:4536/6250 train_loss:2.1569 aux_loss:0.0214 train_time:604927ms step_avg:133.66ms
step:4537/6250 train_loss:2.1515 aux_loss:0.0213 train_time:605065ms step_avg:133.66ms
step:4538/6250 train_loss:2.2152 aux_loss:0.0216 train_time:605202ms step_avg:133.66ms
step:4539/6250 train_loss:2.2347 aux_loss:0.0216 train_time:605341ms step_avg:133.66ms
step:4540/6250 train_loss:2.2001 aux_loss:0.0217 train_time:605479ms step_avg:133.66ms
step:4541/6250 train_loss:2.1287 aux_loss:0.0214 train_time:605617ms step_avg:133.66ms
step:4542/6250 train_loss:2.2406 aux_loss:0.0216 train_time:605756ms step_avg:133.66ms
step:4543/6250 train_loss:2.2198 aux_loss:0.0214 train_time:605894ms step_avg:133.66ms
step:4544/6250 train_loss:2.2165 aux_loss:0.0215 train_time:606032ms step_avg:133.66ms
step:4545/6250 train_loss:2.2521 aux_loss:0.0216 train_time:606171ms step_avg:133.67ms
step:4546/6250 train_loss:2.1923 aux_loss:0.0214 train_time:606309ms step_avg:133.67ms
step:4547/6250 train_loss:2.2192 aux_loss:0.0210 train_time:606446ms step_avg:133.67ms
step:4548/6250 train_loss:2.1556 aux_loss:0.0211 train_time:606584ms step_avg:133.67ms
step:4549/6250 train_loss:2.2283 aux_loss:0.0216 train_time:606724ms step_avg:133.67ms
step:4550/6250 train_loss:2.2126 aux_loss:0.0213 train_time:606862ms step_avg:133.67ms
step:4551/6250 train_loss:2.2159 aux_loss:0.0215 train_time:606999ms step_avg:133.67ms
step:4552/6250 train_loss:2.2104 aux_loss:0.0210 train_time:607137ms step_avg:133.67ms
step:4553/6250 train_loss:2.1024 aux_loss:0.0210 train_time:607275ms step_avg:133.67ms
step:4554/6250 train_loss:2.1238 aux_loss:0.0210 train_time:607413ms step_avg:133.67ms
step:4555/6250 train_loss:2.2142 aux_loss:0.0212 train_time:607551ms step_avg:133.67ms
step:4556/6250 train_loss:2.2536 aux_loss:0.0216 train_time:607689ms step_avg:133.68ms
step:4557/6250 train_loss:2.0782 aux_loss:0.0215 train_time:607828ms step_avg:133.68ms
step:4558/6250 train_loss:2.0697 aux_loss:0.0215 train_time:607966ms step_avg:133.68ms
step:4559/6250 train_loss:2.2341 aux_loss:0.0218 train_time:608104ms step_avg:133.68ms
step:4560/6250 train_loss:2.2270 aux_loss:0.0215 train_time:608243ms step_avg:133.68ms
step:4561/6250 train_loss:2.1706 aux_loss:0.0214 train_time:608381ms step_avg:133.68ms
step:4562/6250 train_loss:2.1825 aux_loss:0.0217 train_time:608519ms step_avg:133.68ms
step:4563/6250 train_loss:2.1454 aux_loss:0.0216 train_time:608657ms step_avg:133.68ms
step:4564/6250 train_loss:2.2426 aux_loss:0.0216 train_time:608797ms step_avg:133.68ms
step:4565/6250 train_loss:2.1975 aux_loss:0.0217 train_time:608934ms step_avg:133.68ms
step:4566/6250 train_loss:2.2474 aux_loss:0.0212 train_time:609073ms step_avg:133.69ms
step:4567/6250 train_loss:2.2781 aux_loss:0.0214 train_time:609211ms step_avg:133.69ms
step:4568/6250 train_loss:2.1164 aux_loss:0.0213 train_time:609350ms step_avg:133.69ms
step:4569/6250 train_loss:2.2309 aux_loss:0.0213 train_time:609488ms step_avg:133.69ms
step:4570/6250 train_loss:2.1814 aux_loss:0.0214 train_time:609626ms step_avg:133.69ms
step:4571/6250 train_loss:2.1765 aux_loss:0.0214 train_time:609765ms step_avg:133.69ms
step:4572/6250 train_loss:2.1078 aux_loss:0.0213 train_time:609903ms step_avg:133.69ms
step:4573/6250 train_loss:2.1849 aux_loss:0.0214 train_time:610041ms step_avg:133.69ms
step:4574/6250 train_loss:2.1856 aux_loss:0.0211 train_time:610180ms step_avg:133.69ms
step:4575/6250 train_loss:2.1117 aux_loss:0.0212 train_time:610318ms step_avg:133.70ms
step:4576/6250 train_loss:2.2385 aux_loss:0.0214 train_time:610456ms step_avg:133.70ms
step:4577/6250 train_loss:2.1530 aux_loss:0.0213 train_time:610594ms step_avg:133.70ms
step:4578/6250 train_loss:2.1124 aux_loss:0.0212 train_time:610949ms step_avg:133.75ms
step:4579/6250 train_loss:2.1205 aux_loss:0.0212 train_time:611087ms step_avg:133.75ms
step:4580/6250 train_loss:2.1791 aux_loss:0.0214 train_time:611223ms step_avg:133.75ms
step:4581/6250 train_loss:2.0757 aux_loss:0.0215 train_time:611359ms step_avg:133.75ms
step:4582/6250 train_loss:2.2314 aux_loss:0.0215 train_time:611498ms step_avg:133.75ms
step:4583/6250 train_loss:2.2846 aux_loss:0.0211 train_time:611635ms step_avg:133.75ms
step:4584/6250 train_loss:2.1231 aux_loss:0.0208 train_time:611775ms step_avg:133.75ms
step:4585/6250 train_loss:2.1597 aux_loss:0.0211 train_time:611914ms step_avg:133.75ms
step:4586/6250 train_loss:2.1868 aux_loss:0.0211 train_time:612052ms step_avg:133.75ms
step:4587/6250 train_loss:2.1514 aux_loss:0.0211 train_time:612189ms step_avg:133.75ms
step:4588/6250 train_loss:2.2173 aux_loss:0.0214 train_time:612327ms step_avg:133.75ms
step:4589/6250 train_loss:2.2132 aux_loss:0.0213 train_time:612464ms step_avg:133.75ms
step:4590/6250 train_loss:2.2154 aux_loss:0.0212 train_time:612601ms step_avg:133.76ms
step:4591/6250 train_loss:2.1322 aux_loss:0.0211 train_time:612740ms step_avg:133.76ms
step:4592/6250 train_loss:2.1701 aux_loss:0.0211 train_time:612879ms step_avg:133.76ms
step:4593/6250 train_loss:2.1690 aux_loss:0.0212 train_time:613017ms step_avg:133.76ms
step:4594/6250 train_loss:2.1258 aux_loss:0.0211 train_time:613156ms step_avg:133.76ms
step:4595/6250 train_loss:2.1969 aux_loss:0.0214 train_time:613293ms step_avg:133.76ms
step:4596/6250 train_loss:2.0923 aux_loss:0.0220 train_time:613431ms step_avg:133.76ms
step:4597/6250 train_loss:2.1428 aux_loss:0.0215 train_time:613570ms step_avg:133.76ms
step:4598/6250 train_loss:2.2472 aux_loss:0.0215 train_time:613708ms step_avg:133.76ms
step:4599/6250 train_loss:2.1607 aux_loss:0.0213 train_time:613846ms step_avg:133.76ms
step:4600/6250 train_loss:2.2359 aux_loss:0.0214 train_time:613985ms step_avg:133.77ms
step:4601/6250 train_loss:2.2189 aux_loss:0.0215 train_time:614122ms step_avg:133.77ms
step:4602/6250 train_loss:2.2124 aux_loss:0.0212 train_time:614259ms step_avg:133.77ms
step:4603/6250 train_loss:2.1790 aux_loss:0.0213 train_time:614398ms step_avg:133.77ms
step:4604/6250 train_loss:2.1461 aux_loss:0.0210 train_time:614536ms step_avg:133.77ms
step:4605/6250 train_loss:2.2320 aux_loss:0.0208 train_time:614673ms step_avg:133.77ms
step:4606/6250 train_loss:2.2445 aux_loss:0.0213 train_time:614812ms step_avg:133.77ms
step:4607/6250 train_loss:2.1351 aux_loss:0.0213 train_time:614949ms step_avg:133.77ms
step:4608/6250 train_loss:2.1336 aux_loss:0.0211 train_time:615088ms step_avg:133.77ms
step:4609/6250 train_loss:2.1331 aux_loss:0.0209 train_time:615227ms step_avg:133.77ms
step:4610/6250 train_loss:2.1843 aux_loss:0.0210 train_time:615364ms step_avg:133.77ms
step:4611/6250 train_loss:2.1583 aux_loss:0.0210 train_time:615502ms step_avg:133.78ms
step:4612/6250 train_loss:2.1972 aux_loss:0.0210 train_time:615640ms step_avg:133.78ms
step:4613/6250 train_loss:2.1785 aux_loss:0.0212 train_time:615779ms step_avg:133.78ms
step:4614/6250 train_loss:2.1118 aux_loss:0.0211 train_time:615918ms step_avg:133.78ms
step:4615/6250 train_loss:2.3147 aux_loss:0.0216 train_time:616056ms step_avg:133.78ms
step:4616/6250 train_loss:2.2058 aux_loss:0.0211 train_time:616194ms step_avg:133.78ms
step:4617/6250 train_loss:2.1315 aux_loss:0.0208 train_time:616332ms step_avg:133.78ms
step:4618/6250 train_loss:2.1569 aux_loss:0.0209 train_time:616470ms step_avg:133.78ms
step:4619/6250 train_loss:2.1592 aux_loss:0.0213 train_time:616608ms step_avg:133.78ms
step:4620/6250 train_loss:2.2429 aux_loss:0.0213 train_time:616746ms step_avg:133.78ms
step:4621/6250 train_loss:2.1677 aux_loss:0.0211 train_time:616883ms step_avg:133.79ms
step:4622/6250 train_loss:2.1804 aux_loss:0.0208 train_time:617022ms step_avg:133.79ms
step:4623/6250 train_loss:2.1500 aux_loss:0.0210 train_time:617161ms step_avg:133.79ms
step:4624/6250 train_loss:2.1625 aux_loss:0.0211 train_time:617299ms step_avg:133.79ms
step:4625/6250 train_loss:2.0910 aux_loss:0.0209 train_time:617437ms step_avg:133.79ms
step:4626/6250 train_loss:2.1426 aux_loss:0.0209 train_time:617576ms step_avg:133.79ms
step:4627/6250 train_loss:2.1498 aux_loss:0.0209 train_time:617715ms step_avg:133.79ms
step:4628/6250 train_loss:2.1782 aux_loss:0.0208 train_time:617853ms step_avg:133.79ms
step:4629/6250 train_loss:2.1901 aux_loss:0.0210 train_time:617991ms step_avg:133.79ms
step:4630/6250 train_loss:2.2769 aux_loss:0.0208 train_time:618130ms step_avg:133.79ms
step:4631/6250 train_loss:2.1198 aux_loss:0.0205 train_time:618268ms step_avg:133.80ms
step:4632/6250 train_loss:2.1733 aux_loss:0.0206 train_time:618406ms step_avg:133.80ms
step:4633/6250 train_loss:2.0829 aux_loss:0.0208 train_time:618544ms step_avg:133.80ms
step:4634/6250 train_loss:2.1428 aux_loss:0.0210 train_time:618682ms step_avg:133.80ms
step:4635/6250 train_loss:2.2320 aux_loss:0.0209 train_time:618820ms step_avg:133.80ms
step:4636/6250 train_loss:2.1466 aux_loss:0.0209 train_time:618959ms step_avg:133.80ms
step:4637/6250 train_loss:2.1311 aux_loss:0.0206 train_time:619096ms step_avg:133.80ms
step:4638/6250 train_loss:2.1426 aux_loss:0.0204 train_time:619234ms step_avg:133.80ms
step:4639/6250 train_loss:2.1728 aux_loss:0.0205 train_time:619373ms step_avg:133.80ms
step:4640/6250 train_loss:2.1735 aux_loss:0.0209 train_time:619510ms step_avg:133.80ms
step:4641/6250 train_loss:2.0073 aux_loss:0.0210 train_time:619648ms step_avg:133.80ms
step:4642/6250 train_loss:2.2928 aux_loss:0.0212 train_time:619786ms step_avg:133.81ms
step:4643/6250 train_loss:2.1131 aux_loss:0.0211 train_time:619924ms step_avg:133.81ms
step:4644/6250 train_loss:2.1068 aux_loss:0.0209 train_time:620062ms step_avg:133.81ms
step:4645/6250 train_loss:2.2402 aux_loss:0.0210 train_time:620200ms step_avg:133.81ms
step:4646/6250 train_loss:2.1178 aux_loss:0.0215 train_time:620338ms step_avg:133.81ms
step:4647/6250 train_loss:2.0951 aux_loss:0.0211 train_time:620476ms step_avg:133.81ms
step:4648/6250 train_loss:2.1905 aux_loss:0.0215 train_time:620614ms step_avg:133.81ms
step:4649/6250 train_loss:2.1725 aux_loss:0.0210 train_time:620753ms step_avg:133.81ms
step:4650/6250 train_loss:2.1256 aux_loss:0.0211 train_time:620891ms step_avg:133.81ms
step:4651/6250 train_loss:2.1116 aux_loss:0.0211 train_time:621029ms step_avg:133.81ms
step:4652/6250 train_loss:2.2507 aux_loss:0.0211 train_time:621166ms step_avg:133.81ms
step:4653/6250 train_loss:2.2071 aux_loss:0.0208 train_time:621305ms step_avg:133.82ms
step:4654/6250 train_loss:2.2046 aux_loss:0.0210 train_time:621442ms step_avg:133.82ms
step:4655/6250 train_loss:2.1957 aux_loss:0.0210 train_time:621580ms step_avg:133.82ms
step:4656/6250 train_loss:2.2469 aux_loss:0.0212 train_time:621718ms step_avg:133.82ms
step:4657/6250 train_loss:2.1877 aux_loss:0.0209 train_time:621856ms step_avg:133.82ms
step:4658/6250 train_loss:2.1736 aux_loss:0.0206 train_time:621994ms step_avg:133.82ms
step:4659/6250 train_loss:2.1524 aux_loss:0.0210 train_time:622132ms step_avg:133.82ms
step:4660/6250 train_loss:2.1293 aux_loss:0.0209 train_time:622270ms step_avg:133.82ms
step:4661/6250 train_loss:2.1212 aux_loss:0.0208 train_time:622408ms step_avg:133.82ms
step:4662/6250 train_loss:2.1401 aux_loss:0.0210 train_time:622547ms step_avg:133.82ms
step:4663/6250 train_loss:2.2760 aux_loss:0.0209 train_time:622685ms step_avg:133.82ms
step:4664/6250 train_loss:2.1709 aux_loss:0.0212 train_time:622824ms step_avg:133.83ms
step:4665/6250 train_loss:2.0951 aux_loss:0.0208 train_time:622962ms step_avg:133.83ms
step:4666/6250 train_loss:2.1634 aux_loss:0.0209 train_time:623100ms step_avg:133.83ms
step:4667/6250 train_loss:2.1233 aux_loss:0.0209 train_time:623239ms step_avg:133.83ms
step:4668/6250 train_loss:2.1660 aux_loss:0.0206 train_time:623378ms step_avg:133.83ms
step:4669/6250 train_loss:2.1608 aux_loss:0.0207 train_time:623517ms step_avg:133.83ms
step:4670/6250 train_loss:2.1114 aux_loss:0.0207 train_time:623654ms step_avg:133.83ms
step:4671/6250 train_loss:2.1643 aux_loss:0.0209 train_time:623793ms step_avg:133.83ms
step:4672/6250 train_loss:2.0948 aux_loss:0.0206 train_time:623930ms step_avg:133.83ms
step:4673/6250 train_loss:2.2203 aux_loss:0.0216 train_time:624069ms step_avg:133.83ms
step:4674/6250 train_loss:2.1858 aux_loss:0.0205 train_time:624207ms step_avg:133.84ms
step:4675/6250 train_loss:2.2357 aux_loss:0.0205 train_time:624344ms step_avg:133.84ms
step:4676/6250 train_loss:2.2158 aux_loss:0.0209 train_time:624482ms step_avg:133.84ms
step:4677/6250 train_loss:2.2740 aux_loss:0.0212 train_time:624620ms step_avg:133.84ms
step:4678/6250 train_loss:2.1092 aux_loss:0.0207 train_time:624758ms step_avg:133.84ms
step:4679/6250 train_loss:2.1085 aux_loss:0.0206 train_time:624896ms step_avg:133.84ms
step:4680/6250 train_loss:2.2088 aux_loss:0.0208 train_time:625034ms step_avg:133.84ms
step:4681/6250 train_loss:2.1744 aux_loss:0.0205 train_time:625172ms step_avg:133.84ms
step:4682/6250 train_loss:2.1404 aux_loss:0.0207 train_time:625311ms step_avg:133.84ms
step:4683/6250 train_loss:2.1489 aux_loss:0.0208 train_time:625449ms step_avg:133.84ms
step:4684/6250 train_loss:2.0917 aux_loss:0.0209 train_time:625586ms step_avg:133.84ms
step:4685/6250 train_loss:2.1941 aux_loss:0.0210 train_time:625725ms step_avg:133.85ms
step:4686/6250 train_loss:2.1884 aux_loss:0.0203 train_time:625863ms step_avg:133.85ms
step:4687/6250 train_loss:2.1801 aux_loss:0.0205 train_time:626001ms step_avg:133.85ms
step:4688/6250 train_loss:2.2069 aux_loss:0.0207 train_time:626140ms step_avg:133.85ms
step:4689/6250 train_loss:2.2256 aux_loss:0.0207 train_time:626278ms step_avg:133.85ms
step:4690/6250 train_loss:2.1474 aux_loss:0.0206 train_time:626416ms step_avg:133.85ms
step:4691/6250 train_loss:2.1318 aux_loss:0.0204 train_time:626554ms step_avg:133.85ms
step:4692/6250 train_loss:2.1990 aux_loss:0.0204 train_time:626692ms step_avg:133.85ms
step:4693/6250 train_loss:2.1679 aux_loss:0.0206 train_time:626830ms step_avg:133.85ms
step:4694/6250 train_loss:2.1376 aux_loss:0.0209 train_time:626968ms step_avg:133.85ms
step:4695/6250 train_loss:2.2185 aux_loss:0.0206 train_time:627106ms step_avg:133.85ms
step:4696/6250 train_loss:2.2003 aux_loss:0.0209 train_time:627244ms step_avg:133.85ms
step:4697/6250 train_loss:2.1994 aux_loss:0.0209 train_time:627382ms step_avg:133.86ms
step:4698/6250 train_loss:2.2772 aux_loss:0.0212 train_time:627519ms step_avg:133.86ms
step:4699/6250 train_loss:2.2146 aux_loss:0.0208 train_time:627658ms step_avg:133.86ms
step:4700/6250 train_loss:2.2280 aux_loss:0.0210 train_time:627797ms step_avg:133.86ms
step:4701/6250 train_loss:2.2175 aux_loss:0.0209 train_time:627935ms step_avg:133.86ms
step:4702/6250 train_loss:2.0772 aux_loss:0.0209 train_time:628073ms step_avg:133.86ms
step:4703/6250 train_loss:2.1918 aux_loss:0.0208 train_time:628212ms step_avg:133.86ms
step:4704/6250 train_loss:2.2233 aux_loss:0.0206 train_time:628355ms step_avg:133.86ms
step:4705/6250 train_loss:2.1918 aux_loss:0.0208 train_time:628500ms step_avg:133.87ms
step:4706/6250 train_loss:2.3543 aux_loss:0.0207 train_time:628647ms step_avg:133.87ms
step:4707/6250 train_loss:2.2295 aux_loss:0.0204 train_time:628794ms step_avg:133.87ms
step:4708/6250 train_loss:2.2812 aux_loss:0.0208 train_time:628941ms step_avg:133.87ms
step:4709/6250 train_loss:2.1252 aux_loss:0.0206 train_time:629091ms step_avg:133.88ms
step:4710/6250 train_loss:2.0718 aux_loss:0.0208 train_time:629229ms step_avg:133.88ms
step:4711/6250 train_loss:2.1110 aux_loss:0.0208 train_time:629368ms step_avg:133.88ms
step:4712/6250 train_loss:2.1585 aux_loss:0.0206 train_time:629506ms step_avg:133.88ms
step:4713/6250 train_loss:2.2361 aux_loss:0.0208 train_time:629644ms step_avg:133.88ms
step:4714/6250 train_loss:2.1684 aux_loss:0.0208 train_time:629783ms step_avg:133.88ms
step:4715/6250 train_loss:2.2602 aux_loss:0.0206 train_time:629922ms step_avg:133.88ms
step:4716/6250 train_loss:2.1640 aux_loss:0.0206 train_time:630067ms step_avg:133.89ms
step:4717/6250 train_loss:2.1925 aux_loss:0.0205 train_time:630216ms step_avg:133.89ms
step:4718/6250 train_loss:2.0956 aux_loss:0.0205 train_time:630368ms step_avg:133.89ms
step:4719/6250 train_loss:2.2267 aux_loss:0.0207 train_time:630515ms step_avg:133.90ms
step:4720/6250 train_loss:2.3280 aux_loss:0.0206 train_time:630665ms step_avg:133.90ms
step:4721/6250 train_loss:2.1998 aux_loss:0.0208 train_time:630813ms step_avg:133.90ms
step:4722/6250 train_loss:2.2753 aux_loss:0.0207 train_time:630951ms step_avg:133.90ms
step:4723/6250 train_loss:2.2495 aux_loss:0.0207 train_time:631090ms step_avg:133.90ms
step:4724/6250 train_loss:2.1310 aux_loss:0.0205 train_time:631228ms step_avg:133.90ms
step:4725/6250 train_loss:2.1239 aux_loss:0.0204 train_time:631366ms step_avg:133.91ms
step:4726/6250 train_loss:2.1086 aux_loss:0.0202 train_time:631505ms step_avg:133.91ms
step:4727/6250 train_loss:2.1535 aux_loss:0.0205 train_time:631644ms step_avg:133.91ms
step:4728/6250 train_loss:2.1838 aux_loss:0.0204 train_time:631782ms step_avg:133.91ms
step:4729/6250 train_loss:2.1057 aux_loss:0.0206 train_time:631920ms step_avg:133.91ms
step:4730/6250 train_loss:2.1954 aux_loss:0.0205 train_time:632058ms step_avg:133.91ms
step:4731/6250 train_loss:2.1389 aux_loss:0.0206 train_time:632197ms step_avg:133.91ms
step:4732/6250 train_loss:2.0647 aux_loss:0.0207 train_time:632336ms step_avg:133.91ms
step:4733/6250 train_loss:2.2642 aux_loss:0.0208 train_time:632478ms step_avg:133.91ms
step:4734/6250 train_loss:2.0488 aux_loss:0.0207 train_time:632618ms step_avg:133.92ms
step:4735/6250 train_loss:2.2063 aux_loss:0.0206 train_time:632756ms step_avg:133.92ms
step:4736/6250 train_loss:2.0869 aux_loss:0.0205 train_time:632894ms step_avg:133.92ms
step:4737/6250 train_loss:2.2126 aux_loss:0.0209 train_time:633032ms step_avg:133.92ms
step:4738/6250 train_loss:2.1409 aux_loss:0.0208 train_time:633170ms step_avg:133.92ms
step:4739/6250 train_loss:2.1398 aux_loss:0.0207 train_time:633310ms step_avg:133.92ms
step:4740/6250 train_loss:2.0915 aux_loss:0.0207 train_time:633450ms step_avg:133.92ms
step:4741/6250 train_loss:2.1336 aux_loss:0.0206 train_time:633590ms step_avg:133.92ms
step:4742/6250 train_loss:2.2250 aux_loss:0.0208 train_time:633731ms step_avg:133.92ms
step:4743/6250 train_loss:2.1582 aux_loss:0.0205 train_time:633870ms step_avg:133.93ms
step:4744/6250 train_loss:2.3001 aux_loss:0.0206 train_time:634007ms step_avg:133.93ms
step:4745/6250 train_loss:2.3202 aux_loss:0.0208 train_time:634144ms step_avg:133.93ms
step:4746/6250 train_loss:2.1137 aux_loss:0.0205 train_time:634283ms step_avg:133.93ms
step:4747/6250 train_loss:2.1749 aux_loss:0.0203 train_time:634424ms step_avg:133.93ms
step:4748/6250 train_loss:2.1610 aux_loss:0.0203 train_time:634565ms step_avg:133.93ms
step:4749/6250 train_loss:2.2567 aux_loss:0.0206 train_time:634705ms step_avg:133.93ms
step:4750/6250 train_loss:2.1353 aux_loss:0.0205 train_time:634844ms step_avg:133.93ms
step:4751/6250 train_loss:2.1413 aux_loss:0.0205 train_time:634983ms step_avg:133.93ms
step:4752/6250 train_loss:2.1318 aux_loss:0.0207 train_time:635121ms step_avg:133.94ms
step:4753/6250 train_loss:2.1608 aux_loss:0.0207 train_time:635259ms step_avg:133.94ms
step:4754/6250 train_loss:2.1511 aux_loss:0.0203 train_time:635398ms step_avg:133.94ms
step:4755/6250 train_loss:2.1018 aux_loss:0.0202 train_time:635537ms step_avg:133.94ms
step:4756/6250 train_loss:2.1451 aux_loss:0.0205 train_time:635675ms step_avg:133.94ms
step:4757/6250 train_loss:2.1860 aux_loss:0.0203 train_time:635814ms step_avg:133.94ms
step:4758/6250 train_loss:2.1381 aux_loss:0.0205 train_time:635952ms step_avg:133.94ms
step:4759/6250 train_loss:2.1236 aux_loss:0.0203 train_time:636090ms step_avg:133.94ms
step:4760/6250 train_loss:2.1668 aux_loss:0.0206 train_time:636228ms step_avg:133.94ms
step:4761/6250 train_loss:2.0737 aux_loss:0.0204 train_time:636367ms step_avg:133.94ms
step:4762/6250 train_loss:2.1326 aux_loss:0.0204 train_time:636505ms step_avg:133.94ms
step:4763/6250 train_loss:2.1356 aux_loss:0.0206 train_time:636644ms step_avg:133.95ms
step:4764/6250 train_loss:2.2205 aux_loss:0.0206 train_time:636784ms step_avg:133.95ms
step:4765/6250 train_loss:2.1485 aux_loss:0.0203 train_time:636922ms step_avg:133.95ms
step:4766/6250 train_loss:2.0573 aux_loss:0.0201 train_time:637061ms step_avg:133.95ms
step:4767/6250 train_loss:2.1817 aux_loss:0.0206 train_time:637200ms step_avg:133.95ms
step:4768/6250 train_loss:2.2373 aux_loss:0.0205 train_time:637338ms step_avg:133.95ms
step:4769/6250 train_loss:2.1860 aux_loss:0.0207 train_time:637476ms step_avg:133.95ms
step:4770/6250 train_loss:2.0885 aux_loss:0.0200 train_time:637616ms step_avg:133.95ms
step:4771/6250 train_loss:2.2013 aux_loss:0.0201 train_time:637755ms step_avg:133.95ms
step:4772/6250 train_loss:2.1078 aux_loss:0.0202 train_time:637894ms step_avg:133.96ms
step:4773/6250 train_loss:2.2395 aux_loss:0.0204 train_time:638033ms step_avg:133.96ms
step:4774/6250 train_loss:2.1879 aux_loss:0.0207 train_time:638172ms step_avg:133.96ms
step:4775/6250 train_loss:2.1045 aux_loss:0.0207 train_time:638310ms step_avg:133.96ms
step:4776/6250 train_loss:2.0832 aux_loss:0.0206 train_time:638449ms step_avg:133.96ms
step:4777/6250 train_loss:2.1132 aux_loss:0.0204 train_time:638587ms step_avg:133.96ms
step:4778/6250 train_loss:2.1978 aux_loss:0.0206 train_time:638726ms step_avg:133.96ms
step:4779/6250 train_loss:2.1350 aux_loss:0.0205 train_time:638865ms step_avg:133.96ms
step:4780/6250 train_loss:2.1264 aux_loss:0.0204 train_time:639003ms step_avg:133.96ms
step:4781/6250 train_loss:2.1462 aux_loss:0.0203 train_time:639142ms step_avg:133.96ms
step:4782/6250 train_loss:2.1345 aux_loss:0.0203 train_time:639281ms step_avg:133.97ms
step:4783/6250 train_loss:2.1120 aux_loss:0.0201 train_time:639420ms step_avg:133.97ms
step:4784/6250 train_loss:2.1155 aux_loss:0.0200 train_time:639559ms step_avg:133.97ms
step:4785/6250 train_loss:2.1249 aux_loss:0.0200 train_time:639699ms step_avg:133.97ms
step:4786/6250 train_loss:2.0979 aux_loss:0.0201 train_time:639839ms step_avg:133.97ms
step:4787/6250 train_loss:2.1210 aux_loss:0.0205 train_time:639981ms step_avg:133.97ms
step:4788/6250 train_loss:2.1642 aux_loss:0.0201 train_time:640121ms step_avg:133.97ms
step:4789/6250 train_loss:2.1135 aux_loss:0.0204 train_time:640262ms step_avg:133.97ms
step:4790/6250 train_loss:2.1367 aux_loss:0.0200 train_time:640403ms step_avg:133.98ms
step:4791/6250 train_loss:2.2117 aux_loss:0.0201 train_time:640543ms step_avg:133.98ms
step:4792/6250 train_loss:2.2333 aux_loss:0.0202 train_time:640682ms step_avg:133.98ms
step:4793/6250 train_loss:2.2107 aux_loss:0.0202 train_time:640820ms step_avg:133.98ms
step:4794/6250 train_loss:2.1839 aux_loss:0.0201 train_time:640960ms step_avg:133.98ms
step:4795/6250 train_loss:2.1633 aux_loss:0.0200 train_time:641098ms step_avg:133.98ms
step:4796/6250 train_loss:2.0744 aux_loss:0.0201 train_time:641236ms step_avg:133.98ms
step:4797/6250 train_loss:2.2496 aux_loss:0.0199 train_time:641374ms step_avg:133.98ms
step:4798/6250 train_loss:2.2299 aux_loss:0.0199 train_time:641513ms step_avg:133.98ms
step:4799/6250 train_loss:2.1935 aux_loss:0.0200 train_time:641652ms step_avg:133.98ms
step:4800/6250 train_loss:2.0991 aux_loss:0.0202 train_time:641790ms step_avg:133.99ms
step:4801/6250 train_loss:2.1538 aux_loss:0.0200 train_time:641929ms step_avg:133.99ms
step:4802/6250 train_loss:2.1371 aux_loss:0.0202 train_time:642067ms step_avg:133.99ms
step:4803/6250 train_loss:2.1649 aux_loss:0.0201 train_time:642205ms step_avg:133.99ms
step:4804/6250 train_loss:2.0730 aux_loss:0.0199 train_time:642343ms step_avg:133.99ms
step:4805/6250 train_loss:2.2255 aux_loss:0.0204 train_time:642482ms step_avg:133.99ms
step:4806/6250 train_loss:2.1795 aux_loss:0.0203 train_time:642623ms step_avg:133.99ms
step:4807/6250 train_loss:2.1869 aux_loss:0.0202 train_time:642765ms step_avg:133.99ms
step:4808/6250 train_loss:2.1533 aux_loss:0.0200 train_time:642907ms step_avg:133.99ms
step:4809/6250 train_loss:2.1772 aux_loss:0.0202 train_time:643050ms step_avg:134.00ms
step:4810/6250 train_loss:2.2155 aux_loss:0.0205 train_time:643191ms step_avg:134.00ms
step:4811/6250 train_loss:2.1882 aux_loss:0.0200 train_time:643331ms step_avg:134.00ms
step:4812/6250 train_loss:2.1014 aux_loss:0.0202 train_time:643472ms step_avg:134.00ms
step:4813/6250 train_loss:2.1565 aux_loss:0.0200 train_time:643621ms step_avg:134.00ms
step:4814/6250 train_loss:2.1464 aux_loss:0.0203 train_time:643760ms step_avg:134.00ms
step:4815/6250 train_loss:2.2030 aux_loss:0.0205 train_time:643897ms step_avg:134.01ms
step:4816/6250 train_loss:2.1934 aux_loss:0.0204 train_time:644032ms step_avg:134.01ms
step:4817/6250 train_loss:2.1822 aux_loss:0.0202 train_time:644170ms step_avg:134.01ms
step:4818/6250 train_loss:2.1492 aux_loss:0.0204 train_time:644307ms step_avg:134.01ms
step:4819/6250 train_loss:2.2195 aux_loss:0.0205 train_time:644445ms step_avg:134.01ms
step:4820/6250 train_loss:2.2385 aux_loss:0.0204 train_time:644582ms step_avg:134.01ms
step:4821/6250 train_loss:2.1998 aux_loss:0.0203 train_time:644719ms step_avg:134.01ms
step:4822/6250 train_loss:2.1520 aux_loss:0.0202 train_time:644854ms step_avg:134.01ms
step:4823/6250 train_loss:2.1674 aux_loss:0.0202 train_time:644989ms step_avg:134.01ms
step:4824/6250 train_loss:2.1426 aux_loss:0.0203 train_time:645125ms step_avg:134.01ms
step:4825/6250 train_loss:2.1879 aux_loss:0.0201 train_time:645261ms step_avg:134.01ms
step:4826/6250 train_loss:2.1195 aux_loss:0.0202 train_time:645396ms step_avg:134.01ms
step:4827/6250 train_loss:2.1216 aux_loss:0.0201 train_time:645530ms step_avg:134.01ms
step:4828/6250 train_loss:2.1474 aux_loss:0.0200 train_time:645670ms step_avg:134.01ms
step:4829/6250 train_loss:2.1474 aux_loss:0.0197 train_time:645806ms step_avg:134.01ms
step:4830/6250 train_loss:2.2197 aux_loss:0.0201 train_time:645943ms step_avg:134.01ms
step:4831/6250 train_loss:2.1996 aux_loss:0.0201 train_time:646083ms step_avg:134.01ms
step:4832/6250 train_loss:2.2539 aux_loss:0.0205 train_time:646221ms step_avg:134.02ms
step:4833/6250 train_loss:2.1891 aux_loss:0.0200 train_time:646360ms step_avg:134.02ms
step:4834/6250 train_loss:2.1454 aux_loss:0.0202 train_time:646498ms step_avg:134.02ms
step:4835/6250 train_loss:2.2219 aux_loss:0.0200 train_time:646637ms step_avg:134.02ms
step:4836/6250 train_loss:2.1397 aux_loss:0.0197 train_time:646775ms step_avg:134.02ms
step:4837/6250 train_loss:2.1763 aux_loss:0.0199 train_time:646911ms step_avg:134.02ms
step:4838/6250 train_loss:2.1504 aux_loss:0.0198 train_time:647046ms step_avg:134.02ms
step:4839/6250 train_loss:2.1190 aux_loss:0.0199 train_time:647180ms step_avg:134.02ms
step:4840/6250 train_loss:2.1464 aux_loss:0.0195 train_time:647314ms step_avg:134.02ms
step:4841/6250 train_loss:2.1436 aux_loss:0.0198 train_time:647448ms step_avg:134.02ms
step:4842/6250 train_loss:2.2302 aux_loss:0.0200 train_time:647582ms step_avg:134.02ms
step:4843/6250 train_loss:2.1546 aux_loss:0.0200 train_time:647717ms step_avg:134.02ms
step:4844/6250 train_loss:2.2481 aux_loss:0.0201 train_time:647850ms step_avg:134.02ms
step:4845/6250 train_loss:2.1361 aux_loss:0.0199 train_time:647985ms step_avg:134.02ms
step:4846/6250 train_loss:2.0542 aux_loss:0.0198 train_time:648118ms step_avg:134.02ms
step:4847/6250 train_loss:2.1329 aux_loss:0.0200 train_time:648253ms step_avg:134.02ms
step:4848/6250 train_loss:2.0776 aux_loss:0.0199 train_time:648387ms step_avg:134.02ms
step:4849/6250 train_loss:2.1941 aux_loss:0.0201 train_time:648522ms step_avg:134.02ms
step:4850/6250 train_loss:2.0887 aux_loss:0.0198 train_time:648656ms step_avg:134.02ms
step:4851/6250 train_loss:2.2495 aux_loss:0.0200 train_time:648791ms step_avg:134.02ms
step:4852/6250 train_loss:2.1763 aux_loss:0.0202 train_time:648925ms step_avg:134.02ms
step:4853/6250 train_loss:2.0844 aux_loss:0.0201 train_time:649058ms step_avg:134.02ms
step:4854/6250 train_loss:2.1213 aux_loss:0.0199 train_time:649192ms step_avg:134.02ms
step:4855/6250 train_loss:2.1762 aux_loss:0.0197 train_time:649327ms step_avg:134.02ms
step:4856/6250 train_loss:2.1731 aux_loss:0.0199 train_time:649461ms step_avg:134.02ms
step:4857/6250 train_loss:2.1802 aux_loss:0.0200 train_time:649595ms step_avg:134.02ms
step:4858/6250 train_loss:2.0632 aux_loss:0.0200 train_time:649730ms step_avg:134.02ms
step:4859/6250 train_loss:2.2432 aux_loss:0.0200 train_time:649864ms step_avg:134.02ms
step:4860/6250 train_loss:2.2062 aux_loss:0.0199 train_time:649997ms step_avg:134.02ms
step:4861/6250 train_loss:2.0644 aux_loss:0.0198 train_time:650131ms step_avg:134.02ms
step:4862/6250 train_loss:2.0850 aux_loss:0.0199 train_time:650266ms step_avg:134.02ms
step:4863/6250 train_loss:2.1861 aux_loss:0.0203 train_time:650401ms step_avg:134.02ms
step:4864/6250 train_loss:2.1210 aux_loss:0.0202 train_time:650535ms step_avg:134.02ms
step:4865/6250 train_loss:2.2068 aux_loss:0.0201 train_time:650670ms step_avg:134.02ms
step:4866/6250 train_loss:2.1563 aux_loss:0.0199 train_time:650805ms step_avg:134.02ms
step:4867/6250 train_loss:2.1977 aux_loss:0.0200 train_time:650938ms step_avg:134.02ms
step:4868/6250 train_loss:2.1961 aux_loss:0.0201 train_time:651072ms step_avg:134.02ms
step:4869/6250 train_loss:2.2943 aux_loss:0.0204 train_time:651207ms step_avg:134.02ms
step:4870/6250 train_loss:2.1493 aux_loss:0.0201 train_time:651342ms step_avg:134.02ms
step:4871/6250 train_loss:2.1652 aux_loss:0.0202 train_time:651476ms step_avg:134.02ms
step:4872/6250 train_loss:2.1015 aux_loss:0.0201 train_time:651610ms step_avg:134.02ms
step:4873/6250 train_loss:2.1988 aux_loss:0.0200 train_time:651745ms step_avg:134.02ms
step:4874/6250 train_loss:2.1995 aux_loss:0.0202 train_time:651880ms step_avg:134.02ms
step:4875/6250 train_loss:2.1238 aux_loss:0.0200 train_time:652015ms step_avg:134.02ms
step:4876/6250 train_loss:2.1876 aux_loss:0.0199 train_time:652149ms step_avg:134.02ms
step:4877/6250 train_loss:2.1150 aux_loss:0.0199 train_time:652285ms step_avg:134.02ms
step:4878/6250 train_loss:2.1470 aux_loss:0.0201 train_time:652420ms step_avg:134.02ms
step:4879/6250 train_loss:2.2262 aux_loss:0.0203 train_time:652555ms step_avg:134.02ms
step:4880/6250 train_loss:2.1167 aux_loss:0.0202 train_time:652689ms step_avg:134.02ms
step:4881/6250 train_loss:2.2290 aux_loss:0.0201 train_time:652824ms step_avg:134.02ms
step:4882/6250 train_loss:2.1513 aux_loss:0.0199 train_time:652958ms step_avg:134.02ms
step:4883/6250 train_loss:2.1786 aux_loss:0.0201 train_time:653092ms step_avg:134.02ms
step:4884/6250 train_loss:2.2301 aux_loss:0.0199 train_time:653226ms step_avg:134.02ms
step:4885/6250 train_loss:2.1010 aux_loss:0.0198 train_time:653360ms step_avg:134.02ms
step:4886/6250 train_loss:2.0705 aux_loss:0.0199 train_time:653494ms step_avg:134.02ms
step:4887/6250 train_loss:2.1541 aux_loss:0.0198 train_time:653630ms step_avg:134.02ms
step:4888/6250 train_loss:2.1293 aux_loss:0.0200 train_time:653767ms step_avg:134.02ms
step:4889/6250 train_loss:2.1228 aux_loss:0.0197 train_time:653902ms step_avg:134.02ms
step:4890/6250 train_loss:2.0860 aux_loss:0.0199 train_time:654037ms step_avg:134.02ms
step:4891/6250 train_loss:2.0763 aux_loss:0.0198 train_time:654170ms step_avg:134.02ms
step:4892/6250 train_loss:2.2476 aux_loss:0.0199 train_time:654305ms step_avg:134.02ms
step:4893/6250 train_loss:2.1544 aux_loss:0.0199 train_time:654439ms step_avg:134.02ms
step:4894/6250 train_loss:2.1451 aux_loss:0.0199 train_time:654574ms step_avg:134.02ms
step:4895/6250 train_loss:2.1132 aux_loss:0.0196 train_time:654709ms step_avg:134.02ms
step:4896/6250 train_loss:2.0786 aux_loss:0.0199 train_time:654843ms step_avg:134.02ms
step:4897/6250 train_loss:2.2095 aux_loss:0.0198 train_time:654976ms step_avg:134.02ms
step:4898/6250 train_loss:2.1834 aux_loss:0.0196 train_time:655109ms step_avg:134.02ms
step:4899/6250 train_loss:2.2137 aux_loss:0.0200 train_time:655244ms step_avg:134.02ms
step:4900/6250 train_loss:2.1083 aux_loss:0.0197 train_time:655378ms step_avg:134.02ms
step:4901/6250 train_loss:2.2442 aux_loss:0.0201 train_time:655513ms step_avg:134.02ms
step:4902/6250 train_loss:2.1559 aux_loss:0.0201 train_time:655648ms step_avg:134.02ms
step:4903/6250 train_loss:2.1161 aux_loss:0.0197 train_time:655784ms step_avg:134.03ms
step:4904/6250 train_loss:2.2383 aux_loss:0.0201 train_time:655919ms step_avg:134.03ms
step:4905/6250 train_loss:2.1109 aux_loss:0.0197 train_time:656053ms step_avg:134.03ms
step:4906/6250 train_loss:2.1514 aux_loss:0.0201 train_time:656188ms step_avg:134.03ms
step:4907/6250 train_loss:2.1638 aux_loss:0.0197 train_time:656324ms step_avg:134.03ms
step:4908/6250 train_loss:2.1621 aux_loss:0.0199 train_time:656458ms step_avg:134.03ms
step:4909/6250 train_loss:2.1495 aux_loss:0.0201 train_time:656592ms step_avg:134.03ms
step:4910/6250 train_loss:2.1586 aux_loss:0.0199 train_time:656727ms step_avg:134.03ms
step:4911/6250 train_loss:2.1667 aux_loss:0.0198 train_time:656863ms step_avg:134.03ms
step:4912/6250 train_loss:2.0969 aux_loss:0.0197 train_time:656997ms step_avg:134.03ms
step:4913/6250 train_loss:2.1868 aux_loss:0.0199 train_time:657131ms step_avg:134.03ms
step:4914/6250 train_loss:2.1160 aux_loss:0.0197 train_time:657267ms step_avg:134.03ms
step:4915/6250 train_loss:2.1766 aux_loss:0.0200 train_time:657401ms step_avg:134.03ms
step:4916/6250 train_loss:2.0818 aux_loss:0.0199 train_time:657536ms step_avg:134.03ms
step:4917/6250 train_loss:2.1295 aux_loss:0.0198 train_time:657671ms step_avg:134.03ms
step:4918/6250 train_loss:2.1632 aux_loss:0.0199 train_time:657806ms step_avg:134.03ms
step:4919/6250 train_loss:2.2455 aux_loss:0.0200 train_time:657941ms step_avg:134.03ms
step:4920/6250 train_loss:2.1048 aux_loss:0.0199 train_time:658075ms step_avg:134.03ms
step:4921/6250 train_loss:2.1406 aux_loss:0.0200 train_time:658209ms step_avg:134.03ms
step:4922/6250 train_loss:2.1869 aux_loss:0.0198 train_time:658344ms step_avg:134.03ms
step:4923/6250 train_loss:2.1395 aux_loss:0.0197 train_time:658478ms step_avg:134.03ms
step:4924/6250 train_loss:2.1480 aux_loss:0.0198 train_time:658612ms step_avg:134.03ms
step:4925/6250 train_loss:2.1740 aux_loss:0.0200 train_time:658748ms step_avg:134.03ms
step:4926/6250 train_loss:2.1321 aux_loss:0.0199 train_time:658882ms step_avg:134.03ms
step:4927/6250 train_loss:2.1375 aux_loss:0.0199 train_time:659018ms step_avg:134.03ms
step:4928/6250 train_loss:2.1436 aux_loss:0.0199 train_time:659151ms step_avg:134.03ms
step:4929/6250 train_loss:2.1654 aux_loss:0.0195 train_time:659286ms step_avg:134.03ms
step:4930/6250 train_loss:2.1826 aux_loss:0.0194 train_time:659420ms step_avg:134.03ms
step:4931/6250 train_loss:2.2403 aux_loss:0.0195 train_time:659553ms step_avg:134.03ms
step:4932/6250 train_loss:2.2878 aux_loss:0.0196 train_time:659687ms step_avg:134.03ms
step:4933/6250 train_loss:2.1567 aux_loss:0.0195 train_time:659823ms step_avg:134.03ms
step:4934/6250 train_loss:2.1137 aux_loss:0.0199 train_time:659956ms step_avg:134.03ms
step:4935/6250 train_loss:2.1087 aux_loss:0.0196 train_time:660091ms step_avg:134.03ms
step:4936/6250 train_loss:2.1357 aux_loss:0.0199 train_time:660225ms step_avg:134.03ms
step:4937/6250 train_loss:2.1452 aux_loss:0.0196 train_time:660359ms step_avg:134.03ms
step:4938/6250 train_loss:2.2069 aux_loss:0.0199 train_time:660493ms step_avg:134.03ms
step:4939/6250 train_loss:2.1642 aux_loss:0.0201 train_time:660626ms step_avg:134.03ms
step:4940/6250 train_loss:2.0690 aux_loss:0.0197 train_time:660761ms step_avg:134.03ms
step:4941/6250 train_loss:2.1149 aux_loss:0.0196 train_time:660895ms step_avg:134.03ms
step:4942/6250 train_loss:2.1257 aux_loss:0.0197 train_time:661028ms step_avg:134.03ms
step:4943/6250 train_loss:2.1670 aux_loss:0.0198 train_time:661163ms step_avg:134.03ms
step:4944/6250 train_loss:2.0791 aux_loss:0.0195 train_time:661297ms step_avg:134.03ms
step:4945/6250 train_loss:2.3256 aux_loss:0.0200 train_time:661431ms step_avg:134.03ms
step:4946/6250 train_loss:2.2015 aux_loss:0.0197 train_time:661564ms step_avg:134.03ms
step:4947/6250 train_loss:2.2055 aux_loss:0.0199 train_time:661699ms step_avg:134.03ms
step:4948/6250 train_loss:2.1525 aux_loss:0.0199 train_time:661834ms step_avg:134.03ms
step:4949/6250 train_loss:2.0944 aux_loss:0.0197 train_time:661967ms step_avg:134.03ms
step:4950/6250 train_loss:2.0947 aux_loss:0.0197 train_time:662101ms step_avg:134.03ms
step:4951/6250 train_loss:2.2216 aux_loss:0.0199 train_time:662236ms step_avg:134.03ms
step:4952/6250 train_loss:2.2557 aux_loss:0.0201 train_time:662370ms step_avg:134.03ms
step:4953/6250 train_loss:2.1870 aux_loss:0.0200 train_time:662503ms step_avg:134.03ms
step:4954/6250 train_loss:2.2363 aux_loss:0.0201 train_time:662637ms step_avg:134.03ms
step:4955/6250 train_loss:2.1696 aux_loss:0.0200 train_time:662772ms step_avg:134.03ms
step:4956/6250 train_loss:2.2318 aux_loss:0.0198 train_time:662905ms step_avg:134.03ms
step:4957/6250 train_loss:2.1748 aux_loss:0.0200 train_time:663040ms step_avg:134.03ms
step:4958/6250 train_loss:2.1178 aux_loss:0.0200 train_time:663174ms step_avg:134.03ms
step:4959/6250 train_loss:2.1776 aux_loss:0.0197 train_time:663309ms step_avg:134.03ms
step:4960/6250 train_loss:2.1508 aux_loss:0.0193 train_time:663442ms step_avg:134.03ms
step:4961/6250 train_loss:2.2182 aux_loss:0.0196 train_time:663577ms step_avg:134.03ms
step:4962/6250 train_loss:2.1902 aux_loss:0.0196 train_time:663712ms step_avg:134.03ms
step:4963/6250 train_loss:2.1854 aux_loss:0.0197 train_time:663846ms step_avg:134.03ms
step:4964/6250 train_loss:2.1563 aux_loss:0.0197 train_time:663981ms step_avg:134.03ms
step:4965/6250 train_loss:2.1553 aux_loss:0.0196 train_time:664117ms step_avg:134.03ms
step:4966/6250 train_loss:2.1580 aux_loss:0.0196 train_time:664252ms step_avg:134.03ms
step:4967/6250 train_loss:2.1309 aux_loss:0.0199 train_time:664386ms step_avg:134.03ms
step:4968/6250 train_loss:2.2321 aux_loss:0.0199 train_time:664521ms step_avg:134.03ms
step:4969/6250 train_loss:2.2603 aux_loss:0.0198 train_time:664655ms step_avg:134.03ms
step:4970/6250 train_loss:2.2257 aux_loss:0.0198 train_time:664791ms step_avg:134.03ms
step:4971/6250 train_loss:2.0999 aux_loss:0.0198 train_time:664925ms step_avg:134.03ms
step:4972/6250 train_loss:2.1687 aux_loss:0.0198 train_time:665060ms step_avg:134.03ms
step:4973/6250 train_loss:2.1283 aux_loss:0.0195 train_time:665195ms step_avg:134.03ms
step:4974/6250 train_loss:2.1275 aux_loss:0.0195 train_time:665329ms step_avg:134.03ms
step:4975/6250 train_loss:2.2077 aux_loss:0.0198 train_time:665464ms step_avg:134.03ms
step:4976/6250 train_loss:2.2992 aux_loss:0.0201 train_time:665598ms step_avg:134.03ms
step:4977/6250 train_loss:2.1575 aux_loss:0.0202 train_time:665734ms step_avg:134.03ms
step:4978/6250 train_loss:2.1993 aux_loss:0.0201 train_time:665868ms step_avg:134.03ms
step:4979/6250 train_loss:2.1792 aux_loss:0.0199 train_time:666003ms step_avg:134.03ms
step:4980/6250 train_loss:2.1705 aux_loss:0.0201 train_time:666138ms step_avg:134.03ms
step:4981/6250 train_loss:2.0876 aux_loss:0.0198 train_time:666273ms step_avg:134.03ms
step:4982/6250 train_loss:2.1285 aux_loss:0.0196 train_time:666407ms step_avg:134.03ms
step:4983/6250 train_loss:2.1792 aux_loss:0.0198 train_time:666542ms step_avg:134.03ms
step:4984/6250 train_loss:2.1366 aux_loss:0.0198 train_time:666679ms step_avg:134.03ms
step:4985/6250 train_loss:2.1289 aux_loss:0.0198 train_time:666814ms step_avg:134.03ms
step:4986/6250 train_loss:2.2283 aux_loss:0.0197 train_time:666947ms step_avg:134.03ms
step:4987/6250 train_loss:2.1378 aux_loss:0.0196 train_time:667082ms step_avg:134.03ms
step:4988/6250 train_loss:2.1332 aux_loss:0.0195 train_time:667218ms step_avg:134.03ms
step:4989/6250 train_loss:2.1308 aux_loss:0.0192 train_time:667353ms step_avg:134.03ms
step:4990/6250 train_loss:2.1732 aux_loss:0.0196 train_time:667488ms step_avg:134.03ms
step:4991/6250 train_loss:2.1831 aux_loss:0.0196 train_time:667622ms step_avg:134.03ms
step:4992/6250 train_loss:2.1211 aux_loss:0.0195 train_time:667758ms step_avg:134.03ms
step:4993/6250 train_loss:2.1464 aux_loss:0.0196 train_time:667892ms step_avg:134.03ms
step:4994/6250 train_loss:2.0835 aux_loss:0.0194 train_time:668027ms step_avg:134.03ms
step:4995/6250 train_loss:2.2024 aux_loss:0.0196 train_time:668162ms step_avg:134.03ms
step:4996/6250 train_loss:2.2662 aux_loss:0.0198 train_time:668297ms step_avg:134.03ms
step:4997/6250 train_loss:2.2498 aux_loss:0.0200 train_time:668431ms step_avg:134.03ms
step:4998/6250 train_loss:2.1939 aux_loss:0.0199 train_time:668565ms step_avg:134.03ms
step:4999/6250 train_loss:2.1658 aux_loss:0.0197 train_time:668700ms step_avg:134.03ms
step:5000/6250 train_loss:2.1518 aux_loss:0.0197 train_time:668834ms step_avg:134.03ms
step:5001/6250 train_loss:2.1144 aux_loss:0.0199 train_time:668969ms step_avg:134.04ms
step:5002/6250 train_loss:2.2594 aux_loss:0.0198 train_time:669103ms step_avg:134.04ms
step:5003/6250 train_loss:2.1884 aux_loss:0.0196 train_time:669238ms step_avg:134.04ms
step:5004/6250 train_loss:2.1526 aux_loss:0.0191 train_time:669372ms step_avg:134.04ms
step:5005/6250 train_loss:2.1969 aux_loss:0.0191 train_time:669507ms step_avg:134.04ms
step:5006/6250 train_loss:2.2306 aux_loss:0.0194 train_time:669647ms step_avg:134.04ms
step:5007/6250 train_loss:2.1567 aux_loss:0.0193 train_time:669787ms step_avg:134.04ms
step:5008/6250 train_loss:2.1336 aux_loss:0.0192 train_time:669926ms step_avg:134.04ms
step:5009/6250 train_loss:2.1545 aux_loss:0.0195 train_time:670065ms step_avg:134.04ms
step:5010/6250 train_loss:2.2255 aux_loss:0.0194 train_time:670201ms step_avg:134.04ms
step:5011/6250 train_loss:2.0866 aux_loss:0.0195 train_time:670336ms step_avg:134.04ms
step:5012/6250 train_loss:2.1443 aux_loss:0.0199 train_time:670473ms step_avg:134.04ms
step:5013/6250 train_loss:2.1685 aux_loss:0.0200 train_time:670609ms step_avg:134.04ms
step:5014/6250 train_loss:2.1535 aux_loss:0.0201 train_time:670744ms step_avg:134.04ms
step:5015/6250 train_loss:2.2432 aux_loss:0.0198 train_time:670881ms step_avg:134.04ms
step:5016/6250 train_loss:2.1507 aux_loss:0.0196 train_time:671018ms step_avg:134.04ms
step:5017/6250 train_loss:2.1786 aux_loss:0.0196 train_time:671155ms step_avg:134.04ms
step:5018/6250 train_loss:2.2205 aux_loss:0.0193 train_time:671293ms step_avg:134.04ms
step:5019/6250 train_loss:2.2071 aux_loss:0.0194 train_time:671430ms step_avg:134.04ms
step:5020/6250 train_loss:2.1254 aux_loss:0.0194 train_time:671566ms step_avg:134.05ms
step:5021/6250 train_loss:2.2019 aux_loss:0.0194 train_time:671702ms step_avg:134.05ms
step:5022/6250 train_loss:2.1681 aux_loss:0.0194 train_time:671837ms step_avg:134.05ms
step:5023/6250 train_loss:2.0847 aux_loss:0.0194 train_time:671977ms step_avg:134.05ms
step:5024/6250 train_loss:2.1951 aux_loss:0.0197 train_time:672114ms step_avg:134.05ms
step:5025/6250 train_loss:2.1446 aux_loss:0.0198 train_time:672259ms step_avg:134.05ms
step:5026/6250 train_loss:2.1112 aux_loss:0.0196 train_time:672395ms step_avg:134.05ms
step:5027/6250 train_loss:2.1929 aux_loss:0.0193 train_time:672529ms step_avg:134.05ms
step:5028/6250 train_loss:2.1862 aux_loss:0.0194 train_time:672664ms step_avg:134.05ms
step:5029/6250 train_loss:2.1352 aux_loss:0.0192 train_time:672804ms step_avg:134.05ms
step:5030/6250 train_loss:2.1244 aux_loss:0.0193 train_time:672944ms step_avg:134.05ms
step:5031/6250 train_loss:2.1137 aux_loss:0.0193 train_time:673083ms step_avg:134.05ms
step:5032/6250 train_loss:2.1955 aux_loss:0.0197 train_time:673226ms step_avg:134.06ms
step:5033/6250 train_loss:2.0934 aux_loss:0.0194 train_time:673369ms step_avg:134.06ms
step:5034/6250 train_loss:2.1961 aux_loss:0.0193 train_time:673509ms step_avg:134.06ms
step:5035/6250 train_loss:2.2018 aux_loss:0.0193 train_time:673650ms step_avg:134.06ms
step:5036/6250 train_loss:2.1653 aux_loss:0.0193 train_time:673793ms step_avg:134.06ms
step:5037/6250 train_loss:2.1855 aux_loss:0.0191 train_time:673932ms step_avg:134.06ms
step:5038/6250 train_loss:2.1714 aux_loss:0.0190 train_time:674073ms step_avg:134.06ms
step:5039/6250 train_loss:2.1390 aux_loss:0.0189 train_time:674213ms step_avg:134.06ms
step:5040/6250 train_loss:2.1094 aux_loss:0.0191 train_time:674353ms step_avg:134.07ms
step:5041/6250 train_loss:2.0908 aux_loss:0.0192 train_time:674502ms step_avg:134.07ms
step:5042/6250 train_loss:2.1418 aux_loss:0.0191 train_time:674643ms step_avg:134.07ms
step:5043/6250 train_loss:2.1290 aux_loss:0.0190 train_time:674782ms step_avg:134.07ms
step:5044/6250 train_loss:2.1113 aux_loss:0.0193 train_time:674922ms step_avg:134.07ms
step:5045/6250 train_loss:2.1477 aux_loss:0.0189 train_time:675059ms step_avg:134.07ms
step:5046/6250 train_loss:2.0848 aux_loss:0.0189 train_time:675193ms step_avg:134.07ms
step:5047/6250 train_loss:2.1632 aux_loss:0.0189 train_time:675328ms step_avg:134.07ms
step:5048/6250 train_loss:2.1034 aux_loss:0.0189 train_time:675464ms step_avg:134.07ms
step:5049/6250 train_loss:2.1258 aux_loss:0.0188 train_time:675598ms step_avg:134.07ms
step:5050/6250 train_loss:2.0953 aux_loss:0.0190 train_time:675734ms step_avg:134.07ms
step:5051/6250 train_loss:2.1066 aux_loss:0.0185 train_time:675875ms step_avg:134.08ms
step:5052/6250 train_loss:2.1043 aux_loss:0.0192 train_time:676017ms step_avg:134.08ms
step:5053/6250 train_loss:2.1969 aux_loss:0.0193 train_time:676158ms step_avg:134.08ms
step:5054/6250 train_loss:2.1754 aux_loss:0.0192 train_time:676292ms step_avg:134.08ms
step:5055/6250 train_loss:2.1748 aux_loss:0.0188 train_time:676431ms step_avg:134.08ms
step:5056/6250 train_loss:2.1826 aux_loss:0.0189 train_time:676570ms step_avg:134.08ms
step:5057/6250 train_loss:2.1011 aux_loss:0.0192 train_time:676712ms step_avg:134.08ms
step:5058/6250 train_loss:2.0460 aux_loss:0.0193 train_time:676854ms step_avg:134.08ms
step:5059/6250 train_loss:2.1666 aux_loss:0.0193 train_time:676996ms step_avg:134.09ms
step:5060/6250 train_loss:2.1386 aux_loss:0.0195 train_time:677136ms step_avg:134.09ms
step:5061/6250 train_loss:2.1283 aux_loss:0.0197 train_time:677271ms step_avg:134.09ms
step:5062/6250 train_loss:2.1238 aux_loss:0.0194 train_time:677406ms step_avg:134.09ms
step:5063/6250 train_loss:2.1723 aux_loss:0.0199 train_time:677539ms step_avg:134.09ms
step:5064/6250 train_loss:2.0803 aux_loss:0.0196 train_time:677673ms step_avg:134.09ms
step:5065/6250 train_loss:2.1542 aux_loss:0.0197 train_time:677808ms step_avg:134.09ms
step:5066/6250 train_loss:2.1871 aux_loss:0.0192 train_time:677941ms step_avg:134.09ms
step:5067/6250 train_loss:2.2076 aux_loss:0.0192 train_time:678075ms step_avg:134.09ms
step:5068/6250 train_loss:2.1516 aux_loss:0.0190 train_time:678209ms step_avg:134.09ms
step:5069/6250 train_loss:2.1881 aux_loss:0.0189 train_time:678342ms step_avg:134.09ms
step:5070/6250 train_loss:2.1500 aux_loss:0.0186 train_time:678476ms step_avg:134.09ms
step:5071/6250 train_loss:2.0654 aux_loss:0.0190 train_time:678610ms step_avg:134.09ms
step:5072/6250 train_loss:2.1372 aux_loss:0.0189 train_time:678745ms step_avg:134.09ms
step:5073/6250 train_loss:2.1722 aux_loss:0.0192 train_time:678886ms step_avg:134.09ms
step:5074/6250 train_loss:2.1344 aux_loss:0.0193 train_time:679029ms step_avg:134.09ms
step:5075/6250 train_loss:2.1646 aux_loss:0.0190 train_time:679174ms step_avg:134.09ms
step:5076/6250 train_loss:2.1529 aux_loss:0.0192 train_time:679319ms step_avg:134.09ms
step:5077/6250 train_loss:2.1103 aux_loss:0.0191 train_time:679462ms step_avg:134.10ms
step:5078/6250 train_loss:2.2554 aux_loss:0.0193 train_time:679608ms step_avg:134.10ms
step:5079/6250 train_loss:2.1488 aux_loss:0.0191 train_time:679751ms step_avg:134.10ms
step:5080/6250 train_loss:2.2116 aux_loss:0.0191 train_time:679892ms step_avg:134.10ms
step:5081/6250 train_loss:2.0959 aux_loss:0.0189 train_time:680029ms step_avg:134.10ms
step:5082/6250 train_loss:2.0663 aux_loss:0.0189 train_time:680168ms step_avg:134.10ms
step:5083/6250 train_loss:2.1946 aux_loss:0.0190 train_time:680309ms step_avg:134.10ms
step:5084/6250 train_loss:2.1302 aux_loss:0.0193 train_time:680442ms step_avg:134.10ms
step:5085/6250 train_loss:2.1214 aux_loss:0.0193 train_time:680580ms step_avg:134.10ms
step:5086/6250 train_loss:2.2001 aux_loss:0.0192 train_time:680715ms step_avg:134.10ms
step:5087/6250 train_loss:2.1928 aux_loss:0.0193 train_time:680853ms step_avg:134.11ms
step:5088/6250 train_loss:2.2625 aux_loss:0.0196 train_time:680995ms step_avg:134.11ms
step:5089/6250 train_loss:2.0950 aux_loss:0.0191 train_time:681131ms step_avg:134.11ms
step:5090/6250 train_loss:2.2307 aux_loss:0.0190 train_time:681265ms step_avg:134.11ms
step:5091/6250 train_loss:2.1100 aux_loss:0.0193 train_time:681399ms step_avg:134.11ms
step:5092/6250 train_loss:2.1124 aux_loss:0.0193 train_time:681533ms step_avg:134.11ms
step:5093/6250 train_loss:2.1482 aux_loss:0.0195 train_time:681667ms step_avg:134.11ms
step:5094/6250 train_loss:2.1535 aux_loss:0.0195 train_time:681802ms step_avg:134.11ms
step:5095/6250 train_loss:2.1324 aux_loss:0.0193 train_time:681937ms step_avg:134.11ms
step:5096/6250 train_loss:2.1807 aux_loss:0.0190 train_time:682072ms step_avg:134.11ms
step:5097/6250 train_loss:2.1151 aux_loss:0.0191 train_time:682207ms step_avg:134.11ms
step:5098/6250 train_loss:2.1652 aux_loss:0.0191 train_time:682341ms step_avg:134.11ms
step:5099/6250 train_loss:2.1445 aux_loss:0.0193 train_time:682477ms step_avg:134.11ms
step:5100/6250 train_loss:2.2075 aux_loss:0.0194 train_time:682618ms step_avg:134.11ms
step:5101/6250 train_loss:2.1337 aux_loss:0.0192 train_time:682753ms step_avg:134.11ms
step:5102/6250 train_loss:2.1004 aux_loss:0.0192 train_time:682887ms step_avg:134.11ms
step:5103/6250 train_loss:2.1206 aux_loss:0.0191 train_time:683021ms step_avg:134.11ms
step:5104/6250 train_loss:2.1048 aux_loss:0.0192 train_time:683155ms step_avg:134.11ms
step:5105/6250 train_loss:2.1144 aux_loss:0.0193 train_time:683289ms step_avg:134.11ms
step:5106/6250 train_loss:2.1380 aux_loss:0.0191 train_time:683426ms step_avg:134.11ms
step:5107/6250 train_loss:2.2775 aux_loss:0.0193 train_time:683559ms step_avg:134.11ms
step:5108/6250 train_loss:2.1366 aux_loss:0.0193 train_time:683695ms step_avg:134.11ms
step:5109/6250 train_loss:2.1641 aux_loss:0.0192 train_time:683833ms step_avg:134.11ms
step:5110/6250 train_loss:2.1585 aux_loss:0.0189 train_time:683968ms step_avg:134.11ms
step:5111/6250 train_loss:2.1560 aux_loss:0.0189 train_time:684101ms step_avg:134.11ms
step:5112/6250 train_loss:2.1603 aux_loss:0.0189 train_time:684236ms step_avg:134.11ms
step:5113/6250 train_loss:2.1349 aux_loss:0.0190 train_time:684370ms step_avg:134.11ms
step:5114/6250 train_loss:2.2403 aux_loss:0.0188 train_time:684504ms step_avg:134.11ms
step:5115/6250 train_loss:2.2198 aux_loss:0.0188 train_time:684637ms step_avg:134.11ms
step:5116/6250 train_loss:2.1967 aux_loss:0.0193 train_time:684774ms step_avg:134.11ms
step:5117/6250 train_loss:2.2217 aux_loss:0.0194 train_time:684908ms step_avg:134.11ms
step:5118/6250 train_loss:2.1187 aux_loss:0.0194 train_time:685042ms step_avg:134.11ms
step:5119/6250 train_loss:2.1942 aux_loss:0.0195 train_time:685177ms step_avg:134.11ms
step:5120/6250 train_loss:2.1334 aux_loss:0.0194 train_time:685311ms step_avg:134.11ms
step:5121/6250 train_loss:2.1272 aux_loss:0.0195 train_time:685446ms step_avg:134.11ms
step:5122/6250 train_loss:2.2518 aux_loss:0.0198 train_time:685579ms step_avg:134.11ms
step:5123/6250 train_loss:2.1772 aux_loss:0.0195 train_time:685714ms step_avg:134.11ms
step:5124/6250 train_loss:2.0757 aux_loss:0.0190 train_time:685848ms step_avg:134.11ms
step:5125/6250 train_loss:2.1069 aux_loss:0.0190 train_time:685981ms step_avg:134.11ms
step:5126/6250 train_loss:2.1617 aux_loss:0.0191 train_time:686115ms step_avg:134.11ms
step:5127/6250 train_loss:2.0906 aux_loss:0.0192 train_time:686249ms step_avg:134.11ms
step:5128/6250 train_loss:2.1327 aux_loss:0.0191 train_time:686383ms step_avg:134.11ms
step:5129/6250 train_loss:2.1440 aux_loss:0.0193 train_time:686517ms step_avg:134.11ms
step:5130/6250 train_loss:2.1436 aux_loss:0.0193 train_time:686651ms step_avg:134.11ms
step:5131/6250 train_loss:2.1177 aux_loss:0.0192 train_time:686786ms step_avg:134.11ms
step:5132/6250 train_loss:2.1162 aux_loss:0.0191 train_time:686920ms step_avg:134.11ms
step:5133/6250 train_loss:2.1679 aux_loss:0.0192 train_time:687054ms step_avg:134.11ms
step:5134/6250 train_loss:2.1110 aux_loss:0.0192 train_time:687188ms step_avg:134.11ms
step:5135/6250 train_loss:2.1627 aux_loss:0.0191 train_time:687322ms step_avg:134.11ms
step:5136/6250 train_loss:2.0822 aux_loss:0.0189 train_time:687456ms step_avg:134.11ms
step:5137/6250 train_loss:2.1177 aux_loss:0.0188 train_time:687590ms step_avg:134.11ms
step:5138/6250 train_loss:2.1371 aux_loss:0.0191 train_time:687725ms step_avg:134.11ms
step:5139/6250 train_loss:2.1520 aux_loss:0.0188 train_time:687858ms step_avg:134.11ms
step:5140/6250 train_loss:2.0698 aux_loss:0.0186 train_time:687992ms step_avg:134.11ms
step:5141/6250 train_loss:2.1911 aux_loss:0.0190 train_time:688126ms step_avg:134.11ms
step:5142/6250 train_loss:2.1389 aux_loss:0.0190 train_time:688260ms step_avg:134.11ms
step:5143/6250 train_loss:2.1009 aux_loss:0.0190 train_time:688395ms step_avg:134.11ms
step:5144/6250 train_loss:2.1718 aux_loss:0.0189 train_time:688530ms step_avg:134.11ms
step:5145/6250 train_loss:2.1191 aux_loss:0.0187 train_time:688664ms step_avg:134.11ms
step:5146/6250 train_loss:2.1321 aux_loss:0.0188 train_time:688798ms step_avg:134.11ms
step:5147/6250 train_loss:2.1644 aux_loss:0.0190 train_time:688932ms step_avg:134.11ms
step:5148/6250 train_loss:2.2199 aux_loss:0.0193 train_time:689066ms step_avg:134.11ms
step:5149/6250 train_loss:2.0772 aux_loss:0.0188 train_time:689199ms step_avg:134.11ms
step:5150/6250 train_loss:2.0541 aux_loss:0.0187 train_time:689335ms step_avg:134.11ms
step:5151/6250 train_loss:2.0792 aux_loss:0.0189 train_time:689473ms step_avg:134.11ms
step:5152/6250 train_loss:2.1578 aux_loss:0.0189 train_time:689614ms step_avg:134.11ms
step:5153/6250 train_loss:2.1791 aux_loss:0.0191 train_time:689753ms step_avg:134.11ms
step:5154/6250 train_loss:2.2054 aux_loss:0.0194 train_time:689889ms step_avg:134.12ms
step:5155/6250 train_loss:2.1455 aux_loss:0.0187 train_time:690023ms step_avg:134.12ms
step:5156/6250 train_loss:2.1680 aux_loss:0.0187 train_time:690157ms step_avg:134.12ms
step:5157/6250 train_loss:2.1775 aux_loss:0.0187 train_time:690292ms step_avg:134.12ms
step:5158/6250 train_loss:2.1476 aux_loss:0.0187 train_time:690426ms step_avg:134.12ms
step:5159/6250 train_loss:2.2575 aux_loss:0.0188 train_time:690560ms step_avg:134.12ms
step:5160/6250 train_loss:2.1564 aux_loss:0.0188 train_time:690696ms step_avg:134.12ms
step:5161/6250 train_loss:2.1368 aux_loss:0.0187 train_time:690830ms step_avg:134.12ms
step:5162/6250 train_loss:2.2636 aux_loss:0.0189 train_time:690964ms step_avg:134.12ms
step:5163/6250 train_loss:2.2158 aux_loss:0.0189 train_time:691097ms step_avg:134.12ms
step:5164/6250 train_loss:2.2491 aux_loss:0.0191 train_time:691232ms step_avg:134.12ms
step:5165/6250 train_loss:2.1608 aux_loss:0.0188 train_time:691366ms step_avg:134.12ms
step:5166/6250 train_loss:2.1806 aux_loss:0.0187 train_time:691499ms step_avg:134.12ms
step:5167/6250 train_loss:2.1862 aux_loss:0.0187 train_time:691634ms step_avg:134.12ms
step:5168/6250 train_loss:2.0559 aux_loss:0.0185 train_time:691770ms step_avg:134.12ms
step:5169/6250 train_loss:2.1133 aux_loss:0.0186 train_time:691903ms step_avg:134.12ms
step:5170/6250 train_loss:2.1766 aux_loss:0.0188 train_time:692036ms step_avg:134.12ms
step:5171/6250 train_loss:2.0684 aux_loss:0.0182 train_time:692171ms step_avg:134.12ms
step:5172/6250 train_loss:2.1690 aux_loss:0.0186 train_time:692305ms step_avg:134.12ms
step:5173/6250 train_loss:2.2148 aux_loss:0.0187 train_time:692438ms step_avg:134.12ms
step:5174/6250 train_loss:2.1227 aux_loss:0.0185 train_time:692572ms step_avg:134.12ms
step:5175/6250 train_loss:2.1522 aux_loss:0.0187 train_time:692708ms step_avg:134.12ms
step:5176/6250 train_loss:2.0683 aux_loss:0.0185 train_time:692842ms step_avg:134.12ms
step:5177/6250 train_loss:2.0969 aux_loss:0.0188 train_time:692975ms step_avg:134.12ms
step:5178/6250 train_loss:2.0840 aux_loss:0.0188 train_time:693109ms step_avg:134.12ms
step:5179/6250 train_loss:2.1146 aux_loss:0.0190 train_time:693244ms step_avg:134.12ms
step:5180/6250 train_loss:2.1429 aux_loss:0.0190 train_time:693377ms step_avg:134.12ms
step:5181/6250 train_loss:2.1527 aux_loss:0.0189 train_time:693510ms step_avg:134.12ms
step:5182/6250 train_loss:2.0569 aux_loss:0.0189 train_time:693644ms step_avg:134.12ms
step:5183/6250 train_loss:2.2018 aux_loss:0.0188 train_time:693779ms step_avg:134.12ms
step:5184/6250 train_loss:2.0365 aux_loss:0.0185 train_time:693913ms step_avg:134.12ms
step:5185/6250 train_loss:2.0961 aux_loss:0.0188 train_time:694046ms step_avg:134.12ms
step:5186/6250 train_loss:2.0873 aux_loss:0.0187 train_time:694180ms step_avg:134.12ms
step:5187/6250 train_loss:2.1940 aux_loss:0.0189 train_time:694314ms step_avg:134.12ms
step:5188/6250 train_loss:2.1977 aux_loss:0.0189 train_time:694448ms step_avg:134.12ms
step:5189/6250 train_loss:2.1235 aux_loss:0.0189 train_time:694581ms step_avg:134.11ms
step:5190/6250 train_loss:2.1814 aux_loss:0.0188 train_time:694717ms step_avg:134.12ms
step:5191/6250 train_loss:2.0769 aux_loss:0.0189 train_time:694851ms step_avg:134.12ms
step:5192/6250 train_loss:2.1700 aux_loss:0.0188 train_time:694985ms step_avg:134.12ms
step:5193/6250 train_loss:2.2151 aux_loss:0.0192 train_time:695118ms step_avg:134.12ms
step:5194/6250 train_loss:2.0746 aux_loss:0.0191 train_time:695253ms step_avg:134.12ms
step:5195/6250 train_loss:2.0090 aux_loss:0.0187 train_time:695388ms step_avg:134.12ms
step:5196/6250 train_loss:2.1372 aux_loss:0.0186 train_time:695522ms step_avg:134.12ms
step:5197/6250 train_loss:2.2122 aux_loss:0.0184 train_time:695655ms step_avg:134.12ms
step:5198/6250 train_loss:2.0910 aux_loss:0.0185 train_time:695791ms step_avg:134.12ms
step:5199/6250 train_loss:2.1533 aux_loss:0.0189 train_time:695925ms step_avg:134.12ms
step:5200/6250 train_loss:2.1138 aux_loss:0.0187 train_time:696059ms step_avg:134.12ms
step:5201/6250 train_loss:2.1411 aux_loss:0.0187 train_time:696193ms step_avg:134.12ms
step:5202/6250 train_loss:2.2046 aux_loss:0.0188 train_time:696328ms step_avg:134.12ms
step:5203/6250 train_loss:2.1938 aux_loss:0.0185 train_time:696462ms step_avg:134.12ms
step:5204/6250 train_loss:2.1155 aux_loss:0.0187 train_time:696595ms step_avg:134.12ms
step:5205/6250 train_loss:2.1293 aux_loss:0.0186 train_time:696730ms step_avg:134.12ms
step:5206/6250 train_loss:2.0825 aux_loss:0.0187 train_time:696864ms step_avg:134.12ms
step:5207/6250 train_loss:2.3139 aux_loss:0.0190 train_time:696998ms step_avg:134.12ms
step:5208/6250 train_loss:2.1988 aux_loss:0.0188 train_time:697133ms step_avg:134.12ms
step:5209/6250 train_loss:2.2330 aux_loss:0.0189 train_time:697267ms step_avg:134.12ms
step:5210/6250 train_loss:2.1767 aux_loss:0.0187 train_time:697400ms step_avg:134.12ms
step:5211/6250 train_loss:2.1366 aux_loss:0.0189 train_time:697534ms step_avg:134.12ms
step:5212/6250 train_loss:2.1492 aux_loss:0.0187 train_time:697669ms step_avg:134.12ms
step:5213/6250 train_loss:2.0339 aux_loss:0.0190 train_time:697803ms step_avg:134.12ms
step:5214/6250 train_loss:2.1065 aux_loss:0.0190 train_time:697936ms step_avg:134.12ms
step:5215/6250 train_loss:2.0392 aux_loss:0.0188 train_time:698070ms step_avg:134.12ms
step:5216/6250 train_loss:2.1513 aux_loss:0.0186 train_time:698204ms step_avg:134.12ms
step:5217/6250 train_loss:2.1014 aux_loss:0.0188 train_time:698337ms step_avg:134.12ms
step:5218/6250 train_loss:2.1126 aux_loss:0.0186 train_time:698472ms step_avg:134.12ms
step:5219/6250 train_loss:2.1151 aux_loss:0.0186 train_time:698606ms step_avg:134.12ms
step:5220/6250 train_loss:2.1161 aux_loss:0.0189 train_time:698744ms step_avg:134.12ms
step:5221/6250 train_loss:2.1732 aux_loss:0.0191 train_time:698877ms step_avg:134.12ms
step:5222/6250 train_loss:2.1855 aux_loss:0.0190 train_time:699011ms step_avg:134.12ms
step:5223/6250 train_loss:2.2229 aux_loss:0.0189 train_time:699146ms step_avg:134.12ms
step:5224/6250 train_loss:2.1265 aux_loss:0.0190 train_time:699286ms step_avg:134.12ms
step:5225/6250 train_loss:2.1047 aux_loss:0.0188 train_time:699425ms step_avg:134.12ms
step:5226/6250 train_loss:2.1582 aux_loss:0.0186 train_time:699561ms step_avg:134.12ms
step:5227/6250 train_loss:2.2748 aux_loss:0.0187 train_time:699697ms step_avg:134.12ms
step:5228/6250 train_loss:2.1161 aux_loss:0.0188 train_time:699832ms step_avg:134.12ms
step:5229/6250 train_loss:2.2190 aux_loss:0.0184 train_time:699966ms step_avg:134.12ms
step:5230/6250 train_loss:2.1120 aux_loss:0.0183 train_time:700100ms step_avg:134.12ms
step:5231/6250 train_loss:2.1643 aux_loss:0.0184 train_time:700234ms step_avg:134.12ms
step:5232/6250 train_loss:2.1295 aux_loss:0.0184 train_time:700368ms step_avg:134.12ms
step:5233/6250 train_loss:2.1336 aux_loss:0.0184 train_time:700502ms step_avg:134.12ms
step:5234/6250 train_loss:2.1888 aux_loss:0.0186 train_time:700636ms step_avg:134.12ms
step:5235/6250 train_loss:2.1885 aux_loss:0.0185 train_time:700771ms step_avg:134.12ms
step:5236/6250 train_loss:2.2170 aux_loss:0.0189 train_time:700905ms step_avg:134.12ms
step:5237/6250 train_loss:2.0964 aux_loss:0.0190 train_time:701038ms step_avg:134.12ms
step:5238/6250 train_loss:2.1813 aux_loss:0.0189 train_time:701173ms step_avg:134.12ms
step:5239/6250 train_loss:2.1543 aux_loss:0.0191 train_time:701307ms step_avg:134.12ms
step:5240/6250 train_loss:2.1118 aux_loss:0.0190 train_time:701441ms step_avg:134.12ms
step:5241/6250 train_loss:2.0505 aux_loss:0.0187 train_time:701574ms step_avg:134.12ms
step:5242/6250 train_loss:2.1564 aux_loss:0.0187 train_time:701710ms step_avg:134.12ms
step:5243/6250 train_loss:2.1695 aux_loss:0.0188 train_time:701849ms step_avg:134.12ms
step:5244/6250 train_loss:2.1523 aux_loss:0.0187 train_time:701988ms step_avg:134.12ms
step:5245/6250 train_loss:2.1472 aux_loss:0.0186 train_time:702127ms step_avg:134.12ms
step:5246/6250 train_loss:2.1858 aux_loss:0.0184 train_time:702263ms step_avg:134.12ms
step:5247/6250 train_loss:2.1575 aux_loss:0.0183 train_time:702397ms step_avg:134.12ms
step:5248/6250 train_loss:2.1139 aux_loss:0.0183 train_time:702532ms step_avg:134.12ms
step:5249/6250 train_loss:2.1201 aux_loss:0.0185 train_time:702667ms step_avg:134.12ms
step:5250/6250 train_loss:2.1520 aux_loss:0.0187 train_time:702802ms step_avg:134.12ms
step:5251/6250 train_loss:2.2380 aux_loss:0.0190 train_time:702935ms step_avg:134.12ms
step:5252/6250 train_loss:2.1503 aux_loss:0.0184 train_time:703069ms step_avg:134.12ms
step:5253/6250 train_loss:2.1189 aux_loss:0.0184 train_time:703203ms step_avg:134.12ms
step:5254/6250 train_loss:2.0895 aux_loss:0.0184 train_time:703337ms step_avg:134.12ms
step:5255/6250 train_loss:2.1866 aux_loss:0.0185 train_time:703482ms step_avg:134.12ms
step:5256/6250 train_loss:2.1463 aux_loss:0.0187 train_time:703624ms step_avg:134.13ms
step:5257/6250 train_loss:2.0818 aux_loss:0.0185 train_time:703769ms step_avg:134.13ms
step:5258/6250 train_loss:2.2287 aux_loss:0.0185 train_time:703913ms step_avg:134.13ms
step:5259/6250 train_loss:2.2069 aux_loss:0.0186 train_time:704056ms step_avg:134.13ms
step:5260/6250 train_loss:2.1305 aux_loss:0.0183 train_time:704201ms step_avg:134.13ms
step:5261/6250 train_loss:2.0986 aux_loss:0.0182 train_time:704344ms step_avg:134.14ms
step:5262/6250 train_loss:2.1871 aux_loss:0.0185 train_time:704488ms step_avg:134.14ms
step:5263/6250 train_loss:2.1724 aux_loss:0.0186 train_time:704633ms step_avg:134.14ms
step:5264/6250 train_loss:2.1365 aux_loss:0.0188 train_time:704777ms step_avg:134.14ms
step:5265/6250 train_loss:2.2019 aux_loss:0.0188 train_time:704921ms step_avg:134.14ms
step:5266/6250 train_loss:2.1888 aux_loss:0.0186 train_time:705066ms step_avg:134.15ms
step:5267/6250 train_loss:2.1702 aux_loss:0.0190 train_time:705209ms step_avg:134.15ms
step:5268/6250 train_loss:2.2279 aux_loss:0.0187 train_time:705354ms step_avg:134.15ms
step:5269/6250 train_loss:2.1782 aux_loss:0.0183 train_time:705499ms step_avg:134.15ms
step:5270/6250 train_loss:2.1304 aux_loss:0.0182 train_time:705642ms step_avg:134.15ms
step:5271/6250 train_loss:2.1444 aux_loss:0.0184 train_time:705787ms step_avg:134.15ms
step:5272/6250 train_loss:2.1235 aux_loss:0.0185 train_time:705931ms step_avg:134.16ms
step:5273/6250 train_loss:2.1808 aux_loss:0.0187 train_time:706074ms step_avg:134.16ms
step:5274/6250 train_loss:2.1542 aux_loss:0.0188 train_time:706220ms step_avg:134.16ms
step:5275/6250 train_loss:2.1841 aux_loss:0.0189 train_time:706363ms step_avg:134.16ms
step:5276/6250 train_loss:2.0981 aux_loss:0.0184 train_time:706506ms step_avg:134.16ms
step:5277/6250 train_loss:2.0838 aux_loss:0.0185 train_time:706649ms step_avg:134.17ms
step:5278/6250 train_loss:2.2210 aux_loss:0.0189 train_time:706793ms step_avg:134.17ms
step:5279/6250 train_loss:2.1609 aux_loss:0.0185 train_time:706938ms step_avg:134.17ms
step:5280/6250 train_loss:2.0837 aux_loss:0.0183 train_time:707081ms step_avg:134.17ms
step:5281/6250 train_loss:2.1525 aux_loss:0.0187 train_time:707225ms step_avg:134.17ms
step:5282/6250 train_loss:2.1195 aux_loss:0.0187 train_time:707368ms step_avg:134.17ms
step:5283/6250 train_loss:2.1312 aux_loss:0.0186 train_time:707512ms step_avg:134.18ms
step:5284/6250 train_loss:2.1198 aux_loss:0.0188 train_time:707657ms step_avg:134.18ms
step:5285/6250 train_loss:2.0881 aux_loss:0.0187 train_time:707811ms step_avg:134.18ms
step:5286/6250 train_loss:2.1997 aux_loss:0.0186 train_time:707952ms step_avg:134.18ms
step:5287/6250 train_loss:2.1215 aux_loss:0.0187 train_time:708090ms step_avg:134.18ms
step:5288/6250 train_loss:2.2007 aux_loss:0.0188 train_time:708233ms step_avg:134.19ms
step:5289/6250 train_loss:2.3328 aux_loss:0.0192 train_time:708375ms step_avg:134.19ms
step:5290/6250 train_loss:2.2205 aux_loss:0.0191 train_time:708517ms step_avg:134.19ms
step:5291/6250 train_loss:2.1460 aux_loss:0.0189 train_time:708658ms step_avg:134.19ms
step:5292/6250 train_loss:2.0638 aux_loss:0.0187 train_time:708799ms step_avg:134.19ms
step:5293/6250 train_loss:2.1476 aux_loss:0.0185 train_time:708936ms step_avg:134.19ms
step:5294/6250 train_loss:2.2172 aux_loss:0.0185 train_time:709070ms step_avg:134.19ms
step:5295/6250 train_loss:2.0638 aux_loss:0.0185 train_time:709204ms step_avg:134.19ms
step:5296/6250 train_loss:2.1723 aux_loss:0.0188 train_time:709338ms step_avg:134.19ms
step:5297/6250 train_loss:2.1525 aux_loss:0.0188 train_time:709473ms step_avg:134.19ms
step:5298/6250 train_loss:2.1941 aux_loss:0.0186 train_time:709610ms step_avg:134.19ms
step:5299/6250 train_loss:2.1987 aux_loss:0.0185 train_time:709745ms step_avg:134.19ms
step:5300/6250 train_loss:2.1672 aux_loss:0.0182 train_time:709888ms step_avg:134.19ms
step:5301/6250 train_loss:2.0943 aux_loss:0.0183 train_time:710025ms step_avg:134.19ms
step:5302/6250 train_loss:2.1084 aux_loss:0.0182 train_time:710163ms step_avg:134.20ms
step:5303/6250 train_loss:2.2285 aux_loss:0.0184 train_time:710302ms step_avg:134.20ms
step:5304/6250 train_loss:2.1362 aux_loss:0.0185 train_time:710436ms step_avg:134.20ms
step:5305/6250 train_loss:2.1996 aux_loss:0.0185 train_time:710575ms step_avg:134.20ms
step:5306/6250 train_loss:2.1723 aux_loss:0.0184 train_time:710711ms step_avg:134.20ms
step:5307/6250 train_loss:2.0700 aux_loss:0.0183 train_time:710845ms step_avg:134.20ms
step:5308/6250 train_loss:2.0810 aux_loss:0.0184 train_time:710984ms step_avg:134.20ms
step:5309/6250 train_loss:2.1086 aux_loss:0.0182 train_time:711122ms step_avg:134.20ms
step:5310/6250 train_loss:2.2055 aux_loss:0.0185 train_time:711261ms step_avg:134.20ms
step:5311/6250 train_loss:2.1245 aux_loss:0.0185 train_time:711400ms step_avg:134.20ms
step:5312/6250 train_loss:2.1605 aux_loss:0.0185 train_time:711539ms step_avg:134.20ms
step:5313/6250 train_loss:2.1457 aux_loss:0.0183 train_time:711675ms step_avg:134.20ms
step:5314/6250 train_loss:2.1703 aux_loss:0.0181 train_time:711810ms step_avg:134.20ms
step:5315/6250 train_loss:2.0923 aux_loss:0.0183 train_time:711949ms step_avg:134.20ms
step:5316/6250 train_loss:2.1421 aux_loss:0.0184 train_time:712085ms step_avg:134.20ms
step:5317/6250 train_loss:2.1030 aux_loss:0.0184 train_time:712219ms step_avg:134.20ms
step:5318/6250 train_loss:2.2123 aux_loss:0.0185 train_time:712355ms step_avg:134.20ms
step:5319/6250 train_loss:2.0862 aux_loss:0.0182 train_time:712490ms step_avg:134.20ms
step:5320/6250 train_loss:2.1112 aux_loss:0.0181 train_time:712623ms step_avg:134.20ms
step:5321/6250 train_loss:2.1841 aux_loss:0.0186 train_time:712758ms step_avg:134.20ms
step:5322/6250 train_loss:2.1283 aux_loss:0.0184 train_time:712892ms step_avg:134.20ms
step:5323/6250 train_loss:2.1530 aux_loss:0.0184 train_time:713027ms step_avg:134.20ms
step:5324/6250 train_loss:2.1967 aux_loss:0.0183 train_time:713160ms step_avg:134.20ms
step:5325/6250 train_loss:2.0975 aux_loss:0.0187 train_time:713295ms step_avg:134.20ms
step:5326/6250 train_loss:2.2176 aux_loss:0.0186 train_time:713430ms step_avg:134.20ms
step:5327/6250 train_loss:2.1015 aux_loss:0.0181 train_time:713564ms step_avg:134.20ms
step:5328/6250 train_loss:2.1384 aux_loss:0.0184 train_time:713698ms step_avg:134.20ms
step:5329/6250 train_loss:2.1436 aux_loss:0.0187 train_time:713832ms step_avg:134.20ms
step:5330/6250 train_loss:2.1316 aux_loss:0.0184 train_time:713967ms step_avg:134.20ms
step:5331/6250 train_loss:2.1297 aux_loss:0.0182 train_time:714100ms step_avg:134.20ms
step:5332/6250 train_loss:2.1314 aux_loss:0.0183 train_time:714234ms step_avg:134.20ms
step:5333/6250 train_loss:2.1742 aux_loss:0.0184 train_time:714368ms step_avg:134.20ms
step:5334/6250 train_loss:2.1313 aux_loss:0.0182 train_time:714502ms step_avg:134.20ms
step:5335/6250 train_loss:2.1507 aux_loss:0.0183 train_time:714637ms step_avg:134.20ms
step:5336/6250 train_loss:2.1618 aux_loss:0.0186 train_time:714771ms step_avg:134.20ms
step:5337/6250 train_loss:2.1496 aux_loss:0.0182 train_time:714905ms step_avg:134.20ms
step:5338/6250 train_loss:2.1967 aux_loss:0.0184 train_time:715038ms step_avg:134.20ms
step:5339/6250 train_loss:2.1565 aux_loss:0.0187 train_time:715172ms step_avg:134.20ms
step:5340/6250 train_loss:2.2161 aux_loss:0.0186 train_time:715308ms step_avg:134.20ms
step:5341/6250 train_loss:2.2060 aux_loss:0.0189 train_time:715443ms step_avg:134.20ms
step:5342/6250 train_loss:2.1663 aux_loss:0.0187 train_time:715578ms step_avg:134.20ms
step:5343/6250 train_loss:2.1488 aux_loss:0.0183 train_time:715712ms step_avg:134.20ms
step:5344/6250 train_loss:2.1567 aux_loss:0.0183 train_time:715846ms step_avg:134.20ms
step:5345/6250 train_loss:2.1517 aux_loss:0.0186 train_time:715981ms step_avg:134.20ms
step:5346/6250 train_loss:2.1645 aux_loss:0.0189 train_time:716115ms step_avg:134.20ms
step:5347/6250 train_loss:2.0999 aux_loss:0.0181 train_time:716249ms step_avg:134.20ms
step:5348/6250 train_loss:2.1727 aux_loss:0.0181 train_time:716384ms step_avg:134.20ms
step:5349/6250 train_loss:2.1035 aux_loss:0.0182 train_time:716518ms step_avg:134.20ms
step:5350/6250 train_loss:2.1706 aux_loss:0.0183 train_time:716652ms step_avg:134.20ms
step:5351/6250 train_loss:2.1096 aux_loss:0.0180 train_time:716788ms step_avg:134.20ms
step:5352/6250 train_loss:2.1703 aux_loss:0.0185 train_time:716923ms step_avg:134.21ms
step:5353/6250 train_loss:2.0695 aux_loss:0.0183 train_time:717057ms step_avg:134.21ms
step:5354/6250 train_loss:2.1668 aux_loss:0.0184 train_time:717192ms step_avg:134.21ms
step:5355/6250 train_loss:2.0401 aux_loss:0.0181 train_time:717326ms step_avg:134.21ms
step:5356/6250 train_loss:2.1517 aux_loss:0.0184 train_time:717462ms step_avg:134.21ms
step:5357/6250 train_loss:2.1421 aux_loss:0.0185 train_time:717597ms step_avg:134.21ms
step:5358/6250 train_loss:2.0815 aux_loss:0.0181 train_time:717731ms step_avg:134.21ms
step:5359/6250 train_loss:2.0954 aux_loss:0.0181 train_time:717867ms step_avg:134.21ms
step:5360/6250 train_loss:2.1344 aux_loss:0.0184 train_time:718002ms step_avg:134.21ms
step:5361/6250 train_loss:2.0844 aux_loss:0.0183 train_time:718137ms step_avg:134.21ms
step:5362/6250 train_loss:2.1925 aux_loss:0.0181 train_time:718272ms step_avg:134.21ms
step:5363/6250 train_loss:2.1200 aux_loss:0.0179 train_time:718407ms step_avg:134.21ms
step:5364/6250 train_loss:2.0508 aux_loss:0.0176 train_time:718542ms step_avg:134.21ms
step:5365/6250 train_loss:2.1945 aux_loss:0.0178 train_time:718677ms step_avg:134.21ms
step:5366/6250 train_loss:2.1118 aux_loss:0.0178 train_time:718810ms step_avg:134.21ms
step:5367/6250 train_loss:2.2054 aux_loss:0.0178 train_time:718945ms step_avg:134.21ms
step:5368/6250 train_loss:2.0925 aux_loss:0.0177 train_time:719079ms step_avg:134.21ms
step:5369/6250 train_loss:2.1195 aux_loss:0.0180 train_time:719214ms step_avg:134.21ms
step:5370/6250 train_loss:2.1292 aux_loss:0.0181 train_time:719348ms step_avg:134.21ms
step:5371/6250 train_loss:2.0928 aux_loss:0.0181 train_time:719482ms step_avg:134.21ms
step:5372/6250 train_loss:2.1160 aux_loss:0.0180 train_time:719617ms step_avg:134.21ms
step:5373/6250 train_loss:2.0365 aux_loss:0.0180 train_time:719752ms step_avg:134.21ms
step:5374/6250 train_loss:2.1469 aux_loss:0.0182 train_time:719886ms step_avg:134.21ms
step:5375/6250 train_loss:2.1391 aux_loss:0.0182 train_time:720020ms step_avg:134.21ms
step:5376/6250 train_loss:2.1373 aux_loss:0.0179 train_time:720154ms step_avg:134.21ms
step:5377/6250 train_loss:2.1417 aux_loss:0.0180 train_time:720288ms step_avg:134.21ms
step:5378/6250 train_loss:2.1944 aux_loss:0.0180 train_time:720422ms step_avg:134.21ms
step:5379/6250 train_loss:2.1423 aux_loss:0.0179 train_time:720556ms step_avg:134.21ms
step:5380/6250 train_loss:2.0910 aux_loss:0.0176 train_time:720690ms step_avg:134.21ms
step:5381/6250 train_loss:2.0516 aux_loss:0.0175 train_time:720825ms step_avg:134.21ms
step:5382/6250 train_loss:2.0907 aux_loss:0.0178 train_time:720960ms step_avg:134.21ms
step:5383/6250 train_loss:2.2102 aux_loss:0.0179 train_time:721095ms step_avg:134.21ms
step:5384/6250 train_loss:2.1072 aux_loss:0.0181 train_time:721229ms step_avg:134.21ms
step:5385/6250 train_loss:2.1637 aux_loss:0.0179 train_time:721364ms step_avg:134.21ms
step:5386/6250 train_loss:2.0584 aux_loss:0.0180 train_time:721498ms step_avg:134.21ms
step:5387/6250 train_loss:2.1345 aux_loss:0.0179 train_time:721632ms step_avg:134.21ms
step:5388/6250 train_loss:2.2139 aux_loss:0.0184 train_time:721766ms step_avg:134.21ms
step:5389/6250 train_loss:2.2136 aux_loss:0.0183 train_time:721901ms step_avg:134.21ms
step:5390/6250 train_loss:2.0846 aux_loss:0.0180 train_time:722035ms step_avg:134.21ms
step:5391/6250 train_loss:2.1559 aux_loss:0.0179 train_time:722168ms step_avg:134.21ms
step:5392/6250 train_loss:2.1254 aux_loss:0.0181 train_time:722303ms step_avg:134.21ms
step:5393/6250 train_loss:2.1297 aux_loss:0.0181 train_time:722437ms step_avg:134.21ms
step:5394/6250 train_loss:2.0634 aux_loss:0.0179 train_time:722571ms step_avg:134.21ms
step:5395/6250 train_loss:2.2634 aux_loss:0.0184 train_time:722705ms step_avg:134.21ms
step:5396/6250 train_loss:2.1282 aux_loss:0.0183 train_time:722840ms step_avg:134.21ms
step:5397/6250 train_loss:2.1218 aux_loss:0.0185 train_time:722974ms step_avg:134.21ms
step:5398/6250 train_loss:2.1621 aux_loss:0.0182 train_time:723109ms step_avg:134.21ms
step:5399/6250 train_loss:2.1051 aux_loss:0.0179 train_time:723243ms step_avg:134.21ms
step:5400/6250 train_loss:2.1946 aux_loss:0.0179 train_time:723377ms step_avg:134.21ms
step:5401/6250 train_loss:2.1255 aux_loss:0.0181 train_time:723512ms step_avg:134.21ms
step:5402/6250 train_loss:2.0913 aux_loss:0.0180 train_time:723646ms step_avg:134.21ms
step:5403/6250 train_loss:2.1167 aux_loss:0.0181 train_time:723782ms step_avg:134.21ms
step:5404/6250 train_loss:2.2706 aux_loss:0.0187 train_time:723917ms step_avg:134.21ms
step:5405/6250 train_loss:2.0813 aux_loss:0.0180 train_time:724050ms step_avg:134.21ms
step:5406/6250 train_loss:2.1697 aux_loss:0.0182 train_time:724184ms step_avg:134.21ms
step:5407/6250 train_loss:2.0478 aux_loss:0.0180 train_time:724318ms step_avg:134.21ms
step:5408/6250 train_loss:2.1607 aux_loss:0.0183 train_time:724453ms step_avg:134.21ms
step:5409/6250 train_loss:2.2134 aux_loss:0.0185 train_time:724586ms step_avg:134.21ms
step:5410/6250 train_loss:2.0738 aux_loss:0.0183 train_time:724721ms step_avg:134.21ms
step:5411/6250 train_loss:2.1199 aux_loss:0.0186 train_time:724856ms step_avg:134.21ms
step:5412/6250 train_loss:2.1695 aux_loss:0.0185 train_time:724991ms step_avg:134.21ms
step:5413/6250 train_loss:2.0196 aux_loss:0.0183 train_time:725125ms step_avg:134.21ms
step:5414/6250 train_loss:2.2284 aux_loss:0.0183 train_time:725259ms step_avg:134.21ms
step:5415/6250 train_loss:2.1133 aux_loss:0.0181 train_time:725393ms step_avg:134.21ms
step:5416/6250 train_loss:2.0168 aux_loss:0.0186 train_time:725527ms step_avg:134.21ms
step:5417/6250 train_loss:2.2132 aux_loss:0.0187 train_time:725662ms step_avg:134.21ms
step:5418/6250 train_loss:2.0736 aux_loss:0.0182 train_time:725796ms step_avg:134.21ms
step:5419/6250 train_loss:2.1590 aux_loss:0.0183 train_time:725930ms step_avg:134.21ms
step:5420/6250 train_loss:2.1145 aux_loss:0.0180 train_time:726065ms step_avg:134.21ms
step:5421/6250 train_loss:2.1457 aux_loss:0.0180 train_time:726199ms step_avg:134.21ms
step:5422/6250 train_loss:2.0912 aux_loss:0.0180 train_time:726334ms step_avg:134.21ms
step:5423/6250 train_loss:2.2109 aux_loss:0.0184 train_time:726470ms step_avg:134.21ms
step:5424/6250 train_loss:2.1176 aux_loss:0.0182 train_time:726604ms step_avg:134.21ms
step:5425/6250 train_loss:2.1814 aux_loss:0.0183 train_time:726739ms step_avg:134.21ms
step:5426/6250 train_loss:2.1860 aux_loss:0.0183 train_time:726873ms step_avg:134.21ms
step:5427/6250 train_loss:2.1101 aux_loss:0.0184 train_time:727007ms step_avg:134.21ms
step:5428/6250 train_loss:2.1485 aux_loss:0.0185 train_time:727142ms step_avg:134.21ms
step:5429/6250 train_loss:2.1770 aux_loss:0.0183 train_time:727277ms step_avg:134.21ms
step:5430/6250 train_loss:2.1296 aux_loss:0.0181 train_time:727413ms step_avg:134.21ms
step:5431/6250 train_loss:2.0221 aux_loss:0.0179 train_time:727547ms step_avg:134.21ms
step:5432/6250 train_loss:2.1116 aux_loss:0.0180 train_time:727683ms step_avg:134.21ms
step:5433/6250 train_loss:2.0846 aux_loss:0.0182 train_time:727817ms step_avg:134.21ms
step:5434/6250 train_loss:2.2537 aux_loss:0.0184 train_time:727952ms step_avg:134.21ms
step:5435/6250 train_loss:2.1713 aux_loss:0.0181 train_time:728086ms step_avg:134.21ms
step:5436/6250 train_loss:2.1301 aux_loss:0.0182 train_time:728222ms step_avg:134.21ms
step:5437/6250 train_loss:2.1566 aux_loss:0.0180 train_time:728357ms step_avg:134.21ms
step:5438/6250 train_loss:2.1174 aux_loss:0.0179 train_time:728492ms step_avg:134.21ms
step:5439/6250 train_loss:2.0706 aux_loss:0.0179 train_time:728626ms step_avg:134.21ms
step:5440/6250 train_loss:2.1077 aux_loss:0.0181 train_time:728761ms step_avg:134.21ms
step:5441/6250 train_loss:2.1672 aux_loss:0.0184 train_time:728896ms step_avg:134.21ms
step:5442/6250 train_loss:2.1124 aux_loss:0.0185 train_time:729030ms step_avg:134.21ms
step:5443/6250 train_loss:2.0558 aux_loss:0.0181 train_time:729164ms step_avg:134.21ms
step:5444/6250 train_loss:2.1229 aux_loss:0.0181 train_time:729298ms step_avg:134.21ms
step:5445/6250 train_loss:2.1169 aux_loss:0.0180 train_time:729432ms step_avg:134.21ms
step:5446/6250 train_loss:2.1095 aux_loss:0.0178 train_time:729565ms step_avg:134.21ms
step:5447/6250 train_loss:2.1505 aux_loss:0.0180 train_time:729700ms step_avg:134.21ms
step:5448/6250 train_loss:2.1448 aux_loss:0.0181 train_time:729834ms step_avg:134.21ms
step:5449/6250 train_loss:2.1758 aux_loss:0.0183 train_time:729968ms step_avg:134.21ms
step:5450/6250 train_loss:2.2045 aux_loss:0.0183 train_time:730102ms step_avg:134.21ms
step:5451/6250 train_loss:2.1832 aux_loss:0.0180 train_time:730236ms step_avg:134.21ms
step:5452/6250 train_loss:2.1417 aux_loss:0.0181 train_time:730370ms step_avg:134.21ms
step:5453/6250 train_loss:2.1546 aux_loss:0.0180 train_time:730505ms step_avg:134.21ms
step:5454/6250 train_loss:2.0971 aux_loss:0.0182 train_time:730638ms step_avg:134.21ms
step:5455/6250 train_loss:2.0910 aux_loss:0.0178 train_time:730772ms step_avg:134.21ms
step:5456/6250 train_loss:2.0963 aux_loss:0.0179 train_time:730907ms step_avg:134.21ms
step:5457/6250 train_loss:2.1829 aux_loss:0.0179 train_time:731041ms step_avg:134.21ms
step:5458/6250 train_loss:2.0771 aux_loss:0.0180 train_time:731177ms step_avg:134.21ms
step:5459/6250 train_loss:2.1077 aux_loss:0.0182 train_time:731311ms step_avg:134.21ms
step:5460/6250 train_loss:2.1477 aux_loss:0.0184 train_time:731446ms step_avg:134.21ms
step:5461/6250 train_loss:2.1209 aux_loss:0.0186 train_time:731580ms step_avg:134.21ms
step:5462/6250 train_loss:2.1171 aux_loss:0.0185 train_time:731715ms step_avg:134.21ms
step:5463/6250 train_loss:2.1440 aux_loss:0.0182 train_time:731850ms step_avg:134.21ms
step:5464/6250 train_loss:2.1107 aux_loss:0.0183 train_time:731985ms step_avg:134.21ms
step:5465/6250 train_loss:2.1383 aux_loss:0.0184 train_time:732118ms step_avg:134.21ms
step:5466/6250 train_loss:2.0865 aux_loss:0.0185 train_time:732252ms step_avg:134.21ms
step:5467/6250 train_loss:2.2176 aux_loss:0.0185 train_time:732387ms step_avg:134.21ms
step:5468/6250 train_loss:2.0314 aux_loss:0.0183 train_time:732522ms step_avg:134.21ms
step:5469/6250 train_loss:2.1018 aux_loss:0.0181 train_time:732657ms step_avg:134.21ms
step:5470/6250 train_loss:2.1250 aux_loss:0.0182 train_time:732792ms step_avg:134.21ms
step:5471/6250 train_loss:2.1248 aux_loss:0.0181 train_time:732926ms step_avg:134.21ms
step:5472/6250 train_loss:2.1470 aux_loss:0.0182 train_time:733061ms step_avg:134.21ms
step:5473/6250 train_loss:2.1908 aux_loss:0.0180 train_time:733195ms step_avg:134.21ms
step:5474/6250 train_loss:2.1645 aux_loss:0.0181 train_time:733330ms step_avg:134.21ms
step:5475/6250 train_loss:2.1736 aux_loss:0.0182 train_time:733465ms step_avg:134.21ms
step:5476/6250 train_loss:2.1698 aux_loss:0.0180 train_time:733600ms step_avg:134.21ms
step:5477/6250 train_loss:2.1284 aux_loss:0.0177 train_time:733735ms step_avg:134.21ms
step:5478/6250 train_loss:2.2493 aux_loss:0.0178 train_time:733869ms step_avg:134.21ms
step:5479/6250 train_loss:2.1714 aux_loss:0.0179 train_time:734003ms step_avg:134.21ms
step:5480/6250 train_loss:2.1207 aux_loss:0.0180 train_time:734138ms step_avg:134.21ms
step:5481/6250 train_loss:2.1239 aux_loss:0.0178 train_time:734274ms step_avg:134.21ms
step:5482/6250 train_loss:2.1509 aux_loss:0.0175 train_time:734409ms step_avg:134.21ms
step:5483/6250 train_loss:2.1701 aux_loss:0.0177 train_time:734542ms step_avg:134.21ms
step:5484/6250 train_loss:2.0100 aux_loss:0.0175 train_time:734677ms step_avg:134.21ms
step:5485/6250 train_loss:2.0836 aux_loss:0.0178 train_time:734812ms step_avg:134.21ms
step:5486/6250 train_loss:2.0515 aux_loss:0.0179 train_time:734946ms step_avg:134.21ms
step:5487/6250 train_loss:2.2013 aux_loss:0.0179 train_time:735081ms step_avg:134.21ms
step:5488/6250 train_loss:2.0928 aux_loss:0.0174 train_time:735216ms step_avg:134.21ms
step:5489/6250 train_loss:2.0934 aux_loss:0.0177 train_time:735351ms step_avg:134.21ms
step:5490/6250 train_loss:2.1025 aux_loss:0.0177 train_time:735485ms step_avg:134.21ms
step:5491/6250 train_loss:2.1838 aux_loss:0.0180 train_time:735620ms step_avg:134.21ms
step:5492/6250 train_loss:2.1102 aux_loss:0.0180 train_time:735755ms step_avg:134.21ms
step:5493/6250 train_loss:2.1633 aux_loss:0.0180 train_time:735891ms step_avg:134.21ms
step:5494/6250 train_loss:2.1878 aux_loss:0.0179 train_time:736026ms step_avg:134.21ms
step:5495/6250 train_loss:2.0809 aux_loss:0.0179 train_time:736161ms step_avg:134.21ms
step:5496/6250 train_loss:2.1189 aux_loss:0.0177 train_time:736295ms step_avg:134.21ms
step:5497/6250 train_loss:2.1351 aux_loss:0.0180 train_time:736430ms step_avg:134.21ms
step:5498/6250 train_loss:2.0980 aux_loss:0.0177 train_time:736564ms step_avg:134.21ms
step:5499/6250 train_loss:2.0341 aux_loss:0.0177 train_time:736699ms step_avg:134.21ms
step:5500/6250 train_loss:2.2058 aux_loss:0.0179 train_time:736834ms step_avg:134.21ms
step:5501/6250 train_loss:2.2103 aux_loss:0.0182 train_time:736968ms step_avg:134.21ms
step:5502/6250 train_loss:2.1152 aux_loss:0.0178 train_time:737102ms step_avg:134.21ms
step:5503/6250 train_loss:2.2326 aux_loss:0.0179 train_time:737236ms step_avg:134.21ms
step:5504/6250 train_loss:2.1550 aux_loss:0.0177 train_time:737370ms step_avg:134.21ms
step:5505/6250 train_loss:2.1006 aux_loss:0.0177 train_time:737504ms step_avg:134.21ms
step:5506/6250 train_loss:2.1835 aux_loss:0.0179 train_time:737638ms step_avg:134.21ms
step:5507/6250 train_loss:2.1187 aux_loss:0.0179 train_time:737771ms step_avg:134.21ms
step:5508/6250 train_loss:2.1681 aux_loss:0.0176 train_time:737905ms step_avg:134.21ms
step:5509/6250 train_loss:2.1981 aux_loss:0.0177 train_time:738039ms step_avg:134.21ms
step:5510/6250 train_loss:2.0474 aux_loss:0.0177 train_time:738173ms step_avg:134.21ms
step:5511/6250 train_loss:2.2423 aux_loss:0.0179 train_time:738308ms step_avg:134.21ms
step:5512/6250 train_loss:2.1241 aux_loss:0.0179 train_time:738442ms step_avg:134.21ms
step:5513/6250 train_loss:2.0190 aux_loss:0.0182 train_time:738577ms step_avg:134.21ms
step:5514/6250 train_loss:2.2259 aux_loss:0.0182 train_time:738712ms step_avg:134.21ms
step:5515/6250 train_loss:2.1198 aux_loss:0.0177 train_time:738846ms step_avg:134.21ms
step:5516/6250 train_loss:2.1118 aux_loss:0.0178 train_time:738980ms step_avg:134.21ms
step:5517/6250 train_loss:2.2208 aux_loss:0.0178 train_time:739115ms step_avg:134.21ms
step:5518/6250 train_loss:2.0940 aux_loss:0.0175 train_time:739249ms step_avg:134.21ms
step:5519/6250 train_loss:2.2512 aux_loss:0.0175 train_time:739383ms step_avg:134.21ms
step:5520/6250 train_loss:2.1312 aux_loss:0.0174 train_time:739518ms step_avg:134.21ms
step:5521/6250 train_loss:2.1095 aux_loss:0.0175 train_time:739652ms step_avg:134.21ms
step:5522/6250 train_loss:2.1486 aux_loss:0.0175 train_time:739789ms step_avg:134.21ms
step:5523/6250 train_loss:2.0951 aux_loss:0.0173 train_time:739924ms step_avg:134.21ms
step:5524/6250 train_loss:2.1613 aux_loss:0.0177 train_time:740059ms step_avg:134.21ms
step:5525/6250 train_loss:2.1611 aux_loss:0.0178 train_time:740192ms step_avg:134.21ms
step:5526/6250 train_loss:2.1236 aux_loss:0.0176 train_time:740325ms step_avg:134.21ms
step:5527/6250 train_loss:2.1763 aux_loss:0.0178 train_time:740460ms step_avg:134.21ms
step:5528/6250 train_loss:2.1966 aux_loss:0.0180 train_time:740595ms step_avg:134.21ms
step:5529/6250 train_loss:2.0648 aux_loss:0.0177 train_time:740732ms step_avg:134.21ms
step:5530/6250 train_loss:2.0763 aux_loss:0.0178 train_time:740866ms step_avg:134.21ms
step:5531/6250 train_loss:2.0786 aux_loss:0.0180 train_time:741000ms step_avg:134.21ms
step:5532/6250 train_loss:2.1408 aux_loss:0.0183 train_time:741135ms step_avg:134.22ms
step:5533/6250 train_loss:2.1119 aux_loss:0.0177 train_time:741269ms step_avg:134.21ms
step:5534/6250 train_loss:2.1309 aux_loss:0.0176 train_time:741403ms step_avg:134.21ms
step:5535/6250 train_loss:2.1116 aux_loss:0.0176 train_time:741537ms step_avg:134.21ms
step:5536/6250 train_loss:2.1637 aux_loss:0.0179 train_time:741671ms step_avg:134.21ms
step:5537/6250 train_loss:2.1645 aux_loss:0.0180 train_time:741804ms step_avg:134.21ms
step:5538/6250 train_loss:2.1531 aux_loss:0.0175 train_time:741939ms step_avg:134.21ms
step:5539/6250 train_loss:2.1580 aux_loss:0.0175 train_time:742073ms step_avg:134.21ms
step:5540/6250 train_loss:2.2162 aux_loss:0.0178 train_time:742206ms step_avg:134.21ms
step:5541/6250 train_loss:2.1190 aux_loss:0.0178 train_time:742340ms step_avg:134.21ms
step:5542/6250 train_loss:2.1638 aux_loss:0.0176 train_time:742474ms step_avg:134.21ms
step:5543/6250 train_loss:2.1450 aux_loss:0.0173 train_time:742609ms step_avg:134.21ms
step:5544/6250 train_loss:2.1583 aux_loss:0.0175 train_time:742744ms step_avg:134.21ms
step:5545/6250 train_loss:2.1998 aux_loss:0.0178 train_time:742879ms step_avg:134.21ms
step:5546/6250 train_loss:2.1350 aux_loss:0.0177 train_time:743013ms step_avg:134.21ms
step:5547/6250 train_loss:2.1741 aux_loss:0.0176 train_time:743147ms step_avg:134.21ms
step:5548/6250 train_loss:2.1452 aux_loss:0.0177 train_time:743282ms step_avg:134.21ms
step:5549/6250 train_loss:2.1645 aux_loss:0.0180 train_time:743417ms step_avg:134.21ms
step:5550/6250 train_loss:2.1169 aux_loss:0.0177 train_time:743550ms step_avg:134.21ms
step:5551/6250 train_loss:2.2618 aux_loss:0.0175 train_time:743684ms step_avg:134.21ms
step:5552/6250 train_loss:2.1311 aux_loss:0.0175 train_time:743819ms step_avg:134.21ms
step:5553/6250 train_loss:2.1123 aux_loss:0.0177 train_time:743954ms step_avg:134.21ms
step:5554/6250 train_loss:2.1558 aux_loss:0.0176 train_time:744090ms step_avg:134.22ms
step:5555/6250 train_loss:2.0810 aux_loss:0.0178 train_time:744223ms step_avg:134.22ms
step:5556/6250 train_loss:2.1157 aux_loss:0.0177 train_time:744358ms step_avg:134.22ms
step:5557/6250 train_loss:2.2883 aux_loss:0.0182 train_time:744494ms step_avg:134.22ms
step:5558/6250 train_loss:2.1458 aux_loss:0.0182 train_time:744628ms step_avg:134.22ms
step:5559/6250 train_loss:2.1139 aux_loss:0.0180 train_time:744764ms step_avg:134.22ms
step:5560/6250 train_loss:2.1926 aux_loss:0.0180 train_time:744899ms step_avg:134.22ms
step:5561/6250 train_loss:2.1347 aux_loss:0.0177 train_time:745033ms step_avg:134.22ms
step:5562/6250 train_loss:2.1304 aux_loss:0.0177 train_time:745168ms step_avg:134.22ms
step:5563/6250 train_loss:2.2503 aux_loss:0.0178 train_time:745302ms step_avg:134.22ms
step:5564/6250 train_loss:2.1829 aux_loss:0.0184 train_time:745438ms step_avg:134.22ms
step:5565/6250 train_loss:2.1384 aux_loss:0.0178 train_time:745572ms step_avg:134.22ms
step:5566/6250 train_loss:2.1563 aux_loss:0.0180 train_time:745707ms step_avg:134.22ms
step:5567/6250 train_loss:2.1884 aux_loss:0.0180 train_time:745841ms step_avg:134.22ms
step:5568/6250 train_loss:2.1141 aux_loss:0.0175 train_time:745977ms step_avg:134.22ms
step:5569/6250 train_loss:2.1228 aux_loss:0.0179 train_time:746110ms step_avg:134.22ms
step:5570/6250 train_loss:2.1192 aux_loss:0.0176 train_time:746245ms step_avg:134.22ms
step:5571/6250 train_loss:2.1215 aux_loss:0.0176 train_time:746381ms step_avg:134.22ms
step:5572/6250 train_loss:2.1582 aux_loss:0.0178 train_time:746515ms step_avg:134.22ms
step:5573/6250 train_loss:2.1807 aux_loss:0.0175 train_time:746649ms step_avg:134.22ms
step:5574/6250 train_loss:2.2174 aux_loss:0.0176 train_time:746784ms step_avg:134.22ms
step:5575/6250 train_loss:2.0972 aux_loss:0.0174 train_time:746919ms step_avg:134.22ms
step:5576/6250 train_loss:2.1105 aux_loss:0.0177 train_time:747052ms step_avg:134.22ms
step:5577/6250 train_loss:2.1696 aux_loss:0.0177 train_time:747186ms step_avg:134.22ms
step:5578/6250 train_loss:2.1711 aux_loss:0.0176 train_time:747319ms step_avg:134.22ms
step:5579/6250 train_loss:2.0952 aux_loss:0.0176 train_time:747455ms step_avg:134.22ms
step:5580/6250 train_loss:2.2217 aux_loss:0.0178 train_time:747589ms step_avg:134.22ms
step:5581/6250 train_loss:2.1765 aux_loss:0.0178 train_time:747722ms step_avg:134.22ms
step:5582/6250 train_loss:2.1996 aux_loss:0.0181 train_time:747857ms step_avg:134.22ms
step:5583/6250 train_loss:2.0466 aux_loss:0.0180 train_time:747992ms step_avg:134.22ms
step:5584/6250 train_loss:2.2402 aux_loss:0.0178 train_time:748125ms step_avg:134.22ms
step:5585/6250 train_loss:2.1201 aux_loss:0.0178 train_time:748259ms step_avg:134.22ms
step:5586/6250 train_loss:2.1079 aux_loss:0.0176 train_time:748393ms step_avg:134.22ms
step:5587/6250 train_loss:2.1182 aux_loss:0.0176 train_time:748527ms step_avg:134.22ms
step:5588/6250 train_loss:2.1681 aux_loss:0.0175 train_time:748660ms step_avg:134.22ms
step:5589/6250 train_loss:2.0656 aux_loss:0.0176 train_time:748795ms step_avg:134.22ms
step:5590/6250 train_loss:2.1742 aux_loss:0.0179 train_time:748934ms step_avg:134.22ms
step:5591/6250 train_loss:2.0866 aux_loss:0.0175 train_time:749069ms step_avg:134.22ms
step:5592/6250 train_loss:2.1617 aux_loss:0.0174 train_time:749202ms step_avg:134.22ms
step:5593/6250 train_loss:2.1888 aux_loss:0.0175 train_time:749342ms step_avg:134.22ms
step:5594/6250 train_loss:2.1061 aux_loss:0.0175 train_time:749480ms step_avg:134.22ms
step:5595/6250 train_loss:2.1469 aux_loss:0.0174 train_time:749621ms step_avg:134.22ms
step:5596/6250 train_loss:2.1294 aux_loss:0.0173 train_time:749766ms step_avg:134.22ms
step:5597/6250 train_loss:2.1312 aux_loss:0.0175 train_time:749908ms step_avg:134.22ms
step:5598/6250 train_loss:2.1723 aux_loss:0.0177 train_time:750052ms step_avg:134.23ms
step:5599/6250 train_loss:2.1186 aux_loss:0.0176 train_time:750194ms step_avg:134.23ms
step:5600/6250 train_loss:2.0975 aux_loss:0.0175 train_time:750329ms step_avg:134.23ms
step:5601/6250 train_loss:2.0591 aux_loss:0.0175 train_time:750465ms step_avg:134.23ms
step:5602/6250 train_loss:2.0789 aux_loss:0.0175 train_time:750609ms step_avg:134.23ms
step:5603/6250 train_loss:2.0731 aux_loss:0.0175 train_time:750748ms step_avg:134.23ms
step:5604/6250 train_loss:2.2034 aux_loss:0.0179 train_time:750883ms step_avg:134.23ms
step:5605/6250 train_loss:2.0847 aux_loss:0.0180 train_time:751017ms step_avg:134.23ms
step:5606/6250 train_loss:2.1080 aux_loss:0.0177 train_time:751151ms step_avg:134.23ms
step:5607/6250 train_loss:2.1429 aux_loss:0.0176 train_time:751286ms step_avg:134.23ms
step:5608/6250 train_loss:2.1948 aux_loss:0.0172 train_time:751420ms step_avg:134.23ms
step:5609/6250 train_loss:2.1504 aux_loss:0.0174 train_time:751554ms step_avg:134.23ms
step:5610/6250 train_loss:2.1351 aux_loss:0.0175 train_time:751689ms step_avg:134.23ms
step:5611/6250 train_loss:2.1416 aux_loss:0.0178 train_time:751823ms step_avg:134.23ms
step:5612/6250 train_loss:2.1027 aux_loss:0.0173 train_time:751958ms step_avg:134.23ms
step:5613/6250 train_loss:2.1199 aux_loss:0.0173 train_time:752092ms step_avg:134.23ms
step:5614/6250 train_loss:2.0457 aux_loss:0.0172 train_time:752225ms step_avg:134.23ms
step:5615/6250 train_loss:2.0962 aux_loss:0.0173 train_time:752360ms step_avg:134.23ms
step:5616/6250 train_loss:2.0655 aux_loss:0.0174 train_time:752494ms step_avg:134.23ms
step:5617/6250 train_loss:2.1900 aux_loss:0.0174 train_time:752628ms step_avg:134.23ms
step:5618/6250 train_loss:2.1117 aux_loss:0.0174 train_time:752762ms step_avg:134.23ms
step:5619/6250 train_loss:2.1804 aux_loss:0.0172 train_time:752897ms step_avg:134.23ms
step:5620/6250 train_loss:2.1475 aux_loss:0.0176 train_time:753031ms step_avg:134.23ms
step:5621/6250 train_loss:2.1531 aux_loss:0.0176 train_time:753165ms step_avg:134.23ms
step:5622/6250 train_loss:2.1685 aux_loss:0.0179 train_time:753300ms step_avg:134.23ms
step:5623/6250 train_loss:2.1184 aux_loss:0.0176 train_time:753435ms step_avg:134.23ms
step:5624/6250 train_loss:2.0675 aux_loss:0.0175 train_time:753569ms step_avg:134.23ms
step:5625/6250 train_loss:2.0926 aux_loss:0.0178 train_time:753702ms step_avg:134.23ms
step:5626/6250 train_loss:2.1505 aux_loss:0.0176 train_time:753837ms step_avg:134.23ms
step:5627/6250 train_loss:2.1671 aux_loss:0.0176 train_time:753971ms step_avg:134.23ms
step:5628/6250 train_loss:2.1161 aux_loss:0.0176 train_time:754105ms step_avg:134.23ms
step:5629/6250 train_loss:2.1810 aux_loss:0.0175 train_time:754239ms step_avg:134.23ms
step:5630/6250 train_loss:2.1772 aux_loss:0.0174 train_time:754373ms step_avg:134.23ms
step:5631/6250 train_loss:2.1847 aux_loss:0.0173 train_time:754507ms step_avg:134.23ms
step:5632/6250 train_loss:2.0039 aux_loss:0.0172 train_time:754641ms step_avg:134.23ms
step:5633/6250 train_loss:2.0617 aux_loss:0.0175 train_time:754777ms step_avg:134.23ms
step:5634/6250 train_loss:2.0919 aux_loss:0.0172 train_time:754911ms step_avg:134.23ms
step:5635/6250 train_loss:2.0935 aux_loss:0.0171 train_time:755046ms step_avg:134.23ms
step:5636/6250 train_loss:2.1063 aux_loss:0.0173 train_time:755179ms step_avg:134.23ms
step:5637/6250 train_loss:2.0768 aux_loss:0.0174 train_time:755314ms step_avg:134.23ms
step:5638/6250 train_loss:2.1523 aux_loss:0.0173 train_time:755448ms step_avg:134.23ms
step:5639/6250 train_loss:2.0629 aux_loss:0.0176 train_time:755582ms step_avg:134.23ms
step:5640/6250 train_loss:2.1006 aux_loss:0.0177 train_time:755717ms step_avg:134.23ms
step:5641/6250 train_loss:2.2951 aux_loss:0.0179 train_time:755851ms step_avg:134.23ms
step:5642/6250 train_loss:2.1370 aux_loss:0.0176 train_time:755985ms step_avg:134.23ms
step:5643/6250 train_loss:2.1225 aux_loss:0.0174 train_time:756119ms step_avg:134.23ms
step:5644/6250 train_loss:2.1110 aux_loss:0.0173 train_time:756254ms step_avg:134.23ms
step:5645/6250 train_loss:2.1946 aux_loss:0.0170 train_time:756388ms step_avg:134.23ms
step:5646/6250 train_loss:2.1506 aux_loss:0.0176 train_time:756522ms step_avg:134.23ms
step:5647/6250 train_loss:2.0927 aux_loss:0.0174 train_time:756657ms step_avg:134.23ms
step:5648/6250 train_loss:2.1796 aux_loss:0.0172 train_time:756791ms step_avg:134.23ms
step:5649/6250 train_loss:2.0967 aux_loss:0.0174 train_time:756925ms step_avg:134.23ms
step:5650/6250 train_loss:2.1584 aux_loss:0.0174 train_time:757059ms step_avg:134.23ms
step:5651/6250 train_loss:2.1389 aux_loss:0.0175 train_time:757194ms step_avg:134.23ms
step:5652/6250 train_loss:2.0806 aux_loss:0.0176 train_time:757328ms step_avg:134.23ms
step:5653/6250 train_loss:2.1909 aux_loss:0.0177 train_time:757462ms step_avg:134.23ms
step:5654/6250 train_loss:2.1325 aux_loss:0.0175 train_time:757596ms step_avg:134.23ms
step:5655/6250 train_loss:2.1505 aux_loss:0.0173 train_time:757730ms step_avg:134.23ms
step:5656/6250 train_loss:2.1903 aux_loss:0.0176 train_time:757865ms step_avg:134.23ms
step:5657/6250 train_loss:2.1198 aux_loss:0.0175 train_time:757998ms step_avg:134.23ms
step:5658/6250 train_loss:2.1021 aux_loss:0.0174 train_time:758133ms step_avg:134.23ms
step:5659/6250 train_loss:2.0938 aux_loss:0.0176 train_time:758267ms step_avg:134.23ms
step:5660/6250 train_loss:2.1048 aux_loss:0.0172 train_time:758400ms step_avg:134.23ms
step:5661/6250 train_loss:2.1268 aux_loss:0.0172 train_time:758535ms step_avg:134.23ms
step:5662/6250 train_loss:2.2614 aux_loss:0.0171 train_time:758670ms step_avg:134.23ms
step:5663/6250 train_loss:2.0877 aux_loss:0.0170 train_time:758804ms step_avg:134.23ms
step:5664/6250 train_loss:2.1370 aux_loss:0.0171 train_time:758938ms step_avg:134.23ms
step:5665/6250 train_loss:2.1739 aux_loss:0.0172 train_time:759073ms step_avg:134.23ms
step:5666/6250 train_loss:2.0949 aux_loss:0.0166 train_time:759207ms step_avg:134.23ms
step:5667/6250 train_loss:2.0998 aux_loss:0.0169 train_time:759341ms step_avg:134.23ms
step:5668/6250 train_loss:2.1794 aux_loss:0.0168 train_time:759475ms step_avg:134.23ms
step:5669/6250 train_loss:2.1637 aux_loss:0.0169 train_time:759608ms step_avg:134.23ms
step:5670/6250 train_loss:2.0263 aux_loss:0.0170 train_time:759745ms step_avg:134.23ms
step:5671/6250 train_loss:2.0980 aux_loss:0.0168 train_time:759878ms step_avg:134.23ms
step:5672/6250 train_loss:2.1264 aux_loss:0.0171 train_time:760012ms step_avg:134.23ms
step:5673/6250 train_loss:2.1981 aux_loss:0.0173 train_time:760146ms step_avg:134.23ms
step:5674/6250 train_loss:2.1510 aux_loss:0.0173 train_time:760280ms step_avg:134.23ms
step:5675/6250 train_loss:2.1712 aux_loss:0.0177 train_time:760415ms step_avg:134.23ms
step:5676/6250 train_loss:2.1876 aux_loss:0.0173 train_time:760549ms step_avg:134.23ms
step:5677/6250 train_loss:2.1194 aux_loss:0.0171 train_time:760683ms step_avg:134.23ms
step:5678/6250 train_loss:2.1853 aux_loss:0.0177 train_time:760817ms step_avg:134.23ms
step:5679/6250 train_loss:2.1904 aux_loss:0.0171 train_time:760952ms step_avg:134.23ms
step:5680/6250 train_loss:2.1490 aux_loss:0.0172 train_time:761085ms step_avg:134.23ms
step:5681/6250 train_loss:2.1571 aux_loss:0.0173 train_time:761219ms step_avg:134.23ms
step:5682/6250 train_loss:2.0535 aux_loss:0.0168 train_time:761354ms step_avg:134.23ms
step:5683/6250 train_loss:2.1337 aux_loss:0.0170 train_time:761488ms step_avg:134.23ms
step:5684/6250 train_loss:2.1481 aux_loss:0.0169 train_time:761622ms step_avg:134.23ms
step:5685/6250 train_loss:2.2160 aux_loss:0.0172 train_time:761757ms step_avg:134.23ms
step:5686/6250 train_loss:2.1551 aux_loss:0.0173 train_time:761892ms step_avg:134.23ms
step:5687/6250 train_loss:2.2407 aux_loss:0.0171 train_time:762025ms step_avg:134.23ms
step:5688/6250 train_loss:2.1725 aux_loss:0.0171 train_time:762159ms step_avg:134.23ms
step:5689/6250 train_loss:2.0543 aux_loss:0.0171 train_time:762293ms step_avg:134.23ms
step:5690/6250 train_loss:2.2130 aux_loss:0.0173 train_time:762429ms step_avg:134.23ms
step:5691/6250 train_loss:2.2184 aux_loss:0.0172 train_time:762563ms step_avg:134.23ms
step:5692/6250 train_loss:2.1847 aux_loss:0.0170 train_time:762696ms step_avg:134.23ms
step:5693/6250 train_loss:2.1715 aux_loss:0.0169 train_time:762831ms step_avg:134.23ms
step:5694/6250 train_loss:2.1451 aux_loss:0.0169 train_time:762965ms step_avg:134.23ms
step:5695/6250 train_loss:2.1834 aux_loss:0.0167 train_time:763098ms step_avg:134.23ms
step:5696/6250 train_loss:2.1916 aux_loss:0.0167 train_time:763234ms step_avg:134.23ms
step:5697/6250 train_loss:2.1795 aux_loss:0.0165 train_time:763368ms step_avg:134.23ms
step:5698/6250 train_loss:2.1570 aux_loss:0.0166 train_time:763504ms step_avg:134.23ms
step:5699/6250 train_loss:2.1932 aux_loss:0.0168 train_time:763637ms step_avg:134.23ms
step:5700/6250 train_loss:2.1627 aux_loss:0.0166 train_time:763773ms step_avg:134.23ms
step:5701/6250 train_loss:2.1974 aux_loss:0.0170 train_time:763907ms step_avg:134.23ms
step:5702/6250 train_loss:2.1985 aux_loss:0.0171 train_time:764040ms step_avg:134.23ms
step:5703/6250 train_loss:2.1527 aux_loss:0.0172 train_time:764174ms step_avg:134.23ms
step:5704/6250 train_loss:2.2214 aux_loss:0.0172 train_time:764308ms step_avg:134.23ms
step:5705/6250 train_loss:2.1344 aux_loss:0.0174 train_time:764442ms step_avg:134.23ms
step:5706/6250 train_loss:2.0362 aux_loss:0.0170 train_time:764576ms step_avg:134.23ms
step:5707/6250 train_loss:2.1182 aux_loss:0.0173 train_time:764712ms step_avg:134.23ms
step:5708/6250 train_loss:2.2058 aux_loss:0.0172 train_time:764846ms step_avg:134.23ms
step:5709/6250 train_loss:2.1663 aux_loss:0.0171 train_time:764979ms step_avg:134.23ms
step:5710/6250 train_loss:2.1425 aux_loss:0.0171 train_time:765114ms step_avg:134.23ms
step:5711/6250 train_loss:2.1525 aux_loss:0.0172 train_time:765248ms step_avg:134.23ms
step:5712/6250 train_loss:2.0942 aux_loss:0.0171 train_time:765382ms step_avg:134.23ms
step:5713/6250 train_loss:2.2743 aux_loss:0.0173 train_time:765515ms step_avg:134.23ms
step:5714/6250 train_loss:2.1120 aux_loss:0.0171 train_time:765654ms step_avg:134.23ms
step:5715/6250 train_loss:2.1701 aux_loss:0.0171 train_time:765795ms step_avg:134.23ms
step:5716/6250 train_loss:2.2111 aux_loss:0.0171 train_time:765934ms step_avg:134.23ms
step:5717/6250 train_loss:2.1611 aux_loss:0.0175 train_time:766074ms step_avg:134.23ms
step:5718/6250 train_loss:2.1075 aux_loss:0.0171 train_time:766211ms step_avg:134.23ms
step:5719/6250 train_loss:2.1034 aux_loss:0.0170 train_time:766345ms step_avg:134.23ms
step:5720/6250 train_loss:2.2252 aux_loss:0.0170 train_time:766486ms step_avg:134.24ms
step:5721/6250 train_loss:2.1400 aux_loss:0.0173 train_time:766628ms step_avg:134.24ms
step:5722/6250 train_loss:2.0666 aux_loss:0.0170 train_time:766767ms step_avg:134.24ms
step:5723/6250 train_loss:2.2312 aux_loss:0.0172 train_time:766907ms step_avg:134.24ms
step:5724/6250 train_loss:2.1828 aux_loss:0.0176 train_time:767041ms step_avg:134.24ms
step:5725/6250 train_loss:2.0824 aux_loss:0.0175 train_time:767175ms step_avg:134.24ms
step:5726/6250 train_loss:2.0421 aux_loss:0.0172 train_time:767311ms step_avg:134.24ms
step:5727/6250 train_loss:2.1415 aux_loss:0.0176 train_time:767445ms step_avg:134.24ms
step:5728/6250 train_loss:2.0720 aux_loss:0.0171 train_time:767578ms step_avg:134.24ms
step:5729/6250 train_loss:2.0777 aux_loss:0.0175 train_time:767713ms step_avg:134.24ms
step:5730/6250 train_loss:2.0299 aux_loss:0.0172 train_time:767847ms step_avg:134.24ms
step:5731/6250 train_loss:2.1045 aux_loss:0.0172 train_time:767981ms step_avg:134.24ms
step:5732/6250 train_loss:2.0764 aux_loss:0.0169 train_time:768114ms step_avg:134.24ms
step:5733/6250 train_loss:2.1222 aux_loss:0.0170 train_time:768249ms step_avg:134.24ms
step:5734/6250 train_loss:2.1098 aux_loss:0.0170 train_time:768383ms step_avg:134.24ms
step:5735/6250 train_loss:2.0671 aux_loss:0.0171 train_time:768517ms step_avg:134.24ms
step:5736/6250 train_loss:2.1703 aux_loss:0.0169 train_time:768651ms step_avg:134.24ms
step:5737/6250 train_loss:2.1971 aux_loss:0.0170 train_time:768788ms step_avg:134.24ms
step:5738/6250 train_loss:2.0912 aux_loss:0.0171 train_time:768923ms step_avg:134.24ms
step:5739/6250 train_loss:2.2490 aux_loss:0.0175 train_time:769056ms step_avg:134.24ms
step:5740/6250 train_loss:2.1903 aux_loss:0.0175 train_time:769190ms step_avg:134.24ms
step:5741/6250 train_loss:2.1072 aux_loss:0.0173 train_time:769325ms step_avg:134.24ms
step:5742/6250 train_loss:2.0805 aux_loss:0.0172 train_time:769459ms step_avg:134.24ms
step:5743/6250 train_loss:2.1325 aux_loss:0.0173 train_time:769593ms step_avg:134.24ms
step:5744/6250 train_loss:2.1006 aux_loss:0.0173 train_time:769728ms step_avg:134.24ms
step:5745/6250 train_loss:2.1060 aux_loss:0.0171 train_time:769863ms step_avg:134.24ms
step:5746/6250 train_loss:2.0993 aux_loss:0.0170 train_time:769996ms step_avg:134.24ms
step:5747/6250 train_loss:2.1832 aux_loss:0.0170 train_time:770131ms step_avg:134.24ms
step:5748/6250 train_loss:2.1647 aux_loss:0.0170 train_time:770265ms step_avg:134.24ms
step:5749/6250 train_loss:2.1934 aux_loss:0.0169 train_time:770400ms step_avg:134.24ms
step:5750/6250 train_loss:2.1110 aux_loss:0.0168 train_time:770533ms step_avg:134.24ms
step:5751/6250 train_loss:2.1470 aux_loss:0.0168 train_time:770667ms step_avg:134.24ms
step:5752/6250 train_loss:2.1368 aux_loss:0.0170 train_time:770802ms step_avg:134.24ms
step:5753/6250 train_loss:2.0571 aux_loss:0.0167 train_time:770936ms step_avg:134.24ms
step:5754/6250 train_loss:2.0444 aux_loss:0.0170 train_time:771070ms step_avg:134.24ms
step:5755/6250 train_loss:2.1219 aux_loss:0.0171 train_time:771203ms step_avg:134.24ms
step:5756/6250 train_loss:2.1371 aux_loss:0.0174 train_time:771337ms step_avg:134.24ms
step:5757/6250 train_loss:2.0946 aux_loss:0.0175 train_time:771472ms step_avg:134.24ms
step:5758/6250 train_loss:2.2277 aux_loss:0.0180 train_time:771606ms step_avg:134.24ms
step:5759/6250 train_loss:2.1375 aux_loss:0.0176 train_time:771740ms step_avg:134.24ms
step:5760/6250 train_loss:2.1391 aux_loss:0.0172 train_time:771874ms step_avg:134.24ms
step:5761/6250 train_loss:2.2017 aux_loss:0.0176 train_time:772009ms step_avg:134.24ms
step:5762/6250 train_loss:2.1726 aux_loss:0.0173 train_time:772142ms step_avg:134.24ms
step:5763/6250 train_loss:2.1380 aux_loss:0.0171 train_time:772277ms step_avg:134.24ms
step:5764/6250 train_loss:2.0703 aux_loss:0.0169 train_time:772411ms step_avg:134.24ms
step:5765/6250 train_loss:2.0579 aux_loss:0.0168 train_time:772545ms step_avg:134.24ms
step:5766/6250 train_loss:2.1564 aux_loss:0.0171 train_time:772683ms step_avg:134.24ms
step:5767/6250 train_loss:2.0583 aux_loss:0.0173 train_time:772826ms step_avg:134.24ms
step:5768/6250 train_loss:2.1922 aux_loss:0.0170 train_time:772967ms step_avg:134.24ms
step:5769/6250 train_loss:2.0926 aux_loss:0.0171 train_time:773107ms step_avg:134.24ms
step:5770/6250 train_loss:2.1402 aux_loss:0.0172 train_time:773248ms step_avg:134.24ms
step:5771/6250 train_loss:2.1110 aux_loss:0.0173 train_time:773388ms step_avg:134.25ms
step:5772/6250 train_loss:2.1087 aux_loss:0.0172 train_time:773526ms step_avg:134.25ms
step:5773/6250 train_loss:2.2955 aux_loss:0.0175 train_time:773663ms step_avg:134.25ms
step:5774/6250 train_loss:2.1636 aux_loss:0.0173 train_time:773801ms step_avg:134.25ms
step:5775/6250 train_loss:2.0549 aux_loss:0.0171 train_time:773940ms step_avg:134.25ms
step:5776/6250 train_loss:2.0345 aux_loss:0.0168 train_time:774080ms step_avg:134.25ms
step:5777/6250 train_loss:2.1871 aux_loss:0.0170 train_time:774216ms step_avg:134.25ms
step:5778/6250 train_loss:2.1327 aux_loss:0.0169 train_time:774351ms step_avg:134.25ms
step:5779/6250 train_loss:2.1703 aux_loss:0.0172 train_time:774485ms step_avg:134.25ms
step:5780/6250 train_loss:2.0914 aux_loss:0.0171 train_time:774624ms step_avg:134.25ms
step:5781/6250 train_loss:2.1353 aux_loss:0.0170 train_time:774768ms step_avg:134.25ms
step:5782/6250 train_loss:2.1163 aux_loss:0.0170 train_time:774911ms step_avg:134.25ms
step:5783/6250 train_loss:2.0942 aux_loss:0.0168 train_time:775045ms step_avg:134.25ms
step:5784/6250 train_loss:2.0449 aux_loss:0.0165 train_time:775180ms step_avg:134.25ms
step:5785/6250 train_loss:2.1173 aux_loss:0.0166 train_time:775314ms step_avg:134.25ms
step:5786/6250 train_loss:2.1014 aux_loss:0.0168 train_time:775449ms step_avg:134.25ms
step:5787/6250 train_loss:2.1100 aux_loss:0.0170 train_time:775584ms step_avg:134.25ms
step:5788/6250 train_loss:2.0721 aux_loss:0.0167 train_time:775719ms step_avg:134.25ms
step:5789/6250 train_loss:2.1533 aux_loss:0.0166 train_time:775852ms step_avg:134.25ms
step:5790/6250 train_loss:2.0864 aux_loss:0.0167 train_time:775986ms step_avg:134.25ms
step:5791/6250 train_loss:2.1083 aux_loss:0.0171 train_time:776126ms step_avg:134.25ms
step:5792/6250 train_loss:2.2714 aux_loss:0.0172 train_time:776271ms step_avg:134.26ms
step:5793/6250 train_loss:2.1520 aux_loss:0.0169 train_time:776413ms step_avg:134.26ms
step:5794/6250 train_loss:2.2385 aux_loss:0.0169 train_time:776554ms step_avg:134.26ms
step:5795/6250 train_loss:2.1824 aux_loss:0.0170 train_time:776698ms step_avg:134.26ms
step:5796/6250 train_loss:2.1012 aux_loss:0.0170 train_time:776842ms step_avg:134.26ms
step:5797/6250 train_loss:2.1564 aux_loss:0.0171 train_time:776980ms step_avg:134.26ms
step:5798/6250 train_loss:2.1611 aux_loss:0.0171 train_time:777115ms step_avg:134.26ms
step:5799/6250 train_loss:2.1218 aux_loss:0.0165 train_time:777249ms step_avg:134.26ms
step:5800/6250 train_loss:2.1906 aux_loss:0.0168 train_time:777383ms step_avg:134.26ms
step:5801/6250 train_loss:2.1145 aux_loss:0.0168 train_time:777517ms step_avg:134.26ms
step:5802/6250 train_loss:2.1356 aux_loss:0.0167 train_time:777650ms step_avg:134.26ms
step:5803/6250 train_loss:2.0337 aux_loss:0.0167 train_time:777785ms step_avg:134.26ms
step:5804/6250 train_loss:2.1767 aux_loss:0.0169 train_time:777923ms step_avg:134.26ms
step:5805/6250 train_loss:2.1063 aux_loss:0.0168 train_time:778059ms step_avg:134.26ms
step:5806/6250 train_loss:2.2275 aux_loss:0.0171 train_time:778193ms step_avg:134.26ms
step:5807/6250 train_loss:2.1488 aux_loss:0.0168 train_time:778328ms step_avg:134.26ms
step:5808/6250 train_loss:2.2067 aux_loss:0.0167 train_time:778462ms step_avg:134.26ms
step:5809/6250 train_loss:2.1237 aux_loss:0.0171 train_time:778595ms step_avg:134.26ms
step:5810/6250 train_loss:2.0939 aux_loss:0.0169 train_time:778730ms step_avg:134.26ms
step:5811/6250 train_loss:2.1258 aux_loss:0.0169 train_time:778866ms step_avg:134.26ms
step:5812/6250 train_loss:2.1158 aux_loss:0.0170 train_time:779002ms step_avg:134.26ms
step:5813/6250 train_loss:2.1608 aux_loss:0.0172 train_time:779142ms step_avg:134.27ms
step:5814/6250 train_loss:2.1501 aux_loss:0.0171 train_time:779279ms step_avg:134.27ms
step:5815/6250 train_loss:2.1885 aux_loss:0.0171 train_time:779413ms step_avg:134.27ms
step:5816/6250 train_loss:2.2685 aux_loss:0.0170 train_time:779549ms step_avg:134.27ms
step:5817/6250 train_loss:2.1090 aux_loss:0.0168 train_time:779684ms step_avg:134.27ms
step:5818/6250 train_loss:2.1404 aux_loss:0.0169 train_time:779826ms step_avg:134.27ms
step:5819/6250 train_loss:2.1743 aux_loss:0.0174 train_time:779969ms step_avg:134.27ms
step:5820/6250 train_loss:2.1728 aux_loss:0.0172 train_time:780104ms step_avg:134.27ms
step:5821/6250 train_loss:2.0976 aux_loss:0.0169 train_time:780237ms step_avg:134.27ms
step:5822/6250 train_loss:2.1497 aux_loss:0.0171 train_time:780373ms step_avg:134.27ms
step:5823/6250 train_loss:2.1547 aux_loss:0.0172 train_time:780508ms step_avg:134.27ms
step:5824/6250 train_loss:2.0847 aux_loss:0.0168 train_time:780644ms step_avg:134.27ms
step:5825/6250 train_loss:2.1558 aux_loss:0.0169 train_time:780782ms step_avg:134.27ms
step:5826/6250 train_loss:2.1466 aux_loss:0.0172 train_time:780917ms step_avg:134.27ms
step:5827/6250 train_loss:2.0773 aux_loss:0.0168 train_time:781055ms step_avg:134.27ms
step:5828/6250 train_loss:2.1646 aux_loss:0.0166 train_time:781193ms step_avg:134.27ms
step:5829/6250 train_loss:2.0552 aux_loss:0.0169 train_time:781328ms step_avg:134.27ms
step:5830/6250 train_loss:2.1936 aux_loss:0.0169 train_time:781463ms step_avg:134.27ms
step:5831/6250 train_loss:2.1266 aux_loss:0.0169 train_time:781599ms step_avg:134.27ms
step:5832/6250 train_loss:2.1827 aux_loss:0.0170 train_time:781734ms step_avg:134.27ms
step:5833/6250 train_loss:2.1707 aux_loss:0.0171 train_time:781870ms step_avg:134.27ms
step:5834/6250 train_loss:2.0226 aux_loss:0.0168 train_time:782004ms step_avg:134.27ms
step:5835/6250 train_loss:2.0589 aux_loss:0.0169 train_time:782139ms step_avg:134.27ms
step:5836/6250 train_loss:2.1016 aux_loss:0.0169 train_time:782274ms step_avg:134.27ms
step:5837/6250 train_loss:2.0535 aux_loss:0.0167 train_time:782409ms step_avg:134.27ms
step:5838/6250 train_loss:2.1756 aux_loss:0.0170 train_time:782543ms step_avg:134.27ms
step:5839/6250 train_loss:2.1603 aux_loss:0.0170 train_time:782678ms step_avg:134.27ms
step:5840/6250 train_loss:2.2061 aux_loss:0.0167 train_time:782811ms step_avg:134.27ms
step:5841/6250 train_loss:2.0585 aux_loss:0.0166 train_time:782945ms step_avg:134.27ms
step:5842/6250 train_loss:2.0465 aux_loss:0.0165 train_time:783080ms step_avg:134.27ms
step:5843/6250 train_loss:2.1087 aux_loss:0.0168 train_time:783213ms step_avg:134.27ms
step:5844/6250 train_loss:2.0790 aux_loss:0.0165 train_time:783350ms step_avg:134.27ms
step:5845/6250 train_loss:2.1280 aux_loss:0.0167 train_time:783490ms step_avg:134.27ms
step:5846/6250 train_loss:2.1430 aux_loss:0.0166 train_time:783634ms step_avg:134.28ms
step:5847/6250 train_loss:2.1550 aux_loss:0.0168 train_time:783781ms step_avg:134.28ms
step:5848/6250 train_loss:2.0410 aux_loss:0.0166 train_time:783927ms step_avg:134.28ms
step:5849/6250 train_loss:2.0914 aux_loss:0.0169 train_time:784072ms step_avg:134.28ms
step:5850/6250 train_loss:2.0758 aux_loss:0.0170 train_time:784210ms step_avg:134.28ms
step:5851/6250 train_loss:2.1027 aux_loss:0.0168 train_time:784347ms step_avg:134.28ms
step:5852/6250 train_loss:2.1276 aux_loss:0.0169 train_time:784483ms step_avg:134.28ms
step:5853/6250 train_loss:2.1760 aux_loss:0.0171 train_time:784619ms step_avg:134.28ms
step:5854/6250 train_loss:2.1140 aux_loss:0.0168 train_time:784754ms step_avg:134.28ms
step:5855/6250 train_loss:2.1498 aux_loss:0.0169 train_time:784887ms step_avg:134.28ms
step:5856/6250 train_loss:2.1094 aux_loss:0.0166 train_time:785023ms step_avg:134.28ms
step:5857/6250 train_loss:2.1321 aux_loss:0.0166 train_time:785160ms step_avg:134.28ms
step:5858/6250 train_loss:2.1890 aux_loss:0.0169 train_time:785295ms step_avg:134.28ms
step:5859/6250 train_loss:2.1472 aux_loss:0.0168 train_time:785430ms step_avg:134.28ms
step:5860/6250 train_loss:2.1657 aux_loss:0.0169 train_time:785568ms step_avg:134.29ms
step:5861/6250 train_loss:2.0590 aux_loss:0.0166 train_time:785705ms step_avg:134.29ms
step:5862/6250 train_loss:2.0859 aux_loss:0.0164 train_time:785842ms step_avg:134.29ms
step:5863/6250 train_loss:2.2462 aux_loss:0.0168 train_time:785979ms step_avg:134.29ms
step:5864/6250 train_loss:2.1344 aux_loss:0.0171 train_time:786111ms step_avg:134.29ms
step:5865/6250 train_loss:2.1317 aux_loss:0.0169 train_time:786245ms step_avg:134.29ms
step:5866/6250 train_loss:2.1973 aux_loss:0.0169 train_time:786379ms step_avg:134.29ms
step:5867/6250 train_loss:2.0649 aux_loss:0.0166 train_time:786517ms step_avg:134.29ms
step:5868/6250 train_loss:2.0785 aux_loss:0.0167 train_time:786655ms step_avg:134.29ms
step:5869/6250 train_loss:2.1503 aux_loss:0.0168 train_time:786792ms step_avg:134.29ms
step:5870/6250 train_loss:2.1280 aux_loss:0.0168 train_time:786925ms step_avg:134.29ms
step:5871/6250 train_loss:2.1128 aux_loss:0.0167 train_time:787057ms step_avg:134.29ms
step:5872/6250 train_loss:2.1257 aux_loss:0.0165 train_time:787189ms step_avg:134.29ms
step:5873/6250 train_loss:1.9956 aux_loss:0.0165 train_time:787321ms step_avg:134.29ms
step:5874/6250 train_loss:2.1347 aux_loss:0.0165 train_time:787454ms step_avg:134.29ms
step:5875/6250 train_loss:2.0985 aux_loss:0.0167 train_time:787586ms step_avg:134.29ms
step:5876/6250 train_loss:2.1902 aux_loss:0.0168 train_time:787720ms step_avg:134.29ms
step:5877/6250 train_loss:2.1730 aux_loss:0.0168 train_time:787863ms step_avg:134.29ms
step:5878/6250 train_loss:2.0706 aux_loss:0.0168 train_time:788003ms step_avg:134.29ms
step:5879/6250 train_loss:2.0778 aux_loss:0.0169 train_time:788141ms step_avg:134.29ms
step:5880/6250 train_loss:2.0987 aux_loss:0.0184 train_time:788279ms step_avg:134.29ms
step:5881/6250 train_loss:2.0602 aux_loss:0.0170 train_time:788416ms step_avg:134.29ms
step:5882/6250 train_loss:2.1955 aux_loss:0.0170 train_time:788554ms step_avg:134.29ms
step:5883/6250 train_loss:2.1169 aux_loss:0.0166 train_time:788694ms step_avg:134.29ms
step:5884/6250 train_loss:2.0635 aux_loss:0.0171 train_time:788829ms step_avg:134.29ms
step:5885/6250 train_loss:2.1702 aux_loss:0.0167 train_time:788963ms step_avg:134.29ms
step:5886/6250 train_loss:2.1619 aux_loss:0.0173 train_time:789095ms step_avg:134.29ms
step:5887/6250 train_loss:2.0983 aux_loss:0.0165 train_time:789227ms step_avg:134.29ms
step:5888/6250 train_loss:2.1544 aux_loss:0.0167 train_time:789360ms step_avg:134.29ms
step:5889/6250 train_loss:2.1052 aux_loss:0.0168 train_time:789491ms step_avg:134.29ms
step:5890/6250 train_loss:2.0649 aux_loss:0.0168 train_time:789624ms step_avg:134.29ms
step:5891/6250 train_loss:2.1030 aux_loss:0.0170 train_time:789756ms step_avg:134.29ms
step:5892/6250 train_loss:2.1579 aux_loss:0.0170 train_time:789888ms step_avg:134.29ms
step:5893/6250 train_loss:2.0942 aux_loss:0.0168 train_time:790020ms step_avg:134.29ms
step:5894/6250 train_loss:2.2229 aux_loss:0.0169 train_time:790152ms step_avg:134.29ms
step:5895/6250 train_loss:2.1429 aux_loss:0.0171 train_time:790284ms step_avg:134.29ms
step:5896/6250 train_loss:2.2207 aux_loss:0.0173 train_time:790417ms step_avg:134.29ms
step:5897/6250 train_loss:2.1437 aux_loss:0.0169 train_time:790549ms step_avg:134.29ms
step:5898/6250 train_loss:2.1047 aux_loss:0.0169 train_time:790682ms step_avg:134.29ms
step:5899/6250 train_loss:2.1784 aux_loss:0.0168 train_time:790814ms step_avg:134.29ms
step:5900/6250 train_loss:2.0782 aux_loss:0.0166 train_time:790947ms step_avg:134.29ms
step:5901/6250 train_loss:2.0755 aux_loss:0.0171 train_time:791079ms step_avg:134.29ms
step:5902/6250 train_loss:2.1378 aux_loss:0.0167 train_time:791210ms step_avg:134.29ms
step:5903/6250 train_loss:2.1522 aux_loss:0.0166 train_time:791342ms step_avg:134.29ms
step:5904/6250 train_loss:2.1249 aux_loss:0.0166 train_time:791475ms step_avg:134.28ms
step:5905/6250 train_loss:2.0663 aux_loss:0.0165 train_time:791606ms step_avg:134.28ms
step:5906/6250 train_loss:2.0881 aux_loss:0.0164 train_time:791740ms step_avg:134.28ms
step:5907/6250 train_loss:2.2944 aux_loss:0.0167 train_time:791872ms step_avg:134.28ms
step:5908/6250 train_loss:2.0954 aux_loss:0.0163 train_time:792005ms step_avg:134.28ms
step:5909/6250 train_loss:2.1633 aux_loss:0.0165 train_time:792136ms step_avg:134.28ms
step:5910/6250 train_loss:2.0903 aux_loss:0.0166 train_time:792267ms step_avg:134.28ms
step:5911/6250 train_loss:2.0796 aux_loss:0.0164 train_time:792401ms step_avg:134.28ms
step:5912/6250 train_loss:2.1643 aux_loss:0.0163 train_time:792533ms step_avg:134.28ms
step:5913/6250 train_loss:2.0737 aux_loss:0.0166 train_time:792665ms step_avg:134.28ms
step:5914/6250 train_loss:2.1125 aux_loss:0.0164 train_time:792797ms step_avg:134.28ms
step:5915/6250 train_loss:2.1862 aux_loss:0.0164 train_time:792931ms step_avg:134.28ms
step:5916/6250 train_loss:2.1083 aux_loss:0.0168 train_time:793063ms step_avg:134.28ms
step:5917/6250 train_loss:2.1790 aux_loss:0.0165 train_time:793195ms step_avg:134.28ms
step:5918/6250 train_loss:2.1748 aux_loss:0.0168 train_time:793327ms step_avg:134.28ms
step:5919/6250 train_loss:2.2454 aux_loss:0.0166 train_time:793461ms step_avg:134.28ms
step:5920/6250 train_loss:2.0930 aux_loss:0.0167 train_time:793593ms step_avg:134.28ms
step:5921/6250 train_loss:2.0802 aux_loss:0.0176 train_time:793727ms step_avg:134.28ms
step:5922/6250 train_loss:2.0684 aux_loss:0.0163 train_time:793860ms step_avg:134.28ms
step:5923/6250 train_loss:2.1502 aux_loss:0.0163 train_time:793993ms step_avg:134.28ms
step:5924/6250 train_loss:2.0821 aux_loss:0.0164 train_time:794125ms step_avg:134.28ms
step:5925/6250 train_loss:2.1156 aux_loss:0.0164 train_time:794258ms step_avg:134.28ms
step:5926/6250 train_loss:2.0302 aux_loss:0.0162 train_time:794391ms step_avg:134.28ms
step:5927/6250 train_loss:2.1251 aux_loss:0.0163 train_time:794523ms step_avg:134.28ms
step:5928/6250 train_loss:2.1919 aux_loss:0.0166 train_time:794656ms step_avg:134.28ms
step:5929/6250 train_loss:2.1740 aux_loss:0.0166 train_time:794789ms step_avg:134.28ms
step:5930/6250 train_loss:2.1527 aux_loss:0.0166 train_time:794923ms step_avg:134.28ms
step:5931/6250 train_loss:2.1417 aux_loss:0.0168 train_time:795055ms step_avg:134.28ms
step:5932/6250 train_loss:2.1555 aux_loss:0.0165 train_time:795187ms step_avg:134.28ms
step:5933/6250 train_loss:2.0803 aux_loss:0.0166 train_time:795320ms step_avg:134.28ms
step:5934/6250 train_loss:2.1810 aux_loss:0.0168 train_time:795453ms step_avg:134.28ms
step:5935/6250 train_loss:2.0924 aux_loss:0.0167 train_time:795586ms step_avg:134.28ms
step:5936/6250 train_loss:2.0264 aux_loss:0.0163 train_time:795719ms step_avg:134.28ms
step:5937/6250 train_loss:2.1052 aux_loss:0.0165 train_time:795853ms step_avg:134.28ms
step:5938/6250 train_loss:2.1100 aux_loss:0.0163 train_time:795985ms step_avg:134.28ms
step:5939/6250 train_loss:2.1254 aux_loss:0.0164 train_time:796117ms step_avg:134.28ms
step:5940/6250 train_loss:2.2037 aux_loss:0.0166 train_time:796250ms step_avg:134.27ms
step:5941/6250 train_loss:2.1549 aux_loss:0.0166 train_time:796383ms step_avg:134.27ms
step:5942/6250 train_loss:2.2146 aux_loss:0.0164 train_time:796515ms step_avg:134.27ms
step:5943/6250 train_loss:2.1079 aux_loss:0.0164 train_time:796647ms step_avg:134.27ms
step:5944/6250 train_loss:2.2249 aux_loss:0.0164 train_time:796781ms step_avg:134.27ms
step:5945/6250 train_loss:2.0847 aux_loss:0.0162 train_time:796914ms step_avg:134.27ms
step:5946/6250 train_loss:2.2199 aux_loss:0.0164 train_time:797045ms step_avg:134.27ms
step:5947/6250 train_loss:2.1143 aux_loss:0.0162 train_time:797178ms step_avg:134.27ms
step:5948/6250 train_loss:2.1203 aux_loss:0.0162 train_time:797310ms step_avg:134.27ms
step:5949/6250 train_loss:2.1103 aux_loss:0.0162 train_time:797443ms step_avg:134.27ms
step:5950/6250 train_loss:2.0632 aux_loss:0.0162 train_time:797576ms step_avg:134.27ms
step:5951/6250 train_loss:2.1828 aux_loss:0.0161 train_time:797710ms step_avg:134.27ms
step:5952/6250 train_loss:2.1914 aux_loss:0.0163 train_time:797843ms step_avg:134.27ms
step:5953/6250 train_loss:2.1149 aux_loss:0.0166 train_time:797977ms step_avg:134.27ms
step:5954/6250 train_loss:2.1267 aux_loss:0.0168 train_time:798110ms step_avg:134.27ms
step:5955/6250 train_loss:2.0503 aux_loss:0.0165 train_time:798243ms step_avg:134.27ms
step:5956/6250 train_loss:2.1427 aux_loss:0.0167 train_time:798375ms step_avg:134.27ms
step:5957/6250 train_loss:2.2179 aux_loss:0.0166 train_time:798507ms step_avg:134.27ms
step:5958/6250 train_loss:2.1657 aux_loss:0.0165 train_time:798639ms step_avg:134.27ms
step:5959/6250 train_loss:2.0812 aux_loss:0.0164 train_time:798772ms step_avg:134.27ms
step:5960/6250 train_loss:2.0948 aux_loss:0.0167 train_time:798905ms step_avg:134.27ms
step:5961/6250 train_loss:2.1443 aux_loss:0.0163 train_time:799037ms step_avg:134.27ms
step:5962/6250 train_loss:2.2056 aux_loss:0.0165 train_time:799169ms step_avg:134.27ms
step:5963/6250 train_loss:2.1422 aux_loss:0.0165 train_time:799301ms step_avg:134.27ms
step:5964/6250 train_loss:2.1726 aux_loss:0.0161 train_time:799434ms step_avg:134.27ms
step:5965/6250 train_loss:2.1310 aux_loss:0.0159 train_time:799565ms step_avg:134.27ms
step:5966/6250 train_loss:2.1195 aux_loss:0.0161 train_time:799698ms step_avg:134.27ms
step:5967/6250 train_loss:2.1064 aux_loss:0.0162 train_time:799830ms step_avg:134.27ms
step:5968/6250 train_loss:2.1107 aux_loss:0.0163 train_time:799963ms step_avg:134.27ms
step:5969/6250 train_loss:2.1300 aux_loss:0.0161 train_time:800095ms step_avg:134.27ms
step:5970/6250 train_loss:2.2365 aux_loss:0.0165 train_time:800226ms step_avg:134.27ms
step:5971/6250 train_loss:2.1020 aux_loss:0.0160 train_time:800359ms step_avg:134.27ms
step:5972/6250 train_loss:2.1753 aux_loss:0.0166 train_time:800491ms step_avg:134.27ms
step:5973/6250 train_loss:2.1306 aux_loss:0.0161 train_time:800623ms step_avg:134.27ms
step:5974/6250 train_loss:2.1051 aux_loss:0.0166 train_time:800756ms step_avg:134.26ms
step:5975/6250 train_loss:2.2025 aux_loss:0.0164 train_time:800888ms step_avg:134.26ms
step:5976/6250 train_loss:2.1633 aux_loss:0.0164 train_time:801021ms step_avg:134.26ms
step:5977/6250 train_loss:2.1965 aux_loss:0.0166 train_time:801153ms step_avg:134.26ms
step:5978/6250 train_loss:2.1140 aux_loss:0.0163 train_time:801285ms step_avg:134.26ms
step:5979/6250 train_loss:2.1671 aux_loss:0.0163 train_time:801419ms step_avg:134.26ms
step:5980/6250 train_loss:2.1401 aux_loss:0.0164 train_time:801552ms step_avg:134.26ms
step:5981/6250 train_loss:2.1888 aux_loss:0.0165 train_time:801685ms step_avg:134.26ms
step:5982/6250 train_loss:2.1026 aux_loss:0.0164 train_time:801817ms step_avg:134.26ms
step:5983/6250 train_loss:2.0521 aux_loss:0.0164 train_time:801950ms step_avg:134.26ms
step:5984/6250 train_loss:2.1276 aux_loss:0.0162 train_time:802082ms step_avg:134.26ms
step:5985/6250 train_loss:2.0755 aux_loss:0.0165 train_time:802215ms step_avg:134.26ms
step:5986/6250 train_loss:2.0790 aux_loss:0.0163 train_time:802347ms step_avg:134.26ms
step:5987/6250 train_loss:2.1003 aux_loss:0.0166 train_time:802479ms step_avg:134.26ms
step:5988/6250 train_loss:2.0952 aux_loss:0.0168 train_time:802611ms step_avg:134.26ms
step:5989/6250 train_loss:2.1311 aux_loss:0.0168 train_time:802743ms step_avg:134.26ms
step:5990/6250 train_loss:2.0885 aux_loss:0.0166 train_time:802876ms step_avg:134.26ms
step:5991/6250 train_loss:2.2374 aux_loss:0.0168 train_time:803008ms step_avg:134.26ms
step:5992/6250 train_loss:2.0608 aux_loss:0.0165 train_time:803141ms step_avg:134.26ms
step:5993/6250 train_loss:2.1720 aux_loss:0.0166 train_time:803273ms step_avg:134.26ms
step:5994/6250 train_loss:2.1414 aux_loss:0.0164 train_time:803406ms step_avg:134.26ms
step:5995/6250 train_loss:2.1277 aux_loss:0.0164 train_time:803539ms step_avg:134.26ms
step:5996/6250 train_loss:2.1844 aux_loss:0.0164 train_time:803671ms step_avg:134.26ms
step:5997/6250 train_loss:2.1722 aux_loss:0.0166 train_time:803803ms step_avg:134.26ms
step:5998/6250 train_loss:2.1673 aux_loss:0.0164 train_time:803937ms step_avg:134.26ms
step:5999/6250 train_loss:2.0035 aux_loss:0.0164 train_time:804069ms step_avg:134.26ms
step:6000/6250 train_loss:2.1581 aux_loss:0.0165 train_time:804200ms step_avg:134.26ms
step:6000/6250 val_loss:2.1157 val_aux_loss:0.0165 train_time:804201ms step_avg:134.26ms
step:6001/6250 train_loss:2.1148 aux_loss:0.0166 train_time:804334ms step_avg:134.26ms
step:6002/6250 train_loss:2.1231 aux_loss:0.0162 train_time:804467ms step_avg:134.26ms
step:6003/6250 train_loss:2.1252 aux_loss:0.0162 train_time:804600ms step_avg:134.26ms
step:6004/6250 train_loss:2.0900 aux_loss:0.0162 train_time:804733ms step_avg:134.26ms
step:6005/6250 train_loss:2.1953 aux_loss:0.0165 train_time:804866ms step_avg:134.26ms
step:6006/6250 train_loss:2.0725 aux_loss:0.0162 train_time:804998ms step_avg:134.26ms
step:6007/6250 train_loss:2.0980 aux_loss:0.0162 train_time:805130ms step_avg:134.26ms
step:6008/6250 train_loss:2.1604 aux_loss:0.0163 train_time:805262ms step_avg:134.26ms
step:6009/6250 train_loss:2.1439 aux_loss:0.0161 train_time:805395ms step_avg:134.25ms
step:6010/6250 train_loss:2.1510 aux_loss:0.0162 train_time:805528ms step_avg:134.25ms
step:6011/6250 train_loss:2.0461 aux_loss:0.0161 train_time:805661ms step_avg:134.25ms
step:6012/6250 train_loss:2.1268 aux_loss:0.0161 train_time:805794ms step_avg:134.25ms
step:6013/6250 train_loss:2.1261 aux_loss:0.0161 train_time:805927ms step_avg:134.25ms
step:6014/6250 train_loss:2.2109 aux_loss:0.0167 train_time:806060ms step_avg:134.25ms
step:6015/6250 train_loss:2.1050 aux_loss:0.0160 train_time:806192ms step_avg:134.25ms
step:6016/6250 train_loss:2.1628 aux_loss:0.0161 train_time:806325ms step_avg:134.25ms
step:6017/6250 train_loss:2.1450 aux_loss:0.0164 train_time:806459ms step_avg:134.25ms
step:6018/6250 train_loss:2.0957 aux_loss:0.0165 train_time:806592ms step_avg:134.25ms
step:6019/6250 train_loss:2.1593 aux_loss:0.0162 train_time:806725ms step_avg:134.25ms
step:6020/6250 train_loss:2.1699 aux_loss:0.0161 train_time:806859ms step_avg:134.25ms
step:6021/6250 train_loss:2.0965 aux_loss:0.0166 train_time:806992ms step_avg:134.25ms
step:6022/6250 train_loss:2.1650 aux_loss:0.0167 train_time:807123ms step_avg:134.25ms
step:6023/6250 train_loss:2.0897 aux_loss:0.0165 train_time:807255ms step_avg:134.25ms
step:6024/6250 train_loss:2.1067 aux_loss:0.0164 train_time:807388ms step_avg:134.25ms
step:6025/6250 train_loss:2.1038 aux_loss:0.0164 train_time:807521ms step_avg:134.25ms
step:6026/6250 train_loss:2.1880 aux_loss:0.0165 train_time:807655ms step_avg:134.25ms
step:6027/6250 train_loss:2.0662 aux_loss:0.0162 train_time:807787ms step_avg:134.25ms
step:6028/6250 train_loss:2.1204 aux_loss:0.0161 train_time:807919ms step_avg:134.25ms
step:6029/6250 train_loss:2.1103 aux_loss:0.0162 train_time:808051ms step_avg:134.25ms
step:6030/6250 train_loss:2.0619 aux_loss:0.0162 train_time:808183ms step_avg:134.25ms
step:6031/6250 train_loss:2.0048 aux_loss:0.0162 train_time:808315ms step_avg:134.25ms
step:6032/6250 train_loss:2.1813 aux_loss:0.0163 train_time:808448ms step_avg:134.25ms
step:6033/6250 train_loss:2.1485 aux_loss:0.0165 train_time:808581ms step_avg:134.25ms
step:6034/6250 train_loss:2.1292 aux_loss:0.0163 train_time:808713ms step_avg:134.25ms
step:6035/6250 train_loss:2.1347 aux_loss:0.0164 train_time:808846ms step_avg:134.25ms
step:6036/6250 train_loss:2.1679 aux_loss:0.0164 train_time:808977ms step_avg:134.25ms
step:6037/6250 train_loss:2.0832 aux_loss:0.0163 train_time:809110ms step_avg:134.25ms
step:6038/6250 train_loss:2.1560 aux_loss:0.0165 train_time:809241ms step_avg:134.25ms
step:6039/6250 train_loss:2.1552 aux_loss:0.0165 train_time:809375ms step_avg:134.25ms
step:6040/6250 train_loss:2.1342 aux_loss:0.0167 train_time:809507ms step_avg:134.25ms
step:6041/6250 train_loss:2.1870 aux_loss:0.0170 train_time:809640ms step_avg:134.25ms
step:6042/6250 train_loss:2.1809 aux_loss:0.0165 train_time:809771ms step_avg:134.25ms
step:6043/6250 train_loss:2.1540 aux_loss:0.0163 train_time:809903ms step_avg:134.25ms
step:6044/6250 train_loss:2.0764 aux_loss:0.0164 train_time:810035ms step_avg:134.25ms
step:6045/6250 train_loss:1.9816 aux_loss:0.0163 train_time:810167ms step_avg:134.24ms
step:6046/6250 train_loss:2.0689 aux_loss:0.0169 train_time:810299ms step_avg:134.24ms
step:6047/6250 train_loss:2.1077 aux_loss:0.0165 train_time:810431ms step_avg:134.24ms
step:6048/6250 train_loss:2.0523 aux_loss:0.0163 train_time:810564ms step_avg:134.24ms
step:6049/6250 train_loss:2.0635 aux_loss:0.0162 train_time:810697ms step_avg:134.24ms
step:6050/6250 train_loss:2.1426 aux_loss:0.0162 train_time:810830ms step_avg:134.24ms
step:6051/6250 train_loss:2.0663 aux_loss:0.0163 train_time:810963ms step_avg:134.24ms
step:6052/6250 train_loss:2.0053 aux_loss:0.0163 train_time:811096ms step_avg:134.24ms
step:6053/6250 train_loss:2.1107 aux_loss:0.0166 train_time:811228ms step_avg:134.24ms
step:6054/6250 train_loss:2.0766 aux_loss:0.0164 train_time:811361ms step_avg:134.24ms
step:6055/6250 train_loss:2.0908 aux_loss:0.0165 train_time:811493ms step_avg:134.24ms
step:6056/6250 train_loss:2.1006 aux_loss:0.0162 train_time:811626ms step_avg:134.24ms
step:6057/6250 train_loss:2.2013 aux_loss:0.0164 train_time:811758ms step_avg:134.24ms
step:6058/6250 train_loss:2.0084 aux_loss:0.0161 train_time:811890ms step_avg:134.24ms
step:6059/6250 train_loss:2.1240 aux_loss:0.0163 train_time:812022ms step_avg:134.24ms
step:6060/6250 train_loss:2.0708 aux_loss:0.0163 train_time:812154ms step_avg:134.24ms
step:6061/6250 train_loss:2.1427 aux_loss:0.0164 train_time:812287ms step_avg:134.24ms
step:6062/6250 train_loss:2.1631 aux_loss:0.0163 train_time:812419ms step_avg:134.24ms
step:6063/6250 train_loss:2.1306 aux_loss:0.0162 train_time:812551ms step_avg:134.24ms
step:6064/6250 train_loss:2.0453 aux_loss:0.0164 train_time:812683ms step_avg:134.24ms
step:6065/6250 train_loss:2.0658 aux_loss:0.0164 train_time:812816ms step_avg:134.24ms
step:6066/6250 train_loss:2.0661 aux_loss:0.0160 train_time:812948ms step_avg:134.24ms
step:6067/6250 train_loss:2.1498 aux_loss:0.0162 train_time:813079ms step_avg:134.24ms
step:6068/6250 train_loss:2.0085 aux_loss:0.0158 train_time:813211ms step_avg:134.24ms
step:6069/6250 train_loss:2.1734 aux_loss:0.0164 train_time:813344ms step_avg:134.24ms
step:6070/6250 train_loss:2.0428 aux_loss:0.0164 train_time:813476ms step_avg:134.24ms
step:6071/6250 train_loss:2.1353 aux_loss:0.0160 train_time:813609ms step_avg:134.24ms
step:6072/6250 train_loss:2.0953 aux_loss:0.0164 train_time:813740ms step_avg:134.24ms
step:6073/6250 train_loss:2.0745 aux_loss:0.0163 train_time:813874ms step_avg:134.24ms
step:6074/6250 train_loss:2.1353 aux_loss:0.0163 train_time:814005ms step_avg:134.24ms
step:6075/6250 train_loss:2.1421 aux_loss:0.0160 train_time:814137ms step_avg:134.24ms
step:6076/6250 train_loss:2.0657 aux_loss:0.0163 train_time:814269ms step_avg:134.23ms
step:6077/6250 train_loss:2.1447 aux_loss:0.0164 train_time:814402ms step_avg:134.23ms
step:6078/6250 train_loss:2.0698 aux_loss:0.0161 train_time:814534ms step_avg:134.23ms
step:6079/6250 train_loss:2.1661 aux_loss:0.0164 train_time:814667ms step_avg:134.23ms
step:6080/6250 train_loss:2.0945 aux_loss:0.0164 train_time:814799ms step_avg:134.23ms
step:6081/6250 train_loss:2.0940 aux_loss:0.0163 train_time:814931ms step_avg:134.23ms
step:6082/6250 train_loss:2.0460 aux_loss:0.0165 train_time:815063ms step_avg:134.23ms
step:6083/6250 train_loss:2.0870 aux_loss:0.0164 train_time:815195ms step_avg:134.23ms
step:6084/6250 train_loss:2.0655 aux_loss:0.0163 train_time:815328ms step_avg:134.23ms
step:6085/6250 train_loss:2.0235 aux_loss:0.0163 train_time:815459ms step_avg:134.23ms
step:6086/6250 train_loss:2.2125 aux_loss:0.0164 train_time:815592ms step_avg:134.23ms
step:6087/6250 train_loss:2.1729 aux_loss:0.0160 train_time:815724ms step_avg:134.23ms
step:6088/6250 train_loss:2.1008 aux_loss:0.0161 train_time:815857ms step_avg:134.23ms
step:6089/6250 train_loss:2.1189 aux_loss:0.0165 train_time:815989ms step_avg:134.23ms
step:6090/6250 train_loss:2.0230 aux_loss:0.0161 train_time:816120ms step_avg:134.23ms
step:6091/6250 train_loss:2.0738 aux_loss:0.0161 train_time:816253ms step_avg:134.23ms
step:6092/6250 train_loss:2.0644 aux_loss:0.0162 train_time:816384ms step_avg:134.23ms
step:6093/6250 train_loss:2.0565 aux_loss:0.0162 train_time:816517ms step_avg:134.23ms
step:6094/6250 train_loss:2.1203 aux_loss:0.0163 train_time:816649ms step_avg:134.23ms
step:6095/6250 train_loss:2.0655 aux_loss:0.0161 train_time:816781ms step_avg:134.23ms
step:6096/6250 train_loss:2.1670 aux_loss:0.0162 train_time:816916ms step_avg:134.23ms
step:6097/6250 train_loss:2.0406 aux_loss:0.0162 train_time:817048ms step_avg:134.23ms
step:6098/6250 train_loss:2.0214 aux_loss:0.0162 train_time:817180ms step_avg:134.23ms
step:6099/6250 train_loss:2.0821 aux_loss:0.0164 train_time:817312ms step_avg:134.23ms
step:6100/6250 train_loss:2.1071 aux_loss:0.0166 train_time:817444ms step_avg:134.23ms
step:6101/6250 train_loss:2.0592 aux_loss:0.0161 train_time:817577ms step_avg:134.23ms
step:6102/6250 train_loss:2.0861 aux_loss:0.0163 train_time:817709ms step_avg:134.23ms
step:6103/6250 train_loss:1.9967 aux_loss:0.0161 train_time:817968ms step_avg:134.25ms
step:6104/6250 train_loss:2.1020 aux_loss:0.0163 train_time:818098ms step_avg:134.25ms
step:6105/6250 train_loss:2.0989 aux_loss:0.0163 train_time:818230ms step_avg:134.25ms
step:6106/6250 train_loss:2.1699 aux_loss:0.0163 train_time:818362ms step_avg:134.25ms
step:6107/6250 train_loss:2.1946 aux_loss:0.0165 train_time:818493ms step_avg:134.25ms
step:6108/6250 train_loss:2.1392 aux_loss:0.0165 train_time:818625ms step_avg:134.24ms
step:6109/6250 train_loss:2.1388 aux_loss:0.0165 train_time:818757ms step_avg:134.24ms
step:6110/6250 train_loss:2.1165 aux_loss:0.0162 train_time:818891ms step_avg:134.24ms
step:6111/6250 train_loss:2.0674 aux_loss:0.0163 train_time:819025ms step_avg:134.24ms
step:6112/6250 train_loss:2.1215 aux_loss:0.0162 train_time:819156ms step_avg:134.24ms
step:6113/6250 train_loss:2.1393 aux_loss:0.0160 train_time:819289ms step_avg:134.24ms
step:6114/6250 train_loss:2.0625 aux_loss:0.0161 train_time:819422ms step_avg:134.24ms
step:6115/6250 train_loss:2.1677 aux_loss:0.0163 train_time:819554ms step_avg:134.24ms
step:6116/6250 train_loss:2.0246 aux_loss:0.0162 train_time:819686ms step_avg:134.24ms
step:6117/6250 train_loss:2.1774 aux_loss:0.0163 train_time:819819ms step_avg:134.24ms
step:6118/6250 train_loss:2.0585 aux_loss:0.0164 train_time:819953ms step_avg:134.24ms
step:6119/6250 train_loss:2.0624 aux_loss:0.0163 train_time:820086ms step_avg:134.24ms
step:6120/6250 train_loss:2.1104 aux_loss:0.0163 train_time:820219ms step_avg:134.24ms
step:6121/6250 train_loss:2.0634 aux_loss:0.0162 train_time:820351ms step_avg:134.24ms
step:6122/6250 train_loss:2.1314 aux_loss:0.0165 train_time:820482ms step_avg:134.24ms
step:6123/6250 train_loss:2.0590 aux_loss:0.0164 train_time:820615ms step_avg:134.24ms
step:6124/6250 train_loss:2.0773 aux_loss:0.0162 train_time:820748ms step_avg:134.24ms
step:6125/6250 train_loss:1.9725 aux_loss:0.0160 train_time:820880ms step_avg:134.24ms
step:6126/6250 train_loss:2.0772 aux_loss:0.0161 train_time:821013ms step_avg:134.24ms
step:6127/6250 train_loss:2.0155 aux_loss:0.0162 train_time:821145ms step_avg:134.24ms
step:6128/6250 train_loss:2.0882 aux_loss:0.0161 train_time:821278ms step_avg:134.24ms
step:6129/6250 train_loss:2.0577 aux_loss:0.0162 train_time:821409ms step_avg:134.24ms
step:6130/6250 train_loss:2.1743 aux_loss:0.0163 train_time:821541ms step_avg:134.24ms
step:6131/6250 train_loss:2.0802 aux_loss:0.0166 train_time:821674ms step_avg:134.24ms
step:6132/6250 train_loss:2.0663 aux_loss:0.0162 train_time:821806ms step_avg:134.24ms
step:6133/6250 train_loss:2.1079 aux_loss:0.0164 train_time:821939ms step_avg:134.24ms
step:6134/6250 train_loss:1.9899 aux_loss:0.0165 train_time:822073ms step_avg:134.24ms
step:6135/6250 train_loss:1.9987 aux_loss:0.0162 train_time:822207ms step_avg:134.24ms
step:6136/6250 train_loss:2.0847 aux_loss:0.0161 train_time:822339ms step_avg:134.24ms
step:6137/6250 train_loss:2.1007 aux_loss:0.0160 train_time:822472ms step_avg:134.24ms
step:6138/6250 train_loss:2.0446 aux_loss:0.0162 train_time:822606ms step_avg:134.24ms
step:6139/6250 train_loss:2.1726 aux_loss:0.0161 train_time:822738ms step_avg:134.24ms
step:6140/6250 train_loss:2.0640 aux_loss:0.0162 train_time:822871ms step_avg:134.24ms
step:6141/6250 train_loss:2.0194 aux_loss:0.0163 train_time:823003ms step_avg:134.24ms
step:6142/6250 train_loss:2.1046 aux_loss:0.0163 train_time:823136ms step_avg:134.24ms
step:6143/6250 train_loss:2.1094 aux_loss:0.0164 train_time:823268ms step_avg:134.24ms
step:6144/6250 train_loss:2.0740 aux_loss:0.0164 train_time:823400ms step_avg:134.24ms
step:6145/6250 train_loss:2.0685 aux_loss:0.0162 train_time:823533ms step_avg:134.24ms
step:6146/6250 train_loss:2.0273 aux_loss:0.0163 train_time:823666ms step_avg:134.24ms
step:6147/6250 train_loss:2.0140 aux_loss:0.0160 train_time:823799ms step_avg:134.23ms
step:6148/6250 train_loss:2.0678 aux_loss:0.0166 train_time:823931ms step_avg:134.23ms
step:6149/6250 train_loss:1.9326 aux_loss:0.0161 train_time:824064ms step_avg:134.23ms
step:6150/6250 train_loss:2.0961 aux_loss:0.0167 train_time:824196ms step_avg:134.23ms
step:6151/6250 train_loss:2.0598 aux_loss:0.0163 train_time:824328ms step_avg:134.23ms
step:6152/6250 train_loss:1.9441 aux_loss:0.0164 train_time:824461ms step_avg:134.23ms
step:6153/6250 train_loss:2.0239 aux_loss:0.0163 train_time:824594ms step_avg:134.23ms
step:6154/6250 train_loss:2.0452 aux_loss:0.0163 train_time:824726ms step_avg:134.23ms
step:6155/6250 train_loss:2.1504 aux_loss:0.0162 train_time:824858ms step_avg:134.23ms
step:6156/6250 train_loss:2.0523 aux_loss:0.0161 train_time:824991ms step_avg:134.23ms
step:6157/6250 train_loss:2.0710 aux_loss:0.0163 train_time:825123ms step_avg:134.23ms
step:6158/6250 train_loss:2.1227 aux_loss:0.0164 train_time:825256ms step_avg:134.23ms
step:6159/6250 train_loss:2.1271 aux_loss:0.0162 train_time:825388ms step_avg:134.23ms
step:6160/6250 train_loss:2.0678 aux_loss:0.0164 train_time:825519ms step_avg:134.23ms
step:6161/6250 train_loss:2.0564 aux_loss:0.0161 train_time:825652ms step_avg:134.23ms
step:6162/6250 train_loss:2.0514 aux_loss:0.0163 train_time:825786ms step_avg:134.23ms
step:6163/6250 train_loss:2.0815 aux_loss:0.0163 train_time:825918ms step_avg:134.23ms
step:6164/6250 train_loss:2.0895 aux_loss:0.0184 train_time:826050ms step_avg:134.23ms
step:6165/6250 train_loss:2.0100 aux_loss:0.0162 train_time:826183ms step_avg:134.23ms
step:6166/6250 train_loss:2.0122 aux_loss:0.0165 train_time:826315ms step_avg:134.23ms
step:6167/6250 train_loss:2.0055 aux_loss:0.0162 train_time:826447ms step_avg:134.23ms
step:6168/6250 train_loss:2.1394 aux_loss:0.0166 train_time:826579ms step_avg:134.23ms
step:6169/6250 train_loss:2.0004 aux_loss:0.0164 train_time:826712ms step_avg:134.23ms
step:6170/6250 train_loss:1.9810 aux_loss:0.0161 train_time:826844ms step_avg:134.23ms
step:6171/6250 train_loss:2.1906 aux_loss:0.0164 train_time:826975ms step_avg:134.23ms
step:6172/6250 train_loss:2.0432 aux_loss:0.0161 train_time:827107ms step_avg:134.23ms
step:6173/6250 train_loss:2.0662 aux_loss:0.0163 train_time:827240ms step_avg:134.23ms
step:6174/6250 train_loss:2.0283 aux_loss:0.0160 train_time:827372ms step_avg:134.23ms
step:6175/6250 train_loss:2.0311 aux_loss:0.0163 train_time:827504ms step_avg:134.23ms
step:6176/6250 train_loss:2.0257 aux_loss:0.0163 train_time:827637ms step_avg:134.23ms
step:6177/6250 train_loss:2.0593 aux_loss:0.0162 train_time:827771ms step_avg:134.23ms
step:6178/6250 train_loss:2.0299 aux_loss:0.0164 train_time:827910ms step_avg:134.23ms
step:6179/6250 train_loss:2.0062 aux_loss:0.0163 train_time:828054ms step_avg:134.23ms
step:6180/6250 train_loss:2.1475 aux_loss:0.0165 train_time:828198ms step_avg:134.23ms
step:6181/6250 train_loss:2.1063 aux_loss:0.0164 train_time:828342ms step_avg:134.23ms
step:6182/6250 train_loss:2.0834 aux_loss:0.0164 train_time:828483ms step_avg:134.23ms
step:6183/6250 train_loss:2.0820 aux_loss:0.0165 train_time:828624ms step_avg:134.23ms
step:6184/6250 train_loss:2.0209 aux_loss:0.0162 train_time:828759ms step_avg:134.23ms
step:6185/6250 train_loss:1.9959 aux_loss:0.0162 train_time:828893ms step_avg:134.23ms
step:6186/6250 train_loss:1.9903 aux_loss:0.0161 train_time:829025ms step_avg:134.23ms
step:6187/6250 train_loss:2.0186 aux_loss:0.0161 train_time:829156ms step_avg:134.23ms
step:6188/6250 train_loss:2.0120 aux_loss:0.0160 train_time:829291ms step_avg:134.23ms
step:6189/6250 train_loss:1.9955 aux_loss:0.0161 train_time:829434ms step_avg:134.23ms
step:6190/6250 train_loss:2.1538 aux_loss:0.0162 train_time:829576ms step_avg:134.24ms
step:6191/6250 train_loss:2.1117 aux_loss:0.0162 train_time:829717ms step_avg:134.24ms
step:6192/6250 train_loss:2.0272 aux_loss:0.0164 train_time:829861ms step_avg:134.24ms
step:6193/6250 train_loss:2.0740 aux_loss:0.0164 train_time:830003ms step_avg:134.24ms
step:6194/6250 train_loss:2.0809 aux_loss:0.0163 train_time:830143ms step_avg:134.24ms
step:6195/6250 train_loss:2.0316 aux_loss:0.0164 train_time:830277ms step_avg:134.24ms
step:6196/6250 train_loss:2.0873 aux_loss:0.0165 train_time:830410ms step_avg:134.24ms
step:6197/6250 train_loss:2.0578 aux_loss:0.0164 train_time:830543ms step_avg:134.24ms
step:6198/6250 train_loss:2.0497 aux_loss:0.0163 train_time:830675ms step_avg:134.24ms
step:6199/6250 train_loss:2.1286 aux_loss:0.0162 train_time:830811ms step_avg:134.24ms
step:6200/6250 train_loss:2.0358 aux_loss:0.0163 train_time:830943ms step_avg:134.24ms
step:6201/6250 train_loss:2.0636 aux_loss:0.0164 train_time:831075ms step_avg:134.24ms
step:6202/6250 train_loss:2.1215 aux_loss:0.0164 train_time:831208ms step_avg:134.24ms
step:6203/6250 train_loss:2.0877 aux_loss:0.0165 train_time:831340ms step_avg:134.24ms
step:6204/6250 train_loss:2.0562 aux_loss:0.0164 train_time:831472ms step_avg:134.24ms
step:6205/6250 train_loss:1.9588 aux_loss:0.0163 train_time:831605ms step_avg:134.24ms
step:6206/6250 train_loss:2.1127 aux_loss:0.0166 train_time:831737ms step_avg:134.24ms
step:6207/6250 train_loss:2.0795 aux_loss:0.0167 train_time:831870ms step_avg:134.24ms
step:6208/6250 train_loss:2.0776 aux_loss:0.0163 train_time:832002ms step_avg:134.24ms
step:6209/6250 train_loss:2.0062 aux_loss:0.0163 train_time:832135ms step_avg:134.24ms
step:6210/6250 train_loss:2.0716 aux_loss:0.0163 train_time:832269ms step_avg:134.24ms
step:6211/6250 train_loss:2.1000 aux_loss:0.0165 train_time:832402ms step_avg:134.24ms
step:6212/6250 train_loss:2.0217 aux_loss:0.0165 train_time:832539ms step_avg:134.24ms
step:6213/6250 train_loss:2.0690 aux_loss:0.0164 train_time:832679ms step_avg:134.24ms
step:6214/6250 train_loss:2.0323 aux_loss:0.0163 train_time:832817ms step_avg:134.24ms
step:6215/6250 train_loss:2.0306 aux_loss:0.0163 train_time:832956ms step_avg:134.24ms
step:6216/6250 train_loss:2.0512 aux_loss:0.0161 train_time:833094ms step_avg:134.24ms
step:6217/6250 train_loss:2.1144 aux_loss:0.0162 train_time:833232ms step_avg:134.24ms
step:6218/6250 train_loss:2.0206 aux_loss:0.0160 train_time:833371ms step_avg:134.24ms
step:6219/6250 train_loss:2.0418 aux_loss:0.0161 train_time:833510ms step_avg:134.24ms
step:6220/6250 train_loss:2.0516 aux_loss:0.0162 train_time:833648ms step_avg:134.24ms
step:6221/6250 train_loss:2.0728 aux_loss:0.0164 train_time:833786ms step_avg:134.24ms
step:6222/6250 train_loss:2.0917 aux_loss:0.0162 train_time:833925ms step_avg:134.24ms
step:6223/6250 train_loss:1.9908 aux_loss:0.0162 train_time:834063ms step_avg:134.24ms
step:6224/6250 train_loss:2.0928 aux_loss:0.0165 train_time:834201ms step_avg:134.25ms
step:6225/6250 train_loss:2.0338 aux_loss:0.0164 train_time:834340ms step_avg:134.25ms
step:6226/6250 train_loss:2.0667 aux_loss:0.0166 train_time:834478ms step_avg:134.25ms
step:6227/6250 train_loss:2.0589 aux_loss:0.0167 train_time:834617ms step_avg:134.25ms
step:6228/6250 train_loss:1.9663 aux_loss:0.0163 train_time:834759ms step_avg:134.25ms
step:6229/6250 train_loss:2.0177 aux_loss:0.0162 train_time:834899ms step_avg:134.25ms
step:6230/6250 train_loss:1.9384 aux_loss:0.0161 train_time:835039ms step_avg:134.25ms
step:6231/6250 train_loss:1.9821 aux_loss:0.0162 train_time:835177ms step_avg:134.25ms
step:6232/6250 train_loss:2.1111 aux_loss:0.0164 train_time:835315ms step_avg:134.25ms
step:6233/6250 train_loss:2.0882 aux_loss:0.0166 train_time:835453ms step_avg:134.25ms
step:6234/6250 train_loss:2.1102 aux_loss:0.0165 train_time:835591ms step_avg:134.25ms
step:6235/6250 train_loss:1.9645 aux_loss:0.0165 train_time:835730ms step_avg:134.25ms
step:6236/6250 train_loss:2.1383 aux_loss:0.0164 train_time:835868ms step_avg:134.25ms
step:6237/6250 train_loss:2.0741 aux_loss:0.0162 train_time:836005ms step_avg:134.25ms
step:6238/6250 train_loss:2.0930 aux_loss:0.0165 train_time:836142ms step_avg:134.26ms
step:6239/6250 train_loss:2.0668 aux_loss:0.0164 train_time:836280ms step_avg:134.26ms
step:6240/6250 train_loss:1.9656 aux_loss:0.0169 train_time:836418ms step_avg:134.26ms
step:6241/6250 train_loss:2.0673 aux_loss:0.0162 train_time:836556ms step_avg:134.26ms
step:6242/6250 train_loss:2.0257 aux_loss:0.0161 train_time:836695ms step_avg:134.26ms
step:6243/6250 train_loss:2.0842 aux_loss:0.0166 train_time:836833ms step_avg:134.26ms
step:6244/6250 train_loss:2.0345 aux_loss:0.0162 train_time:836970ms step_avg:134.26ms
step:6245/6250 train_loss:1.9964 aux_loss:0.0161 train_time:837108ms step_avg:134.26ms
step:6246/6250 train_loss:2.0452 aux_loss:0.0162 train_time:837246ms step_avg:134.26ms
step:6247/6250 train_loss:1.9651 aux_loss:0.0162 train_time:837384ms step_avg:134.26ms
step:6248/6250 train_loss:2.0093 aux_loss:0.0162 train_time:837522ms step_avg:134.26ms
step:6249/6250 train_loss:2.1451 aux_loss:0.0163 train_time:837660ms step_avg:134.26ms
step:6250/6250 train_loss:2.0355 aux_loss:0.0162 train_time:837797ms step_avg:134.26ms
step:6250/6250 val_loss:2.0390 val_aux_loss:0.0163 train_time:837798ms step_avg:134.26ms