Upload . with huggingface_hub

.gitattributes

@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+replay.mp4 filter=lfs diff=lfs merge=lfs -text

.summary/0/events.out.tfevents.1686707529.arkark

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf996dbea29d3b9158dc0596984d0e73abe9707cce33c6bc5b7799f9ffb45919
+size 374892

README.md

@@ -0,0 +1,56 @@
+---
+library_name: sample-factory
+tags:
+- deep-reinforcement-learning
+- reinforcement-learning
+- sample-factory
+model-index:
+- name: APPO
+  results:
+  - task:
+      type: reinforcement-learning
+      name: reinforcement-learning
+    dataset:
+      name: doom_health_gathering_supreme
+      type: doom_health_gathering_supreme
+    metrics:
+    - type: mean_reward
+      value: 11.33 +/- 4.52
+      name: mean_reward
+      verified: false
+---
+
+A(n) **APPO** model trained on the **doom_health_gathering_supreme** environment.
+
+This model was trained using Sample-Factory 2.0: https://github.com/alex-petrenko/sample-factory.
+Documentation for how to use Sample-Factory can be found at https://www.samplefactory.dev/
+
+
+## Downloading the model
+
+After installing Sample-Factory, download the model with:
+```
+python -m sample_factory.huggingface.load_from_hub -r arkadyark/rl_course_vizdoom_health_gathering_supreme
+```
+
+    
+## Using the model
+
+To run the model after download, use the `enjoy` script corresponding to this environment:
+```
+python -m <path.to.enjoy.module> --algo=APPO --env=doom_health_gathering_supreme --train_dir=./train_dir --experiment=rl_course_vizdoom_health_gathering_supreme
+```
+
+
+You can also upload models to the Hugging Face Hub using the same script with the `--push_to_hub` flag.
+See https://www.samplefactory.dev/10-huggingface/huggingface/ for more details
+    
+## Training with this model
+
+To continue training with this model, use the `train` script corresponding to this environment:
+```
+python -m <path.to.train.module> --algo=APPO --env=doom_health_gathering_supreme --train_dir=./train_dir --experiment=rl_course_vizdoom_health_gathering_supreme --restart_behavior=resume --train_for_env_steps=10000000000
+```
+
+Note, you may have to adjust `--train_for_env_steps` to a suitably high number as the experiment will resume at the number of steps it concluded at.
+    

checkpoint_p0/best_000001861_7622656_reward_27.864.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f88dc3faedbf6db368882bfd86f5c70662c14dac4fd8424dd41a3c8fa447b1c
+size 34924044

checkpoint_p0/checkpoint_000001696_6946816.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75c28105d6149676ed8d9722cefb9e58bf8feb985120c16a460e69010a8e0fe3
+size 34924044

checkpoint_p0/checkpoint_000001955_8007680.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:774103284f9b5cc129d719419571fb284a7d35d2977ec398f08b5f2126195dae
+size 34924044

config.json

@@ -0,0 +1,142 @@
+{
+  "help": false,
+  "algo": "APPO",
+  "env": "doom_health_gathering_supreme",
+  "experiment": "default_experiment",
+  "train_dir": "/home/ark/projects/deep-rl-course/unit-8-p2/train_dir",
+  "restart_behavior": "resume",
+  "device": "gpu",
+  "seed": null,
+  "num_policies": 1,
+  "async_rl": true,
+  "serial_mode": false,
+  "batched_sampling": false,
+  "num_batches_to_accumulate": 2,
+  "worker_num_splits": 2,
+  "policy_workers_per_policy": 1,
+  "max_policy_lag": 1000,
+  "num_workers": 8,
+  "num_envs_per_worker": 4,
+  "batch_size": 1024,
+  "num_batches_per_epoch": 1,
+  "num_epochs": 1,
+  "rollout": 32,
+  "recurrence": 32,
+  "shuffle_minibatches": false,
+  "gamma": 0.99,
+  "reward_scale": 1.0,
+  "reward_clip": 1000.0,
+  "value_bootstrap": false,
+  "normalize_returns": true,
+  "exploration_loss_coeff": 0.001,
+  "value_loss_coeff": 0.5,
+  "kl_loss_coeff": 0.0,
+  "exploration_loss": "symmetric_kl",
+  "gae_lambda": 0.95,
+  "ppo_clip_ratio": 0.1,
+  "ppo_clip_value": 0.2,
+  "with_vtrace": false,
+  "vtrace_rho": 1.0,
+  "vtrace_c": 1.0,
+  "optimizer": "adam",
+  "adam_eps": 1e-06,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.999,
+  "max_grad_norm": 4.0,
+  "learning_rate": 0.0001,
+  "lr_schedule": "constant",
+  "lr_schedule_kl_threshold": 0.008,
+  "lr_adaptive_min": 1e-06,
+  "lr_adaptive_max": 0.01,
+  "obs_subtract_mean": 0.0,
+  "obs_scale": 255.0,
+  "normalize_input": true,
+  "normalize_input_keys": null,
+  "decorrelate_experience_max_seconds": 0,
+  "decorrelate_envs_on_one_worker": true,
+  "actor_worker_gpus": [],
+  "set_workers_cpu_affinity": true,
+  "force_envs_single_thread": false,
+  "default_niceness": 0,
+  "log_to_file": true,
+  "experiment_summaries_interval": 10,
+  "flush_summaries_interval": 30,
+  "stats_avg": 100,
+  "summaries_use_frameskip": true,
+  "heartbeat_interval": 20,
+  "heartbeat_reporting_interval": 600,
+  "train_for_env_steps": 8000000,
+  "train_for_seconds": 10000000000,
+  "save_every_sec": 120,
+  "keep_checkpoints": 2,
+  "load_checkpoint_kind": "latest",
+  "save_milestones_sec": -1,
+  "save_best_every_sec": 5,
+  "save_best_metric": "reward",
+  "save_best_after": 100000,
+  "benchmark": false,
+  "encoder_mlp_layers": [
+    512,
+    512
+  ],
+  "encoder_conv_architecture": "convnet_simple",
+  "encoder_conv_mlp_layers": [
+    512
+  ],
+  "use_rnn": true,
+  "rnn_size": 512,
+  "rnn_type": "gru",
+  "rnn_num_layers": 1,
+  "decoder_mlp_layers": [],
+  "nonlinearity": "elu",
+  "policy_initialization": "orthogonal",
+  "policy_init_gain": 1.0,
+  "actor_critic_share_weights": true,
+  "adaptive_stddev": true,
+  "continuous_tanh_scale": 0.0,
+  "initial_stddev": 1.0,
+  "use_env_info_cache": false,
+  "env_gpu_actions": false,
+  "env_gpu_observations": true,
+  "env_frameskip": 4,
+  "env_framestack": 1,
+  "pixel_format": "CHW",
+  "use_record_episode_statistics": false,
+  "with_wandb": false,
+  "wandb_user": null,
+  "wandb_project": "sample_factory",
+  "wandb_group": null,
+  "wandb_job_type": "SF",
+  "wandb_tags": [],
+  "with_pbt": false,
+  "pbt_mix_policies_in_one_env": true,
+  "pbt_period_env_steps": 5000000,
+  "pbt_start_mutation": 20000000,
+  "pbt_replace_fraction": 0.3,
+  "pbt_mutation_rate": 0.15,
+  "pbt_replace_reward_gap": 0.1,
+  "pbt_replace_reward_gap_absolute": 1e-06,
+  "pbt_optimize_gamma": false,
+  "pbt_target_objective": "true_objective",
+  "pbt_perturb_min": 1.1,
+  "pbt_perturb_max": 1.5,
+  "num_agents": -1,
+  "num_humans": 0,
+  "num_bots": -1,
+  "start_bot_difficulty": null,
+  "timelimit": null,
+  "res_w": 128,
+  "res_h": 72,
+  "wide_aspect_ratio": false,
+  "eval_env_frameskip": 1,
+  "fps": 35,
+  "command_line": "--env=doom_health_gathering_supreme --num_workers=8 --num_envs_per_worker=4 --train_for_env_steps=8000000",
+  "cli_args": {
+    "env": "doom_health_gathering_supreme",
+    "num_workers": 8,
+    "num_envs_per_worker": 4,
+    "train_for_env_steps": 8000000
+  },
+  "git_hash": "20c803507bca6e18d0b086b975dd370898588bc9",
+  "git_repo_name": "[email protected]:arkadyark/hf-deep-rl.git"
+}

git.diff

@@ -0,0 +1,116 @@
+diff --git a/unit-4/main.py b/unit-4/main.py
+index 347c250..834b615 100644
+--- a/unit-4/main.py
++++ b/unit-4/main.py
+@@ -69,7 +69,7 @@ class CartpolePolicy(nn.Module):
+ 
+ class PixelcopterPolicy(nn.Module):
+     def __init__(self, s_size, a_size, h_size, device):
+-        super(Policy, self).__init__()
++        super(PixelcopterPolicy, self).__init__()
+         self.fc1 = nn.Linear(s_size, h_size)
+         self.fc2 = nn.Linear(h_size, h_size * 2)
+         self.fc3 = nn.Linear(h_size * 2, a_size)
+@@ -170,8 +170,29 @@ def reinforce(policy, env, optimizer, n_training_episodes, max_t, gamma, print_e
+ 
+     return scores
+ 
+-
+-def push_to_hub(repo_id, model, hyperparameters, eval_env, video_fps=30):
++def record_video(env, policy, out_directory, fps=30):
++    """
++    Generate a replay video of the agent
++    :param env
++    :param Qtable: Qtable of our agent
++    :param out_directory
++    :param fps: how many frame per seconds (with taxi-v3 and frozenlake-v1 we use 1)
++    """
++    images = []
++    done = False
++    state = env.reset()
++    img = env.render(mode="rgb_array")
++    images.append(img)
++    while not done:
++        # Take the action (index) that have the maximum expected future reward given that state
++        action, _ = policy.act(state)
++        state, reward, done, info = env.step(action)  # We directly put next_state = state for recording logic
++        img = env.render(mode="rgb_array")
++        images.append(img)
++    imageio.mimsave(out_directory, [np.array(img) for i, img in enumerate(images)], fps=fps)
++
++
++def push_to_hub(repo_id, model, hparams, eval_env, video_fps=30):
+     """
+     Evaluate, Generate a video and Upload a model to Hugging Face Hub.
+     This method does the complete pipeline:
+@@ -182,7 +203,7 @@ def push_to_hub(repo_id, model, hyperparameters, eval_env, video_fps=30):
+ 
+     :param repo_id: repo_id: id of the model repository from the Hugging Face Hub
+     :param model: the pytorch model we want to save
+-    :param hyperparameters: training hyperparameters
++    :param hparams: training hparams
+     :param eval_env: evaluation environment
+     :param video_fps: how many frame per seconds to record our video replay
+     """
+@@ -202,15 +223,15 @@ def push_to_hub(repo_id, model, hyperparameters, eval_env, video_fps=30):
+         # Step 2: Save the model
+         torch.save(model, local_directory / "model.pt")
+ 
+-        # Step 3: Save the hyperparameters to JSON
+-        with open(local_directory / "hyperparameters.json", "w") as outfile:
+-            json.dump(hyperparameters, outfile)
++        # Step 3: Save the hparams to JSON
++        with open(local_directory / "hparams.json", "w") as outfile:
++            json.dump(hparams, outfile)
+ 
+         # Step 4: Evaluate the model and build JSON
+         mean_reward, std_reward = evaluate_agent(
+             eval_env,
+-            hyperparameters["max_t"],
+-            hyperparameters["n_evaluation_episodes"],
++            hparams["max_t"],
++            hparams["n_evaluation_episodes"],
+             model,
+         )
+         # Get datetime
+@@ -218,9 +239,9 @@ def push_to_hub(repo_id, model, hyperparameters, eval_env, video_fps=30):
+         eval_form_datetime = eval_datetime.isoformat()
+ 
+         evaluate_data = {
+-            "env_id": hyperparameters["env_id"],
++            "env_id": hparams["env_id"],
+             "mean_reward": mean_reward,
+-            "n_evaluation_episodes": hyperparameters["n_evaluation_episodes"],
++            "n_evaluation_episodes": hparams["n_evaluation_episodes"],
+             "eval_datetime": eval_form_datetime,
+         }
+ 
+@@ -229,7 +250,7 @@ def push_to_hub(repo_id, model, hyperparameters, eval_env, video_fps=30):
+             json.dump(evaluate_data, outfile)
+ 
+         # Step 5: Create the model card
+-        env_name = hyperparameters["env_id"]
++        env_name = hparams["env_id"]
+ 
+         metadata = {}
+         metadata["tags"] = [
+@@ -256,8 +277,8 @@ def push_to_hub(repo_id, model, hyperparameters, eval_env, video_fps=30):
+         metadata = {**metadata, **eval}
+ 
+         model_card = f"""
+-  # **Reinforce** Agent playing **{env_id}**
+-  This is a trained model of a **Reinforce** agent playing **{env_id}** .
++  # **Reinforce** Agent playing **{env_name}**
++  This is a trained model of a **Reinforce** agent playing **{env_name}** .
+   To learn to use this model and train yours check Unit 4 of the Deep Reinforcement Learning Course: https://huggingface.co/deep-rl-course/unit4/introduction
+   """
+ 
+@@ -277,7 +298,7 @@ def push_to_hub(repo_id, model, hyperparameters, eval_env, video_fps=30):
+ 
+         # Step 6: Record a video
+         video_path = local_directory / "replay.mp4"
+-        record_video(env, model, video_path, video_fps)
++        record_video(eval_env, model, video_path, video_fps)
+ 
+         # Step 7. Push everything to the Hub
+         api.upload_folder(

replay.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:864c3254d5d9d8485cfc00199cecce00f0800e52e59c386da18cbd4614e83e1b
+size 21834334

sf_log.txt

@@ -0,0 +1,890 @@
+[2023-06-13 21:52:11,005][939011] Saving configuration to /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/config.json...
+[2023-06-13 21:52:11,026][939011] Rollout worker 0 uses device cpu
+[2023-06-13 21:52:11,027][939011] Rollout worker 1 uses device cpu
+[2023-06-13 21:52:11,027][939011] Rollout worker 2 uses device cpu
+[2023-06-13 21:52:11,027][939011] Rollout worker 3 uses device cpu
+[2023-06-13 21:52:11,027][939011] Rollout worker 4 uses device cpu
+[2023-06-13 21:52:11,027][939011] Rollout worker 5 uses device cpu
+[2023-06-13 21:52:11,027][939011] Rollout worker 6 uses device cpu
+[2023-06-13 21:52:11,027][939011] Rollout worker 7 uses device cpu
+[2023-06-13 21:52:11,062][939011] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-06-13 21:52:11,062][939011] InferenceWorker_p0-w0: min num requests: 2
+[2023-06-13 21:52:11,080][939011] Starting all processes...
+[2023-06-13 21:52:11,080][939011] Starting process learner_proc0
+[2023-06-13 21:52:11,739][939011] Starting all processes...
+[2023-06-13 21:52:11,742][939084] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-06-13 21:52:11,742][939084] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for learning process 0
+[2023-06-13 21:52:11,752][939011] Starting process inference_proc0-0
+[2023-06-13 21:52:11,754][939084] Num visible devices: 1
+[2023-06-13 21:52:11,752][939011] Starting process rollout_proc0
+[2023-06-13 21:52:11,752][939011] Starting process rollout_proc1
+[2023-06-13 21:52:11,752][939011] Starting process rollout_proc2
+[2023-06-13 21:52:11,753][939011] Starting process rollout_proc3
+[2023-06-13 21:52:11,753][939011] Starting process rollout_proc4
+[2023-06-13 21:52:11,769][939084] Starting seed is not provided
+[2023-06-13 21:52:11,770][939084] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-06-13 21:52:11,770][939084] Initializing actor-critic model on device cuda:0
+[2023-06-13 21:52:11,770][939084] RunningMeanStd input shape: (3, 72, 128)
+[2023-06-13 21:52:11,770][939084] RunningMeanStd input shape: (1,)
+[2023-06-13 21:52:11,755][939011] Starting process rollout_proc5
+[2023-06-13 21:52:11,779][939084] ConvEncoder: input_channels=3
+[2023-06-13 21:52:11,756][939011] Starting process rollout_proc6
+[2023-06-13 21:52:11,762][939011] Starting process rollout_proc7
+[2023-06-13 21:52:11,880][939084] Conv encoder output size: 512
+[2023-06-13 21:52:11,880][939084] Policy head output size: 512
+[2023-06-13 21:52:11,889][939084] Created Actor Critic model with architecture:
+[2023-06-13 21:52:11,889][939084] ActorCriticSharedWeights(
+  (obs_normalizer): ObservationNormalizer(
+    (running_mean_std): RunningMeanStdDictInPlace(
+      (running_mean_std): ModuleDict(
+        (obs): RunningMeanStdInPlace()
+      )
+    )
+  )
+  (returns_normalizer): RecursiveScriptModule(original_name=RunningMeanStdInPlace)
+  (encoder): VizdoomEncoder(
+    (basic_encoder): ConvEncoder(
+      (enc): RecursiveScriptModule(
+        original_name=ConvEncoderImpl
+        (conv_head): RecursiveScriptModule(
+          original_name=Sequential
+          (0): RecursiveScriptModule(original_name=Conv2d)
+          (1): RecursiveScriptModule(original_name=ELU)
+          (2): RecursiveScriptModule(original_name=Conv2d)
+          (3): RecursiveScriptModule(original_name=ELU)
+          (4): RecursiveScriptModule(original_name=Conv2d)
+          (5): RecursiveScriptModule(original_name=ELU)
+        )
+        (mlp_layers): RecursiveScriptModule(
+          original_name=Sequential
+          (0): RecursiveScriptModule(original_name=Linear)
+          (1): RecursiveScriptModule(original_name=ELU)
+        )
+      )
+    )
+  )
+  (core): ModelCoreRNN(
+    (core): GRU(512, 512)
+  )
+  (decoder): MlpDecoder(
+    (mlp): Identity()
+  )
+  (critic_linear): Linear(in_features=512, out_features=1, bias=True)
+  (action_parameterization): ActionParameterizationDefault(
+    (distribution_linear): Linear(in_features=512, out_features=5, bias=True)
+  )
+)
+[2023-06-13 21:52:12,821][939130] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-06-13 21:52:12,821][939130] Set environment var CUDA_VISIBLE_DEVICES to '0' (GPU indices [0]) for inference process 0
+[2023-06-13 21:52:12,826][939130] Num visible devices: 1
+[2023-06-13 21:52:12,838][939133] Worker 0 uses CPU cores [0, 1]
+[2023-06-13 21:52:12,851][939134] Worker 2 uses CPU cores [4, 5]
+[2023-06-13 21:52:12,871][939131] Worker 1 uses CPU cores [2, 3]
+[2023-06-13 21:52:12,943][939135] Worker 4 uses CPU cores [8, 9]
+[2023-06-13 21:52:12,943][939136] Worker 3 uses CPU cores [6, 7]
+[2023-06-13 21:52:12,968][939137] Worker 5 uses CPU cores [10, 11]
+[2023-06-13 21:52:13,084][939139] Worker 7 uses CPU cores [14, 15]
+[2023-06-13 21:52:13,132][939138] Worker 6 uses CPU cores [12, 13]
+[2023-06-13 21:52:15,189][939084] Using optimizer <class 'torch.optim.adam.Adam'>
+[2023-06-13 21:52:15,190][939084] No checkpoints found
+[2023-06-13 21:52:15,190][939084] Did not load from checkpoint, starting from scratch!
+[2023-06-13 21:52:15,190][939084] Initialized policy 0 weights for model version 0
+[2023-06-13 21:52:15,194][939084] LearnerWorker_p0 finished initialization!
+[2023-06-13 21:52:15,194][939084] Using GPUs [0] for process 0 (actually maps to GPUs [0])
+[2023-06-13 21:52:15,279][939130] RunningMeanStd input shape: (3, 72, 128)
+[2023-06-13 21:52:15,279][939130] RunningMeanStd input shape: (1,)
+[2023-06-13 21:52:15,286][939130] ConvEncoder: input_channels=3
+[2023-06-13 21:52:15,358][939130] Conv encoder output size: 512
+[2023-06-13 21:52:15,359][939130] Policy head output size: 512
+[2023-06-13 21:52:18,124][939011] Inference worker 0-0 is ready!
+[2023-06-13 21:52:18,124][939011] All inference workers are ready! Signal rollout workers to start!
+[2023-06-13 21:52:18,164][939136] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,164][939134] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,164][939138] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,167][939135] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,168][939139] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,169][939133] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,171][939131] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,171][939137] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:52:18,256][939134] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process...
+[2023-06-13 21:52:18,256][939133] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process...
+[2023-06-13 21:52:18,256][939136] VizDoom game.init() threw an exception ViZDoomUnexpectedExitException('Controlled ViZDoom instance exited unexpectedly.'). Terminate process...
+[2023-06-13 21:52:18,256][939134] EvtLoop [rollout_proc2_evt_loop, process=rollout_proc2] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init
+    self.game.init()
+vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly.
+
+During handling of the above exception, another exception occurred:
+
+Traceback (most recent call last):
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 379, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset
+    self._ensure_initialized()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized
+    self.initialize()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize
+    self._game_init()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init
+    raise EnvCriticalError()
+sample_factory.envs.env_utils.EnvCriticalError
+[2023-06-13 21:52:18,256][939133] EvtLoop [rollout_proc0_evt_loop, process=rollout_proc0] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init
+    self.game.init()
+vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly.
+
+During handling of the above exception, another exception occurred:
+
+Traceback (most recent call last):
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 379, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset
+    self._ensure_initialized()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized
+    self.initialize()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize
+    self._game_init()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init
+    raise EnvCriticalError()
+sample_factory.envs.env_utils.EnvCriticalError
+[2023-06-13 21:52:18,257][939134] Unhandled exception  in evt loop rollout_proc2_evt_loop
+[2023-06-13 21:52:18,257][939133] Unhandled exception  in evt loop rollout_proc0_evt_loop
+[2023-06-13 21:52:18,256][939136] EvtLoop [rollout_proc3_evt_loop, process=rollout_proc3] unhandled exception in slot='init' connected to emitter=Emitter(object_id='Sampler', signal_name='_inference_workers_initialized'), args=()
+Traceback (most recent call last):
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 228, in _game_init
+    self.game.init()
+vizdoom.vizdoom.ViZDoomUnexpectedExitException: Controlled ViZDoom instance exited unexpectedly.
+
+During handling of the above exception, another exception occurred:
+
+Traceback (most recent call last):
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/signal_slot/signal_slot.py", line 355, in _process_signal
+    slot_callable(*args)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/rollout_worker.py", line 150, in init
+    env_runner.init(self.timing)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 418, in init
+    self._reset()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/sampling/non_batched_sampling.py", line 430, in _reset
+    observations, info = e.reset(seed=seed)  # new way of doing seeding since Gym 0.26.0
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 125, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/algo/utils/make_env.py", line 110, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/scenario_wrappers/gathering_reward_shaping.py", line 30, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 379, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sample_factory/envs/env_wrappers.py", line 84, in reset
+    obs, info = self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/gym/core.py", line 323, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/wrappers/multiplayer_stats.py", line 51, in reset
+    return self.env.reset(**kwargs)
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 323, in reset
+    self._ensure_initialized()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 274, in _ensure_initialized
+    self.initialize()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 269, in initialize
+    self._game_init()
+  File "/home/ark/.miniconda3/envs/deep-rl/lib/python3.9/site-packages/sf_examples/vizdoom/doom/doom_gym.py", line 244, in _game_init
+    raise EnvCriticalError()
+sample_factory.envs.env_utils.EnvCriticalError
+[2023-06-13 21:52:18,257][939136] Unhandled exception  in evt loop rollout_proc3_evt_loop
+[2023-06-13 21:52:18,399][939137] Decorrelating experience for 0 frames...
+[2023-06-13 21:52:18,399][939139] Decorrelating experience for 0 frames...
+[2023-06-13 21:52:18,400][939138] Decorrelating experience for 0 frames...
+[2023-06-13 21:52:18,457][939135] Decorrelating experience for 0 frames...
+[2023-06-13 21:52:18,557][939139] Decorrelating experience for 32 frames...
+[2023-06-13 21:52:18,558][939138] Decorrelating experience for 32 frames...
+[2023-06-13 21:52:18,583][939137] Decorrelating experience for 32 frames...
+[2023-06-13 21:52:18,585][939131] Decorrelating experience for 0 frames...
+[2023-06-13 21:52:18,618][939135] Decorrelating experience for 32 frames...
+[2023-06-13 21:52:18,748][939138] Decorrelating experience for 64 frames...
+[2023-06-13 21:52:18,749][939139] Decorrelating experience for 64 frames...
+[2023-06-13 21:52:18,796][939135] Decorrelating experience for 64 frames...
+[2023-06-13 21:52:18,827][939137] Decorrelating experience for 64 frames...
+[2023-06-13 21:52:18,850][939131] Decorrelating experience for 32 frames...
+[2023-06-13 21:52:18,920][939138] Decorrelating experience for 96 frames...
+[2023-06-13 21:52:18,928][939139] Decorrelating experience for 96 frames...
+[2023-06-13 21:52:19,006][939137] Decorrelating experience for 96 frames...
+[2023-06-13 21:52:19,018][939135] Decorrelating experience for 96 frames...
+[2023-06-13 21:52:19,109][939131] Decorrelating experience for 64 frames...
+[2023-06-13 21:52:19,324][939131] Decorrelating experience for 96 frames...
+[2023-06-13 21:52:19,582][939084] Signal inference workers to stop experience collection...
+[2023-06-13 21:52:19,584][939130] InferenceWorker_p0-w0: stopping experience collection
+[2023-06-13 21:52:19,617][939011] Fps is (10 sec: nan, 60 sec: nan, 300 sec: nan). Total num frames: 0. Throughput: 0: nan. Samples: 0. Policy #0 lag: (min: -1.0, avg: -1.0, max: -1.0)
+[2023-06-13 21:52:19,617][939011] Avg episode reward: [(0, '3.026')]
+[2023-06-13 21:52:19,769][939084] Signal inference workers to resume experience collection...
+[2023-06-13 21:52:19,770][939130] InferenceWorker_p0-w0: resuming experience collection
+[2023-06-13 21:52:21,744][939130] Updated weights for policy 0, policy_version 10 (0.0188)
+[2023-06-13 21:52:23,748][939130] Updated weights for policy 0, policy_version 20 (0.0006)
+[2023-06-13 21:52:24,616][939011] Fps is (10 sec: 19660.9, 60 sec: 19660.9, 300 sec: 19660.9). Total num frames: 98304. Throughput: 0: 4553.2. Samples: 22766. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:52:24,617][939011] Avg episode reward: [(0, '4.413')]
+[2023-06-13 21:52:25,701][939130] Updated weights for policy 0, policy_version 30 (0.0006)
+[2023-06-13 21:52:27,706][939130] Updated weights for policy 0, policy_version 40 (0.0006)
+[2023-06-13 21:52:29,616][939011] Fps is (10 sec: 20070.4, 60 sec: 20070.4, 300 sec: 20070.4). Total num frames: 200704. Throughput: 0: 3837.2. Samples: 38372. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
+[2023-06-13 21:52:29,617][939011] Avg episode reward: [(0, '4.413')]
+[2023-06-13 21:52:29,628][939084] Saving new best policy, reward=4.413!
+[2023-06-13 21:52:29,712][939130] Updated weights for policy 0, policy_version 50 (0.0007)
+[2023-06-13 21:52:31,057][939011] Heartbeat connected on Batcher_0
+[2023-06-13 21:52:31,059][939011] Heartbeat connected on LearnerWorker_p0
+[2023-06-13 21:52:31,065][939011] Heartbeat connected on InferenceWorker_p0-w0
+[2023-06-13 21:52:31,067][939011] Heartbeat connected on RolloutWorker_w1
+[2023-06-13 21:52:31,075][939011] Heartbeat connected on RolloutWorker_w4
+[2023-06-13 21:52:31,076][939011] Heartbeat connected on RolloutWorker_w5
+[2023-06-13 21:52:31,077][939011] Heartbeat connected on RolloutWorker_w6
+[2023-06-13 21:52:31,079][939011] Heartbeat connected on RolloutWorker_w7
+[2023-06-13 21:52:31,720][939130] Updated weights for policy 0, policy_version 60 (0.0007)
+[2023-06-13 21:52:33,758][939130] Updated weights for policy 0, policy_version 70 (0.0006)
+[2023-06-13 21:52:34,617][939011] Fps is (10 sec: 20479.5, 60 sec: 20206.6, 300 sec: 20206.6). Total num frames: 303104. Throughput: 0: 4591.4. Samples: 68872. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:52:34,617][939011] Avg episode reward: [(0, '4.716')]
+[2023-06-13 21:52:34,625][939084] Saving new best policy, reward=4.716!
+[2023-06-13 21:52:35,813][939130] Updated weights for policy 0, policy_version 80 (0.0006)
+[2023-06-13 21:52:37,804][939130] Updated weights for policy 0, policy_version 90 (0.0007)
+[2023-06-13 21:52:39,616][939011] Fps is (10 sec: 20070.6, 60 sec: 20070.5, 300 sec: 20070.5). Total num frames: 401408. Throughput: 0: 4959.9. Samples: 99198. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:52:39,617][939011] Avg episode reward: [(0, '4.360')]
+[2023-06-13 21:52:39,838][939130] Updated weights for policy 0, policy_version 100 (0.0006)
+[2023-06-13 21:52:41,812][939130] Updated weights for policy 0, policy_version 110 (0.0006)
+[2023-06-13 21:52:43,799][939130] Updated weights for policy 0, policy_version 120 (0.0006)
+[2023-06-13 21:52:44,617][939011] Fps is (10 sec: 20070.7, 60 sec: 20152.3, 300 sec: 20152.3). Total num frames: 503808. Throughput: 0: 4585.3. Samples: 114632. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:52:44,617][939011] Avg episode reward: [(0, '4.265')]
+[2023-06-13 21:52:45,851][939130] Updated weights for policy 0, policy_version 130 (0.0007)
+[2023-06-13 21:52:47,807][939130] Updated weights for policy 0, policy_version 140 (0.0007)
+[2023-06-13 21:52:49,617][939011] Fps is (10 sec: 20479.5, 60 sec: 20206.9, 300 sec: 20206.9). Total num frames: 606208. Throughput: 0: 4843.2. Samples: 145296. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:52:49,617][939011] Avg episode reward: [(0, '4.741')]
+[2023-06-13 21:52:49,636][939084] Saving new best policy, reward=4.741!
+[2023-06-13 21:52:49,871][939130] Updated weights for policy 0, policy_version 150 (0.0007)
+[2023-06-13 21:52:51,840][939130] Updated weights for policy 0, policy_version 160 (0.0007)
+[2023-06-13 21:52:53,835][939130] Updated weights for policy 0, policy_version 170 (0.0007)
+[2023-06-13 21:52:54,617][939011] Fps is (10 sec: 20480.0, 60 sec: 20245.9, 300 sec: 20245.9). Total num frames: 708608. Throughput: 0: 5027.1. Samples: 175950. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:52:54,617][939011] Avg episode reward: [(0, '4.804')]
+[2023-06-13 21:52:54,639][939084] Saving new best policy, reward=4.804!
+[2023-06-13 21:52:55,868][939130] Updated weights for policy 0, policy_version 180 (0.0007)
+[2023-06-13 21:52:57,849][939130] Updated weights for policy 0, policy_version 190 (0.0007)
+[2023-06-13 21:52:59,616][939011] Fps is (10 sec: 20480.3, 60 sec: 20275.2, 300 sec: 20275.2). Total num frames: 811008. Throughput: 0: 4782.4. Samples: 191294. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:52:59,617][939011] Avg episode reward: [(0, '4.770')]
+[2023-06-13 21:52:59,844][939130] Updated weights for policy 0, policy_version 200 (0.0007)
+[2023-06-13 21:53:01,856][939130] Updated weights for policy 0, policy_version 210 (0.0006)
+[2023-06-13 21:53:03,931][939130] Updated weights for policy 0, policy_version 220 (0.0007)
+[2023-06-13 21:53:04,616][939011] Fps is (10 sec: 20480.3, 60 sec: 20298.0, 300 sec: 20298.0). Total num frames: 913408. Throughput: 0: 4926.6. Samples: 221696. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:53:04,617][939011] Avg episode reward: [(0, '5.059')]
+[2023-06-13 21:53:04,617][939084] Saving new best policy, reward=5.059!
+[2023-06-13 21:53:05,980][939130] Updated weights for policy 0, policy_version 230 (0.0007)
+[2023-06-13 21:53:07,997][939130] Updated weights for policy 0, policy_version 240 (0.0007)
+[2023-06-13 21:53:09,616][939011] Fps is (10 sec: 20070.3, 60 sec: 20234.2, 300 sec: 20234.2). Total num frames: 1011712. Throughput: 0: 5087.6. Samples: 251708. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:53:09,617][939011] Avg episode reward: [(0, '5.930')]
+[2023-06-13 21:53:09,619][939084] Saving new best policy, reward=5.930!
+[2023-06-13 21:53:10,032][939130] Updated weights for policy 0, policy_version 250 (0.0006)
+[2023-06-13 21:53:12,084][939130] Updated weights for policy 0, policy_version 260 (0.0006)
+[2023-06-13 21:53:14,136][939130] Updated weights for policy 0, policy_version 270 (0.0007)
+[2023-06-13 21:53:14,617][939011] Fps is (10 sec: 20070.2, 60 sec: 20256.6, 300 sec: 20256.6). Total num frames: 1114112. Throughput: 0: 5075.0. Samples: 266746. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:53:14,617][939011] Avg episode reward: [(0, '7.275')]
+[2023-06-13 21:53:14,617][939084] Saving new best policy, reward=7.275!
+[2023-06-13 21:53:16,186][939130] Updated weights for policy 0, policy_version 280 (0.0007)
+[2023-06-13 21:53:18,233][939130] Updated weights for policy 0, policy_version 290 (0.0006)
+[2023-06-13 21:53:19,617][939011] Fps is (10 sec: 20070.2, 60 sec: 20206.9, 300 sec: 20206.9). Total num frames: 1212416. Throughput: 0: 5067.4. Samples: 296906. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:53:19,617][939011] Avg episode reward: [(0, '8.478')]
+[2023-06-13 21:53:19,628][939084] Saving new best policy, reward=8.478!
+[2023-06-13 21:53:20,288][939130] Updated weights for policy 0, policy_version 300 (0.0006)
+[2023-06-13 21:53:22,308][939130] Updated weights for policy 0, policy_version 310 (0.0007)
+[2023-06-13 21:53:24,363][939130] Updated weights for policy 0, policy_version 320 (0.0007)
+[2023-06-13 21:53:24,617][939011] Fps is (10 sec: 20070.3, 60 sec: 20275.2, 300 sec: 20227.9). Total num frames: 1314816. Throughput: 0: 5060.7. Samples: 326930. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:53:24,617][939011] Avg episode reward: [(0, '8.446')]
+[2023-06-13 21:53:26,556][939130] Updated weights for policy 0, policy_version 330 (0.0007)
+[2023-06-13 21:53:28,505][939130] Updated weights for policy 0, policy_version 340 (0.0007)
+[2023-06-13 21:53:29,616][939011] Fps is (10 sec: 20070.7, 60 sec: 20206.9, 300 sec: 20187.4). Total num frames: 1413120. Throughput: 0: 5037.3. Samples: 341308. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:53:29,617][939011] Avg episode reward: [(0, '9.765')]
+[2023-06-13 21:53:29,619][939084] Saving new best policy, reward=9.765!
+[2023-06-13 21:53:30,540][939130] Updated weights for policy 0, policy_version 350 (0.0007)
+[2023-06-13 21:53:32,579][939130] Updated weights for policy 0, policy_version 360 (0.0006)
+[2023-06-13 21:53:34,617][939011] Fps is (10 sec: 19660.8, 60 sec: 20138.7, 300 sec: 20152.3). Total num frames: 1511424. Throughput: 0: 5035.9. Samples: 371912. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:53:34,617][939011] Avg episode reward: [(0, '9.763')]
+[2023-06-13 21:53:34,697][939130] Updated weights for policy 0, policy_version 370 (0.0007)
+[2023-06-13 21:53:37,006][939130] Updated weights for policy 0, policy_version 380 (0.0007)
+[2023-06-13 21:53:39,148][939130] Updated weights for policy 0, policy_version 390 (0.0007)
+[2023-06-13 21:53:39,616][939011] Fps is (10 sec: 19251.2, 60 sec: 20070.4, 300 sec: 20070.4). Total num frames: 1605632. Throughput: 0: 4974.9. Samples: 399822. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:53:39,617][939011] Avg episode reward: [(0, '11.339')]
+[2023-06-13 21:53:39,619][939084] Saving new best policy, reward=11.339!
+[2023-06-13 21:53:41,214][939130] Updated weights for policy 0, policy_version 400 (0.0007)
+[2023-06-13 21:53:43,262][939130] Updated weights for policy 0, policy_version 410 (0.0007)
+[2023-06-13 21:53:44,616][939011] Fps is (10 sec: 19251.3, 60 sec: 20002.2, 300 sec: 20046.3). Total num frames: 1703936. Throughput: 0: 4959.2. Samples: 414456. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:53:44,617][939011] Avg episode reward: [(0, '12.911')]
+[2023-06-13 21:53:44,617][939084] Saving new best policy, reward=12.911!
+[2023-06-13 21:53:45,309][939130] Updated weights for policy 0, policy_version 420 (0.0007)
+[2023-06-13 21:53:47,377][939130] Updated weights for policy 0, policy_version 430 (0.0007)
+[2023-06-13 21:53:49,389][939130] Updated weights for policy 0, policy_version 440 (0.0007)
+[2023-06-13 21:53:49,616][939011] Fps is (10 sec: 20070.4, 60 sec: 20002.2, 300 sec: 20070.4). Total num frames: 1806336. Throughput: 0: 4950.7. Samples: 444480. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:53:49,617][939011] Avg episode reward: [(0, '15.000')]
+[2023-06-13 21:53:49,619][939084] Saving new best policy, reward=15.000!
+[2023-06-13 21:53:51,589][939130] Updated weights for policy 0, policy_version 450 (0.0007)
+[2023-06-13 21:53:53,619][939130] Updated weights for policy 0, policy_version 460 (0.0007)
+[2023-06-13 21:53:54,616][939011] Fps is (10 sec: 19660.8, 60 sec: 19865.6, 300 sec: 20005.7). Total num frames: 1900544. Throughput: 0: 4936.7. Samples: 473858. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:53:54,617][939011] Avg episode reward: [(0, '15.611')]
+[2023-06-13 21:53:54,617][939084] Saving new best policy, reward=15.611!
+[2023-06-13 21:53:55,650][939130] Updated weights for policy 0, policy_version 470 (0.0007)
+[2023-06-13 21:53:57,704][939130] Updated weights for policy 0, policy_version 480 (0.0007)
+[2023-06-13 21:53:59,616][939011] Fps is (10 sec: 19660.7, 60 sec: 19865.6, 300 sec: 20029.4). Total num frames: 2002944. Throughput: 0: 4945.7. Samples: 489302. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:53:59,617][939011] Avg episode reward: [(0, '15.169')]
+[2023-06-13 21:53:59,726][939130] Updated weights for policy 0, policy_version 490 (0.0006)
+[2023-06-13 21:54:01,734][939130] Updated weights for policy 0, policy_version 500 (0.0007)
+[2023-06-13 21:54:03,709][939130] Updated weights for policy 0, policy_version 510 (0.0007)
+[2023-06-13 21:54:04,616][939011] Fps is (10 sec: 20480.0, 60 sec: 19865.6, 300 sec: 20050.9). Total num frames: 2105344. Throughput: 0: 4947.7. Samples: 519550. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:54:04,617][939011] Avg episode reward: [(0, '13.910')]
+[2023-06-13 21:54:05,717][939130] Updated weights for policy 0, policy_version 520 (0.0006)
+[2023-06-13 21:54:07,710][939130] Updated weights for policy 0, policy_version 530 (0.0006)
+[2023-06-13 21:54:09,617][939011] Fps is (10 sec: 20479.9, 60 sec: 19933.8, 300 sec: 20070.4). Total num frames: 2207744. Throughput: 0: 4966.1. Samples: 550404. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:54:09,617][939011] Avg episode reward: [(0, '15.506')]
+[2023-06-13 21:54:09,620][939084] Saving /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000000539_2207744.pth...
+[2023-06-13 21:54:09,698][939130] Updated weights for policy 0, policy_version 540 (0.0006)
+[2023-06-13 21:54:11,694][939130] Updated weights for policy 0, policy_version 550 (0.0007)
+[2023-06-13 21:54:13,667][939130] Updated weights for policy 0, policy_version 560 (0.0006)
+[2023-06-13 21:54:14,616][939011] Fps is (10 sec: 20480.0, 60 sec: 19933.9, 300 sec: 20088.2). Total num frames: 2310144. Throughput: 0: 4989.8. Samples: 565850. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:54:14,617][939011] Avg episode reward: [(0, '18.241')]
+[2023-06-13 21:54:14,617][939084] Saving new best policy, reward=18.241!
+[2023-06-13 21:54:15,722][939130] Updated weights for policy 0, policy_version 570 (0.0006)
+[2023-06-13 21:54:17,799][939130] Updated weights for policy 0, policy_version 580 (0.0007)
+[2023-06-13 21:54:19,616][939011] Fps is (10 sec: 20480.2, 60 sec: 20002.2, 300 sec: 20104.5). Total num frames: 2412544. Throughput: 0: 4979.7. Samples: 595998. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:54:19,617][939011] Avg episode reward: [(0, '16.196')]
+[2023-06-13 21:54:19,798][939130] Updated weights for policy 0, policy_version 590 (0.0007)
+[2023-06-13 21:54:21,840][939130] Updated weights for policy 0, policy_version 600 (0.0007)
+[2023-06-13 21:54:23,964][939130] Updated weights for policy 0, policy_version 610 (0.0007)
+[2023-06-13 21:54:24,617][939011] Fps is (10 sec: 20070.3, 60 sec: 19933.9, 300 sec: 20086.8). Total num frames: 2510848. Throughput: 0: 5023.2. Samples: 625864. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:54:24,617][939011] Avg episode reward: [(0, '18.045')]
+[2023-06-13 21:54:26,184][939130] Updated weights for policy 0, policy_version 620 (0.0007)
+[2023-06-13 21:54:28,223][939130] Updated weights for policy 0, policy_version 630 (0.0007)
+[2023-06-13 21:54:29,616][939011] Fps is (10 sec: 19251.3, 60 sec: 19865.6, 300 sec: 20038.9). Total num frames: 2605056. Throughput: 0: 5010.8. Samples: 639940. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:54:29,617][939011] Avg episode reward: [(0, '18.442')]
+[2023-06-13 21:54:29,619][939084] Saving new best policy, reward=18.442!
+[2023-06-13 21:54:30,452][939130] Updated weights for policy 0, policy_version 640 (0.0007)
+[2023-06-13 21:54:32,489][939130] Updated weights for policy 0, policy_version 650 (0.0006)
+[2023-06-13 21:54:34,617][939011] Fps is (10 sec: 18841.6, 60 sec: 19797.3, 300 sec: 19994.5). Total num frames: 2699264. Throughput: 0: 4987.7. Samples: 668926. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:54:34,617][939011] Avg episode reward: [(0, '20.413')]
+[2023-06-13 21:54:34,629][939084] Saving new best policy, reward=20.413!
+[2023-06-13 21:54:34,631][939130] Updated weights for policy 0, policy_version 660 (0.0006)
+[2023-06-13 21:54:36,671][939130] Updated weights for policy 0, policy_version 670 (0.0007)
+[2023-06-13 21:54:38,906][939130] Updated weights for policy 0, policy_version 680 (0.0007)
+[2023-06-13 21:54:39,617][939011] Fps is (10 sec: 19251.0, 60 sec: 19865.6, 300 sec: 19982.6). Total num frames: 2797568. Throughput: 0: 4973.5. Samples: 697666. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:54:39,617][939011] Avg episode reward: [(0, '21.822')]
+[2023-06-13 21:54:39,620][939084] Saving new best policy, reward=21.822!
+[2023-06-13 21:54:40,981][939130] Updated weights for policy 0, policy_version 690 (0.0007)
+[2023-06-13 21:54:43,040][939130] Updated weights for policy 0, policy_version 700 (0.0007)
+[2023-06-13 21:54:44,616][939011] Fps is (10 sec: 19660.9, 60 sec: 19865.6, 300 sec: 19971.5). Total num frames: 2895872. Throughput: 0: 4963.4. Samples: 712654. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:54:44,617][939011] Avg episode reward: [(0, '16.995')]
+[2023-06-13 21:54:45,106][939130] Updated weights for policy 0, policy_version 710 (0.0007)
+[2023-06-13 21:54:47,142][939130] Updated weights for policy 0, policy_version 720 (0.0007)
+[2023-06-13 21:54:49,131][939130] Updated weights for policy 0, policy_version 730 (0.0006)
+[2023-06-13 21:54:49,617][939011] Fps is (10 sec: 20070.5, 60 sec: 19865.6, 300 sec: 19988.5). Total num frames: 2998272. Throughput: 0: 4959.1. Samples: 742708. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:54:49,617][939011] Avg episode reward: [(0, '19.847')]
+[2023-06-13 21:54:51,210][939130] Updated weights for policy 0, policy_version 740 (0.0007)
+[2023-06-13 21:54:53,222][939130] Updated weights for policy 0, policy_version 750 (0.0006)
+[2023-06-13 21:54:54,617][939011] Fps is (10 sec: 20070.2, 60 sec: 19933.8, 300 sec: 19977.9). Total num frames: 3096576. Throughput: 0: 4945.6. Samples: 772954. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:54:54,617][939011] Avg episode reward: [(0, '20.540')]
+[2023-06-13 21:54:55,240][939130] Updated weights for policy 0, policy_version 760 (0.0006)
+[2023-06-13 21:54:57,465][939130] Updated weights for policy 0, policy_version 770 (0.0007)
+[2023-06-13 21:54:59,616][939011] Fps is (10 sec: 19251.3, 60 sec: 19797.3, 300 sec: 19942.4). Total num frames: 3190784. Throughput: 0: 4917.8. Samples: 787150. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:54:59,617][939011] Avg episode reward: [(0, '18.935')]
+[2023-06-13 21:54:59,629][939130] Updated weights for policy 0, policy_version 780 (0.0007)
+[2023-06-13 21:55:01,744][939130] Updated weights for policy 0, policy_version 790 (0.0006)
+[2023-06-13 21:55:03,886][939130] Updated weights for policy 0, policy_version 800 (0.0007)
+[2023-06-13 21:55:04,616][939011] Fps is (10 sec: 19251.4, 60 sec: 19729.1, 300 sec: 19933.9). Total num frames: 3289088. Throughput: 0: 4887.4. Samples: 815932. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:55:04,617][939011] Avg episode reward: [(0, '22.243')]
+[2023-06-13 21:55:04,617][939084] Saving new best policy, reward=22.243!
+[2023-06-13 21:55:05,927][939130] Updated weights for policy 0, policy_version 810 (0.0007)
+[2023-06-13 21:55:08,039][939130] Updated weights for policy 0, policy_version 820 (0.0007)
+[2023-06-13 21:55:09,616][939011] Fps is (10 sec: 19660.7, 60 sec: 19660.8, 300 sec: 19925.8). Total num frames: 3387392. Throughput: 0: 4884.0. Samples: 845644. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:55:09,617][939011] Avg episode reward: [(0, '20.575')]
+[2023-06-13 21:55:10,040][939130] Updated weights for policy 0, policy_version 830 (0.0007)
+[2023-06-13 21:55:12,124][939130] Updated weights for policy 0, policy_version 840 (0.0007)
+[2023-06-13 21:55:14,178][939130] Updated weights for policy 0, policy_version 850 (0.0007)
+[2023-06-13 21:55:14,617][939011] Fps is (10 sec: 19660.4, 60 sec: 19592.5, 300 sec: 19918.2). Total num frames: 3485696. Throughput: 0: 4901.8. Samples: 860524. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:55:14,617][939011] Avg episode reward: [(0, '18.917')]
+[2023-06-13 21:55:16,347][939130] Updated weights for policy 0, policy_version 860 (0.0007)
+[2023-06-13 21:55:18,416][939130] Updated weights for policy 0, policy_version 870 (0.0007)
+[2023-06-13 21:55:19,616][939011] Fps is (10 sec: 19660.8, 60 sec: 19524.3, 300 sec: 19911.1). Total num frames: 3584000. Throughput: 0: 4907.7. Samples: 889774. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:55:19,617][939011] Avg episode reward: [(0, '22.941')]
+[2023-06-13 21:55:19,620][939084] Saving new best policy, reward=22.941!
+[2023-06-13 21:55:20,490][939130] Updated weights for policy 0, policy_version 880 (0.0007)
+[2023-06-13 21:55:22,575][939130] Updated weights for policy 0, policy_version 890 (0.0006)
+[2023-06-13 21:55:24,585][939130] Updated weights for policy 0, policy_version 900 (0.0007)
+[2023-06-13 21:55:24,617][939011] Fps is (10 sec: 20070.7, 60 sec: 19592.5, 300 sec: 19926.5). Total num frames: 3686400. Throughput: 0: 4933.3. Samples: 919662. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:55:24,617][939011] Avg episode reward: [(0, '21.077')]
+[2023-06-13 21:55:26,633][939130] Updated weights for policy 0, policy_version 910 (0.0006)
+[2023-06-13 21:55:28,668][939130] Updated weights for policy 0, policy_version 920 (0.0007)
+[2023-06-13 21:55:29,617][939011] Fps is (10 sec: 20070.3, 60 sec: 19660.8, 300 sec: 19919.5). Total num frames: 3784704. Throughput: 0: 4935.1. Samples: 934734. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:55:29,617][939011] Avg episode reward: [(0, '21.238')]
+[2023-06-13 21:55:30,728][939130] Updated weights for policy 0, policy_version 930 (0.0006)
+[2023-06-13 21:55:32,743][939130] Updated weights for policy 0, policy_version 940 (0.0006)
+[2023-06-13 21:55:34,616][939011] Fps is (10 sec: 20070.5, 60 sec: 19797.3, 300 sec: 19933.9). Total num frames: 3887104. Throughput: 0: 4937.7. Samples: 964904. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:55:34,617][939011] Avg episode reward: [(0, '22.372')]
+[2023-06-13 21:55:34,727][939130] Updated weights for policy 0, policy_version 950 (0.0006)
+[2023-06-13 21:55:36,795][939130] Updated weights for policy 0, policy_version 960 (0.0007)
+[2023-06-13 21:55:38,848][939130] Updated weights for policy 0, policy_version 970 (0.0007)
+[2023-06-13 21:55:39,616][939011] Fps is (10 sec: 20070.7, 60 sec: 19797.4, 300 sec: 19927.1). Total num frames: 3985408. Throughput: 0: 4937.8. Samples: 995152. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:55:39,617][939011] Avg episode reward: [(0, '20.133')]
+[2023-06-13 21:55:40,893][939130] Updated weights for policy 0, policy_version 980 (0.0007)
+[2023-06-13 21:55:42,950][939130] Updated weights for policy 0, policy_version 990 (0.0007)
+[2023-06-13 21:55:44,616][939011] Fps is (10 sec: 20070.4, 60 sec: 19865.6, 300 sec: 19940.5). Total num frames: 4087808. Throughput: 0: 4953.1. Samples: 1010040. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:55:44,617][939011] Avg episode reward: [(0, '21.818')]
+[2023-06-13 21:55:45,039][939130] Updated weights for policy 0, policy_version 1000 (0.0007)
+[2023-06-13 21:55:47,075][939130] Updated weights for policy 0, policy_version 1010 (0.0007)
+[2023-06-13 21:55:49,137][939130] Updated weights for policy 0, policy_version 1020 (0.0006)
+[2023-06-13 21:55:49,617][939011] Fps is (10 sec: 20070.1, 60 sec: 19797.3, 300 sec: 19933.9). Total num frames: 4186112. Throughput: 0: 4978.3. Samples: 1039958. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:55:49,617][939011] Avg episode reward: [(0, '20.995')]
+[2023-06-13 21:55:51,168][939130] Updated weights for policy 0, policy_version 1030 (0.0007)
+[2023-06-13 21:55:53,465][939130] Updated weights for policy 0, policy_version 1040 (0.0007)
+[2023-06-13 21:55:54,617][939011] Fps is (10 sec: 19251.1, 60 sec: 19729.1, 300 sec: 19908.5). Total num frames: 4280320. Throughput: 0: 4950.6. Samples: 1068422. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:55:54,617][939011] Avg episode reward: [(0, '21.696')]
+[2023-06-13 21:55:55,570][939130] Updated weights for policy 0, policy_version 1050 (0.0007)
+[2023-06-13 21:55:57,688][939130] Updated weights for policy 0, policy_version 1060 (0.0006)
+[2023-06-13 21:55:59,616][939011] Fps is (10 sec: 19251.3, 60 sec: 19797.3, 300 sec: 19902.8). Total num frames: 4378624. Throughput: 0: 4943.2. Samples: 1082966. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:55:59,617][939011] Avg episode reward: [(0, '23.069')]
+[2023-06-13 21:55:59,620][939084] Saving new best policy, reward=23.069!
+[2023-06-13 21:55:59,788][939130] Updated weights for policy 0, policy_version 1070 (0.0007)
+[2023-06-13 21:56:01,864][939130] Updated weights for policy 0, policy_version 1080 (0.0007)
+[2023-06-13 21:56:03,963][939130] Updated weights for policy 0, policy_version 1090 (0.0007)
+[2023-06-13 21:56:04,617][939011] Fps is (10 sec: 19660.8, 60 sec: 19797.3, 300 sec: 19897.5). Total num frames: 4476928. Throughput: 0: 4947.6. Samples: 1112416. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:56:04,617][939011] Avg episode reward: [(0, '21.870')]
+[2023-06-13 21:56:06,099][939130] Updated weights for policy 0, policy_version 1100 (0.0007)
+[2023-06-13 21:56:08,185][939130] Updated weights for policy 0, policy_version 1110 (0.0007)
+[2023-06-13 21:56:09,617][939011] Fps is (10 sec: 19251.1, 60 sec: 19729.1, 300 sec: 19874.5). Total num frames: 4571136. Throughput: 0: 4934.4. Samples: 1141710. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:56:09,617][939011] Avg episode reward: [(0, '20.885')]
+[2023-06-13 21:56:09,628][939084] Saving /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000001117_4575232.pth...
+[2023-06-13 21:56:10,257][939130] Updated weights for policy 0, policy_version 1120 (0.0007)
+[2023-06-13 21:56:12,373][939130] Updated weights for policy 0, policy_version 1130 (0.0007)
+[2023-06-13 21:56:14,616][939011] Fps is (10 sec: 18841.7, 60 sec: 19660.9, 300 sec: 19852.5). Total num frames: 4665344. Throughput: 0: 4925.8. Samples: 1156396. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:56:14,617][939011] Avg episode reward: [(0, '22.998')]
+[2023-06-13 21:56:14,629][939130] Updated weights for policy 0, policy_version 1140 (0.0006)
+[2023-06-13 21:56:16,762][939130] Updated weights for policy 0, policy_version 1150 (0.0007)
+[2023-06-13 21:56:19,023][939130] Updated weights for policy 0, policy_version 1160 (0.0007)
+[2023-06-13 21:56:19,617][939011] Fps is (10 sec: 18841.5, 60 sec: 19592.5, 300 sec: 19831.5). Total num frames: 4759552. Throughput: 0: 4871.0. Samples: 1184098. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:56:19,617][939011] Avg episode reward: [(0, '24.154')]
+[2023-06-13 21:56:19,627][939084] Saving new best policy, reward=24.154!
+[2023-06-13 21:56:21,056][939130] Updated weights for policy 0, policy_version 1170 (0.0007)
+[2023-06-13 21:56:23,139][939130] Updated weights for policy 0, policy_version 1180 (0.0007)
+[2023-06-13 21:56:24,618][939011] Fps is (10 sec: 19657.8, 60 sec: 19592.1, 300 sec: 19844.6). Total num frames: 4861952. Throughput: 0: 4851.9. Samples: 1213496. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:56:24,619][939011] Avg episode reward: [(0, '23.396')]
+[2023-06-13 21:56:25,243][939130] Updated weights for policy 0, policy_version 1190 (0.0007)
+[2023-06-13 21:56:27,321][939130] Updated weights for policy 0, policy_version 1200 (0.0007)
+[2023-06-13 21:56:29,388][939130] Updated weights for policy 0, policy_version 1210 (0.0007)
+[2023-06-13 21:56:29,617][939011] Fps is (10 sec: 20070.5, 60 sec: 19592.5, 300 sec: 19841.0). Total num frames: 4960256. Throughput: 0: 4848.4. Samples: 1228218. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:56:29,617][939011] Avg episode reward: [(0, '23.236')]
+[2023-06-13 21:56:31,426][939130] Updated weights for policy 0, policy_version 1220 (0.0007)
+[2023-06-13 21:56:33,495][939130] Updated weights for policy 0, policy_version 1230 (0.0007)
+[2023-06-13 21:56:34,616][939011] Fps is (10 sec: 19663.7, 60 sec: 19524.3, 300 sec: 19837.5). Total num frames: 5058560. Throughput: 0: 4847.2. Samples: 1258080. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:56:34,617][939011] Avg episode reward: [(0, '21.353')]
+[2023-06-13 21:56:35,508][939130] Updated weights for policy 0, policy_version 1240 (0.0007)
+[2023-06-13 21:56:37,518][939130] Updated weights for policy 0, policy_version 1250 (0.0006)
+[2023-06-13 21:56:39,566][939130] Updated weights for policy 0, policy_version 1260 (0.0006)
+[2023-06-13 21:56:39,616][939011] Fps is (10 sec: 20070.5, 60 sec: 19592.5, 300 sec: 19849.8). Total num frames: 5160960. Throughput: 0: 4887.0. Samples: 1288338. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:56:39,617][939011] Avg episode reward: [(0, '23.808')]
+[2023-06-13 21:56:41,579][939130] Updated weights for policy 0, policy_version 1270 (0.0007)
+[2023-06-13 21:56:43,596][939130] Updated weights for policy 0, policy_version 1280 (0.0006)
+[2023-06-13 21:56:44,616][939011] Fps is (10 sec: 20070.4, 60 sec: 19524.3, 300 sec: 19846.3). Total num frames: 5259264. Throughput: 0: 4904.5. Samples: 1303670. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:56:44,617][939011] Avg episode reward: [(0, '20.961')]
+[2023-06-13 21:56:45,629][939130] Updated weights for policy 0, policy_version 1290 (0.0007)
+[2023-06-13 21:56:47,710][939130] Updated weights for policy 0, policy_version 1300 (0.0007)
+[2023-06-13 21:56:49,616][939011] Fps is (10 sec: 20070.4, 60 sec: 19592.5, 300 sec: 19858.0). Total num frames: 5361664. Throughput: 0: 4916.2. Samples: 1333646. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:56:49,617][939011] Avg episode reward: [(0, '22.300')]
+[2023-06-13 21:56:49,739][939130] Updated weights for policy 0, policy_version 1310 (0.0007)
+[2023-06-13 21:56:51,757][939130] Updated weights for policy 0, policy_version 1320 (0.0007)
+[2023-06-13 21:56:53,817][939130] Updated weights for policy 0, policy_version 1330 (0.0006)
+[2023-06-13 21:56:54,617][939011] Fps is (10 sec: 20070.2, 60 sec: 19660.8, 300 sec: 19854.4). Total num frames: 5459968. Throughput: 0: 4935.6. Samples: 1363812. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:56:54,617][939011] Avg episode reward: [(0, '21.960')]
+[2023-06-13 21:56:55,854][939130] Updated weights for policy 0, policy_version 1340 (0.0007)
+[2023-06-13 21:56:57,867][939130] Updated weights for policy 0, policy_version 1350 (0.0007)
+[2023-06-13 21:56:59,616][939011] Fps is (10 sec: 20070.4, 60 sec: 19729.1, 300 sec: 19865.6). Total num frames: 5562368. Throughput: 0: 4947.3. Samples: 1379026. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:56:59,617][939011] Avg episode reward: [(0, '24.849')]
+[2023-06-13 21:56:59,620][939084] Saving new best policy, reward=24.849!
+[2023-06-13 21:56:59,923][939130] Updated weights for policy 0, policy_version 1360 (0.0007)
+[2023-06-13 21:57:01,966][939130] Updated weights for policy 0, policy_version 1370 (0.0006)
+[2023-06-13 21:57:04,008][939130] Updated weights for policy 0, policy_version 1380 (0.0007)
+[2023-06-13 21:57:04,617][939011] Fps is (10 sec: 20070.6, 60 sec: 19729.1, 300 sec: 19862.0). Total num frames: 5660672. Throughput: 0: 4999.0. Samples: 1409052. Policy #0 lag: (min: 0.0, avg: 0.5, max: 1.0)
+[2023-06-13 21:57:04,617][939011] Avg episode reward: [(0, '23.593')]
+[2023-06-13 21:57:06,084][939130] Updated weights for policy 0, policy_version 1390 (0.0007)
+[2023-06-13 21:57:08,094][939130] Updated weights for policy 0, policy_version 1400 (0.0007)
+[2023-06-13 21:57:09,616][939011] Fps is (10 sec: 20070.4, 60 sec: 19865.6, 300 sec: 19872.7). Total num frames: 5763072. Throughput: 0: 5013.8. Samples: 1439108. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:57:09,617][939011] Avg episode reward: [(0, '25.375')]
+[2023-06-13 21:57:09,620][939084] Saving new best policy, reward=25.375!
+[2023-06-13 21:57:10,147][939130] Updated weights for policy 0, policy_version 1410 (0.0007)
+[2023-06-13 21:57:12,168][939130] Updated weights for policy 0, policy_version 1420 (0.0007)
+[2023-06-13 21:57:14,185][939130] Updated weights for policy 0, policy_version 1430 (0.0006)
+[2023-06-13 21:57:14,617][939011] Fps is (10 sec: 20480.0, 60 sec: 20002.1, 300 sec: 19883.0). Total num frames: 5865472. Throughput: 0: 5024.7. Samples: 1454328. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:57:14,617][939011] Avg episode reward: [(0, '23.377')]
+[2023-06-13 21:57:16,309][939130] Updated weights for policy 0, policy_version 1440 (0.0007)
+[2023-06-13 21:57:18,507][939130] Updated weights for policy 0, policy_version 1450 (0.0007)
+[2023-06-13 21:57:19,617][939011] Fps is (10 sec: 19660.6, 60 sec: 20002.1, 300 sec: 19869.1). Total num frames: 5959680. Throughput: 0: 5011.8. Samples: 1483612. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:57:19,617][939011] Avg episode reward: [(0, '23.753')]
+[2023-06-13 21:57:20,646][939130] Updated weights for policy 0, policy_version 1460 (0.0007)
+[2023-06-13 21:57:22,776][939130] Updated weights for policy 0, policy_version 1470 (0.0007)
+[2023-06-13 21:57:24,617][939011] Fps is (10 sec: 19251.1, 60 sec: 19934.3, 300 sec: 19855.2). Total num frames: 6057984. Throughput: 0: 4980.0. Samples: 1512440. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:57:24,617][939011] Avg episode reward: [(0, '22.807')]
+[2023-06-13 21:57:24,850][939130] Updated weights for policy 0, policy_version 1480 (0.0007)
+[2023-06-13 21:57:26,977][939130] Updated weights for policy 0, policy_version 1490 (0.0007)
+[2023-06-13 21:57:29,018][939130] Updated weights for policy 0, policy_version 1500 (0.0007)
+[2023-06-13 21:57:29,616][939011] Fps is (10 sec: 19251.4, 60 sec: 19865.6, 300 sec: 19827.4). Total num frames: 6152192. Throughput: 0: 4961.6. Samples: 1526942. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:57:29,617][939011] Avg episode reward: [(0, '25.046')]
+[2023-06-13 21:57:31,074][939130] Updated weights for policy 0, policy_version 1510 (0.0007)
+[2023-06-13 21:57:33,103][939130] Updated weights for policy 0, policy_version 1520 (0.0006)
+[2023-06-13 21:57:34,617][939011] Fps is (10 sec: 19660.9, 60 sec: 19933.9, 300 sec: 19841.3). Total num frames: 6254592. Throughput: 0: 4961.3. Samples: 1556906. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:57:34,617][939011] Avg episode reward: [(0, '25.530')]
+[2023-06-13 21:57:34,617][939084] Saving new best policy, reward=25.530!
+[2023-06-13 21:57:35,163][939130] Updated weights for policy 0, policy_version 1530 (0.0006)
+[2023-06-13 21:57:37,163][939130] Updated weights for policy 0, policy_version 1540 (0.0006)
+[2023-06-13 21:57:39,163][939130] Updated weights for policy 0, policy_version 1550 (0.0006)
+[2023-06-13 21:57:39,617][939011] Fps is (10 sec: 20479.5, 60 sec: 19933.8, 300 sec: 19841.3). Total num frames: 6356992. Throughput: 0: 4968.3. Samples: 1587386. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:57:39,617][939011] Avg episode reward: [(0, '25.690')]
+[2023-06-13 21:57:39,620][939084] Saving new best policy, reward=25.690!
+[2023-06-13 21:57:41,192][939130] Updated weights for policy 0, policy_version 1560 (0.0007)
+[2023-06-13 21:57:43,363][939130] Updated weights for policy 0, policy_version 1570 (0.0007)
+[2023-06-13 21:57:44,616][939011] Fps is (10 sec: 20070.5, 60 sec: 19933.9, 300 sec: 19827.4). Total num frames: 6455296. Throughput: 0: 4961.6. Samples: 1602298. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:57:44,617][939011] Avg episode reward: [(0, '23.340')]
+[2023-06-13 21:57:45,354][939130] Updated weights for policy 0, policy_version 1580 (0.0006)
+[2023-06-13 21:57:47,458][939130] Updated weights for policy 0, policy_version 1590 (0.0007)
+[2023-06-13 21:57:49,497][939130] Updated weights for policy 0, policy_version 1600 (0.0007)
+[2023-06-13 21:57:49,617][939011] Fps is (10 sec: 19661.1, 60 sec: 19865.6, 300 sec: 19813.5). Total num frames: 6553600. Throughput: 0: 4951.6. Samples: 1631876. Policy #0 lag: (min: 0.0, avg: 0.2, max: 1.0)
+[2023-06-13 21:57:49,617][939011] Avg episode reward: [(0, '23.274')]
+[2023-06-13 21:57:51,574][939130] Updated weights for policy 0, policy_version 1610 (0.0007)
+[2023-06-13 21:57:53,851][939130] Updated weights for policy 0, policy_version 1620 (0.0007)
+[2023-06-13 21:57:54,616][939011] Fps is (10 sec: 19251.2, 60 sec: 19797.4, 300 sec: 19785.8). Total num frames: 6647808. Throughput: 0: 4924.7. Samples: 1660718. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:57:54,617][939011] Avg episode reward: [(0, '20.124')]
+[2023-06-13 21:57:55,866][939130] Updated weights for policy 0, policy_version 1630 (0.0007)
+[2023-06-13 21:57:57,947][939130] Updated weights for policy 0, policy_version 1640 (0.0006)
+[2023-06-13 21:57:59,617][939011] Fps is (10 sec: 19251.2, 60 sec: 19729.0, 300 sec: 19771.9). Total num frames: 6746112. Throughput: 0: 4918.9. Samples: 1675678. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:57:59,617][939011] Avg episode reward: [(0, '21.282')]
+[2023-06-13 21:58:00,030][939130] Updated weights for policy 0, policy_version 1650 (0.0007)
+[2023-06-13 21:58:02,090][939130] Updated weights for policy 0, policy_version 1660 (0.0007)
+[2023-06-13 21:58:04,100][939130] Updated weights for policy 0, policy_version 1670 (0.0007)
+[2023-06-13 21:58:04,617][939011] Fps is (10 sec: 20070.3, 60 sec: 19797.3, 300 sec: 19785.8). Total num frames: 6848512. Throughput: 0: 4931.6. Samples: 1705532. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:04,617][939011] Avg episode reward: [(0, '23.178')]
+[2023-06-13 21:58:06,210][939130] Updated weights for policy 0, policy_version 1680 (0.0007)
+[2023-06-13 21:58:08,338][939130] Updated weights for policy 0, policy_version 1690 (0.0007)
+[2023-06-13 21:58:09,617][939011] Fps is (10 sec: 20070.5, 60 sec: 19729.1, 300 sec: 19771.9). Total num frames: 6946816. Throughput: 0: 4943.6. Samples: 1734900. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:09,617][939011] Avg episode reward: [(0, '24.549')]
+[2023-06-13 21:58:09,620][939084] Saving /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000001696_6946816.pth...
+[2023-06-13 21:58:09,672][939084] Removing /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000000539_2207744.pth
+[2023-06-13 21:58:10,427][939130] Updated weights for policy 0, policy_version 1700 (0.0007)
+[2023-06-13 21:58:12,467][939130] Updated weights for policy 0, policy_version 1710 (0.0007)
+[2023-06-13 21:58:14,528][939130] Updated weights for policy 0, policy_version 1720 (0.0007)
+[2023-06-13 21:58:14,617][939011] Fps is (10 sec: 19660.7, 60 sec: 19660.8, 300 sec: 19771.9). Total num frames: 7045120. Throughput: 0: 4951.0. Samples: 1749736. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:14,617][939011] Avg episode reward: [(0, '23.787')]
+[2023-06-13 21:58:16,602][939130] Updated weights for policy 0, policy_version 1730 (0.0007)
+[2023-06-13 21:58:18,696][939130] Updated weights for policy 0, policy_version 1740 (0.0007)
+[2023-06-13 21:58:19,617][939011] Fps is (10 sec: 19660.7, 60 sec: 19729.1, 300 sec: 19758.0). Total num frames: 7143424. Throughput: 0: 4943.7. Samples: 1779372. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:19,617][939011] Avg episode reward: [(0, '22.990')]
+[2023-06-13 21:58:20,758][939130] Updated weights for policy 0, policy_version 1750 (0.0007)
+[2023-06-13 21:58:22,835][939130] Updated weights for policy 0, policy_version 1760 (0.0007)
+[2023-06-13 21:58:24,617][939011] Fps is (10 sec: 19660.9, 60 sec: 19729.1, 300 sec: 19758.0). Total num frames: 7241728. Throughput: 0: 4922.2. Samples: 1808886. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:24,617][939011] Avg episode reward: [(0, '26.268')]
+[2023-06-13 21:58:24,618][939084] Saving new best policy, reward=26.268!
+[2023-06-13 21:58:24,957][939130] Updated weights for policy 0, policy_version 1770 (0.0007)
+[2023-06-13 21:58:27,035][939130] Updated weights for policy 0, policy_version 1780 (0.0007)
+[2023-06-13 21:58:29,232][939130] Updated weights for policy 0, policy_version 1790 (0.0007)
+[2023-06-13 21:58:29,617][939011] Fps is (10 sec: 19660.2, 60 sec: 19797.2, 300 sec: 19758.0). Total num frames: 7340032. Throughput: 0: 4914.9. Samples: 1823470. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:29,617][939011] Avg episode reward: [(0, '24.855')]
+[2023-06-13 21:58:31,502][939130] Updated weights for policy 0, policy_version 1800 (0.0007)
+[2023-06-13 21:58:33,641][939130] Updated weights for policy 0, policy_version 1810 (0.0007)
+[2023-06-13 21:58:34,617][939011] Fps is (10 sec: 18841.4, 60 sec: 19592.5, 300 sec: 19744.1). Total num frames: 7430144. Throughput: 0: 4873.5. Samples: 1851186. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:34,617][939011] Avg episode reward: [(0, '26.337')]
+[2023-06-13 21:58:34,617][939084] Saving new best policy, reward=26.337!
+[2023-06-13 21:58:35,729][939130] Updated weights for policy 0, policy_version 1820 (0.0007)
+[2023-06-13 21:58:37,890][939130] Updated weights for policy 0, policy_version 1830 (0.0006)
+[2023-06-13 21:58:39,617][939011] Fps is (10 sec: 18842.2, 60 sec: 19524.3, 300 sec: 19744.1). Total num frames: 7528448. Throughput: 0: 4883.2. Samples: 1880460. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:39,617][939011] Avg episode reward: [(0, '26.105')]
+[2023-06-13 21:58:39,935][939130] Updated weights for policy 0, policy_version 1840 (0.0006)
+[2023-06-13 21:58:42,066][939130] Updated weights for policy 0, policy_version 1850 (0.0007)
+[2023-06-13 21:58:44,348][939130] Updated weights for policy 0, policy_version 1860 (0.0007)
+[2023-06-13 21:58:44,617][939011] Fps is (10 sec: 19251.3, 60 sec: 19456.0, 300 sec: 19716.3). Total num frames: 7622656. Throughput: 0: 4876.7. Samples: 1895132. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:44,617][939011] Avg episode reward: [(0, '27.864')]
+[2023-06-13 21:58:44,617][939084] Saving new best policy, reward=27.864!
+[2023-06-13 21:58:46,738][939130] Updated weights for policy 0, policy_version 1870 (0.0007)
+[2023-06-13 21:58:49,037][939130] Updated weights for policy 0, policy_version 1880 (0.0007)
+[2023-06-13 21:58:49,617][939011] Fps is (10 sec: 18022.3, 60 sec: 19251.2, 300 sec: 19688.6). Total num frames: 7708672. Throughput: 0: 4799.2. Samples: 1921498. Policy #0 lag: (min: 0.0, avg: 0.3, max: 1.0)
+[2023-06-13 21:58:49,617][939011] Avg episode reward: [(0, '27.040')]
+[2023-06-13 21:58:51,335][939130] Updated weights for policy 0, policy_version 1890 (0.0007)
+[2023-06-13 21:58:53,651][939130] Updated weights for policy 0, policy_version 1900 (0.0008)
+[2023-06-13 21:58:54,617][939011] Fps is (10 sec: 17203.1, 60 sec: 19114.6, 300 sec: 19633.0). Total num frames: 7794688. Throughput: 0: 4733.8. Samples: 1947922. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:58:54,617][939011] Avg episode reward: [(0, '22.124')]
+[2023-06-13 21:58:56,039][939130] Updated weights for policy 0, policy_version 1910 (0.0008)
+[2023-06-13 21:58:58,450][939130] Updated weights for policy 0, policy_version 1920 (0.0008)
+[2023-06-13 21:58:59,617][939011] Fps is (10 sec: 17203.1, 60 sec: 18909.8, 300 sec: 19577.5). Total num frames: 7880704. Throughput: 0: 4685.9. Samples: 1960600. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:58:59,617][939011] Avg episode reward: [(0, '24.145')]
+[2023-06-13 21:59:00,803][939130] Updated weights for policy 0, policy_version 1930 (0.0007)
+[2023-06-13 21:59:03,120][939130] Updated weights for policy 0, policy_version 1940 (0.0007)
+[2023-06-13 21:59:04,617][939011] Fps is (10 sec: 17612.7, 60 sec: 18705.0, 300 sec: 19535.8). Total num frames: 7970816. Throughput: 0: 4612.2. Samples: 1986920. Policy #0 lag: (min: 0.0, avg: 0.4, max: 1.0)
+[2023-06-13 21:59:04,617][939011] Avg episode reward: [(0, '24.831')]
+[2023-06-13 21:59:05,399][939130] Updated weights for policy 0, policy_version 1950 (0.0007)
+[2023-06-13 21:59:06,549][939011] Component Batcher_0 stopped!
+[2023-06-13 21:59:06,549][939084] Stopping Batcher_0...
+[2023-06-13 21:59:06,549][939084] Loop batcher_evt_loop terminating...
+[2023-06-13 21:59:06,549][939011] Component RolloutWorker_w0 process died already! Don't wait for it.
+[2023-06-13 21:59:06,549][939084] Saving /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth...
+[2023-06-13 21:59:06,550][939011] Component RolloutWorker_w2 process died already! Don't wait for it.
+[2023-06-13 21:59:06,550][939011] Component RolloutWorker_w3 process died already! Don't wait for it.
+[2023-06-13 21:59:06,561][939135] Stopping RolloutWorker_w4...
+[2023-06-13 21:59:06,561][939137] Stopping RolloutWorker_w5...
+[2023-06-13 21:59:06,561][939131] Stopping RolloutWorker_w1...
+[2023-06-13 21:59:06,561][939138] Stopping RolloutWorker_w6...
+[2023-06-13 21:59:06,561][939011] Component RolloutWorker_w6 stopped!
+[2023-06-13 21:59:06,562][939011] Component RolloutWorker_w4 stopped!
+[2023-06-13 21:59:06,562][939135] Loop rollout_proc4_evt_loop terminating...
+[2023-06-13 21:59:06,562][939011] Component RolloutWorker_w5 stopped!
+[2023-06-13 21:59:06,562][939137] Loop rollout_proc5_evt_loop terminating...
+[2023-06-13 21:59:06,561][939139] Stopping RolloutWorker_w7...
+[2023-06-13 21:59:06,562][939131] Loop rollout_proc1_evt_loop terminating...
+[2023-06-13 21:59:06,562][939138] Loop rollout_proc6_evt_loop terminating...
+[2023-06-13 21:59:06,562][939011] Component RolloutWorker_w1 stopped!
+[2023-06-13 21:59:06,562][939011] Component RolloutWorker_w7 stopped!
+[2023-06-13 21:59:06,562][939139] Loop rollout_proc7_evt_loop terminating...
+[2023-06-13 21:59:06,568][939130] Weights refcount: 2 0
+[2023-06-13 21:59:06,570][939130] Stopping InferenceWorker_p0-w0...
+[2023-06-13 21:59:06,570][939130] Loop inference_proc0-0_evt_loop terminating...
+[2023-06-13 21:59:06,570][939011] Component InferenceWorker_p0-w0 stopped!
+[2023-06-13 21:59:06,622][939084] Removing /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000001117_4575232.pth
+[2023-06-13 21:59:06,629][939084] Saving /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth...
+[2023-06-13 21:59:06,723][939084] Stopping LearnerWorker_p0...
+[2023-06-13 21:59:06,723][939084] Loop learner_proc0_evt_loop terminating...
+[2023-06-13 21:59:06,723][939011] Component LearnerWorker_p0 stopped!
+[2023-06-13 21:59:06,723][939011] Waiting for process learner_proc0 to stop...
+[2023-06-13 21:59:07,334][939011] Waiting for process inference_proc0-0 to join...
+[2023-06-13 21:59:07,334][939011] Waiting for process rollout_proc0 to join...
+[2023-06-13 21:59:07,334][939011] Waiting for process rollout_proc1 to join...
+[2023-06-13 21:59:07,334][939011] Waiting for process rollout_proc2 to join...
+[2023-06-13 21:59:07,334][939011] Waiting for process rollout_proc3 to join...
+[2023-06-13 21:59:07,334][939011] Waiting for process rollout_proc4 to join...
+[2023-06-13 21:59:07,334][939011] Waiting for process rollout_proc5 to join...
+[2023-06-13 21:59:07,334][939011] Waiting for process rollout_proc6 to join...
+[2023-06-13 21:59:07,335][939011] Waiting for process rollout_proc7 to join...
+[2023-06-13 21:59:07,335][939011] Batcher 0 profile tree view:
+batching: 17.7602, releasing_batches: 0.0403
+[2023-06-13 21:59:07,335][939011] InferenceWorker_p0-w0 profile tree view:
+wait_policy: 0.0000
+  wait_policy_total: 5.2519
+update_model: 5.9176
+  weight_update: 0.0007
+one_step: 0.0017
+  handle_policy_step: 370.6976
+    deserialize: 14.7161, stack: 2.3144, obs_to_device_normalize: 83.9145, forward: 181.7526, send_messages: 19.1489
+    prepare_outputs: 51.5523
+      to_cpu: 31.3785
+[2023-06-13 21:59:07,335][939011] Learner 0 profile tree view:
+misc: 0.0118, prepare_batch: 8.2374
+train: 27.2126
+  epoch_init: 0.0098, minibatch_init: 0.0108, losses_postprocess: 0.4967, kl_divergence: 0.3587, after_optimizer: 6.7019
+  calculate_losses: 10.1412
+    losses_init: 0.0056, forward_head: 0.9810, bptt_initial: 5.5283, tail: 0.7455, advantages_returns: 0.2198, losses: 1.1784
+    bptt: 1.2402
+      bptt_forward_core: 1.1827
+  update: 8.9267
+    clip: 1.2964
+[2023-06-13 21:59:07,335][939011] RolloutWorker_w7 profile tree view:
+wait_for_trajectories: 0.3107, enqueue_policy_requests: 18.0286, env_step: 244.1739, overhead: 17.8636, complete_rollouts: 0.7567
+save_policy_outputs: 17.4264
+  split_output_tensors: 8.4588
+[2023-06-13 21:59:07,335][939011] Loop Runner_EvtLoop terminating...
+[2023-06-13 21:59:07,336][939011] Runner profile tree view:
+main_loop: 416.2561
+[2023-06-13 21:59:07,336][939011] Collected {0: 8007680}, FPS: 19237.4
+[2023-06-13 21:59:07,365][939011] Loading existing experiment configuration from /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/config.json
+[2023-06-13 21:59:07,365][939011] Overriding arg 'num_workers' with value 1 passed from command line
+[2023-06-13 21:59:07,365][939011] Adding new argument 'no_render'=True that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'save_video'=True that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'video_frames'=1000000000.0 that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'video_name'=None that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'max_num_frames'=100000 that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'max_num_episodes'=10 that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'push_to_hub'=True that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'hf_repository'='arkadyark/rl_course_vizdoom_health_gathering_supreme' that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'policy_index'=0 that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'eval_deterministic'=False that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'train_script'=None that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Adding new argument 'enjoy_script'=None that is not in the saved config file!
+[2023-06-13 21:59:07,366][939011] Using frameskip 1 and render_action_repeat=4 for evaluation
+[2023-06-13 21:59:07,377][939011] Doom resolution: 160x120, resize resolution: (128, 72)
+[2023-06-13 21:59:07,378][939011] RunningMeanStd input shape: (3, 72, 128)
+[2023-06-13 21:59:07,378][939011] RunningMeanStd input shape: (1,)
+[2023-06-13 21:59:07,387][939011] ConvEncoder: input_channels=3
+[2023-06-13 21:59:07,460][939011] Conv encoder output size: 512
+[2023-06-13 21:59:07,461][939011] Policy head output size: 512
+[2023-06-13 21:59:10,683][939011] Loading state from checkpoint /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/checkpoint_p0/checkpoint_000001955_8007680.pth...
+[2023-06-13 21:59:11,159][939011] Num frames 100...
+[2023-06-13 21:59:11,239][939011] Num frames 200...
+[2023-06-13 21:59:11,318][939011] Num frames 300...
+[2023-06-13 21:59:11,400][939011] Num frames 400...
+[2023-06-13 21:59:11,482][939011] Num frames 500...
+[2023-06-13 21:59:11,566][939011] Num frames 600...
+[2023-06-13 21:59:11,648][939011] Num frames 700...
+[2023-06-13 21:59:11,729][939011] Num frames 800...
+[2023-06-13 21:59:11,809][939011] Num frames 900...
+[2023-06-13 21:59:11,889][939011] Num frames 1000...
+[2023-06-13 21:59:11,970][939011] Num frames 1100...
+[2023-06-13 21:59:12,052][939011] Num frames 1200...
+[2023-06-13 21:59:12,132][939011] Num frames 1300...
+[2023-06-13 21:59:12,194][939011] Avg episode rewards: #0: 33.120, true rewards: #0: 13.120
+[2023-06-13 21:59:12,194][939011] Avg episode reward: 33.120, avg true_objective: 13.120
+[2023-06-13 21:59:12,265][939011] Num frames 1400...
+[2023-06-13 21:59:12,346][939011] Num frames 1500...
+[2023-06-13 21:59:12,427][939011] Num frames 1600...
+[2023-06-13 21:59:12,511][939011] Num frames 1700...
+[2023-06-13 21:59:12,590][939011] Num frames 1800...
+[2023-06-13 21:59:12,672][939011] Num frames 1900...
+[2023-06-13 21:59:12,751][939011] Num frames 2000...
+[2023-06-13 21:59:12,832][939011] Num frames 2100...
+[2023-06-13 21:59:12,915][939011] Num frames 2200...
+[2023-06-13 21:59:12,998][939011] Num frames 2300...
+[2023-06-13 21:59:13,083][939011] Num frames 2400...
+[2023-06-13 21:59:13,169][939011] Avg episode rewards: #0: 28.190, true rewards: #0: 12.190
+[2023-06-13 21:59:13,169][939011] Avg episode reward: 28.190, avg true_objective: 12.190
+[2023-06-13 21:59:13,221][939011] Num frames 2500...
+[2023-06-13 21:59:13,307][939011] Num frames 2600...
+[2023-06-13 21:59:13,399][939011] Num frames 2700...
+[2023-06-13 21:59:13,479][939011] Num frames 2800...
+[2023-06-13 21:59:13,563][939011] Num frames 2900...
+[2023-06-13 21:59:13,646][939011] Num frames 3000...
+[2023-06-13 21:59:13,732][939011] Num frames 3100...
+[2023-06-13 21:59:13,847][939011] Avg episode rewards: #0: 23.920, true rewards: #0: 10.587
+[2023-06-13 21:59:13,847][939011] Avg episode reward: 23.920, avg true_objective: 10.587
+[2023-06-13 21:59:13,871][939011] Num frames 3200...
+[2023-06-13 21:59:13,953][939011] Num frames 3300...
+[2023-06-13 21:59:14,035][939011] Num frames 3400...
+[2023-06-13 21:59:14,120][939011] Num frames 3500...
+[2023-06-13 21:59:14,206][939011] Num frames 3600...
+[2023-06-13 21:59:14,290][939011] Num frames 3700...
+[2023-06-13 21:59:14,375][939011] Num frames 3800...
+[2023-06-13 21:59:14,463][939011] Num frames 3900...
+[2023-06-13 21:59:14,549][939011] Num frames 4000...
+[2023-06-13 21:59:14,608][939011] Avg episode rewards: #0: 22.770, true rewards: #0: 10.020
+[2023-06-13 21:59:14,608][939011] Avg episode reward: 22.770, avg true_objective: 10.020
+[2023-06-13 21:59:14,683][939011] Num frames 4100...
+[2023-06-13 21:59:14,765][939011] Num frames 4200...
+[2023-06-13 21:59:14,848][939011] Num frames 4300...
+[2023-06-13 21:59:14,931][939011] Num frames 4400...
+[2023-06-13 21:59:15,013][939011] Num frames 4500...
+[2023-06-13 21:59:15,083][939011] Avg episode rewards: #0: 20.240, true rewards: #0: 9.040
+[2023-06-13 21:59:15,084][939011] Avg episode reward: 20.240, avg true_objective: 9.040
+[2023-06-13 21:59:15,150][939011] Num frames 4600...
+[2023-06-13 21:59:15,230][939011] Num frames 4700...
+[2023-06-13 21:59:15,314][939011] Num frames 4800...
+[2023-06-13 21:59:15,403][939011] Num frames 4900...
+[2023-06-13 21:59:15,487][939011] Num frames 5000...
+[2023-06-13 21:59:15,573][939011] Num frames 5100...
+[2023-06-13 21:59:15,655][939011] Num frames 5200...
+[2023-06-13 21:59:15,737][939011] Num frames 5300...
+[2023-06-13 21:59:15,820][939011] Num frames 5400...
+[2023-06-13 21:59:15,905][939011] Num frames 5500...
+[2023-06-13 21:59:15,988][939011] Num frames 5600...
+[2023-06-13 21:59:16,073][939011] Num frames 5700...
+[2023-06-13 21:59:16,158][939011] Num frames 5800...
+[2023-06-13 21:59:16,209][939011] Avg episode rewards: #0: 22.500, true rewards: #0: 9.667
+[2023-06-13 21:59:16,209][939011] Avg episode reward: 22.500, avg true_objective: 9.667
+[2023-06-13 21:59:16,292][939011] Num frames 5900...
+[2023-06-13 21:59:16,370][939011] Num frames 6000...
+[2023-06-13 21:59:16,449][939011] Num frames 6100...
+[2023-06-13 21:59:16,530][939011] Num frames 6200...
+[2023-06-13 21:59:16,610][939011] Num frames 6300...
+[2023-06-13 21:59:16,690][939011] Num frames 6400...
+[2023-06-13 21:59:16,774][939011] Num frames 6500...
+[2023-06-13 21:59:16,857][939011] Num frames 6600...
+[2023-06-13 21:59:16,940][939011] Num frames 6700...
+[2023-06-13 21:59:17,066][939011] Avg episode rewards: #0: 22.417, true rewards: #0: 9.703
+[2023-06-13 21:59:17,067][939011] Avg episode reward: 22.417, avg true_objective: 9.703
+[2023-06-13 21:59:17,075][939011] Num frames 6800...
+[2023-06-13 21:59:17,154][939011] Num frames 6900...
+[2023-06-13 21:59:17,232][939011] Num frames 7000...
+[2023-06-13 21:59:17,314][939011] Num frames 7100...
+[2023-06-13 21:59:17,396][939011] Num frames 7200...
+[2023-06-13 21:59:17,477][939011] Num frames 7300...
+[2023-06-13 21:59:17,557][939011] Num frames 7400...
+[2023-06-13 21:59:17,637][939011] Num frames 7500...
+[2023-06-13 21:59:17,718][939011] Num frames 7600...
+[2023-06-13 21:59:17,799][939011] Num frames 7700...
+[2023-06-13 21:59:17,880][939011] Num frames 7800...
+[2023-06-13 21:59:17,963][939011] Num frames 7900...
+[2023-06-13 21:59:18,046][939011] Num frames 8000...
+[2023-06-13 21:59:18,130][939011] Num frames 8100...
+[2023-06-13 21:59:18,219][939011] Num frames 8200...
+[2023-06-13 21:59:18,303][939011] Num frames 8300...
+[2023-06-13 21:59:18,391][939011] Num frames 8400...
+[2023-06-13 21:59:18,481][939011] Num frames 8500...
+[2023-06-13 21:59:18,567][939011] Num frames 8600...
+[2023-06-13 21:59:18,655][939011] Num frames 8700...
+[2023-06-13 21:59:18,738][939011] Num frames 8800...
+[2023-06-13 21:59:18,850][939011] Avg episode rewards: #0: 25.465, true rewards: #0: 11.090
+[2023-06-13 21:59:18,850][939011] Avg episode reward: 25.465, avg true_objective: 11.090
+[2023-06-13 21:59:18,874][939011] Num frames 8900...
+[2023-06-13 21:59:18,956][939011] Num frames 9000...
+[2023-06-13 21:59:19,037][939011] Num frames 9100...
+[2023-06-13 21:59:19,119][939011] Num frames 9200...
+[2023-06-13 21:59:19,200][939011] Num frames 9300...
+[2023-06-13 21:59:19,284][939011] Num frames 9400...
+[2023-06-13 21:59:19,369][939011] Num frames 9500...
+[2023-06-13 21:59:19,460][939011] Num frames 9600...
+[2023-06-13 21:59:19,548][939011] Avg episode rewards: #0: 24.378, true rewards: #0: 10.711
+[2023-06-13 21:59:19,549][939011] Avg episode reward: 24.378, avg true_objective: 10.711
+[2023-06-13 21:59:19,601][939011] Num frames 9700...
+[2023-06-13 21:59:19,685][939011] Num frames 9800...
+[2023-06-13 21:59:19,768][939011] Num frames 9900...
+[2023-06-13 21:59:19,850][939011] Num frames 10000...
+[2023-06-13 21:59:19,929][939011] Num frames 10100...
+[2023-06-13 21:59:20,010][939011] Num frames 10200...
+[2023-06-13 21:59:20,092][939011] Num frames 10300...
+[2023-06-13 21:59:20,176][939011] Num frames 10400...
+[2023-06-13 21:59:20,261][939011] Num frames 10500...
+[2023-06-13 21:59:20,345][939011] Num frames 10600...
+[2023-06-13 21:59:20,435][939011] Num frames 10700...
+[2023-06-13 21:59:20,521][939011] Num frames 10800...
+[2023-06-13 21:59:20,604][939011] Num frames 10900...
+[2023-06-13 21:59:20,691][939011] Num frames 11000...
+[2023-06-13 21:59:20,773][939011] Num frames 11100...
+[2023-06-13 21:59:20,859][939011] Num frames 11200...
+[2023-06-13 21:59:20,941][939011] Num frames 11300...
+[2023-06-13 21:59:21,017][939011] Avg episode rewards: #0: 26.026, true rewards: #0: 11.326
+[2023-06-13 21:59:21,017][939011] Avg episode reward: 26.026, avg true_objective: 11.326
+[2023-06-13 21:59:34,750][939011] Replay video saved to /home/ark/projects/deep-rl-course/unit-8-p2/train_dir/default_experiment/replay.mp4!