| { | |
| "best_metric": 0.8639652677279306, | |
| "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-MM_Classification_base_V10/checkpoint-100", | |
| "epoch": 6.686567164179104, | |
| "eval_steps": 500, | |
| "global_step": 112, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 3.4567720890045166, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.9254, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.9552238805970149, | |
| "eval_accuracy": 0.8133140376266281, | |
| "eval_loss": 0.4842127859592438, | |
| "eval_runtime": 115.4192, | |
| "eval_samples_per_second": 5.987, | |
| "eval_steps_per_second": 0.052, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.1940298507462686, | |
| "grad_norm": 2.1373002529144287, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5607, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.7910447761194028, | |
| "grad_norm": 1.4010789394378662, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.4552, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9701492537313432, | |
| "eval_accuracy": 0.849493487698987, | |
| "eval_loss": 0.3855249881744385, | |
| "eval_runtime": 118.0615, | |
| "eval_samples_per_second": 5.853, | |
| "eval_steps_per_second": 0.051, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.388059701492537, | |
| "grad_norm": 3.225425958633423, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.4006, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "grad_norm": 1.8630387783050537, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.4034, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "eval_accuracy": 0.8610709117221418, | |
| "eval_loss": 0.34522491693496704, | |
| "eval_runtime": 76.1197, | |
| "eval_samples_per_second": 9.078, | |
| "eval_steps_per_second": 0.079, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.582089552238806, | |
| "grad_norm": 2.277456521987915, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.3583, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8581765557163531, | |
| "eval_loss": 0.33568304777145386, | |
| "eval_runtime": 76.3565, | |
| "eval_samples_per_second": 9.05, | |
| "eval_steps_per_second": 0.079, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 4.17910447761194, | |
| "grad_norm": 1.8838659524917603, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.3683, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.776119402985074, | |
| "grad_norm": 2.815551996231079, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.353, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.955223880597015, | |
| "eval_accuracy": 0.8625180897250362, | |
| "eval_loss": 0.32812753319740295, | |
| "eval_runtime": 76.4607, | |
| "eval_samples_per_second": 9.037, | |
| "eval_steps_per_second": 0.078, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 5.373134328358209, | |
| "grad_norm": 2.2589757442474365, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.3218, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.970149253731344, | |
| "grad_norm": 2.3260316848754883, | |
| "learning_rate": 6e-06, | |
| "loss": 0.3387, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.970149253731344, | |
| "eval_accuracy": 0.8639652677279306, | |
| "eval_loss": 0.3239884674549103, | |
| "eval_runtime": 76.5373, | |
| "eval_samples_per_second": 9.028, | |
| "eval_steps_per_second": 0.078, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.567164179104478, | |
| "grad_norm": 2.35009765625, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.3157, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.686567164179104, | |
| "eval_accuracy": 0.8639652677279306, | |
| "eval_loss": 0.325300008058548, | |
| "eval_runtime": 76.553, | |
| "eval_samples_per_second": 9.026, | |
| "eval_steps_per_second": 0.078, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.686567164179104, | |
| "step": 112, | |
| "total_flos": 4.4574447113624494e+18, | |
| "train_loss": 0.43422421866229605, | |
| "train_runtime": 7488.3728, | |
| "train_samples_per_second": 7.947, | |
| "train_steps_per_second": 0.015 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 112, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.4574447113624494e+18, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |