| { | |
| "best_metric": 0.8968, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0043/checkpoints/checkpoint-2997", | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 2997, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 123.35450744628906, | |
| "learning_rate": 8.728616793536588e-05, | |
| "loss": 3.0448, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6962666666666667, | |
| "eval_loss": 1.222732424736023, | |
| "eval_runtime": 15.5011, | |
| "eval_samples_per_second": 241.919, | |
| "eval_steps_per_second": 3.806, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 80.11041259765625, | |
| "learning_rate": 7.947199994035401e-05, | |
| "loss": 0.7978, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.84, | |
| "eval_loss": 0.5508497357368469, | |
| "eval_runtime": 15.5222, | |
| "eval_samples_per_second": 241.59, | |
| "eval_steps_per_second": 3.801, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 128.939453125, | |
| "learning_rate": 6.75e-05, | |
| "loss": 0.4997, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8674666666666667, | |
| "eval_loss": 0.42465877532958984, | |
| "eval_runtime": 15.5411, | |
| "eval_samples_per_second": 241.296, | |
| "eval_steps_per_second": 3.796, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 94.61632537841797, | |
| "learning_rate": 5.2814167995011876e-05, | |
| "loss": 0.3898, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8749333333333333, | |
| "eval_loss": 0.40273380279541016, | |
| "eval_runtime": 15.6198, | |
| "eval_samples_per_second": 240.079, | |
| "eval_steps_per_second": 3.777, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 75.70044708251953, | |
| "learning_rate": 3.7185832004988136e-05, | |
| "loss": 0.3129, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8896, | |
| "eval_loss": 0.3701862692832947, | |
| "eval_runtime": 15.6553, | |
| "eval_samples_per_second": 239.536, | |
| "eval_steps_per_second": 3.769, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 176.26866149902344, | |
| "learning_rate": 2.250000000000001e-05, | |
| "loss": 0.2698, | |
| "step": 1998 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8882666666666666, | |
| "eval_loss": 0.35964110493659973, | |
| "eval_runtime": 15.491, | |
| "eval_samples_per_second": 242.076, | |
| "eval_steps_per_second": 3.809, | |
| "step": 1998 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 115.3402328491211, | |
| "learning_rate": 1.0528000059645995e-05, | |
| "loss": 0.2436, | |
| "step": 2331 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8957333333333334, | |
| "eval_loss": 0.34972500801086426, | |
| "eval_runtime": 15.6167, | |
| "eval_samples_per_second": 240.127, | |
| "eval_steps_per_second": 3.778, | |
| "step": 2331 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 155.9450225830078, | |
| "learning_rate": 2.7138320646341257e-06, | |
| "loss": 0.231, | |
| "step": 2664 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8954666666666666, | |
| "eval_loss": 0.34319230914115906, | |
| "eval_runtime": 15.541, | |
| "eval_samples_per_second": 241.297, | |
| "eval_steps_per_second": 3.796, | |
| "step": 2664 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 72.65001678466797, | |
| "learning_rate": 0.0, | |
| "loss": 0.2138, | |
| "step": 2997 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8968, | |
| "eval_loss": 0.3348565101623535, | |
| "eval_runtime": 15.5786, | |
| "eval_samples_per_second": 240.715, | |
| "eval_steps_per_second": 3.787, | |
| "step": 2997 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2997, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "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": 7.3587863155968e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |