| { | |
| "best_metric": 0.8992, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0840/checkpoints/checkpoint-1665", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1665, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 32.83119583129883, | |
| "learning_rate": 0.0003, | |
| "loss": 1.3859, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8533333333333334, | |
| "eval_loss": 0.47526314854621887, | |
| "eval_runtime": 7.1032, | |
| "eval_samples_per_second": 527.928, | |
| "eval_steps_per_second": 8.306, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 68.54553985595703, | |
| "learning_rate": 0.0003, | |
| "loss": 0.368, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8888, | |
| "eval_loss": 0.3685701787471771, | |
| "eval_runtime": 7.1033, | |
| "eval_samples_per_second": 527.927, | |
| "eval_steps_per_second": 8.306, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 35.07218933105469, | |
| "learning_rate": 0.0003, | |
| "loss": 0.2386, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8930666666666667, | |
| "eval_loss": 0.34429270029067993, | |
| "eval_runtime": 7.082, | |
| "eval_samples_per_second": 529.515, | |
| "eval_steps_per_second": 8.331, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 37.94849395751953, | |
| "learning_rate": 0.0003, | |
| "loss": 0.1602, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8909333333333334, | |
| "eval_loss": 0.38455480337142944, | |
| "eval_runtime": 7.1119, | |
| "eval_samples_per_second": 527.289, | |
| "eval_steps_per_second": 8.296, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 51.23830795288086, | |
| "learning_rate": 0.0003, | |
| "loss": 0.1179, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8992, | |
| "eval_loss": 0.3650992810726166, | |
| "eval_runtime": 7.1008, | |
| "eval_samples_per_second": 528.11, | |
| "eval_steps_per_second": 8.309, | |
| "step": 1665 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1665, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 4.088214619776e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |