| { | |
| "best_metric": 0.8872, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0335/checkpoints/checkpoint-1332", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 1332, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 28.45093536376953, | |
| "learning_rate": 0.000375, | |
| "loss": 1.291, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8277333333333333, | |
| "eval_loss": 0.5559366941452026, | |
| "eval_runtime": 12.4551, | |
| "eval_samples_per_second": 301.082, | |
| "eval_steps_per_second": 4.737, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 38.268035888671875, | |
| "learning_rate": 0.00025, | |
| "loss": 0.4042, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8677333333333334, | |
| "eval_loss": 0.44679224491119385, | |
| "eval_runtime": 12.6212, | |
| "eval_samples_per_second": 297.119, | |
| "eval_steps_per_second": 4.675, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 51.983463287353516, | |
| "learning_rate": 0.000125, | |
| "loss": 0.2328, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8834666666666666, | |
| "eval_loss": 0.3962169587612152, | |
| "eval_runtime": 12.4138, | |
| "eval_samples_per_second": 302.083, | |
| "eval_steps_per_second": 4.753, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 37.18984603881836, | |
| "learning_rate": 0.0, | |
| "loss": 0.1307, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8872, | |
| "eval_loss": 0.3714931011199951, | |
| "eval_runtime": 12.4217, | |
| "eval_samples_per_second": 301.892, | |
| "eval_steps_per_second": 4.75, | |
| "step": 1332 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1332, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 3.2705716958208e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |