| { | |
| "best_metric": 0.8765333333333334, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0557/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": Infinity, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8646, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8704, | |
| "eval_loss": 0.41381436586380005, | |
| "eval_runtime": 4.3233, | |
| "eval_samples_per_second": 867.402, | |
| "eval_steps_per_second": 13.647, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.122258186340332, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2713, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8610666666666666, | |
| "eval_loss": 0.4730485677719116, | |
| "eval_runtime": 4.389, | |
| "eval_samples_per_second": 854.405, | |
| "eval_steps_per_second": 13.443, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 47.51023864746094, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1653, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8592, | |
| "eval_loss": 0.47206926345825195, | |
| "eval_runtime": 4.5766, | |
| "eval_samples_per_second": 819.377, | |
| "eval_steps_per_second": 12.892, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.467433214187622, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1141, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8613333333333333, | |
| "eval_loss": 0.5148317813873291, | |
| "eval_runtime": 5.6365, | |
| "eval_samples_per_second": 665.308, | |
| "eval_steps_per_second": 10.468, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.6832707524299622, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0904, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8765333333333334, | |
| "eval_loss": 0.4857804775238037, | |
| "eval_runtime": 4.7015, | |
| "eval_samples_per_second": 797.615, | |
| "eval_steps_per_second": 12.549, | |
| "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": 8.2370660311296e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |