| { | |
| "best_metric": 0.9610666666666666, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0610/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": 0.4199051856994629, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.7542, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9464, | |
| "eval_loss": 0.19837306439876556, | |
| "eval_runtime": 11.0175, | |
| "eval_samples_per_second": 340.368, | |
| "eval_steps_per_second": 5.355, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.1362549513578415, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0868, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9530666666666666, | |
| "eval_loss": 0.17300479114055634, | |
| "eval_runtime": 10.9033, | |
| "eval_samples_per_second": 343.932, | |
| "eval_steps_per_second": 5.411, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.11241993308067322, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.0237, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9597333333333333, | |
| "eval_loss": 0.15057295560836792, | |
| "eval_runtime": 10.9142, | |
| "eval_samples_per_second": 343.59, | |
| "eval_steps_per_second": 5.406, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.24373456835746765, | |
| "learning_rate": 0.0, | |
| "loss": 0.0098, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9610666666666666, | |
| "eval_loss": 0.14674106240272522, | |
| "eval_runtime": 10.9502, | |
| "eval_samples_per_second": 342.461, | |
| "eval_steps_per_second": 5.388, | |
| "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": 6.58965282490368e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |