| { | |
| "best_metric": 0.8701333333333333, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0501/checkpoints/checkpoint-999", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 999, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.10642699152231216, | |
| "learning_rate": 0.0003, | |
| "loss": 0.63, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8469333333333333, | |
| "eval_loss": 0.5062747001647949, | |
| "eval_runtime": 6.1476, | |
| "eval_samples_per_second": 609.992, | |
| "eval_steps_per_second": 9.597, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2296755611896515, | |
| "learning_rate": 0.0003, | |
| "loss": 0.2915, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8584, | |
| "eval_loss": 0.4851309359073639, | |
| "eval_runtime": 6.1859, | |
| "eval_samples_per_second": 606.221, | |
| "eval_steps_per_second": 9.538, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.11353299021720886, | |
| "learning_rate": 0.0003, | |
| "loss": 0.209, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8701333333333333, | |
| "eval_loss": 0.46878987550735474, | |
| "eval_runtime": 6.1581, | |
| "eval_samples_per_second": 608.959, | |
| "eval_steps_per_second": 9.581, | |
| "step": 999 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.94223961867776e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |