| { | |
| "best_metric": 0.6752970218658447, | |
| "best_model_checkpoint": "resnet50_4a/checkpoint-2250", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 2250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 2.259565830230713, | |
| "learning_rate": 7.954545454545454e-07, | |
| "loss": 0.6891, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.59125, | |
| "eval_loss": 0.6828036308288574, | |
| "eval_runtime": 217.1108, | |
| "eval_samples_per_second": 110.543, | |
| "eval_steps_per_second": 1.727, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 2.487384796142578, | |
| "learning_rate": 5.681818181818182e-07, | |
| "loss": 0.6834, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.575913906097412, | |
| "learning_rate": 3.4090909090909085e-07, | |
| "loss": 0.6789, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6325, | |
| "eval_loss": 0.677152693271637, | |
| "eval_runtime": 111.7965, | |
| "eval_samples_per_second": 214.676, | |
| "eval_steps_per_second": 3.354, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 2.801647663116455, | |
| "learning_rate": 1.1363636363636363e-07, | |
| "loss": 0.6766, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6425833333333333, | |
| "eval_loss": 0.6752970218658447, | |
| "eval_runtime": 91.513, | |
| "eval_samples_per_second": 262.258, | |
| "eval_steps_per_second": 4.098, | |
| "step": 2250 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 6.11579618721792e+18, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |