| { | |
| "best_metric": 0.9758454106280193, | |
| "best_model_checkpoint": "./vit-base-patch16-finetuned-beans/checkpoint-39", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 5.000819206237793, | |
| "learning_rate": 4.1428571428571437e-05, | |
| "loss": 0.3493, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9420289855072463, | |
| "eval_loss": 0.28053051233291626, | |
| "eval_runtime": 3.9125, | |
| "eval_samples_per_second": 52.908, | |
| "eval_steps_per_second": 3.323, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 4.752105236053467, | |
| "learning_rate": 2.714285714285714e-05, | |
| "loss": 0.2435, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9468599033816425, | |
| "eval_loss": 0.22151030600070953, | |
| "eval_runtime": 3.4713, | |
| "eval_samples_per_second": 59.633, | |
| "eval_steps_per_second": 3.745, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 4.369238376617432, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 0.1721, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9758454106280193, | |
| "eval_loss": 0.16175661981105804, | |
| "eval_runtime": 4.0571, | |
| "eval_samples_per_second": 51.021, | |
| "eval_steps_per_second": 3.204, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 39, | |
| "total_flos": 1.922593493780951e+17, | |
| "train_loss": 0.23679116139045128, | |
| "train_runtime": 141.695, | |
| "train_samples_per_second": 17.509, | |
| "train_steps_per_second": 0.275 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 39, | |
| "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": 1.922593493780951e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |