| { | |
| "best_metric": 0.9261333333333334, | |
| "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0208/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": 0.42911025881767273, | |
| "learning_rate": 9e-05, | |
| "loss": 0.635, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9256, | |
| "eval_loss": 0.26288625597953796, | |
| "eval_runtime": 7.2354, | |
| "eval_samples_per_second": 518.285, | |
| "eval_steps_per_second": 8.154, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2592241168022156, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1466, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9170666666666667, | |
| "eval_loss": 0.2746236324310303, | |
| "eval_runtime": 7.2179, | |
| "eval_samples_per_second": 519.541, | |
| "eval_steps_per_second": 8.174, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.4936254620552063, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0797, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9157333333333333, | |
| "eval_loss": 0.3118854761123657, | |
| "eval_runtime": 8.0569, | |
| "eval_samples_per_second": 465.442, | |
| "eval_steps_per_second": 7.323, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.009943166747689247, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0498, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9184, | |
| "eval_loss": 0.33027541637420654, | |
| "eval_runtime": 6.7831, | |
| "eval_samples_per_second": 552.846, | |
| "eval_steps_per_second": 8.698, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.3855164051055908, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0426, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9261333333333334, | |
| "eval_loss": 0.3112790286540985, | |
| "eval_runtime": 6.7529, | |
| "eval_samples_per_second": 555.317, | |
| "eval_steps_per_second": 8.737, | |
| "step": 1665 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2331, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 8.2370660311296e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |