| { | |
| "best_metric": 0.8936, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0363/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": 27.471256256103516, | |
| "learning_rate": 0.00042677669529663686, | |
| "loss": 1.3041, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8296, | |
| "eval_loss": 0.5335186123847961, | |
| "eval_runtime": 7.3462, | |
| "eval_samples_per_second": 510.467, | |
| "eval_steps_per_second": 8.031, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 32.49036407470703, | |
| "learning_rate": 0.00025, | |
| "loss": 0.38, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8594666666666667, | |
| "eval_loss": 0.4390268623828888, | |
| "eval_runtime": 7.2913, | |
| "eval_samples_per_second": 514.313, | |
| "eval_steps_per_second": 8.092, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 28.340110778808594, | |
| "learning_rate": 7.322330470336314e-05, | |
| "loss": 0.1574, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8805333333333333, | |
| "eval_loss": 0.39576002955436707, | |
| "eval_runtime": 7.412, | |
| "eval_samples_per_second": 505.935, | |
| "eval_steps_per_second": 7.96, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 18.1268310546875, | |
| "learning_rate": 0.0, | |
| "loss": 0.064, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8936, | |
| "eval_loss": 0.37900441884994507, | |
| "eval_runtime": 7.433, | |
| "eval_samples_per_second": 504.505, | |
| "eval_steps_per_second": 7.938, | |
| "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": 3.2705716958208e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |