| { | |
| "best_metric": 0.8821333333333333, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0725/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": 172.63401794433594, | |
| "learning_rate": 0.0001, | |
| "loss": 2.827, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7432, | |
| "eval_loss": 0.9977997541427612, | |
| "eval_runtime": 7.4331, | |
| "eval_samples_per_second": 504.497, | |
| "eval_steps_per_second": 7.937, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 62.385231018066406, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7044, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8445333333333334, | |
| "eval_loss": 0.5167447328567505, | |
| "eval_runtime": 7.5036, | |
| "eval_samples_per_second": 499.762, | |
| "eval_steps_per_second": 7.863, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 52.03400802612305, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4232, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8685333333333334, | |
| "eval_loss": 0.4251783788204193, | |
| "eval_runtime": 7.5892, | |
| "eval_samples_per_second": 494.121, | |
| "eval_steps_per_second": 7.774, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 109.6050033569336, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2958, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8821333333333333, | |
| "eval_loss": 0.39940476417541504, | |
| "eval_runtime": 7.5347, | |
| "eval_samples_per_second": 497.698, | |
| "eval_steps_per_second": 7.83, | |
| "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 | |
| } | |