| { | |
| "best_metric": 0.8784, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0424/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": 188.19430541992188, | |
| "learning_rate": 0.0001, | |
| "loss": 2.8304, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7456, | |
| "eval_loss": 0.960559606552124, | |
| "eval_runtime": 7.557, | |
| "eval_samples_per_second": 496.229, | |
| "eval_steps_per_second": 7.807, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 104.47616577148438, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6854, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8413333333333334, | |
| "eval_loss": 0.5277605056762695, | |
| "eval_runtime": 7.4061, | |
| "eval_samples_per_second": 506.339, | |
| "eval_steps_per_second": 7.966, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 84.31538391113281, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4342, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8570666666666666, | |
| "eval_loss": 0.4485729932785034, | |
| "eval_runtime": 7.4366, | |
| "eval_samples_per_second": 504.265, | |
| "eval_steps_per_second": 7.934, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 44.102787017822266, | |
| "learning_rate": 0.0001, | |
| "loss": 0.307, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8784, | |
| "eval_loss": 0.40127813816070557, | |
| "eval_runtime": 7.5391, | |
| "eval_samples_per_second": 497.404, | |
| "eval_steps_per_second": 7.826, | |
| "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 | |
| } | |