| { | |
| "best_metric": 0.9058666666666667, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0426/checkpoints/checkpoint-999", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 999, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 29.101356506347656, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 1.3494, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8570666666666666, | |
| "eval_loss": 0.482969731092453, | |
| "eval_runtime": 15.9672, | |
| "eval_samples_per_second": 234.856, | |
| "eval_steps_per_second": 3.695, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 51.21794128417969, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 0.3889, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8925333333333333, | |
| "eval_loss": 0.36371445655822754, | |
| "eval_runtime": 15.6539, | |
| "eval_samples_per_second": 239.558, | |
| "eval_steps_per_second": 3.769, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 32.8780632019043, | |
| "learning_rate": 0.0, | |
| "loss": 0.2182, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9058666666666667, | |
| "eval_loss": 0.3130524754524231, | |
| "eval_runtime": 15.7979, | |
| "eval_samples_per_second": 237.374, | |
| "eval_steps_per_second": 3.735, | |
| "step": 999 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 999, | |
| "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": 2.4529287718656e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |