| { | |
| "best_metric": 0.7933333333333333, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0493/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": 218.59780883789062, | |
| "learning_rate": 6.75e-05, | |
| "loss": 3.0602, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6626666666666666, | |
| "eval_loss": 1.4252634048461914, | |
| "eval_runtime": 7.7428, | |
| "eval_samples_per_second": 484.324, | |
| "eval_steps_per_second": 7.62, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 95.0203628540039, | |
| "learning_rate": 2.250000000000001e-05, | |
| "loss": 0.9362, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7786666666666666, | |
| "eval_loss": 0.751481831073761, | |
| "eval_runtime": 7.9847, | |
| "eval_samples_per_second": 469.646, | |
| "eval_steps_per_second": 7.389, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 77.612548828125, | |
| "learning_rate": 0.0, | |
| "loss": 0.6847, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7933333333333333, | |
| "eval_loss": 0.7068875432014465, | |
| "eval_runtime": 7.8448, | |
| "eval_samples_per_second": 478.025, | |
| "eval_steps_per_second": 7.521, | |
| "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 | |
| } | |