| { | |
| "best_metric": 0.6453333333333333, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0438/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": 140.0988006591797, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 3.694, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.46773333333333333, | |
| "eval_loss": 3.1674306392669678, | |
| "eval_runtime": 15.6941, | |
| "eval_samples_per_second": 238.943, | |
| "eval_steps_per_second": 3.759, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 58.13117218017578, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 2.3285, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6285333333333334, | |
| "eval_loss": 1.76164710521698, | |
| "eval_runtime": 15.4804, | |
| "eval_samples_per_second": 242.242, | |
| "eval_steps_per_second": 3.811, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 179.87008666992188, | |
| "learning_rate": 0.0, | |
| "loss": 1.6748, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6453333333333333, | |
| "eval_loss": 1.6105812788009644, | |
| "eval_runtime": 15.7426, | |
| "eval_samples_per_second": 238.207, | |
| "eval_steps_per_second": 3.748, | |
| "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 | |
| } | |