| { | |
| "best_metric": 0.88, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0635/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": 47.46843338012695, | |
| "learning_rate": 0.0005, | |
| "loss": 1.272, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8325333333333333, | |
| "eval_loss": 0.5606372356414795, | |
| "eval_runtime": 8.9939, | |
| "eval_samples_per_second": 416.947, | |
| "eval_steps_per_second": 6.56, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 22.392839431762695, | |
| "learning_rate": 0.0005, | |
| "loss": 0.4107, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8584, | |
| "eval_loss": 0.4872633218765259, | |
| "eval_runtime": 7.8267, | |
| "eval_samples_per_second": 479.131, | |
| "eval_steps_per_second": 7.538, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 17.348373413085938, | |
| "learning_rate": 0.0005, | |
| "loss": 0.2552, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8597333333333333, | |
| "eval_loss": 0.49062782526016235, | |
| "eval_runtime": 8.5472, | |
| "eval_samples_per_second": 438.742, | |
| "eval_steps_per_second": 6.903, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 27.852598190307617, | |
| "learning_rate": 0.0005, | |
| "loss": 0.1806, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.88, | |
| "eval_loss": 0.42726242542266846, | |
| "eval_runtime": 7.9499, | |
| "eval_samples_per_second": 471.704, | |
| "eval_steps_per_second": 7.421, | |
| "step": 1332 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1665, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.2705716958208e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |