| { |
| "best_metric": 0.8757333333333334, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0816/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": 239.9533233642578, |
| "learning_rate": 8e-05, |
| "loss": 2.8205, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.7408, |
| "eval_loss": 1.056075930595398, |
| "eval_runtime": 7.4924, |
| "eval_samples_per_second": 500.51, |
| "eval_steps_per_second": 7.875, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 278.7412414550781, |
| "learning_rate": 6e-05, |
| "loss": 0.714, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8370666666666666, |
| "eval_loss": 0.5275999307632446, |
| "eval_runtime": 7.5306, |
| "eval_samples_per_second": 497.969, |
| "eval_steps_per_second": 7.835, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 152.28260803222656, |
| "learning_rate": 4e-05, |
| "loss": 0.4369, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8642666666666666, |
| "eval_loss": 0.44530943036079407, |
| "eval_runtime": 7.4313, |
| "eval_samples_per_second": 504.624, |
| "eval_steps_per_second": 7.939, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 82.85832214355469, |
| "learning_rate": 2e-05, |
| "loss": 0.3373, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8757333333333334, |
| "eval_loss": 0.41544368863105774, |
| "eval_runtime": 7.5048, |
| "eval_samples_per_second": 499.678, |
| "eval_steps_per_second": 7.862, |
| "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 |
| } |
|
|
