| { |
| "best_metric": 0.8997333333333334, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0595/checkpoints/checkpoint-2997", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 2997, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 31.006071090698242, |
| "learning_rate": 0.0004444444444444444, |
| "loss": 1.2234, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8344, |
| "eval_loss": 0.5291814804077148, |
| "eval_runtime": 7.9123, |
| "eval_samples_per_second": 473.943, |
| "eval_steps_per_second": 7.457, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 45.96661376953125, |
| "learning_rate": 0.0003888888888888889, |
| "loss": 0.3672, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8650666666666667, |
| "eval_loss": 0.4304977357387543, |
| "eval_runtime": 7.6252, |
| "eval_samples_per_second": 491.789, |
| "eval_steps_per_second": 7.737, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 26.87983512878418, |
| "learning_rate": 0.0003333333333333333, |
| "loss": 0.1957, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8682666666666666, |
| "eval_loss": 0.4820173680782318, |
| "eval_runtime": 7.5754, |
| "eval_samples_per_second": 495.022, |
| "eval_steps_per_second": 7.788, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 24.122703552246094, |
| "learning_rate": 0.0002777777777777778, |
| "loss": 0.1146, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8808, |
| "eval_loss": 0.44177883863449097, |
| "eval_runtime": 7.6164, |
| "eval_samples_per_second": 492.358, |
| "eval_steps_per_second": 7.746, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 22.64092445373535, |
| "learning_rate": 0.0002222222222222222, |
| "loss": 0.0535, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8861333333333333, |
| "eval_loss": 0.45425719022750854, |
| "eval_runtime": 7.7481, |
| "eval_samples_per_second": 483.991, |
| "eval_steps_per_second": 7.615, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 43.797035217285156, |
| "learning_rate": 0.00016666666666666666, |
| "loss": 0.0353, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8962666666666667, |
| "eval_loss": 0.44793862104415894, |
| "eval_runtime": 7.6442, |
| "eval_samples_per_second": 490.569, |
| "eval_steps_per_second": 7.718, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 33.60854721069336, |
| "learning_rate": 0.0001111111111111111, |
| "loss": 0.0257, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8973333333333333, |
| "eval_loss": 0.4569612443447113, |
| "eval_runtime": 7.6977, |
| "eval_samples_per_second": 487.159, |
| "eval_steps_per_second": 7.665, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 36.8309326171875, |
| "learning_rate": 5.555555555555555e-05, |
| "loss": 0.0193, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8976, |
| "eval_loss": 0.49047908186912537, |
| "eval_runtime": 7.6427, |
| "eval_samples_per_second": 490.663, |
| "eval_steps_per_second": 7.72, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 29.092975616455078, |
| "learning_rate": 0.0, |
| "loss": 0.0089, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8997333333333334, |
| "eval_loss": 0.46292173862457275, |
| "eval_runtime": 7.6217, |
| "eval_samples_per_second": 492.014, |
| "eval_steps_per_second": 7.741, |
| "step": 2997 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2997, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "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": 7.3587863155968e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
