| { |
| "best_metric": 0.8474666666666667, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0184/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": 390.371337890625, |
| "learning_rate": 6.788924172750679e-05, |
| "loss": 3.4478, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.5453333333333333, |
| "eval_loss": 2.4156854152679443, |
| "eval_runtime": 13.6037, |
| "eval_samples_per_second": 275.661, |
| "eval_steps_per_second": 4.337, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 56.958866119384766, |
| "learning_rate": 6.181155550916422e-05, |
| "loss": 1.3814, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.7584, |
| "eval_loss": 0.8664318323135376, |
| "eval_runtime": 13.6737, |
| "eval_samples_per_second": 274.249, |
| "eval_steps_per_second": 4.315, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 87.05301666259766, |
| "learning_rate": 5.2499999999999995e-05, |
| "loss": 0.7397, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7994666666666667, |
| "eval_loss": 0.6592377424240112, |
| "eval_runtime": 13.2082, |
| "eval_samples_per_second": 283.915, |
| "eval_steps_per_second": 4.467, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 344.10015869140625, |
| "learning_rate": 4.1077686218342566e-05, |
| "loss": 0.5906, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8234666666666667, |
| "eval_loss": 0.5729377865791321, |
| "eval_runtime": 13.4053, |
| "eval_samples_per_second": 279.741, |
| "eval_steps_per_second": 4.401, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 101.65443420410156, |
| "learning_rate": 2.8922313781657438e-05, |
| "loss": 0.4966, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8370666666666666, |
| "eval_loss": 0.5418789982795715, |
| "eval_runtime": 13.3324, |
| "eval_samples_per_second": 281.269, |
| "eval_steps_per_second": 4.425, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 140.08969116210938, |
| "learning_rate": 1.7500000000000005e-05, |
| "loss": 0.4416, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8384, |
| "eval_loss": 0.5170932412147522, |
| "eval_runtime": 13.3713, |
| "eval_samples_per_second": 280.451, |
| "eval_steps_per_second": 4.412, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 113.3970718383789, |
| "learning_rate": 8.188444490835772e-06, |
| "loss": 0.4003, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8432, |
| "eval_loss": 0.5022388696670532, |
| "eval_runtime": 14.265, |
| "eval_samples_per_second": 262.881, |
| "eval_steps_per_second": 4.136, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 110.79833984375, |
| "learning_rate": 2.1107582724932087e-06, |
| "loss": 0.3913, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8445333333333334, |
| "eval_loss": 0.495644211769104, |
| "eval_runtime": 13.2817, |
| "eval_samples_per_second": 282.344, |
| "eval_steps_per_second": 4.442, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 186.59793090820312, |
| "learning_rate": 0.0, |
| "loss": 0.3834, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8474666666666667, |
| "eval_loss": 0.4916774332523346, |
| "eval_runtime": 13.5755, |
| "eval_samples_per_second": 276.233, |
| "eval_steps_per_second": 4.346, |
| "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 |
| } |
|
|
