| { | |
| "best_metric": 0.8858666666666667, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0558/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": 121.92276000976562, | |
| "learning_rate": 6.788924172750679e-05, | |
| "loss": 3.416, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5818666666666666, | |
| "eval_loss": 2.2763755321502686, | |
| "eval_runtime": 8.035, | |
| "eval_samples_per_second": 466.71, | |
| "eval_steps_per_second": 7.343, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 82.4437026977539, | |
| "learning_rate": 6.181155550916422e-05, | |
| "loss": 1.1406, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7973333333333333, | |
| "eval_loss": 0.6625077724456787, | |
| "eval_runtime": 8.2783, | |
| "eval_samples_per_second": 452.99, | |
| "eval_steps_per_second": 7.127, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 77.45076751708984, | |
| "learning_rate": 5.2499999999999995e-05, | |
| "loss": 0.5919, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8493333333333334, | |
| "eval_loss": 0.4886707663536072, | |
| "eval_runtime": 8.2342, | |
| "eval_samples_per_second": 455.416, | |
| "eval_steps_per_second": 7.165, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 140.9535369873047, | |
| "learning_rate": 4.1077686218342566e-05, | |
| "loss": 0.4463, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8610666666666666, | |
| "eval_loss": 0.43152865767478943, | |
| "eval_runtime": 8.1653, | |
| "eval_samples_per_second": 459.261, | |
| "eval_steps_per_second": 7.226, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 63.2565803527832, | |
| "learning_rate": 2.8922313781657438e-05, | |
| "loss": 0.3675, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8738666666666667, | |
| "eval_loss": 0.4110257625579834, | |
| "eval_runtime": 8.194, | |
| "eval_samples_per_second": 457.651, | |
| "eval_steps_per_second": 7.2, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 79.78846740722656, | |
| "learning_rate": 1.7500000000000005e-05, | |
| "loss": 0.3204, | |
| "step": 1998 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8821333333333333, | |
| "eval_loss": 0.3883652985095978, | |
| "eval_runtime": 8.5128, | |
| "eval_samples_per_second": 440.515, | |
| "eval_steps_per_second": 6.931, | |
| "step": 1998 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 288.0624084472656, | |
| "learning_rate": 8.188444490835772e-06, | |
| "loss": 0.2964, | |
| "step": 2331 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8832, | |
| "eval_loss": 0.3787687420845032, | |
| "eval_runtime": 8.2506, | |
| "eval_samples_per_second": 454.514, | |
| "eval_steps_per_second": 7.151, | |
| "step": 2331 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 131.0572052001953, | |
| "learning_rate": 2.1107582724932087e-06, | |
| "loss": 0.2814, | |
| "step": 2664 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8845333333333333, | |
| "eval_loss": 0.3687947392463684, | |
| "eval_runtime": 8.1973, | |
| "eval_samples_per_second": 457.468, | |
| "eval_steps_per_second": 7.197, | |
| "step": 2664 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 210.7683563232422, | |
| "learning_rate": 0.0, | |
| "loss": 0.2767, | |
| "step": 2997 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8858666666666667, | |
| "eval_loss": 0.3655661642551422, | |
| "eval_runtime": 8.2341, | |
| "eval_samples_per_second": 455.426, | |
| "eval_steps_per_second": 7.165, | |
| "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 | |
| } | |