| { |
| "best_metric": 0.9130666666666667, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0787/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": 31.821407318115234, |
| "learning_rate": 0.0004849231551964771, |
| "loss": 1.3563, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8357333333333333, |
| "eval_loss": 0.5577715635299683, |
| "eval_runtime": 15.3678, |
| "eval_samples_per_second": 244.017, |
| "eval_steps_per_second": 3.839, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 34.75889587402344, |
| "learning_rate": 0.0004415111107797445, |
| "loss": 0.4236, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8613333333333333, |
| "eval_loss": 0.4780004620552063, |
| "eval_runtime": 15.3214, |
| "eval_samples_per_second": 244.755, |
| "eval_steps_per_second": 3.851, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 13.803163528442383, |
| "learning_rate": 0.000375, |
| "loss": 0.2538, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8845333333333333, |
| "eval_loss": 0.45998311042785645, |
| "eval_runtime": 15.3158, |
| "eval_samples_per_second": 244.846, |
| "eval_steps_per_second": 3.852, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 26.248886108398438, |
| "learning_rate": 0.00029341204441673266, |
| "loss": 0.1492, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8834666666666666, |
| "eval_loss": 0.47012877464294434, |
| "eval_runtime": 15.3237, |
| "eval_samples_per_second": 244.718, |
| "eval_steps_per_second": 3.85, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 33.17631149291992, |
| "learning_rate": 0.00020658795558326743, |
| "loss": 0.0915, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8978666666666667, |
| "eval_loss": 0.41796088218688965, |
| "eval_runtime": 15.7468, |
| "eval_samples_per_second": 238.143, |
| "eval_steps_per_second": 3.747, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 21.727567672729492, |
| "learning_rate": 0.00012500000000000006, |
| "loss": 0.0513, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.9072, |
| "eval_loss": 0.40345847606658936, |
| "eval_runtime": 15.2861, |
| "eval_samples_per_second": 245.32, |
| "eval_steps_per_second": 3.86, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 81.20289611816406, |
| "learning_rate": 5.848888922025553e-05, |
| "loss": 0.0367, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.9088, |
| "eval_loss": 0.4345357120037079, |
| "eval_runtime": 15.9591, |
| "eval_samples_per_second": 234.976, |
| "eval_steps_per_second": 3.697, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 22.956645965576172, |
| "learning_rate": 1.5076844803522921e-05, |
| "loss": 0.0198, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.9104, |
| "eval_loss": 0.4132912755012512, |
| "eval_runtime": 15.3487, |
| "eval_samples_per_second": 244.32, |
| "eval_steps_per_second": 3.844, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 36.78782272338867, |
| "learning_rate": 0.0, |
| "loss": 0.0167, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.9130666666666667, |
| "eval_loss": 0.40943896770477295, |
| "eval_runtime": 15.2069, |
| "eval_samples_per_second": 246.598, |
| "eval_steps_per_second": 3.88, |
| "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 |
| } |
|
|
