| { | |
| "best_metric": 0.6992, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0316/checkpoints/checkpoint-1665", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1665, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 69.52816009521484, | |
| "learning_rate": 2.4e-05, | |
| "loss": 3.8413, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3797333333333333, | |
| "eval_loss": 3.7060070037841797, | |
| "eval_runtime": 8.0722, | |
| "eval_samples_per_second": 464.557, | |
| "eval_steps_per_second": 7.309, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 169.8428497314453, | |
| "learning_rate": 1.8e-05, | |
| "loss": 3.3805, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5656, | |
| "eval_loss": 3.14346981048584, | |
| "eval_runtime": 7.8524, | |
| "eval_samples_per_second": 477.563, | |
| "eval_steps_per_second": 7.514, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 127.0908203125, | |
| "learning_rate": 1.2e-05, | |
| "loss": 2.4977, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6325333333333333, | |
| "eval_loss": 2.0015251636505127, | |
| "eval_runtime": 7.8645, | |
| "eval_samples_per_second": 476.827, | |
| "eval_steps_per_second": 7.502, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 164.16636657714844, | |
| "learning_rate": 6e-06, | |
| "loss": 1.7382, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6930666666666667, | |
| "eval_loss": 1.562028169631958, | |
| "eval_runtime": 7.8608, | |
| "eval_samples_per_second": 477.049, | |
| "eval_steps_per_second": 7.506, | |
| "step": 1332 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 90.84521484375, | |
| "learning_rate": 0.0, | |
| "loss": 1.4621, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6992, | |
| "eval_loss": 1.528403878211975, | |
| "eval_runtime": 7.886, | |
| "eval_samples_per_second": 475.525, | |
| "eval_steps_per_second": 7.482, | |
| "step": 1665 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.088214619776e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |