|
{ |
|
"best_metric": 0.8674666666666667, |
|
"best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0158/checkpoints/checkpoint-1332", |
|
"epoch": 4.0, |
|
"eval_steps": 500, |
|
"global_step": 1332, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 29.299392700195312, |
|
"learning_rate": 0.0005, |
|
"loss": 1.3433, |
|
"step": 333 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.8048, |
|
"eval_loss": 0.646480143070221, |
|
"eval_runtime": 9.2071, |
|
"eval_samples_per_second": 407.293, |
|
"eval_steps_per_second": 6.408, |
|
"step": 333 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 19.392345428466797, |
|
"learning_rate": 0.0005, |
|
"loss": 0.4646, |
|
"step": 666 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.856, |
|
"eval_loss": 0.4834669828414917, |
|
"eval_runtime": 9.2158, |
|
"eval_samples_per_second": 406.911, |
|
"eval_steps_per_second": 6.402, |
|
"step": 666 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 24.09852409362793, |
|
"learning_rate": 0.0005, |
|
"loss": 0.2995, |
|
"step": 999 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8517333333333333, |
|
"eval_loss": 0.4870469272136688, |
|
"eval_runtime": 9.1986, |
|
"eval_samples_per_second": 407.671, |
|
"eval_steps_per_second": 6.414, |
|
"step": 999 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 18.58635139465332, |
|
"learning_rate": 0.0005, |
|
"loss": 0.2108, |
|
"step": 1332 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.8674666666666667, |
|
"eval_loss": 0.4608042240142822, |
|
"eval_runtime": 9.2259, |
|
"eval_samples_per_second": 406.466, |
|
"eval_steps_per_second": 6.395, |
|
"step": 1332 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 2664, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 8, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 3.2705716958208e+18, |
|
"train_batch_size": 64, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|