| { | |
| "best_metric": 0.8314666666666667, | |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0535/checkpoints/checkpoint-999", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 999, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 151.10972595214844, | |
| "learning_rate": 5.2499999999999995e-05, | |
| "loss": 3.2538, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6949333333333333, | |
| "eval_loss": 1.676459789276123, | |
| "eval_runtime": 8.0391, | |
| "eval_samples_per_second": 466.471, | |
| "eval_steps_per_second": 7.339, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 303.08892822265625, | |
| "learning_rate": 1.7500000000000005e-05, | |
| "loss": 1.0426, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8149333333333333, | |
| "eval_loss": 0.7393606901168823, | |
| "eval_runtime": 7.626, | |
| "eval_samples_per_second": 491.741, | |
| "eval_steps_per_second": 7.737, | |
| "step": 666 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 107.206787109375, | |
| "learning_rate": 0.0, | |
| "loss": 0.7023, | |
| "step": 999 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8314666666666667, | |
| "eval_loss": 0.6715682148933411, | |
| "eval_runtime": 7.6372, | |
| "eval_samples_per_second": 491.015, | |
| "eval_steps_per_second": 7.725, | |
| "step": 999 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 999, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 2.4529287718656e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |