| { |
| "best_metric": 0.5994666666666667, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0844/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": 79.6295166015625, |
| "learning_rate": 2.25e-05, |
| "loss": 3.8478, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.3848, |
| "eval_loss": 3.7185258865356445, |
| "eval_runtime": 8.5262, |
| "eval_samples_per_second": 439.821, |
| "eval_steps_per_second": 6.92, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 58.328369140625, |
| "learning_rate": 1.5e-05, |
| "loss": 3.4138, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.5384, |
| "eval_loss": 3.1679365634918213, |
| "eval_runtime": 8.5142, |
| "eval_samples_per_second": 440.441, |
| "eval_steps_per_second": 6.93, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 241.73202514648438, |
| "learning_rate": 7.5e-06, |
| "loss": 2.8013, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.5693333333333334, |
| "eval_loss": 2.5446224212646484, |
| "eval_runtime": 8.4376, |
| "eval_samples_per_second": 444.439, |
| "eval_steps_per_second": 6.993, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 157.76280212402344, |
| "learning_rate": 0.0, |
| "loss": 2.3922, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.5994666666666667, |
| "eval_loss": 2.2678139209747314, |
| "eval_runtime": 8.6433, |
| "eval_samples_per_second": 433.863, |
| "eval_steps_per_second": 6.826, |
| "step": 1332 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1332, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "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": 3.2705716958208e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
