| { |
| "best_metric": 0.8376, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0792/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": 279.99774169921875, |
| "learning_rate": 2.9095389311788626e-05, |
| "loss": 3.8227, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.4504, |
| "eval_loss": 3.7041821479797363, |
| "eval_runtime": 8.3567, |
| "eval_samples_per_second": 448.74, |
| "eval_steps_per_second": 7.06, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 137.0520782470703, |
| "learning_rate": 2.649066664678467e-05, |
| "loss": 2.9449, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6245333333333334, |
| "eval_loss": 2.086500644683838, |
| "eval_runtime": 7.8051, |
| "eval_samples_per_second": 480.455, |
| "eval_steps_per_second": 7.559, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 228.76365661621094, |
| "learning_rate": 2.25e-05, |
| "loss": 1.4457, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7648, |
| "eval_loss": 1.0245957374572754, |
| "eval_runtime": 7.3817, |
| "eval_samples_per_second": 508.01, |
| "eval_steps_per_second": 7.993, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 213.8609161376953, |
| "learning_rate": 1.760472266500396e-05, |
| "loss": 0.8729, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8098666666666666, |
| "eval_loss": 0.7323598265647888, |
| "eval_runtime": 8.4215, |
| "eval_samples_per_second": 445.289, |
| "eval_steps_per_second": 7.006, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 225.816650390625, |
| "learning_rate": 1.2395277334996045e-05, |
| "loss": 0.679, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8189333333333333, |
| "eval_loss": 0.6465139389038086, |
| "eval_runtime": 7.3525, |
| "eval_samples_per_second": 510.029, |
| "eval_steps_per_second": 8.024, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 150.20758056640625, |
| "learning_rate": 7.500000000000004e-06, |
| "loss": 0.5952, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.8322666666666667, |
| "eval_loss": 0.5934996008872986, |
| "eval_runtime": 7.3181, |
| "eval_samples_per_second": 512.431, |
| "eval_steps_per_second": 8.062, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 109.96058654785156, |
| "learning_rate": 3.5093333532153316e-06, |
| "loss": 0.5438, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8333333333333334, |
| "eval_loss": 0.5958553552627563, |
| "eval_runtime": 7.2961, |
| "eval_samples_per_second": 513.971, |
| "eval_steps_per_second": 8.086, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 230.1630401611328, |
| "learning_rate": 9.046106882113753e-07, |
| "loss": 0.5218, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.8357333333333333, |
| "eval_loss": 0.5596023201942444, |
| "eval_runtime": 7.4148, |
| "eval_samples_per_second": 505.747, |
| "eval_steps_per_second": 7.957, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 131.81619262695312, |
| "learning_rate": 0.0, |
| "loss": 0.5175, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8376, |
| "eval_loss": 0.5719669461250305, |
| "eval_runtime": 7.2357, |
| "eval_samples_per_second": 518.264, |
| "eval_steps_per_second": 8.154, |
| "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 |
| } |
|
|
