trainer_state.json

{
  "best_metric": 0.8888,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0606/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 124.18888092041016,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.8223,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7173333333333334,
      "eval_loss": 1.0415351390838623,
      "eval_runtime": 9.509,
      "eval_samples_per_second": 394.364,
      "eval_steps_per_second": 6.205,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 47.08750915527344,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.7521,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5404305458068848,
      "eval_runtime": 9.668,
      "eval_samples_per_second": 387.877,
      "eval_steps_per_second": 6.103,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 95.7308349609375,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.4542,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.4508105516433716,
      "eval_runtime": 9.2586,
      "eval_samples_per_second": 405.028,
      "eval_steps_per_second": 6.372,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 79.94267272949219,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.3325,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.41041216254234314,
      "eval_runtime": 9.5799,
      "eval_samples_per_second": 391.443,
      "eval_steps_per_second": 6.159,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 83.37353515625,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.2591,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.40017572045326233,
      "eval_runtime": 9.2755,
      "eval_samples_per_second": 404.289,
      "eval_steps_per_second": 6.361,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 115.26484680175781,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.2058,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8832,
      "eval_loss": 0.3955419361591339,
      "eval_runtime": 9.611,
      "eval_samples_per_second": 390.178,
      "eval_steps_per_second": 6.139,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 76.40304565429688,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.1777,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.3982413411140442,
      "eval_runtime": 9.5173,
      "eval_samples_per_second": 394.018,
      "eval_steps_per_second": 6.199,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 72.4686508178711,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.1588,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8888,
      "eval_loss": 0.37332624197006226,
      "eval_runtime": 9.516,
      "eval_samples_per_second": 394.072,
      "eval_steps_per_second": 6.2,
      "step": 2664
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}