trainer_state.json

{
  "best_metric": 0.8970666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0683/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": 160.28533935546875,
      "learning_rate": 9e-05,
      "loss": 2.9311,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7157333333333333,
      "eval_loss": 1.0953688621520996,
      "eval_runtime": 8.1059,
      "eval_samples_per_second": 462.628,
      "eval_steps_per_second": 7.279,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 40.13555908203125,
      "learning_rate": 9e-05,
      "loss": 0.7322,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8397333333333333,
      "eval_loss": 0.5361349582672119,
      "eval_runtime": 8.7266,
      "eval_samples_per_second": 429.719,
      "eval_steps_per_second": 6.761,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 59.52849197387695,
      "learning_rate": 9e-05,
      "loss": 0.4547,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.44500553607940674,
      "eval_runtime": 8.4718,
      "eval_samples_per_second": 442.645,
      "eval_steps_per_second": 6.964,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 112.37177276611328,
      "learning_rate": 9e-05,
      "loss": 0.332,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.4220704138278961,
      "eval_runtime": 8.1091,
      "eval_samples_per_second": 462.444,
      "eval_steps_per_second": 7.276,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 148.5243377685547,
      "learning_rate": 9e-05,
      "loss": 0.238,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.38942885398864746,
      "eval_runtime": 8.0428,
      "eval_samples_per_second": 466.256,
      "eval_steps_per_second": 7.336,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 121.89383697509766,
      "learning_rate": 9e-05,
      "loss": 0.1848,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.39422407746315,
      "eval_runtime": 9.3214,
      "eval_samples_per_second": 402.3,
      "eval_steps_per_second": 6.33,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 95.5490493774414,
      "learning_rate": 9e-05,
      "loss": 0.1403,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.40979328751564026,
      "eval_runtime": 8.0348,
      "eval_samples_per_second": 466.72,
      "eval_steps_per_second": 7.343,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 48.0687255859375,
      "learning_rate": 9e-05,
      "loss": 0.1222,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8970666666666667,
      "eval_loss": 0.405814528465271,
      "eval_runtime": 8.0672,
      "eval_samples_per_second": 464.845,
      "eval_steps_per_second": 7.314,
      "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
}