trainer_state.json

{
  "best_metric": 0.8954666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0144/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 304.1365051269531,
      "learning_rate": 9e-05,
      "loss": 2.804,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6909333333333333,
      "eval_loss": 1.1068755388259888,
      "eval_runtime": 8.1498,
      "eval_samples_per_second": 460.136,
      "eval_steps_per_second": 7.239,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 46.08738708496094,
      "learning_rate": 9e-05,
      "loss": 0.6792,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.4752102196216583,
      "eval_runtime": 9.1896,
      "eval_samples_per_second": 408.071,
      "eval_steps_per_second": 6.42,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 124.31500244140625,
      "learning_rate": 9e-05,
      "loss": 0.4222,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8714666666666666,
      "eval_loss": 0.4164305329322815,
      "eval_runtime": 8.0368,
      "eval_samples_per_second": 466.601,
      "eval_steps_per_second": 7.341,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 108.91909790039062,
      "learning_rate": 9e-05,
      "loss": 0.3,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.36473527550697327,
      "eval_runtime": 8.461,
      "eval_samples_per_second": 443.212,
      "eval_steps_per_second": 6.973,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 78.62613677978516,
      "learning_rate": 9e-05,
      "loss": 0.225,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8906666666666667,
      "eval_loss": 0.371341735124588,
      "eval_runtime": 8.0968,
      "eval_samples_per_second": 463.148,
      "eval_steps_per_second": 7.287,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 102.33200073242188,
      "learning_rate": 9e-05,
      "loss": 0.1744,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.39523717761039734,
      "eval_runtime": 8.1785,
      "eval_samples_per_second": 458.518,
      "eval_steps_per_second": 7.214,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 78.21224212646484,
      "learning_rate": 9e-05,
      "loss": 0.1403,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8954666666666666,
      "eval_loss": 0.36337125301361084,
      "eval_runtime": 8.0689,
      "eval_samples_per_second": 464.75,
      "eval_steps_per_second": 7.312,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}