trainer_state.json

{
  "best_metric": 0.8850666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0343/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": 175.85565185546875,
      "learning_rate": 7e-05,
      "loss": 3.4229,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5584,
      "eval_loss": 2.038343667984009,
      "eval_runtime": 15.2282,
      "eval_samples_per_second": 246.254,
      "eval_steps_per_second": 3.874,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 132.67327880859375,
      "learning_rate": 7e-05,
      "loss": 1.1419,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8066666666666666,
      "eval_loss": 0.6681300401687622,
      "eval_runtime": 15.2577,
      "eval_samples_per_second": 245.778,
      "eval_steps_per_second": 3.867,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 74.82705688476562,
      "learning_rate": 7e-05,
      "loss": 0.5915,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8410666666666666,
      "eval_loss": 0.5087078213691711,
      "eval_runtime": 15.2393,
      "eval_samples_per_second": 246.074,
      "eval_steps_per_second": 3.872,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 191.7124481201172,
      "learning_rate": 7e-05,
      "loss": 0.4231,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8552,
      "eval_loss": 0.4694192707538605,
      "eval_runtime": 15.1446,
      "eval_samples_per_second": 247.613,
      "eval_steps_per_second": 3.896,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 71.66416931152344,
      "learning_rate": 7e-05,
      "loss": 0.3387,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.4305684268474579,
      "eval_runtime": 15.2152,
      "eval_samples_per_second": 246.463,
      "eval_steps_per_second": 3.878,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 131.28231811523438,
      "learning_rate": 7e-05,
      "loss": 0.2666,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.39838552474975586,
      "eval_runtime": 15.1641,
      "eval_samples_per_second": 247.295,
      "eval_steps_per_second": 3.891,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 137.51133728027344,
      "learning_rate": 7e-05,
      "loss": 0.2201,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.3964410722255707,
      "eval_runtime": 15.1992,
      "eval_samples_per_second": 246.723,
      "eval_steps_per_second": 3.882,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}