trainer_state.json

{
  "best_metric": 0.8757333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0642/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 144.13609313964844,
      "learning_rate": 9e-05,
      "loss": 3.0185,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6592,
      "eval_loss": 1.340995192527771,
      "eval_runtime": 7.4741,
      "eval_samples_per_second": 501.733,
      "eval_steps_per_second": 7.894,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 90.15927124023438,
      "learning_rate": 9e-05,
      "loss": 0.8791,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.6486772894859314,
      "eval_runtime": 7.3525,
      "eval_samples_per_second": 510.029,
      "eval_steps_per_second": 8.024,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 130.17623901367188,
      "learning_rate": 9e-05,
      "loss": 0.5502,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.48257994651794434,
      "eval_runtime": 7.5103,
      "eval_samples_per_second": 499.313,
      "eval_steps_per_second": 7.856,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 48.438663482666016,
      "learning_rate": 9e-05,
      "loss": 0.4,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.4664623737335205,
      "eval_runtime": 7.8123,
      "eval_samples_per_second": 480.014,
      "eval_steps_per_second": 7.552,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 98.97806549072266,
      "learning_rate": 9e-05,
      "loss": 0.3052,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.427635133266449,
      "eval_runtime": 7.3323,
      "eval_samples_per_second": 511.438,
      "eval_steps_per_second": 8.047,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 104.73067474365234,
      "learning_rate": 9e-05,
      "loss": 0.2339,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.4297317564487457,
      "eval_runtime": 7.7114,
      "eval_samples_per_second": 486.296,
      "eval_steps_per_second": 7.651,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}