trainer_state.json

{
  "best_metric": 0.8586666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0068/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": 130.54525756835938,
      "learning_rate": 5e-05,
      "loss": 3.6862,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4882666666666667,
      "eval_loss": 3.050694227218628,
      "eval_runtime": 7.5189,
      "eval_samples_per_second": 498.741,
      "eval_steps_per_second": 7.847,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 147.956298828125,
      "learning_rate": 5e-05,
      "loss": 1.8216,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7514666666666666,
      "eval_loss": 1.0035070180892944,
      "eval_runtime": 7.4707,
      "eval_samples_per_second": 501.961,
      "eval_steps_per_second": 7.898,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 112.0032730102539,
      "learning_rate": 5e-05,
      "loss": 0.8,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.6631394624710083,
      "eval_runtime": 7.7219,
      "eval_samples_per_second": 485.63,
      "eval_steps_per_second": 7.641,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 405.89263916015625,
      "learning_rate": 5e-05,
      "loss": 0.5457,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8378666666666666,
      "eval_loss": 0.5548146367073059,
      "eval_runtime": 7.5821,
      "eval_samples_per_second": 494.585,
      "eval_steps_per_second": 7.781,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 125.52337646484375,
      "learning_rate": 5e-05,
      "loss": 0.438,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8408,
      "eval_loss": 0.5259442329406738,
      "eval_runtime": 7.5998,
      "eval_samples_per_second": 493.432,
      "eval_steps_per_second": 7.763,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 108.82524108886719,
      "learning_rate": 5e-05,
      "loss": 0.3544,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.46336066722869873,
      "eval_runtime": 7.6629,
      "eval_samples_per_second": 489.372,
      "eval_steps_per_second": 7.699,
      "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
}