trainer_state.json

{
  "best_metric": 0.9138666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0555/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": 39.76778030395508,
      "learning_rate": 0.000288581929876693,
      "loss": 1.4114,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8685333333333334,
      "eval_loss": 0.43541938066482544,
      "eval_runtime": 7.3232,
      "eval_samples_per_second": 512.071,
      "eval_steps_per_second": 8.057,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 46.03316879272461,
      "learning_rate": 0.00025606601717798207,
      "loss": 0.3431,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.3915860950946808,
      "eval_runtime": 7.1631,
      "eval_samples_per_second": 523.513,
      "eval_steps_per_second": 8.237,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 55.15711212158203,
      "learning_rate": 0.00020740251485476345,
      "loss": 0.1769,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8938666666666667,
      "eval_loss": 0.38031789660453796,
      "eval_runtime": 8.3955,
      "eval_samples_per_second": 446.67,
      "eval_steps_per_second": 7.028,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 50.75686264038086,
      "learning_rate": 0.00015,
      "loss": 0.0946,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9042666666666667,
      "eval_loss": 0.37791457772254944,
      "eval_runtime": 7.2535,
      "eval_samples_per_second": 516.992,
      "eval_steps_per_second": 8.134,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 39.59269332885742,
      "learning_rate": 9.259748514523653e-05,
      "loss": 0.0551,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9128,
      "eval_loss": 0.3770175278186798,
      "eval_runtime": 7.3274,
      "eval_samples_per_second": 511.778,
      "eval_steps_per_second": 8.052,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 29.820220947265625,
      "learning_rate": 4.3933982822017876e-05,
      "loss": 0.0246,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9138666666666667,
      "eval_loss": 0.37077876925468445,
      "eval_runtime": 7.1849,
      "eval_samples_per_second": 521.931,
      "eval_steps_per_second": 8.212,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}