trainer_state.json

{
  "best_metric": 0.9010666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0781/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": 84.94281005859375,
      "learning_rate": 9e-05,
      "loss": 2.9546,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6818666666666666,
      "eval_loss": 1.2200342416763306,
      "eval_runtime": 15.4776,
      "eval_samples_per_second": 242.286,
      "eval_steps_per_second": 3.812,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 51.79869842529297,
      "learning_rate": 9e-05,
      "loss": 0.7779,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8482666666666666,
      "eval_loss": 0.49968305230140686,
      "eval_runtime": 15.5256,
      "eval_samples_per_second": 241.536,
      "eval_steps_per_second": 3.8,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 79.82996368408203,
      "learning_rate": 9e-05,
      "loss": 0.4642,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.3955487012863159,
      "eval_runtime": 15.3406,
      "eval_samples_per_second": 244.45,
      "eval_steps_per_second": 3.846,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 76.92202758789062,
      "learning_rate": 9e-05,
      "loss": 0.3345,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.3851511776447296,
      "eval_runtime": 15.5244,
      "eval_samples_per_second": 241.556,
      "eval_steps_per_second": 3.8,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 39.38869857788086,
      "learning_rate": 9e-05,
      "loss": 0.26,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8946666666666667,
      "eval_loss": 0.3458930552005768,
      "eval_runtime": 15.1889,
      "eval_samples_per_second": 246.891,
      "eval_steps_per_second": 3.884,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 58.503822326660156,
      "learning_rate": 9e-05,
      "loss": 0.1975,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8856,
      "eval_loss": 0.3628941774368286,
      "eval_runtime": 14.7247,
      "eval_samples_per_second": 254.674,
      "eval_steps_per_second": 4.007,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 53.8726692199707,
      "learning_rate": 9e-05,
      "loss": 0.1651,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.3518194556236267,
      "eval_runtime": 14.6229,
      "eval_samples_per_second": 256.447,
      "eval_steps_per_second": 4.035,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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
}