trainer_state.json

{
  "best_metric": 0.8936,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0813/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": 64.3273696899414,
      "learning_rate": 0.0003,
      "loss": 1.4719,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8410666666666666,
      "eval_loss": 0.5281460285186768,
      "eval_runtime": 12.6828,
      "eval_samples_per_second": 295.676,
      "eval_steps_per_second": 4.652,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 38.49894714355469,
      "learning_rate": 0.0003,
      "loss": 0.4261,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8725333333333334,
      "eval_loss": 0.4123291075229645,
      "eval_runtime": 12.7538,
      "eval_samples_per_second": 294.03,
      "eval_steps_per_second": 4.626,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 54.19624328613281,
      "learning_rate": 0.0003,
      "loss": 0.2462,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.4482128620147705,
      "eval_runtime": 12.6132,
      "eval_samples_per_second": 297.309,
      "eval_steps_per_second": 4.678,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 43.20565414428711,
      "learning_rate": 0.0003,
      "loss": 0.1681,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.4265141189098358,
      "eval_runtime": 12.8454,
      "eval_samples_per_second": 291.934,
      "eval_steps_per_second": 4.593,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 39.015960693359375,
      "learning_rate": 0.0003,
      "loss": 0.116,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8898666666666667,
      "eval_loss": 0.44538813829421997,
      "eval_runtime": 12.4705,
      "eval_samples_per_second": 300.71,
      "eval_steps_per_second": 4.731,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 55.14019012451172,
      "learning_rate": 0.0003,
      "loss": 0.0946,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.4770936071872711,
      "eval_runtime": 12.4761,
      "eval_samples_per_second": 300.574,
      "eval_steps_per_second": 4.729,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 35.80834197998047,
      "learning_rate": 0.0003,
      "loss": 0.075,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.4459027945995331,
      "eval_runtime": 12.9614,
      "eval_samples_per_second": 289.321,
      "eval_steps_per_second": 4.552,
      "step": 2331
    }
  ],
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}