trainer_state.json

{
  "best_metric": 0.8346666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0641/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": 99.98721313476562,
      "learning_rate": 5e-05,
      "loss": 3.7526,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4053333333333333,
      "eval_loss": 3.353086471557617,
      "eval_runtime": 13.4319,
      "eval_samples_per_second": 279.187,
      "eval_steps_per_second": 4.393,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 137.9910430908203,
      "learning_rate": 5e-05,
      "loss": 1.9694,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6792,
      "eval_loss": 1.122665286064148,
      "eval_runtime": 13.531,
      "eval_samples_per_second": 277.142,
      "eval_steps_per_second": 4.36,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 229.35693359375,
      "learning_rate": 5e-05,
      "loss": 0.9356,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7578666666666667,
      "eval_loss": 0.7803103923797607,
      "eval_runtime": 13.3829,
      "eval_samples_per_second": 280.207,
      "eval_steps_per_second": 4.409,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 74.74185943603516,
      "learning_rate": 5e-05,
      "loss": 0.6794,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7914666666666667,
      "eval_loss": 0.6452983617782593,
      "eval_runtime": 14.0865,
      "eval_samples_per_second": 266.212,
      "eval_steps_per_second": 4.188,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 172.08258056640625,
      "learning_rate": 5e-05,
      "loss": 0.5702,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8136,
      "eval_loss": 0.5748082399368286,
      "eval_runtime": 13.5584,
      "eval_samples_per_second": 276.581,
      "eval_steps_per_second": 4.352,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 98.72806549072266,
      "learning_rate": 5e-05,
      "loss": 0.4844,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8298666666666666,
      "eval_loss": 0.5223501920700073,
      "eval_runtime": 13.6079,
      "eval_samples_per_second": 275.575,
      "eval_steps_per_second": 4.336,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 149.6796417236328,
      "learning_rate": 5e-05,
      "loss": 0.4213,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8346666666666667,
      "eval_loss": 0.5169291496276855,
      "eval_runtime": 13.3602,
      "eval_samples_per_second": 280.684,
      "eval_steps_per_second": 4.416,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}