trainer_state.json

{
  "best_metric": 0.9565333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0265/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.5619161128997803,
      "learning_rate": 8.140576474687264e-05,
      "loss": 0.5811,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9426666666666667,
      "eval_loss": 0.22165745496749878,
      "eval_runtime": 5.8316,
      "eval_samples_per_second": 643.053,
      "eval_steps_per_second": 10.117,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.03518875688314438,
      "learning_rate": 5.890576474687264e-05,
      "loss": 0.0841,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9368,
      "eval_loss": 0.22407181560993195,
      "eval_runtime": 5.9801,
      "eval_samples_per_second": 627.082,
      "eval_steps_per_second": 9.866,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 10.616680145263672,
      "learning_rate": 3.1094235253127374e-05,
      "loss": 0.0231,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9525333333333333,
      "eval_loss": 0.19126087427139282,
      "eval_runtime": 6.1809,
      "eval_samples_per_second": 606.704,
      "eval_steps_per_second": 9.545,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.34493646025657654,
      "learning_rate": 8.59423525312737e-06,
      "loss": 0.0056,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.956,
      "eval_loss": 0.17310093343257904,
      "eval_runtime": 5.9427,
      "eval_samples_per_second": 631.024,
      "eval_steps_per_second": 9.928,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03603304922580719,
      "learning_rate": 0.0,
      "loss": 0.0025,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9565333333333333,
      "eval_loss": 0.17338140308856964,
      "eval_runtime": 6.1597,
      "eval_samples_per_second": 608.796,
      "eval_steps_per_second": 9.578,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}