trainer_state.json

{
  "best_metric": 0.9549333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0220/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": 3.7468011379241943,
      "learning_rate": 3e-05,
      "loss": 1.1195,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9424,
      "eval_loss": 0.2539292871952057,
      "eval_runtime": 5.6536,
      "eval_samples_per_second": 663.29,
      "eval_steps_per_second": 10.436,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.390302836894989,
      "learning_rate": 3e-05,
      "loss": 0.1365,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9421333333333334,
      "eval_loss": 0.20439672470092773,
      "eval_runtime": 6.2365,
      "eval_samples_per_second": 601.294,
      "eval_steps_per_second": 9.46,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 8.59044361114502,
      "learning_rate": 3e-05,
      "loss": 0.0496,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9485333333333333,
      "eval_loss": 0.19204457104206085,
      "eval_runtime": 5.8129,
      "eval_samples_per_second": 645.113,
      "eval_steps_per_second": 10.15,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.06902637332677841,
      "learning_rate": 3e-05,
      "loss": 0.0188,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9512,
      "eval_loss": 0.19098326563835144,
      "eval_runtime": 5.3442,
      "eval_samples_per_second": 701.7,
      "eval_steps_per_second": 11.04,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.02616540528833866,
      "learning_rate": 3e-05,
      "loss": 0.0081,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9536,
      "eval_loss": 0.20332421362400055,
      "eval_runtime": 5.8163,
      "eval_samples_per_second": 644.736,
      "eval_steps_per_second": 10.144,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05017227306962013,
      "learning_rate": 3e-05,
      "loss": 0.0055,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9549333333333333,
      "eval_loss": 0.2062648981809616,
      "eval_runtime": 6.6637,
      "eval_samples_per_second": 562.751,
      "eval_steps_per_second": 8.854,
      "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}