trainer_state.json

{
  "best_metric": 0.9202666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0404/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": 6.360915660858154,
      "learning_rate": 9e-05,
      "loss": 0.6628,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9178666666666667,
      "eval_loss": 0.2721829116344452,
      "eval_runtime": 12.5931,
      "eval_samples_per_second": 297.783,
      "eval_steps_per_second": 4.685,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9845744967460632,
      "learning_rate": 9e-05,
      "loss": 0.1471,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.28605976700782776,
      "eval_runtime": 12.4885,
      "eval_samples_per_second": 300.275,
      "eval_steps_per_second": 4.724,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.0756,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9125333333333333,
      "eval_loss": 0.3181639313697815,
      "eval_runtime": 12.6562,
      "eval_samples_per_second": 296.298,
      "eval_steps_per_second": 4.662,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.09268773347139359,
      "learning_rate": 9e-05,
      "loss": 0.0522,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9184,
      "eval_loss": 0.32669752836227417,
      "eval_runtime": 12.6483,
      "eval_samples_per_second": 296.483,
      "eval_steps_per_second": 4.665,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.04,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9202666666666667,
      "eval_loss": 0.3123162090778351,
      "eval_runtime": 12.579,
      "eval_samples_per_second": 298.115,
      "eval_steps_per_second": 4.69,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}