trainer_state.json

{
  "best_metric": 0.9061333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0432/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": Infinity,
      "learning_rate": 8.753753753753754e-05,
      "loss": 1.6476,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7898666666666667,
      "eval_loss": 0.7555978894233704,
      "eval_runtime": 13.2542,
      "eval_samples_per_second": 282.93,
      "eval_steps_per_second": 4.451,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.507507507507507e-05,
      "loss": 0.5141,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8504,
      "eval_loss": 0.5017176270484924,
      "eval_runtime": 13.2674,
      "eval_samples_per_second": 282.648,
      "eval_steps_per_second": 4.447,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8462406396865845,
      "learning_rate": 6.257507507507508e-05,
      "loss": 0.2777,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4625999927520752,
      "eval_runtime": 13.3001,
      "eval_samples_per_second": 281.952,
      "eval_steps_per_second": 4.436,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 5.011261261261262e-05,
      "loss": 0.158,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.43361231684684753,
      "eval_runtime": 13.7122,
      "eval_samples_per_second": 273.479,
      "eval_steps_per_second": 4.303,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.626383066177368,
      "learning_rate": 3.7612612612612614e-05,
      "loss": 0.0807,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8890666666666667,
      "eval_loss": 0.41183263063430786,
      "eval_runtime": 13.3753,
      "eval_samples_per_second": 280.367,
      "eval_steps_per_second": 4.411,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 71.33343505859375,
      "learning_rate": 2.5112612612612612e-05,
      "loss": 0.0468,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.4409084618091583,
      "eval_runtime": 13.3037,
      "eval_samples_per_second": 281.877,
      "eval_steps_per_second": 4.435,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.6109321117401123,
      "learning_rate": 1.2612612612612611e-05,
      "loss": 0.0169,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9061333333333333,
      "eval_loss": 0.3964722752571106,
      "eval_runtime": 13.5289,
      "eval_samples_per_second": 277.184,
      "eval_steps_per_second": 4.361,
      "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}