trainer_state.json

{
  "best_metric": 0.8893333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0440/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 2.626126126126126e-05,
      "loss": 2.1149,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7922666666666667,
      "eval_loss": 0.938521683216095,
      "eval_runtime": 12.2938,
      "eval_samples_per_second": 305.032,
      "eval_steps_per_second": 4.799,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 19.071985244750977,
      "learning_rate": 2.251126126126126e-05,
      "loss": 0.6478,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5619415640830994,
      "eval_runtime": 10.2572,
      "eval_samples_per_second": 365.598,
      "eval_steps_per_second": 5.752,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.385346412658691,
      "learning_rate": 1.876126126126126e-05,
      "loss": 0.3419,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4606999456882477,
      "eval_runtime": 10.236,
      "eval_samples_per_second": 366.353,
      "eval_steps_per_second": 5.764,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9545121192932129,
      "learning_rate": 1.5011261261261261e-05,
      "loss": 0.1962,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.42097240686416626,
      "eval_runtime": 10.0175,
      "eval_samples_per_second": 374.346,
      "eval_steps_per_second": 5.89,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.220184087753296,
      "learning_rate": 1.1261261261261261e-05,
      "loss": 0.1055,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.41826948523521423,
      "eval_runtime": 10.5949,
      "eval_samples_per_second": 353.945,
      "eval_steps_per_second": 5.569,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.4259107112884521,
      "learning_rate": 7.511261261261261e-06,
      "loss": 0.0547,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8832,
      "eval_loss": 0.41788020730018616,
      "eval_runtime": 10.2333,
      "eval_samples_per_second": 366.451,
      "eval_steps_per_second": 5.766,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.7584707140922546,
      "learning_rate": 3.7612612612612617e-06,
      "loss": 0.026,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.884,
      "eval_loss": 0.42027005553245544,
      "eval_runtime": 10.5515,
      "eval_samples_per_second": 355.399,
      "eval_steps_per_second": 5.592,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.2030571699142456,
      "learning_rate": 1.1261261261261262e-08,
      "loss": 0.0159,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.42002514004707336,
      "eval_runtime": 10.1737,
      "eval_samples_per_second": 368.596,
      "eval_steps_per_second": 5.799,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}