trainer_state.json

{
  "best_metric": 0.916,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0243/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": Infinity,
      "learning_rate": 8.55698766265509e-05,
      "loss": 1.5752,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8194666666666667,
      "eval_loss": 0.6849744915962219,
      "eval_runtime": 6.7488,
      "eval_samples_per_second": 555.658,
      "eval_steps_per_second": 8.742,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.31517728663805e-05,
      "loss": 0.4999,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8517333333333333,
      "eval_loss": 0.5312510132789612,
      "eval_runtime": 8.3351,
      "eval_samples_per_second": 449.906,
      "eval_steps_per_second": 7.079,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 5.5190743462682014e-05,
      "loss": 0.2922,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.4102298319339752,
      "eval_runtime": 7.2233,
      "eval_samples_per_second": 519.151,
      "eval_steps_per_second": 8.168,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 47.75657653808594,
      "learning_rate": 3.516402310353216e-05,
      "loss": 0.1584,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8933333333333333,
      "eval_loss": 0.3752487003803253,
      "eval_runtime": 6.8307,
      "eval_samples_per_second": 548.994,
      "eval_steps_per_second": 8.638,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5082464814186096,
      "learning_rate": 1.708543859906803e-05,
      "loss": 0.084,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9042666666666667,
      "eval_loss": 0.34613215923309326,
      "eval_runtime": 7.4443,
      "eval_samples_per_second": 503.74,
      "eval_steps_per_second": 7.926,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0815901905298233,
      "learning_rate": 4.53567532968734e-06,
      "loss": 0.037,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.916,
      "eval_loss": 0.30998340249061584,
      "eval_runtime": 6.7879,
      "eval_samples_per_second": 552.454,
      "eval_steps_per_second": 8.692,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}