trainer_state.json

{
  "best_metric": 0.9109333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0920/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": 4.8108254489432715e-05,
      "loss": 1.9474,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8093333333333333,
      "eval_loss": 0.8223518133163452,
      "eval_runtime": 4.7409,
      "eval_samples_per_second": 790.987,
      "eval_steps_per_second": 12.445,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.5536627769470215,
      "learning_rate": 4.2698504093359784e-05,
      "loss": 0.5925,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.5446879863739014,
      "eval_runtime": 4.6029,
      "eval_samples_per_second": 814.696,
      "eval_steps_per_second": 12.818,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 3.4621534620947785e-05,
      "loss": 0.3219,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.4275277256965637,
      "eval_runtime": 4.9603,
      "eval_samples_per_second": 755.999,
      "eval_steps_per_second": 11.894,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.386879563331604,
      "learning_rate": 2.5058963771413844e-05,
      "loss": 0.1807,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8869333333333334,
      "eval_loss": 0.41011252999305725,
      "eval_runtime": 4.413,
      "eval_samples_per_second": 849.758,
      "eval_steps_per_second": 13.37,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.816536545753479,
      "learning_rate": 1.5487416222190068e-05,
      "loss": 0.0953,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9008,
      "eval_loss": 0.3531179428100586,
      "eval_runtime": 4.4183,
      "eval_samples_per_second": 848.748,
      "eval_steps_per_second": 13.354,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.5180375576019287,
      "learning_rate": 7.36407332135117e-06,
      "loss": 0.048,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9109333333333334,
      "eval_loss": 0.3344250023365021,
      "eval_runtime": 4.418,
      "eval_samples_per_second": 848.798,
      "eval_steps_per_second": 13.354,
      "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
}