trainer_state.json

{
  "best_metric": 0.8698666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0018/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": Infinity,
      "learning_rate": 2.7151869176294504e-05,
      "loss": 2.0722,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7792,
      "eval_loss": 0.9068673253059387,
      "eval_runtime": 6.4024,
      "eval_samples_per_second": 585.716,
      "eval_steps_per_second": 9.215,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.28916597366333,
      "learning_rate": 1.9662164055386567e-05,
      "loss": 0.655,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8352,
      "eval_loss": 0.5665485858917236,
      "eval_runtime": 6.3547,
      "eval_samples_per_second": 590.11,
      "eval_steps_per_second": 9.284,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.016693115234375,
      "learning_rate": 1.0391670726449015e-05,
      "loss": 0.3547,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.46642571687698364,
      "eval_runtime": 7.578,
      "eval_samples_per_second": 494.854,
      "eval_steps_per_second": 7.786,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 2.89810315724514e-06,
      "loss": 0.2135,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.42681270837783813,
      "eval_runtime": 6.8019,
      "eval_samples_per_second": 551.32,
      "eval_steps_per_second": 8.674,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.4178959131240845,
      "learning_rate": 1.0680510444061043e-10,
      "loss": 0.1524,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.41647571325302124,
      "eval_runtime": 6.1647,
      "eval_samples_per_second": 608.304,
      "eval_steps_per_second": 9.571,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}