trainer_state.json

{
  "best_metric": 0.5978666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0994/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": 0.0004003003003003003,
      "loss": 3.1489,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.26586666666666664,
      "eval_loss": 2.7554843425750732,
      "eval_runtime": 6.4954,
      "eval_samples_per_second": 577.336,
      "eval_steps_per_second": 9.083,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003006006006006006,
      "loss": 2.4207,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.40026666666666666,
      "eval_loss": 2.163839340209961,
      "eval_runtime": 6.5371,
      "eval_samples_per_second": 573.651,
      "eval_steps_per_second": 9.025,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.577508926391602,
      "learning_rate": 0.00020060060060060058,
      "loss": 1.8475,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.492,
      "eval_loss": 1.8106557130813599,
      "eval_runtime": 6.5739,
      "eval_samples_per_second": 570.44,
      "eval_steps_per_second": 8.975,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 20.615957260131836,
      "learning_rate": 0.0001006006006006006,
      "loss": 1.3093,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5624,
      "eval_loss": 1.5229171514511108,
      "eval_runtime": 6.5033,
      "eval_samples_per_second": 576.63,
      "eval_steps_per_second": 9.072,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 9.00900900900901e-07,
      "loss": 0.7432,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5978666666666667,
      "eval_loss": 1.4133386611938477,
      "eval_runtime": 8.5809,
      "eval_samples_per_second": 437.018,
      "eval_steps_per_second": 6.876,
      "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
}