trainer_state.json

{
  "best_metric": 0.8808,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0344/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": 2.572715572715573e-05,
      "loss": 2.2345,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7592,
      "eval_loss": 1.065805435180664,
      "eval_runtime": 5.6797,
      "eval_samples_per_second": 660.241,
      "eval_steps_per_second": 10.388,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.930996894836426,
      "learning_rate": 2.1441441441441442e-05,
      "loss": 0.7356,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.603091299533844,
      "eval_runtime": 5.6695,
      "eval_samples_per_second": 661.436,
      "eval_steps_per_second": 10.407,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.765143632888794,
      "learning_rate": 1.715572715572716e-05,
      "loss": 0.3789,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8584,
      "eval_loss": 0.5110082626342773,
      "eval_runtime": 6.6051,
      "eval_samples_per_second": 567.745,
      "eval_steps_per_second": 8.933,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7575669884681702,
      "learning_rate": 1.287001287001287e-05,
      "loss": 0.2178,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.4275243580341339,
      "eval_runtime": 7.8717,
      "eval_samples_per_second": 476.388,
      "eval_steps_per_second": 7.495,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.868784189224243,
      "learning_rate": 8.584298584298585e-06,
      "loss": 0.1268,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.418069452047348,
      "eval_runtime": 5.7164,
      "eval_samples_per_second": 656.004,
      "eval_steps_per_second": 10.321,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.8718852996826172,
      "learning_rate": 4.298584298584299e-06,
      "loss": 0.0757,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.4062211215496063,
      "eval_runtime": 5.7778,
      "eval_samples_per_second": 649.039,
      "eval_steps_per_second": 10.212,
      "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
}