trainer_state.json

{
  "best_metric": 0.9,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0002/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 8.753753753753754e-05,
      "loss": 1.4401,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8130666666666667,
      "eval_loss": 0.6409623622894287,
      "eval_runtime": 6.4991,
      "eval_samples_per_second": 577.0,
      "eval_steps_per_second": 9.078,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.507507507507507e-05,
      "loss": 0.4216,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.4884535074234009,
      "eval_runtime": 6.5849,
      "eval_samples_per_second": 569.486,
      "eval_steps_per_second": 8.96,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.767101764678955,
      "learning_rate": 6.257507507507508e-05,
      "loss": 0.2073,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.49698111414909363,
      "eval_runtime": 7.1137,
      "eval_samples_per_second": 527.154,
      "eval_steps_per_second": 8.294,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6268244981765747,
      "learning_rate": 5.0075075075075074e-05,
      "loss": 0.1047,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.45239993929862976,
      "eval_runtime": 6.6086,
      "eval_samples_per_second": 567.442,
      "eval_steps_per_second": 8.928,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.11518193781375885,
      "learning_rate": 3.7575075075075075e-05,
      "loss": 0.0492,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.45672422647476196,
      "eval_runtime": 6.985,
      "eval_samples_per_second": 536.863,
      "eval_steps_per_second": 8.447,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.1365586668252945,
      "learning_rate": 2.5075075075075076e-05,
      "loss": 0.0179,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8965333333333333,
      "eval_loss": 0.43156230449676514,
      "eval_runtime": 8.7116,
      "eval_samples_per_second": 430.461,
      "eval_steps_per_second": 6.773,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.020841768011450768,
      "learning_rate": 1.2575075075075077e-05,
      "loss": 0.007,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.4165920913219452,
      "eval_runtime": 6.6709,
      "eval_samples_per_second": 562.14,
      "eval_steps_per_second": 8.844,
      "step": 2331
    }
  ],
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}