trainer_state.json

{
  "best_metric": 0.8157333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0380/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": 0.0003,
      "loss": 1.912,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7234666666666667,
      "eval_loss": 0.8676877617835999,
      "eval_runtime": 6.4055,
      "eval_samples_per_second": 585.436,
      "eval_steps_per_second": 9.211,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.31269341707229614,
      "learning_rate": 0.0003,
      "loss": 0.6812,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7778666666666667,
      "eval_loss": 0.7085742354393005,
      "eval_runtime": 8.3736,
      "eval_samples_per_second": 447.834,
      "eval_steps_per_second": 7.046,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8178706169128418,
      "learning_rate": 0.0003,
      "loss": 0.4142,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7810666666666667,
      "eval_loss": 0.7226043343544006,
      "eval_runtime": 6.6156,
      "eval_samples_per_second": 566.843,
      "eval_steps_per_second": 8.918,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.784552574157715,
      "learning_rate": 0.0003,
      "loss": 0.2912,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8061333333333334,
      "eval_loss": 0.6577738523483276,
      "eval_runtime": 7.2206,
      "eval_samples_per_second": 519.346,
      "eval_steps_per_second": 8.171,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.5618534088134766,
      "learning_rate": 0.0003,
      "loss": 0.211,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.6421175599098206,
      "eval_runtime": 6.2541,
      "eval_samples_per_second": 599.608,
      "eval_steps_per_second": 9.434,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.41744890809059143,
      "learning_rate": 0.0003,
      "loss": 0.1705,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8018666666666666,
      "eval_loss": 0.743894636631012,
      "eval_runtime": 6.8981,
      "eval_samples_per_second": 543.625,
      "eval_steps_per_second": 8.553,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.027371443808078766,
      "learning_rate": 0.0003,
      "loss": 0.1492,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8157333333333333,
      "eval_loss": 0.7092033624649048,
      "eval_runtime": 6.6121,
      "eval_samples_per_second": 567.141,
      "eval_steps_per_second": 8.923,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}