trainer_state.json

{
  "best_metric": 0.9378666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0951/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": 2.6133041381835938,
      "learning_rate": 5e-05,
      "loss": 0.8514,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9309333333333333,
      "eval_loss": 0.2516824007034302,
      "eval_runtime": 7.1185,
      "eval_samples_per_second": 526.793,
      "eval_steps_per_second": 8.288,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.14062505960464478,
      "learning_rate": 5e-05,
      "loss": 0.136,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9344,
      "eval_loss": 0.2230706363916397,
      "eval_runtime": 7.274,
      "eval_samples_per_second": 515.532,
      "eval_steps_per_second": 8.111,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.296933174133301,
      "learning_rate": 5e-05,
      "loss": 0.07,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9293333333333333,
      "eval_loss": 0.24532681703567505,
      "eval_runtime": 7.1827,
      "eval_samples_per_second": 522.087,
      "eval_steps_per_second": 8.214,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.10744459927082062,
      "learning_rate": 5e-05,
      "loss": 0.0454,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9314666666666667,
      "eval_loss": 0.2619159519672394,
      "eval_runtime": 7.1885,
      "eval_samples_per_second": 521.666,
      "eval_steps_per_second": 8.208,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.0356,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9336,
      "eval_loss": 0.2516113519668579,
      "eval_runtime": 7.2703,
      "eval_samples_per_second": 515.8,
      "eval_steps_per_second": 8.115,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.5025848746299744,
      "learning_rate": 5e-05,
      "loss": 0.0202,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9378666666666666,
      "eval_loss": 0.2641214430332184,
      "eval_runtime": 7.1756,
      "eval_samples_per_second": 522.604,
      "eval_steps_per_second": 8.222,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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
}