trainer_state.json

{
  "best_metric": 0.8704,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0907/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": 2.1542859077453613,
      "learning_rate": 0.0003,
      "loss": 0.651,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.45952558517456055,
      "eval_runtime": 5.3021,
      "eval_samples_per_second": 707.271,
      "eval_steps_per_second": 11.128,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.114544630050659,
      "learning_rate": 0.0003,
      "loss": 0.2999,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4802402853965759,
      "eval_runtime": 5.3335,
      "eval_samples_per_second": 703.106,
      "eval_steps_per_second": 11.062,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.679951548576355,
      "learning_rate": 0.0003,
      "loss": 0.2074,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.4928929805755615,
      "eval_runtime": 5.267,
      "eval_samples_per_second": 711.983,
      "eval_steps_per_second": 11.202,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.046160340309143,
      "learning_rate": 0.0003,
      "loss": 0.1649,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8538666666666667,
      "eval_loss": 0.5773593187332153,
      "eval_runtime": 5.2367,
      "eval_samples_per_second": 716.097,
      "eval_steps_per_second": 11.267,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.024483617395162582,
      "learning_rate": 0.0003,
      "loss": 0.1452,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8613333333333333,
      "eval_loss": 0.5466427803039551,
      "eval_runtime": 5.7259,
      "eval_samples_per_second": 654.915,
      "eval_steps_per_second": 10.304,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.482271671295166,
      "learning_rate": 0.0003,
      "loss": 0.1142,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8704,
      "eval_loss": 0.5548595786094666,
      "eval_runtime": 5.3074,
      "eval_samples_per_second": 706.559,
      "eval_steps_per_second": 11.117,
      "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
}