trainer_state.json

{
  "best_metric": 0.9429333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0660/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": 1.9028676748275757,
      "learning_rate": 6.125e-05,
      "loss": 0.6884,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9242666666666667,
      "eval_loss": 0.28298017382621765,
      "eval_runtime": 11.2817,
      "eval_samples_per_second": 332.396,
      "eval_steps_per_second": 5.23,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.14824776351451874,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.1158,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9338666666666666,
      "eval_loss": 0.24205492436885834,
      "eval_runtime": 11.9301,
      "eval_samples_per_second": 314.33,
      "eval_steps_per_second": 4.945,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6641613245010376,
      "learning_rate": 4.374999999999999e-05,
      "loss": 0.0425,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9328,
      "eval_loss": 0.2528708577156067,
      "eval_runtime": 11.857,
      "eval_samples_per_second": 316.269,
      "eval_steps_per_second": 4.976,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05375959351658821,
      "learning_rate": 3.5e-05,
      "loss": 0.0184,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9381333333333334,
      "eval_loss": 0.2356237918138504,
      "eval_runtime": 11.474,
      "eval_samples_per_second": 326.825,
      "eval_steps_per_second": 5.142,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03574265539646149,
      "learning_rate": 2.6249999999999998e-05,
      "loss": 0.0084,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9408,
      "eval_loss": 0.2524432837963104,
      "eval_runtime": 11.4852,
      "eval_samples_per_second": 326.507,
      "eval_steps_per_second": 5.137,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.04126298055052757,
      "learning_rate": 1.75e-05,
      "loss": 0.0037,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9429333333333333,
      "eval_loss": 0.23425869643688202,
      "eval_runtime": 11.2158,
      "eval_samples_per_second": 334.35,
      "eval_steps_per_second": 5.26,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}