trainer_state.json

{
  "best_metric": 0.9472,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0319/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": 6.190037727355957,
      "learning_rate": 6.653391037658466e-05,
      "loss": 0.7084,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9112,
      "eval_loss": 0.30569204688072205,
      "eval_runtime": 5.7313,
      "eval_samples_per_second": 654.305,
      "eval_steps_per_second": 10.294,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.05386636406183243,
      "learning_rate": 5.682214306505567e-05,
      "loss": 0.1466,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9317333333333333,
      "eval_loss": 0.23556813597679138,
      "eval_runtime": 5.7017,
      "eval_samples_per_second": 657.696,
      "eval_steps_per_second": 10.348,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4644635021686554,
      "learning_rate": 4.2788232688471e-05,
      "loss": 0.0686,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9285333333333333,
      "eval_loss": 0.2650826871395111,
      "eval_runtime": 5.8537,
      "eval_samples_per_second": 640.626,
      "eval_steps_per_second": 10.079,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.013232639990746975,
      "learning_rate": 2.7211767311528996e-05,
      "loss": 0.027,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.94,
      "eval_loss": 0.23186801373958588,
      "eval_runtime": 5.7058,
      "eval_samples_per_second": 657.231,
      "eval_steps_per_second": 10.34,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03421556204557419,
      "learning_rate": 1.3177856934944328e-05,
      "loss": 0.0109,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9416,
      "eval_loss": 0.2150125950574875,
      "eval_runtime": 5.6766,
      "eval_samples_per_second": 660.61,
      "eval_steps_per_second": 10.394,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.02700868248939514,
      "learning_rate": 3.4660896234153336e-06,
      "loss": 0.0041,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9472,
      "eval_loss": 0.21619658172130585,
      "eval_runtime": 5.724,
      "eval_samples_per_second": 655.133,
      "eval_steps_per_second": 10.307,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}