trainer_state.json

{
  "best_metric": 0.9306666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0911/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 5.046912670135498,
      "learning_rate": 0.0002625,
      "loss": 0.6394,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5181965231895447,
      "eval_runtime": 5.9643,
      "eval_samples_per_second": 628.736,
      "eval_steps_per_second": 9.892,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1149674654006958,
      "learning_rate": 0.000225,
      "loss": 0.2862,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.4562033712863922,
      "eval_runtime": 5.7629,
      "eval_samples_per_second": 650.72,
      "eval_steps_per_second": 10.238,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.05627270042896271,
      "learning_rate": 0.00018749999999999998,
      "loss": 0.1601,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9008,
      "eval_loss": 0.3592537045478821,
      "eval_runtime": 6.3371,
      "eval_samples_per_second": 591.757,
      "eval_steps_per_second": 9.31,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.12092157453298569,
      "learning_rate": 0.00015,
      "loss": 0.0945,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9066666666666666,
      "eval_loss": 0.3700014650821686,
      "eval_runtime": 5.7913,
      "eval_samples_per_second": 647.52,
      "eval_steps_per_second": 10.188,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.008114288561046124,
      "learning_rate": 0.0001125,
      "loss": 0.0512,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9114666666666666,
      "eval_loss": 0.34400323033332825,
      "eval_runtime": 6.2246,
      "eval_samples_per_second": 602.452,
      "eval_steps_per_second": 9.479,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.008365815505385399,
      "learning_rate": 7.5e-05,
      "loss": 0.0225,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9274666666666667,
      "eval_loss": 0.3100140392780304,
      "eval_runtime": 5.8104,
      "eval_samples_per_second": 645.395,
      "eval_steps_per_second": 10.154,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 3.7612612612612614e-05,
      "loss": 0.0078,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9264,
      "eval_loss": 0.334543377161026,
      "eval_runtime": 5.734,
      "eval_samples_per_second": 653.992,
      "eval_steps_per_second": 10.289,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.2952848672866821,
      "learning_rate": 1.1261261261261261e-07,
      "loss": 0.0033,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9306666666666666,
      "eval_loss": 0.30983737111091614,
      "eval_runtime": 5.8411,
      "eval_samples_per_second": 642.004,
      "eval_steps_per_second": 10.101,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}