trainer_state.json

{
  "best_metric": 0.662251655629139,
  "best_model_checkpoint": "outputs/indobert-base-uncased-twitter-indonesia-sarcastic/checkpoint-236",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 413,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.997697221833061e-06,
      "loss": 0.5531,
      "step": 59
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7723880597014925,
      "eval_f1": 0.4077669902912622,
      "eval_loss": 0.4976803660392761,
      "eval_precision": 0.5833333333333334,
      "eval_recall": 0.31343283582089554,
      "eval_runtime": 1.068,
      "eval_samples_per_second": 250.933,
      "eval_steps_per_second": 4.682,
      "step": 59
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.990791008447151e-06,
      "loss": 0.4992,
      "step": 118
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7723880597014925,
      "eval_f1": 0.3440860215053763,
      "eval_loss": 0.47853246331214905,
      "eval_precision": 0.6153846153846154,
      "eval_recall": 0.23880597014925373,
      "eval_runtime": 1.066,
      "eval_samples_per_second": 251.4,
      "eval_steps_per_second": 4.69,
      "step": 118
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.979044933876465e-06,
      "loss": 0.44,
      "step": 177
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7798507462686567,
      "eval_f1": 0.3655913978494624,
      "eval_loss": 0.48192882537841797,
      "eval_precision": 0.6538461538461539,
      "eval_recall": 0.2537313432835821,
      "eval_runtime": 1.0667,
      "eval_samples_per_second": 251.249,
      "eval_steps_per_second": 4.687,
      "step": 177
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.962224338961997e-06,
      "loss": 0.3815,
      "step": 236
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8097014925373134,
      "eval_f1": 0.662251655629139,
      "eval_loss": 0.4523766338825226,
      "eval_precision": 0.5952380952380952,
      "eval_recall": 0.746268656716418,
      "eval_runtime": 1.0729,
      "eval_samples_per_second": 249.795,
      "eval_steps_per_second": 4.66,
      "step": 236
    },
    {
      "epoch": 5.0,
      "learning_rate": 9.940506627721576e-06,
      "loss": 0.3104,
      "step": 295
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8171641791044776,
      "eval_f1": 0.5420560747663551,
      "eval_loss": 0.4547075927257538,
      "eval_precision": 0.725,
      "eval_recall": 0.43283582089552236,
      "eval_runtime": 1.0733,
      "eval_samples_per_second": 249.7,
      "eval_steps_per_second": 4.659,
      "step": 295
    },
    {
      "epoch": 6.0,
      "learning_rate": 9.913913232914188e-06,
      "loss": 0.2592,
      "step": 354
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8171641791044776,
      "eval_f1": 0.5663716814159292,
      "eval_loss": 0.40578439831733704,
      "eval_precision": 0.6956521739130435,
      "eval_recall": 0.47761194029850745,
      "eval_runtime": 1.0671,
      "eval_samples_per_second": 251.149,
      "eval_steps_per_second": 4.686,
      "step": 354
    },
    {
      "epoch": 7.0,
      "learning_rate": 9.882470399009847e-06,
      "loss": 0.2083,
      "step": 413
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8059701492537313,
      "eval_f1": 0.5737704918032788,
      "eval_loss": 0.4358327090740204,
      "eval_precision": 0.6363636363636364,
      "eval_recall": 0.5223880597014925,
      "eval_runtime": 1.0662,
      "eval_samples_per_second": 251.356,
      "eval_steps_per_second": 4.689,
      "step": 413
    },
    {
      "epoch": 7.0,
      "step": 413,
      "total_flos": 864714483440640.0,
      "train_loss": 0.37881522374926696,
      "train_runtime": 182.4457,
      "train_samples_per_second": 1029.347,
      "train_steps_per_second": 32.338
    }
  ],
  "logging_steps": 500,
  "max_steps": 5900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 864714483440640.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}