trainer_state.json

{
  "best_metric": 0.7537109966676764,
  "best_model_checkpoint": "models/nominal-groups-recognition-bert-base-spanish-wwm-cased/checkpoint-456",
  "epoch": 5.0,
  "global_step": 570,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.451,
      "step": 114
    },
    {
      "epoch": 1.0,
      "eval_NG_f1": 0.7225241545893719,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.698482194979568,
      "eval_NG_recall": 0.7482801751094434,
      "eval_loss": 0.28818875551223755,
      "eval_overall_accuracy": 0.8899181258590808,
      "eval_overall_f1": 0.7225241545893719,
      "eval_overall_precision": 0.698482194979568,
      "eval_overall_recall": 0.7482801751094434,
      "eval_runtime": 2.2675,
      "eval_samples_per_second": 172.436,
      "eval_steps_per_second": 11.025,
      "step": 114
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.2e-05,
      "loss": 0.2429,
      "step": 228
    },
    {
      "epoch": 2.0,
      "eval_NG_f1": 0.7386942429387251,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7293508076805851,
      "eval_NG_recall": 0.7482801751094434,
      "eval_loss": 0.29169175028800964,
      "eval_overall_accuracy": 0.8932050439251777,
      "eval_overall_f1": 0.7386942429387251,
      "eval_overall_precision": 0.7293508076805851,
      "eval_overall_recall": 0.7482801751094434,
      "eval_runtime": 2.3333,
      "eval_samples_per_second": 167.577,
      "eval_steps_per_second": 10.715,
      "step": 228
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.193,
      "step": 342
    },
    {
      "epoch": 3.0,
      "eval_NG_f1": 0.751063829787234,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7306327616794795,
      "eval_NG_recall": 0.7726704190118824,
      "eval_loss": 0.2864079773426056,
      "eval_overall_accuracy": 0.9000179286439969,
      "eval_overall_f1": 0.751063829787234,
      "eval_overall_precision": 0.7306327616794795,
      "eval_overall_recall": 0.7726704190118824,
      "eval_runtime": 2.3686,
      "eval_samples_per_second": 165.074,
      "eval_steps_per_second": 10.555,
      "step": 342
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.1586,
      "step": 456
    },
    {
      "epoch": 4.0,
      "eval_NG_f1": 0.7537109966676764,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7309048178613397,
      "eval_NG_recall": 0.7779862414008756,
      "eval_loss": 0.29878920316696167,
      "eval_overall_accuracy": 0.9019303173369988,
      "eval_overall_f1": 0.7537109966676764,
      "eval_overall_precision": 0.7309048178613397,
      "eval_overall_recall": 0.7779862414008756,
      "eval_runtime": 2.3061,
      "eval_samples_per_second": 169.554,
      "eval_steps_per_second": 10.841,
      "step": 456
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.1386,
      "step": 570
    },
    {
      "epoch": 5.0,
      "eval_NG_f1": 0.7514363471424251,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7274590163934426,
      "eval_NG_recall": 0.7770481550969356,
      "eval_loss": 0.3115690052509308,
      "eval_overall_accuracy": 0.9003167393772784,
      "eval_overall_f1": 0.7514363471424251,
      "eval_overall_precision": 0.7274590163934426,
      "eval_overall_recall": 0.7770481550969356,
      "eval_runtime": 2.3108,
      "eval_samples_per_second": 169.206,
      "eval_steps_per_second": 10.819,
      "step": 570
    },
    {
      "epoch": 5.0,
      "step": 570,
      "total_flos": 2383047966228480.0,
      "train_loss": 0.23679824628328022,
      "train_runtime": 150.2286,
      "train_samples_per_second": 60.707,
      "train_steps_per_second": 3.794
    }
  ],
  "max_steps": 570,
  "num_train_epochs": 5,
  "total_flos": 2383047966228480.0,
  "trial_name": null,
  "trial_params": null
}