trainer_state.json

{
  "best_metric": 0.751588502269289,
  "best_model_checkpoint": "models/nominal-groups-recognition-bert-base-spanish-wwm-cased/checkpoint-1140",
  "epoch": 5.0,
  "global_step": 1140,
  "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": 0.3955,
      "step": 228
    },
    {
      "epoch": 1.0,
      "eval_NG_f1": 0.7305991767037658,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7128830705147278,
      "eval_NG_recall": 0.7492182614133833,
      "eval_loss": 0.2777566611766815,
      "eval_overall_accuracy": 0.8924281360186458,
      "eval_overall_f1": 0.7305991767037658,
      "eval_overall_precision": 0.7128830705147278,
      "eval_overall_recall": 0.7492182614133833,
      "eval_runtime": 13.4558,
      "eval_samples_per_second": 29.058,
      "eval_steps_per_second": 3.642,
      "step": 228
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.2e-05,
      "loss": 0.2186,
      "step": 456
    },
    {
      "epoch": 2.0,
      "eval_NG_f1": 0.7509135200974422,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7317507418397626,
      "eval_NG_recall": 0.7711069418386491,
      "eval_loss": 0.2762524485588074,
      "eval_overall_accuracy": 0.8990019721508397,
      "eval_overall_f1": 0.7509135200974422,
      "eval_overall_precision": 0.7317507418397626,
      "eval_overall_recall": 0.7711069418386491,
      "eval_runtime": 13.7791,
      "eval_samples_per_second": 28.376,
      "eval_steps_per_second": 3.556,
      "step": 456
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1586,
      "step": 684
    },
    {
      "epoch": 3.0,
      "eval_NG_f1": 0.7496210973022128,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7273529411764705,
      "eval_NG_recall": 0.7732958098811757,
      "eval_loss": 0.29604893922805786,
      "eval_overall_accuracy": 0.8992410207374649,
      "eval_overall_f1": 0.7496210973022128,
      "eval_overall_precision": 0.7273529411764705,
      "eval_overall_recall": 0.7732958098811757,
      "eval_runtime": 13.0377,
      "eval_samples_per_second": 29.99,
      "eval_steps_per_second": 3.758,
      "step": 684
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.119,
      "step": 912
    },
    {
      "epoch": 4.0,
      "eval_NG_f1": 0.7498103474434835,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7282640730916593,
      "eval_NG_recall": 0.7726704190118824,
      "eval_loss": 0.3329506516456604,
      "eval_overall_accuracy": 0.8981653020976513,
      "eval_overall_f1": 0.7498103474434835,
      "eval_overall_precision": 0.7282640730916593,
      "eval_overall_recall": 0.7726704190118824,
      "eval_runtime": 12.649,
      "eval_samples_per_second": 30.912,
      "eval_steps_per_second": 3.874,
      "step": 912
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.0943,
      "step": 1140
    },
    {
      "epoch": 5.0,
      "eval_NG_f1": 0.751588502269289,
      "eval_NG_number": 3198,
      "eval_NG_precision": 0.7280187573270809,
      "eval_NG_recall": 0.776735459662289,
      "eval_loss": 0.3567813038825989,
      "eval_overall_accuracy": 0.8991812585908086,
      "eval_overall_f1": 0.751588502269289,
      "eval_overall_precision": 0.7280187573270809,
      "eval_overall_recall": 0.776735459662289,
      "eval_runtime": 12.6742,
      "eval_samples_per_second": 30.85,
      "eval_steps_per_second": 3.866,
      "step": 1140
    },
    {
      "epoch": 5.0,
      "step": 1140,
      "total_flos": 2383047966228480.0,
      "train_loss": 0.19723336571141292,
      "train_runtime": 917.7763,
      "train_samples_per_second": 9.937,
      "train_steps_per_second": 1.242
    }
  ],
  "max_steps": 1140,
  "num_train_epochs": 5,
  "total_flos": 2383047966228480.0,
  "trial_name": null,
  "trial_params": null
}