trainer_state.json

{
  "best_metric": 0.13593988120555878,
  "best_model_checkpoint": "/home/sdonoso/data/all_results/ner-c/distillbert/epochs_4_bs_64_lr_5e-5/checkpoint-500",
  "epoch": 4.0,
  "global_step": 524,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.76,
      "eval_accuracy": 0.9476202751909258,
      "eval_f1": 0.7172177028136428,
      "eval_loss": 0.1729145497083664,
      "eval_precision": 0.7087020648967551,
      "eval_recall": 0.7259404743919021,
      "eval_runtime": 1.7788,
      "eval_samples_per_second": 1077.132,
      "eval_steps_per_second": 16.865,
      "step": 100
    },
    {
      "epoch": 1.53,
      "eval_accuracy": 0.9522766087713079,
      "eval_f1": 0.7523824162311713,
      "eval_loss": 0.1584220975637436,
      "eval_precision": 0.765681213827624,
      "eval_recall": 0.7395376945157879,
      "eval_runtime": 1.776,
      "eval_samples_per_second": 1078.832,
      "eval_steps_per_second": 16.892,
      "step": 200
    },
    {
      "epoch": 2.29,
      "eval_accuracy": 0.9581574453001643,
      "eval_f1": 0.7900313386061781,
      "eval_loss": 0.13973243534564972,
      "eval_precision": 0.7804806132979507,
      "eval_recall": 0.7998187037316815,
      "eval_runtime": 1.7908,
      "eval_samples_per_second": 1069.888,
      "eval_steps_per_second": 16.752,
      "step": 300
    },
    {
      "epoch": 3.05,
      "eval_accuracy": 0.9592047175587278,
      "eval_f1": 0.7948698717467937,
      "eval_loss": 0.1360633671283722,
      "eval_precision": 0.7892463509085493,
      "eval_recall": 0.8005741048496752,
      "eval_runtime": 1.7307,
      "eval_samples_per_second": 1107.051,
      "eval_steps_per_second": 17.334,
      "step": 400
    },
    {
      "epoch": 3.82,
      "learning_rate": 2.5763358778625957e-06,
      "loss": 0.1581,
      "step": 500
    },
    {
      "epoch": 3.82,
      "eval_accuracy": 0.9594786195340445,
      "eval_f1": 0.7985358930305521,
      "eval_loss": 0.13593988120555878,
      "eval_precision": 0.7897458628841607,
      "eval_recall": 0.8075237951352168,
      "eval_runtime": 1.769,
      "eval_samples_per_second": 1083.1,
      "eval_steps_per_second": 16.959,
      "step": 500
    },
    {
      "epoch": 4.0,
      "step": 524,
      "total_flos": 997670842354368.0,
      "train_loss": 0.15398232195213551,
      "train_runtime": 729.0203,
      "train_samples_per_second": 45.672,
      "train_steps_per_second": 0.719
    }
  ],
  "max_steps": 524,
  "num_train_epochs": 4,
  "total_flos": 997670842354368.0,
  "trial_name": null,
  "trial_params": null
}