models/bert_ner/checkpoint-4796/trainer_state.json

{
  "best_metric": 0.08877604454755783,
  "best_model_checkpoint": "/content/drive/MyDrive/chatbot_info/models/bert_ner/checkpoint-4796",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 4796,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2085070892410342,
      "grad_norm": 1.3864336013793945,
      "learning_rate": 1.8609952738393106e-05,
      "loss": 0.217,
      "step": 500
    },
    {
      "epoch": 0.4170141784820684,
      "grad_norm": 0.8502457737922668,
      "learning_rate": 1.721990547678621e-05,
      "loss": 0.1147,
      "step": 1000
    },
    {
      "epoch": 0.6255212677231026,
      "grad_norm": 1.1136561632156372,
      "learning_rate": 1.5829858215179316e-05,
      "loss": 0.1029,
      "step": 1500
    },
    {
      "epoch": 0.8340283569641368,
      "grad_norm": 0.9568729400634766,
      "learning_rate": 1.4439810953572422e-05,
      "loss": 0.1039,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9708774583963692,
      "eval_f1": 0.8271461197438539,
      "eval_loss": 0.09369628131389618,
      "eval_precision": 0.8277762839472976,
      "eval_recall": 0.8265169142652586,
      "eval_runtime": 32.2988,
      "eval_samples_per_second": 296.915,
      "eval_steps_per_second": 18.577,
      "step": 2398
    },
    {
      "epoch": 1.042535446205171,
      "grad_norm": 1.206861972808838,
      "learning_rate": 1.3049763691965527e-05,
      "loss": 0.093,
      "step": 2500
    },
    {
      "epoch": 1.2510425354462051,
      "grad_norm": 0.7828037142753601,
      "learning_rate": 1.1659716430358635e-05,
      "loss": 0.0807,
      "step": 3000
    },
    {
      "epoch": 1.4595496246872393,
      "grad_norm": 0.5838351845741272,
      "learning_rate": 1.026966916875174e-05,
      "loss": 0.0777,
      "step": 3500
    },
    {
      "epoch": 1.6680567139282736,
      "grad_norm": 0.8385056853294373,
      "learning_rate": 8.879621907144844e-06,
      "loss": 0.0758,
      "step": 4000
    },
    {
      "epoch": 1.8765638031693077,
      "grad_norm": 0.9494819045066833,
      "learning_rate": 7.4895746455379494e-06,
      "loss": 0.076,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9721030659338555,
      "eval_f1": 0.83379981340797,
      "eval_loss": 0.08877604454755783,
      "eval_precision": 0.8278782531980591,
      "eval_recall": 0.8398066941113299,
      "eval_runtime": 31.2589,
      "eval_samples_per_second": 306.793,
      "eval_steps_per_second": 19.195,
      "step": 4796
    }
  ],
  "logging_steps": 500,
  "max_steps": 7194,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1797436693363512.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}