models/bilstm_ner/checkpoint-9592/trainer_state.json

{
  "best_metric": 0.09198188781738281,
  "best_model_checkpoint": "/content/drive/MyDrive/chatbot_info/models/bilstm_ner/checkpoint-7194",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 9592,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2085070892410342,
      "grad_norm": 0.9963818788528442,
      "learning_rate": 1.9165971643035866e-05,
      "loss": 0.3555,
      "step": 500
    },
    {
      "epoch": 0.4170141784820684,
      "grad_norm": 0.8802713751792908,
      "learning_rate": 1.8331943286071728e-05,
      "loss": 0.1417,
      "step": 1000
    },
    {
      "epoch": 0.6255212677231026,
      "grad_norm": 0.7941210269927979,
      "learning_rate": 1.7497914929107593e-05,
      "loss": 0.1213,
      "step": 1500
    },
    {
      "epoch": 0.8340283569641368,
      "grad_norm": 0.6393231153488159,
      "learning_rate": 1.6663886572143454e-05,
      "loss": 0.117,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9699247400371512,
      "eval_f1": 0.8206677407805227,
      "eval_loss": 0.10422153025865555,
      "eval_precision": 0.8208146821844226,
      "eval_recall": 0.8205208519778057,
      "eval_runtime": 32.7374,
      "eval_samples_per_second": 292.937,
      "eval_steps_per_second": 18.328,
      "step": 2398
    },
    {
      "epoch": 1.042535446205171,
      "grad_norm": 0.7787706851959229,
      "learning_rate": 1.5829858215179316e-05,
      "loss": 0.1043,
      "step": 2500
    },
    {
      "epoch": 1.2510425354462051,
      "grad_norm": 0.5809928178787231,
      "learning_rate": 1.499582985821518e-05,
      "loss": 0.0923,
      "step": 3000
    },
    {
      "epoch": 1.4595496246872393,
      "grad_norm": 0.8074227571487427,
      "learning_rate": 1.4161801501251045e-05,
      "loss": 0.0885,
      "step": 3500
    },
    {
      "epoch": 1.6680567139282736,
      "grad_norm": 0.8078889846801758,
      "learning_rate": 1.3327773144286907e-05,
      "loss": 0.0866,
      "step": 4000
    },
    {
      "epoch": 1.8765638031693077,
      "grad_norm": 0.8388847708702087,
      "learning_rate": 1.249374478732277e-05,
      "loss": 0.0865,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9717775139317106,
      "eval_f1": 0.83455259166741,
      "eval_loss": 0.0938563197851181,
      "eval_precision": 0.8319548203486303,
      "eval_recall": 0.837166636835511,
      "eval_runtime": 33.2641,
      "eval_samples_per_second": 288.299,
      "eval_steps_per_second": 18.037,
      "step": 4796
    },
    {
      "epoch": 2.085070892410342,
      "grad_norm": 0.901061475276947,
      "learning_rate": 1.1659716430358635e-05,
      "loss": 0.0816,
      "step": 5000
    },
    {
      "epoch": 2.293577981651376,
      "grad_norm": 1.0216280221939087,
      "learning_rate": 1.0825688073394496e-05,
      "loss": 0.0707,
      "step": 5500
    },
    {
      "epoch": 2.5020850708924103,
      "grad_norm": 0.6803075075149536,
      "learning_rate": 9.99165971643036e-06,
      "loss": 0.0711,
      "step": 6000
    },
    {
      "epoch": 2.7105921601334444,
      "grad_norm": 0.9469559788703918,
      "learning_rate": 9.157631359466222e-06,
      "loss": 0.0708,
      "step": 6500
    },
    {
      "epoch": 2.9190992493744785,
      "grad_norm": 0.7367610931396484,
      "learning_rate": 8.323603002502085e-06,
      "loss": 0.0685,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9728738581742278,
      "eval_f1": 0.836874388834563,
      "eval_loss": 0.09198188781738281,
      "eval_precision": 0.8313316849169905,
      "eval_recall": 0.8424914981206372,
      "eval_runtime": 31.4043,
      "eval_samples_per_second": 305.372,
      "eval_steps_per_second": 19.106,
      "step": 7194
    },
    {
      "epoch": 3.127606338615513,
      "grad_norm": 0.6511647701263428,
      "learning_rate": 7.4895746455379494e-06,
      "loss": 0.0613,
      "step": 7500
    },
    {
      "epoch": 3.336113427856547,
      "grad_norm": 0.5995446443557739,
      "learning_rate": 6.655546288573812e-06,
      "loss": 0.0583,
      "step": 8000
    },
    {
      "epoch": 3.5446205170975813,
      "grad_norm": 0.8369386196136475,
      "learning_rate": 5.821517931609675e-06,
      "loss": 0.0581,
      "step": 8500
    },
    {
      "epoch": 3.7531276063386154,
      "grad_norm": 0.8538610339164734,
      "learning_rate": 4.987489574645538e-06,
      "loss": 0.058,
      "step": 9000
    },
    {
      "epoch": 3.96163469557965,
      "grad_norm": 0.8798556923866272,
      "learning_rate": 4.153461217681402e-06,
      "loss": 0.0551,
      "step": 9500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9729456711158774,
      "eval_f1": 0.8398951297547104,
      "eval_loss": 0.09327501803636551,
      "eval_precision": 0.8341129744042366,
      "eval_recall": 0.8457580096652945,
      "eval_runtime": 31.6141,
      "eval_samples_per_second": 303.346,
      "eval_steps_per_second": 18.979,
      "step": 9592
    }
  ],
  "logging_steps": 500,
  "max_steps": 11990,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}