output / README.md
ginami's picture
Training completed!
5dba4ab verified
metadata
license: mit
base_model: beomi/KcELECTRA-base-v2022
tags:
  - generated_from_trainer
model-index:
  - name: kcelectra_base_ner_law_v2
    results: []

kcelectra_base_ner_law_v2

This model is a fine-tuned version of beomi/KcELECTRA-base-v2022 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0009

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 3e-05
  • train_batch_size: 8
  • eval_batch_size: 64
  • seed: 15
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
0.0105 0.11 1000 0.0022
0.0028 0.22 2000 0.0019
0.0024 0.33 3000 0.0016
0.0014 0.44 4000 0.0027
0.0014 0.56 5000 0.0015
0.0016 0.67 6000 0.0013
0.0012 0.78 7000 0.0013
0.0016 0.89 8000 0.0011
0.0011 1.0 9000 0.0009
0.0006 1.11 10000 0.0013
0.0005 1.22 11000 0.0014

Framework versions

  • Transformers 4.37.2
  • Pytorch 2.1.0+cu121
  • Tokenizers 0.15.1