kiyeon1221's picture
Update README.md
555c596
metadata
license: cc-by-nc-4.0
base_model: KT-AI/midm-bitext-S-7B-inst-v1
tags:
  - generated_from_trainer
model-index:
  - name: lora-midm-7b-food-order-understanding
    results: []

์‹คํ—˜๋‚ด์šฉ๊ณผ ํ…Œ์ŠคํŠธ ๋ฐ์ดํ„ฐ์— ๋Œ€ํ•œ ๋ถ„๋ฅ˜ ๊ฒฐ๊ณผ ๋ฆฌํฌํŠธ

์ฃผ๋ฌธ ๋ฌธ์žฅ์— ์˜ํ•ด ํ•™์Šต๋œ midm์„ nsmc (์˜ํ™” ๋ฆฌ๋ทฐ ๋ฐ์ดํ„ฐ์…‹) train dataset 3000๊ฐœ๋กœ ํ•™์Šต์„ ์‹œ์ผฐ๋‹ค. ์ฒ˜์Œ์—๋Š” 2000๊ฐœ๋ฅผ ํ•™์Šต์‹œ์ผฐ์œผ๋‚˜ ์ •ํ™•๋„๊ฐ€ ์˜ˆ์ƒํ•œ ๊ฒƒ ๋งŒํผ ๋‚˜์˜ค์ง€ ์•Š์•„ 1000๊ฐœ๋ฅผ ๋” ํ•™์Šต์‹œ์ผฐ๋”๋‹ˆ ์•ฝ 2%์˜ ์ •ํ™•๋„๊ฐ€ ์˜ฌ๋ผ๊ฐ”๋‹ค. ๊ทธ๋ฆฌ๊ณ  1000๊ฐœ์˜ test dataset์œผ๋กœ ํ…Œ์ŠคํŠธ๋ฅผ ํ•ด๋ณด์•˜๋‹ค. ์ •ํ™•๋„๋Š” 89.30 % ๊ฐ€ ๋‚˜์™”์œผ๋ฉฐ ํ‘œ๋Š” ๋ฐ‘ ์‚ฌ์ง„์œผ๋กœ ํ™•์ธํ•  ์ˆ˜ ์žˆ๋‹ค. ์ด ์‹คํ—˜์€ ๋‹ค๋ฅธ ๋ฐ์ดํ„ฐ ์…‹๊ณผ ๋‹ค๋ฅธ ์š”๊ตฌ๋กœ ํ•™์Šต๋˜์–ด์žˆ๋˜ LLM์„ ์ƒˆ ๋ฐ์ดํ„ฐ ์…‹๊ณผ ์ƒˆ ์š”๊ตฌ๋กœ ๋ฏธ์„ธํŠœ๋‹ ํ•˜์˜€์„ ๋•Œ ์ •ํ™•๋„๊ฐ€ ์–ผ๋งŒํผ ๋‚˜์˜ค๋Š”์ง€๋ฅผ ๋ณด์—ฌ์ฃผ๋Š” ๊ฒƒ์ด๋‹ค.

ํ‘œ

image/jpeg

lora-midm-7b-food-order-understanding

This model is a fine-tuned version of KT-AI/midm-bitext-S-7B-inst-v1 on an unknown dataset.

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: 0.0001
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 2
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.03
  • training_steps: 300
  • mixed_precision_training: Native AMP

Training results

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.1.0+cu118
  • Datasets 2.15.0
  • Tokenizers 0.15.0