STS Task with finetuned model Web Prototype
pip install -r requiremnts.txt
streamlit run main.py
Semantic Text Similarity
monologg/koelectra-base-v3-discriminator
1) Pre-trained model with a regression layer
2) Fine-tuned with 9325 train samples
3) Output scales : 0-5 float to two decimal points
Streamlit Library