msh1273 / jina-shortcode-covidQA-frontend

shortcode로 Jina framework사용(frontend: Streamlit)

jina-shortcode-front(covidQA)

This is a simple example using the Jina neural search framework.
Data set used: Covid-QA from kaggle

If you want to run it in local

Instructions

This is a shortcode using jinahub's indexer and encoder. You can try running this project with the Run on ainize button.

---

이것은 jina를 쉽게 이해할 수 있도록 jina프레임워크를 사용하여 shortcode로 구성한 프로젝트입니다. covidQA의 news 데이터 셋을 사용하였으며 jinahub에 만들어져있는 encoder와 indexer를 가져와 사용했습니다.
jinahub의 pytorchtransformerencoder는 distilbert-base-nli-stsb-mean-tokens모델을 사용하였으며 오타를 입력하더라도 유사한 질문과 답변을 확인하실 수 있습니다. simpleindexer에는 5개의 top_k를 보여줍니다.

Clone this repo

git clone https://github.com/msh1273/jina-shortcode-covidQA-frontend.git
cd jina-shortcode-covidQA-frontend

Create a virtual environment (optional)

We wouldn't want our project clashing with our system libraries, now would we?

virtualenv env --python=python3.8 # Python versions >= 3.7 work fine
source env/bin/activate

Install everything

Make sure you're in your virtual environment first!

pip install -r requirements.txt

Run your app

streamlit run front.py

FAQ

Why this dataset?

Ask any questions you have about Corona. jina will show you the desired result even if you type a typo.