jcharlet/rimay_search

fix dockerfile, install git lfs

fix parquet files

fix openai key bug

fix unit tests with new sources format

upd: Victory, deploy successful! update readme

fix: python version and server port in dockerfile

git add chromadb embedding in git (to be used in docker container)

fix req: missing lib

fix: add dockerfile

upd: minor, style and function definition

fix: undo local package install for huggingface

initial commit (merged with huggingface initial commit)

upd python notebook (minor)

add: yogi simple UI + remove_embeddings when openapi unset

add openai_key input in streamlit app + upd UI accordingly

fix cost format

upd: improve display of sources + cost

fix style

upd: provide formatted sources in streamlit app + minor fixes

add streamlit app to run queries

upd: major refactoring + add get_doc_by_id

add: cost tracking + choose response size + documentation

add documentation and click commands on predict_model

fix style and unit test

upd: qa chain for mindfulness contents

upd: refactor embedding and qa chain + added unit tests

upd: improve text preparation

add: clean dataset, add features and balance a bit more article text length

fix quality issues: run black, fix flake8 issues

fix: makefile and cleaning of scraped dataset

fix bugs in scraping, add caching by storing html files

add command to scrap whole website, improve scraping

add collect_toc_links function

add scrap_page function

add aws s3 configuration

add langchain retrieval + QA chain sample

fix requirements install

add: cookie cutter datascience template

Initial commit