Added kb_id filter to knn. Fix #3458 (#3513)

Move settings initialization after module init phase (#3438)

Use consistent log file names, introduced initLogger (#3403)

Rework logging (#3358)

bigger resolution for OCR (#2919)

fix: torch dependency start error (#2777)

Fix: renrank_model and pdf_parser bugs | Update: session API (#2601)

add lighten control (#2567)

fix parsing spaces in russian language PDFs (#1987) (#2427)

Fix docx parser line bug (#1715)

fix: When parsing the bold content in PDF, the result is duplicated. (#1729)

Fix pdfparser content confusion (#1700)

pypdf2 to pypdf (#1684)

fix bug about divided by zero (#1482)

fix: Delete hardcode (#1464)

fix pdf_paser char content confusion (#1462)

fix pdf_parser content confusion (#1458)

Fix occasional errors in pdf table recognition (#1277)

add self-rag (#1070)

Update readme and add license (#1018)

fix bug in pdf parser (#986)

fix #917 #915 (#946)

fixbug for computing 'not concating feature' (#896)

fix coordinate error (#686)

remove PyMuPDF (#618)

refine code (#595)

fix exception in pdf parser (#584)

refactor code (#583)

Refactor (#537)

enlarge docker memory usage (#501)

fix divide by zero bug (#447)

Bug fix pdf parse index out of range (#440)

rm page number exception for pdf parser (#424)

make sure the models will not be load twice (#422)

let's load model from local (#163)

apply pep8 formalize (#155)

support snapshot download from local (#153)

fix plainPdf bugs (#152)

refine page ranges (#147)

add use layout or not option (#145)

refine manual parser (#140)

refine for English corpus (#135)

fix github account login issue (#132)

refine manul parser (#131)

add dockerfile for cuda envirement. Refine table search strategy, (#123)

refine pdf parser, add time zone to userinfo (#112)

fix task cancling bug (#98)

change callback strategy, add timezone to docker (#96)

fix position extraction bug (#93)

fix table desc bugs, add positions to chunks (#91)