Spaces:
Running
Running
Commit History
bugfix: prevent model load on every extraction
ff75fe0
chore: address review points
2691833
fix: lint
f48b4d7
add: ContrieverRetriever
b82a487
upfate: CalPaliRetriever methods to align with other retrievers
c063304
update: dependencies
9595675
update: docs with lib sources and notes
e6f2eb8
add: docs for pymupdf and fitzpil
04ea7bb
add: two modules on fitz to handle img extractions
f9d44bd
temp: attempt - force to png with pillow
3d948a1
temp: attempt - all format img extraction from pdf
5406446
fix: catch byaldi import
3bd446c
fix: catch byaldi import
353a440
remove: notebook
dc4625d
fix: lint
f673b7f
update: decouple Byaldi from pyproject.toml because of conflict with adapters and mentioned the same in relevant docs as well
8f9e28d
add: basic workflow to check code format and lint
4069faf
fix: bug wrt result unpacking in BM25sRetriever.retrieve
86ac070
update: docs
e2274d5
add: docs for BM25sRetriever
b5a3ebb
add: docs for pdfplumber image loader
e19286a
add: hacky impl of img extraction with pdfplumber
4fd52cf
add: example usage for marker and pdf2img loaders
bf0f2e5
add: marker image loader + docs + corrections
331f289
chore: improve doc + code formatting
f37090a
add: docs for base img loader + pdf2image
cc5cebc
add: base image loader + pdf2img from load_image
5c74069
update: BM25sRetriever
7f98acf
update: BM25sRetriever
88a5bcf
add: BM25sRetriever
4ea2b30
refactor: colpali retrieval
21537b7
Merge pull request #11 from soumik12345/feat/semantic-chunking
694a076
unverified
Atanu Sarkar
commited on