chardet
filetype
python-magic
lxml
nltk
requests
beautifulsoup4
emoji
dataclasses-json
python-iso639
langdetect
numpy<2
rapidfuzz
backoff
typing-extensions
unstructured-client
wrapt
tqdm
psutil
python-oxmsg
html5lib

[all-docs]
python-docx>=1.1.2
google-cloud-vision
effdet
pdfminer.six
pypdf
pi_heif
unstructured-inference>=0.8.7
openpyxl
markdown
pdf2image
networkx
pikepdf
python-pptx>=1.0.1
onnx
unstructured.pytesseract>=0.3.12
pypandoc
xlrd
pandas

[csv]
pandas

[doc]
python-docx>=1.1.2

[docx]
python-docx>=1.1.2

[epub]
pypandoc

[huggingface]
langdetect
sacremoses
sentencepiece
torch
transformers

[image]
onnx
pdf2image
pdfminer.six
pikepdf
pi_heif
pypdf
google-cloud-vision
effdet
unstructured-inference>=0.8.7
unstructured.pytesseract>=0.3.12

[local-inference]
python-docx>=1.1.2
google-cloud-vision
effdet
pdfminer.six
pypdf
pi_heif
unstructured-inference>=0.8.7
openpyxl
markdown
pdf2image
networkx
pikepdf
python-pptx>=1.0.1
onnx
unstructured.pytesseract>=0.3.12
pypandoc
xlrd
pandas

[md]
markdown

[odt]
python-docx>=1.1.2
pypandoc

[org]
pypandoc

[paddleocr]
paddlepaddle==3.0.0b1
unstructured.paddleocr==2.8.1.0

[pdf]
onnx
pdf2image
pdfminer.six
pikepdf
pi_heif
pypdf
google-cloud-vision
effdet
unstructured-inference>=0.8.7
unstructured.pytesseract>=0.3.12

[ppt]
python-pptx>=1.0.1

[pptx]
python-pptx>=1.0.1

[rst]
pypandoc

[rtf]
pypandoc

[tsv]
pandas

[xlsx]
openpyxl
pandas
xlrd
networkx
