masadonline commited on
Commit
6ed44f6
·
verified ·
1 Parent(s): 2e93654

Update requirements.txt

Browse files
Files changed (1) hide show
  1. requirements.txt +7 -30
requirements.txt CHANGED
@@ -1,35 +1,12 @@
1
  streamlit
2
  langchain
3
- langchain-community
4
- langchain-huggingface
5
  langchain-groq
6
- faiss-cpu # For CPU-based FAISS, use faiss-gpu if you have a compatible GPU environment
 
 
7
  sentence-transformers
 
 
 
8
  python-dotenv
9
- unstructured # Base library for UnstructuredFileLoader
10
- pypdf # Dependency for UnstructuredFileLoader for PDFs
11
- python-docx # Dependency for UnstructuredFileLoader for DOCX
12
- openpyxl # Dependency for UnstructuredFileLoader for XLSX
13
- xlrd # Dependency for UnstructuredFileLoader for older XLS files
14
- python-pptx # Dependency for UnstructuredFileLoader for PPTX
15
- beautifulsoup4 # Dependency for UnstructuredFileLoader for HTML
16
- lxml # Often a useful parser for unstructured
17
- markdown # For .md file processing by unstructured
18
- pymagic # For file type detection, unstructured might need it
19
- libmagic # System dependency for pymagic, often needs manual install (e.g., brew install libmagic / sudo apt-get install libmagic1) - for HF Spaces, usually pre-installed or handled by base image.
20
-
21
- # Optional, for advanced PDF table/layout analysis with unstructured (strategy="hi_res"):
22
- # detectron2 # This has more complex installation steps (PyTorch, torchvision, etc.)
23
- # # Refer to official detectron2 installation guide. May not be straightforward on Hugging Face free tier.
24
- # Pillow # Usually a dependency of detectron2 or unstructured itself
25
-
26
- # Other potential dependencies for unstructured depending on file types:
27
- # tabulate # For markdown table conversion
28
- # pandas # Often used by excel/csv loaders
29
-
30
- # Note on libmagic:
31
- # For Hugging Face Spaces, common system libraries are often available.
32
- # If you encounter issues with 'libmagic' locally, install it using your system's package manager:
33
- # macOS: brew install libmagic
34
- # Debian/Ubuntu: sudo apt-get install libmagic1
35
- # Windows: Requires a different approach, often involving manual DLL placement or using a pre-compiled package.
 
1
  streamlit
2
  langchain
 
 
3
  langchain-groq
4
+ langchain-community
5
+ langchain-text-splitters
6
+ faiss-cpu
7
  sentence-transformers
8
+ unstructured[all-docs] # Includes pypdf, python-docx, openpyxl, Pillow, pytesseract, lxml, etc.
9
+ # pdf2image # often a dependency for unstructured PDF processing with images
10
+ # pytesseract # If you have PDFs with scanned images that need OCR
11
  python-dotenv
12
+ groq