Sparrow Parse refactoring and cleanup from dependencies

katanaml · Nov 8, 2024 · 2fb649f · 2fb649f
1 parent 98a7ec9
commit 2fb649f
Show file tree

Hide file tree

Showing 10 changed files with 8 additions and 959 deletions.
diff --git a/sparrow-data/parse/requirements.txt b/sparrow-data/parse/requirements.txt
@@ -1,14 +1,11 @@
-torch==2.2.2
-unstructured[all-docs]==0.14.5
-unstructured-inference==0.7.33
 rich
-pymupdf4llm==0.0.9
 transformers==4.41.2
 sentence-transformers==3.0.1
 numpy==1.26.4
 pypdf==4.3.0
 easyocr==1.7.1
 gradio_client
+pdf2image
 
 
 # Force reinstall:

diff --git a/sparrow-data/parse/setup.py b/sparrow-data/parse/setup.py
@@ -8,10 +8,10 @@
 
 setup(
     name="sparrow-parse",
-    version="0.3.4",
+    version="0.3.5",
     author="Andrej Baranovskij",
     author_email="[email protected]",
-    description="Sparrow Parse is a Python package for parsing and extracting information from documents.",
+    description="Sparrow Parse is a Python package (part of Sparrow) for parsing and extracting information from documents.",
     long_description=long_description,
     long_description_content_type="text/markdown",
     url="https://github.com/katanaml/sparrow/tree/main/sparrow-data/parse",
@@ -30,7 +30,7 @@
             'sparrow-parse=sparrow_parse:main',
         ],
     },
-    keywords="llm, rag, vision",
+    keywords="llm, vllm, ocr, vision",
     packages=find_packages(),
     python_requires='>=3.10',
     install_requires=requirements,

diff --git a/sparrow-data/parse/sparrow_parse/__init__.py b/sparrow-data/parse/sparrow_parse/__init__.py
@@ -1 +1 @@
-__version__ = '0.3.4'
+__version__ = '0.3.5'
diff --git a/sparrow-data/parse/sparrow_parse/data/invoice_1_table.txt b/sparrow-data/parse/sparrow_parse/data/invoice_1_table.txt
diff --git a/sparrow-data/parse/sparrow_parse/extractors/html_extractor.py b/sparrow-data/parse/sparrow_parse/extractors/html_extractor.py