dependency cleanup

llmware-ai · Dec 14, 2023 · 487855b · 487855b
1 parent c974df9
commit 487855b
Show file tree

Hide file tree

Showing 4 changed files with 12 additions and 14 deletions.
diff --git a/NOTICE b/NOTICE
@@ -22,9 +22,6 @@ Open-Source dependencies for the llmware package:
    * Software: libzip (https://libzip.org/)
    * Software: lxml (https://github.com/lxml/lxml)
    * Software: numpy (https://github.com/numpy/numpy)
-   * Software: pandas (https://github.com/pandas-dev/pandas)
-   * Software: scikit-learn (https://github.com/scikit-learn/scikit-learn)
-   * Software: scipy (https://github.com/scipy/scipy)
    * Software: torch (https://github.com/pytorch/pytorch)
    * Software: Werkzeug (https://github.com/pallets/werkzeug/)
 

diff --git a/llmware/parsers.py b/llmware/parsers.py
@@ -27,8 +27,7 @@
 import requests
 from bs4 import BeautifulSoup
 from urllib.request import urlopen, Request
-from PIL import Image
-from PIL.ExifTags import TAGS, GPSTAGS
+
 import pytesseract
 from pytesseract.pytesseract import TesseractNotFoundError
 from pdf2image import convert_from_path
@@ -2937,6 +2936,15 @@ def exif_extractor(self, fp):
         exif_table = {}
         creator_device = {}
         time_stamps = {}
+
+        #   PIL/Pillow required for EXIF image processing - must be installed separately
+
+        try:
+            from PIL import Image
+            from PIL.ExifTags import TAGS, GPSTAGS
+
+        except:
+            raise DependencyNotInstalledException("PIL")
 
         try:
             img = Image.open(fp)

diff --git a/llmware/requirements.txt b/llmware/requirements.txt
@@ -7,14 +7,12 @@ datasets==2.15.0
 faiss-cpu==1.7.4
 huggingface-hub==0.19.4
 lxml==4.9.3
-numpy==1.23.2
+numpy>=1.23.2
 openai==0.27.7
 pdf2image==1.16.0
-Pillow==10.0.1
 pymilvus==2.3.0
 pymongo==4.5.0
 pytesseract==0.3.10
-scipy==1.11.2
 sentence-transformers==2.2.2
 tabulate==0.9.0
 tokenizers==0.15.0

diff --git a/setup.py b/setup.py
@@ -60,17 +60,12 @@ def glob_fix(package_name, glob):
         'faiss-cpu==1.7.4',
         'huggingface-hub==0.19.4',
         'lxml==4.9.3',
-        'numpy==1.23.2',
+        'numpy>=1.23.2',
         'openai==0.27.7',
-        'pandas==2.1.1',
         'pdf2image==1.16.0',
-        'Pillow==10.0.1',
         'pymilvus==2.3.0',
         'pymongo==4.5.0',
         'pytesseract==0.3.10',
-        'python-on-whales==0.64.3',
-        'scipy==1.11.2',
-        'scikit-learn==1.3.1',
         'sentence-transformers==2.2.2',
         'tabulate==0.9.0',
         'tokenizers>=0.15.0',