@@ -56,7 +56,6 @@ def doc_analyze(pdf_bytes: bytes, ocr: bool = False, show_log: bool = False):
page_height = img_dict['height']
result = ocr_engine(img)
spans = []
- need_remove = []
for line in result:
line.pop('img')
'''
@@ -15,3 +15,5 @@ scikit-learn>=1.0.2
nltk==3.8.1
s3pathlib>=2.1.1
pytest
+paddlepaddle
+paddleocr>=2.6.0.3