yujunjun/olmocr
1
0
Fork 0
mirror of https://github.com/allenai/olmocr.git synced 2025-11-14 01:19:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix for some math equations stuff

Browse Source
This commit is contained in:
Jake Poznanski 2025-08-22 20:52:41 +00:00
parent d36357f3db
commit 5c6225b227
1 changed files with 2 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
olmocr/bench/synth/mine_html_templates.py
View File

@ -875,14 +875,7 @@ def generate_tests_from_html(html_content: str, pdf_id: str, page_num: int, verb
if num_order_tests > 5: if num_order_tests > 5:
break break
# Step 4: Generate Math tests for LaTeX equations # Step 4: Generate Math tests for LaTeX equations from the markdown
# Get only the body content as a string to search for math patterns
body = soup.find('body')
if body:
body_html = str(body)
else:
# If no body tag, use the whole soup
body_html = str(soup)
# Define math patterns to search for # Define math patterns to search for
math_patterns = [ math_patterns = [
@ -893,7 +886,7 @@ def generate_tests_from_html(html_content: str, pdf_id: str, page_num: int, verb
math_equations = [] math_equations = []
for pattern, flags in math_patterns: for pattern, flags in math_patterns:
matches = re.findall(pattern, body_html, flags) matches = re.findall(pattern, markdown_content, flags)
for match in matches: for match in matches:
# Clean up the match - remove extra whitespace and newlines # Clean up the match - remove extra whitespace and newlines
equation = match.strip() equation = match.strip()