# Python
__pycache__/
*.py[cod]
.pytest_cache/
.mypy_cache/
.DS_Store

# Virtual environments
.venv/
venv/
env/
**/.venv_labelstudio/

# Ignore unrelated nested repositories
AI-Scientist/
superpowers_framework/
.agents/

# Model weights and large artifacts
*.pth
*.pt
*.ckpt
*.onnx
*.bin

# Archives and temporary bundles
*.zip
*.tar
*.tar.gz
*.tgz
*.7z

# LaTeX intermediates
*.aux
*.bbl
*.bcf
*.blg
*.fdb_latexmk
*.fls
*.log
*.out
*.run.xml
*.synctex.gz
*.toc

# Experiment outputs and caches
research_mongolian/runs/
research_mongolian/page_level_ocr/results/
research_mongolian/page_level_ocr/runs/
research_mongolian/page_level_ocr/page_level_yolo_dataset_105/
research_mongolian/page_level_ocr/page_level_yolo_dataset/
research_mongolian/page_level_ocr/page_split_manifest.csv*
research_mongolian/page_level_ocr/*.bak

# Large page-level OCR workspace: ignore by default, then opt back in for
# top-level public scripts/docs/metadata only.
research_mongolian/page_level_ocr/**
!research_mongolian/page_level_ocr/
!research_mongolian/page_level_ocr/*.py
!research_mongolian/page_level_ocr/*.md
!research_mongolian/page_level_ocr/*.json
research_mongolian/page_level_ocr/page_level_annotations_*/
research_mongolian/page_level_ocr/page_level_annotations*.json
!research_mongolian/page_level_ocr/page_level_annotations.example.json

# Local env or project-local secrets/config
research_mongolian/page_level_ocr/label_studio_local.env

# Unrelated legacy OCR scratch files in the workspace root
OCR.pdf
OCR_Final_Corrected.pdf
OCR_Final_Corrected.tex
OCR_paragraphs.txt
OCR_textutil.html
OCR_verbatim_text.txt
copy_apacite.py
parse_images.py
patch_latex.py
test_images2.py
test_images3.py
images/

# Keep final manuscript PDFs only if intentionally tracked.
# Intermediate PDFs can be removed from the index later if needed.
