Skip to content

feat: Add optional rect parameter to extract method for constrained table detection #67

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
conjuncts merged 1 commit into from
Jul 18, 2025
Merged

feat: Add optional rect parameter to extract method for constrained table detection #67

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 5 additions & 3 deletions gmft/detectors/tatr.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@
from gmft.core._dataclasses import with_config
from gmft.core.ml import _resolve_device
from gmft.detectors.base import BaseDetector, CroppedTable, RotatedCroppedTable
from gmft.base import Rect

from gmft.impl.tatr.config import TATRDetectorConfig
from gmft.pdf_bindings.base import BasePage
Expand Down Expand Up @@ -54,19 +55,20 @@ def __init__(self, config: TATRDetectorConfig = None, default_implementation=Tru
self.config = config

def extract(
self, page: BasePage, config_overrides: TATRDetectorConfig = None
self, page: BasePage, config_overrides: TATRDetectorConfig = None, rect: Rect = None
) -> list[CroppedTable]:
"""
Detect tables in a page.

:param page: BasePage
:param config_overrides: override the config for this call only
:param config_overrides: Optional config overrides for this extraction
:param rect: Optional Rect to constrain detection within given dimensions
:return: list of CroppedTable objects
"""
config = with_config(self.config, config_overrides)

img = page.get_image(
72
72, rect=rect
) # use standard dpi = 72, which means we don't need any scaling
encoding = self.image_processor(img, return_tensors="pt").to(
_resolve_device(self.config.torch_device)
Expand Down