pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 20:42:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

refine admin initialization (#75)

Browse Source
This commit is contained in:
KevinHuSh
2024-02-27 14:57:34 +08:00
committed by GitHub
parent d1c600d5d3
commit 4568a4b2cb
13 changed files with 91 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
deepdoc/vision/t_recognizer.py
View File

@ -74,6 +74,7 @@ def get_table_html(img, tb_cpns, ocr):
clmns = sorted([r for r in tb_cpns if re.match(
r"table column$", r["label"])], key=lambda x: x["x0"])
clmns = Recognizer.layouts_cleanup(boxes, clmns, 5, 0.5)
for b in boxes:
ii = Recognizer.find_overlapped_with_threashold(b, rows, thr=0.3)
if ii is not None:
@ -89,7 +90,7 @@ def get_table_html(img, tb_cpns, ocr):
b["H_right"] = headers[ii]["x1"]
b["H"] = ii
ii = Recognizer.find_overlapped_with_threashold(b, clmns, thr=0.3)
ii = Recognizer.find_horizontally_tightest_fit(b, clmns)
if ii is not None:
b["C"] = ii
b["C_left"] = clmns[ii]["x0"]
@ -102,6 +103,7 @@ def get_table_html(img, tb_cpns, ocr):
b["H_left"] = spans[ii]["x0"]
b["H_right"] = spans[ii]["x1"]
b["SP"] = ii
html = """
<html>
<head>