AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-14 17:45:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix:t_recognizer TypeError: 'super' object is not callable (#4404)

Browse Source 
### What problem does this PR solve?

[Bug]: layout recognizer failed for wrong boxes class type #4230
(https://github.com/infiniflow/ragflow/issues/4230)

### Type of change

- [✅ ] Bug Fix (non-breaking change which fixes an issue)

---------

Co-authored-by: youzhiqiang <zhiqiang.you@aminer.com>
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
This commit is contained in:
Zhi-Qiang You 2025-01-08 10:59:35 +08:00 committed by GitHub
parent 5e64d79587
commit b7ce4e7e62
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
deepdoc/vision/layout_recognizer.py
View File

@ -153,6 +153,8 @@ class LayoutRecognizer(Recognizer):
ocr_res = [b for b in ocr_res if b["text"].strip() not in garbag_set] ocr_res = [b for b in ocr_res if b["text"].strip() not in garbag_set]
return ocr_res, page_layout return ocr_res, page_layout
def forward(self, image_list, thr=0.7, batch_size=16):
return super().__call__(image_list, thr, batch_size)
class LayoutRecognizer4YOLOv10(LayoutRecognizer): class LayoutRecognizer4YOLOv10(LayoutRecognizer):
labels = [ labels = [

7
deepdoc/vision/t_recognizer.py
View File

@ -23,7 +23,7 @@ sys.path.insert(
'../../'))) '../../')))
from deepdoc.vision.seeit import draw_box from deepdoc.vision.seeit import draw_box
from deepdoc.vision import LayoutRecognizer, TableStructureRecognizer, OCR, init_in_out, Recognizer from deepdoc.vision import LayoutRecognizer, TableStructureRecognizer, OCR, init_in_out
import argparse import argparse
import re import re
import numpy as np import numpy as np
@ -33,9 +33,8 @@ def main(args):
images, outputs = init_in_out(args) images, outputs = init_in_out(args)
if args.mode.lower() == "layout": if args.mode.lower() == "layout":
detr = LayoutRecognizer("layout") detr = LayoutRecognizer("layout")
layouts = super(Recognizer, detr)(images, thr=float(args.threshold)) layouts = detr.forward(images, thr=float(args.threshold))
if args.mode.lower() == "tsr": if args.mode.lower() == "tsr":
labels = TableStructureRecognizer.labels
detr = TableStructureRecognizer() detr = TableStructureRecognizer()
ocr = OCR() ocr = OCR()
layouts = detr(images, thr=float(args.threshold)) layouts = detr(images, thr=float(args.threshold))
@ -50,7 +49,7 @@ def main(args):
"bbox": [t["x0"], t["top"], t["x1"], t["bottom"]], "bbox": [t["x0"], t["top"], t["x1"], t["bottom"]],
"score": t["score"] "score": t["score"]
} for t in lyt] } for t in lyt]
img = draw_box(images[i], lyt, labels, float(args.threshold)) img = draw_box(images[i], lyt, detr.labels, float(args.threshold))
img.save(outputs[i], quality=95) img.save(outputs[i], quality=95)
logging.info("save result to: " + outputs[i]) logging.info("save result to: " + outputs[i])