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

Optimized Recognizer.sort_X_firstly and Recognizer.sort_Y_firstly (#5182)

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

Optimized Recognizer.sort_X_firstly and Recognizer.sort_Y_firstly

### Type of change

- [x] Performance Improvement
This commit is contained in:
Zhichang Yu
2025-02-20 15:41:12 +08:00
committed by GitHub
parent 07ddb8fcff
commit c326f14fed
3 changed files with 16 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
deepdoc/vision/recognizer.py
View File

@ -19,7 +19,7 @@ import os
import math
import numpy as np
import cv2
from copy import deepcopy
from functools import cmp_to_key
import onnxruntime as ort
from huggingface_hub import snapshot_download
@ -99,30 +99,22 @@ class Recognizer(object):
@staticmethod
def sort_Y_firstly(arr, threashold):
# sort using y1 first and then x1
arr = sorted(arr, key=lambda r: (r["top"], r["x0"]))
for i in range(len(arr) - 1):
for j in range(i, -1, -1):
# restore the order using th
if abs(arr[j + 1]["top"] - arr[j]["top"]) < threashold \
and arr[j + 1]["x0"] < arr[j]["x0"]:
tmp = deepcopy(arr[j])
arr[j] = deepcopy(arr[j + 1])
arr[j + 1] = deepcopy(tmp)
def cmp(c1, c2):
diff = c1["top"] - c2["top"]
if abs(diff) < threashold:
diff = c1["x0"] - c2["x0"]
return diff
arr = sorted(arr, key=cmp_to_key(cmp))
return arr
@staticmethod
def sort_X_firstly(arr, threashold, copy=True):
# sort using y1 first and then x1
arr = sorted(arr, key=lambda r: (r["x0"], r["top"]))
for i in range(len(arr) - 1):
for j in range(i, -1, -1):
# restore the order using th
if abs(arr[j + 1]["x0"] - arr[j]["x0"]) < threashold \
and arr[j + 1]["top"] < arr[j]["top"]:
tmp = deepcopy(arr[j]) if copy else arr[j]
arr[j] = deepcopy(arr[j + 1]) if copy else arr[j + 1]
arr[j + 1] = deepcopy(tmp) if copy else tmp
def sort_X_firstly(arr, threashold):
def cmp(c1, c2):
diff = c1["x0"] - c2["x0"]
if abs(diff) < threashold:
diff = c1["top"] - c2["top"]
return diff
arr = sorted(arr, key=cmp_to_key(cmp))
return arr
@staticmethod
@ -145,8 +137,6 @@ class Recognizer(object):
arr[j + 1] = tmp
return arr
return sorted(arr, key=lambda r: (r.get("C", r["x0"]), r["top"]))
@staticmethod
def sort_R_firstly(arr, thr=0):
# sort using y1 first and then x1