赵小蒙 hai 1 ano
pai
achega
3ed288b60c
Modificáronse 4 ficheiros con 24 adicións e 4 borrados
  1. 2 1
      layout/layout_spiler_recog.py
  2. 2 2
      libs/commons.py
  3. 1 1
      pre_proc/equations_replace.py
  4. 19 0
      requirements.txt

+ 2 - 1
layout/layout_spiler_recog.py

@@ -2,7 +2,8 @@
 找到能分割布局的水平的横线、色块
 找到能分割布局的水平的横线、色块
 """
 """
 
 
-import os,  fitz
+import os
+from libs.commons import fitz
 from libs.boxbase import _is_in_or_part_overlap
 from libs.boxbase import _is_in_or_part_overlap
 
 
 
 

+ 2 - 2
libs/commons.py

@@ -7,8 +7,8 @@ from loguru import logger
 from boto3.s3.transfer import TransferConfig
 from boto3.s3.transfer import TransferConfig
 from botocore.config import Config
 from botocore.config import Config
 
 
-import fitz  # 1.23.9中已经切换到rebase
-# import fitz_new as fitz  # 使用rebased的新版pymupdf库
+import fitz # 1.23.9中已经切换到rebase
+# import fitz_old as fitz  # 使用1.23.9之前的pymupdf库
 
 
 def get_delta_time(input_time):
 def get_delta_time(input_time):
     return round(time.time() - input_time, 2)
     return round(time.time() - input_time, 2)

+ 1 - 1
pre_proc/equations_replace.py

@@ -1,7 +1,7 @@
 """
 """
 对pymupdf返回的结构里的公式进行替换,替换为模型识别的公式结果
 对pymupdf返回的结构里的公式进行替换,替换为模型识别的公式结果
 """
 """
-import fitz
+from libs.commons import fitz
 import json
 import json
 import os
 import os
 from pathlib import Path
 from pathlib import Path

+ 19 - 0
requirements.txt

@@ -0,0 +1,19 @@
+boto3==1.34.52
+botocore==1.34.52
+Brotli==1.1.0
+click==8.1.7
+Distance==0.1.3
+PyMuPDF==1.23.25
+loguru==0.7.2
+matplotlib==3.8.3
+mmeval==0.2.1
+numpy==1.26.4
+pandas==2.2.1
+pycld2==0.41
+pytest==8.0.2
+regex==2023.12.25
+spacy==3.7.4
+termcolor==2.4.0
+tqdm==4.66.2
+en_core_web_sm @ https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.7.1/en_core_web_sm-3.7.1-py3-none-any.whl
+zh_core_web_sm @ https://github.com/explosion/spacy-models/releases/download/zh_core_web_sm-3.7.0/zh_core_web_sm-3.7.0-py3-none-any.whl