qurator-spk/page2tsv
1
0
Fork 0
mirror of https://github.com/qurator-spk/page2tsv.git synced 2025-07-11 19:19:52 +02:00
Code Issues Releases Wiki Activity

store OCR or NED confidences in tsv file

Browse source
This commit is contained in:
Kai 2021-02-26 12:18:10 +01:00
parent 5d55ba24a3
commit 900015da61
2 changed files with 13 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

5
tsvtools/cli.py
View file

@ -75,9 +75,9 @@ def page2tsv(page_xml_file, tsv_out_file, purpose, image_url, ner_rest_endpoint,
ned_threshold, min_confidence, max_confidence):
if purpose == "NERD":
out_columns = ['No.', 'TOKEN', 'NE-TAG', 'NE-EMB', 'ID', 'url_id', 'left', 'right', 'top', 'bottom']
out_columns = ['No.', 'TOKEN', 'NE-TAG', 'NE-EMB', 'ID', 'url_id', 'left', 'right', 'top', 'bottom', 'conf']
elif purpose == "OCR":
out_columns = ['TEXT', 'url_id', 'left', 'right', 'top', 'bottom']
out_columns = ['TEXT', 'url_id', 'left', 'right', 'top', 'bottom', 'conf']
if min_confidence is not None and max_confidence is not None:
out_columns += ['ocrconf']
@ -168,6 +168,7 @@ def page2tsv(page_xml_file, tsv_out_file, purpose, image_url, ner_rest_endpoint,
tsv['NE-TAG'] = 'O'
tsv['NE-EMB'] = 'O'
tsv['ID'] = '-'
tsv['conf'] = '-'
tsv = tsv.rename(columns={'TEXT': 'TOKEN'})
elif purpose == 'OCR':