mirrors/mmocr
1
0
Fork 0
mirror of https://github.com/open-mmlab/mmocr.git synced 2025-06-03 21:54:47 +08:00
Code Issues Projects Releases Wiki Activity

[Update] Update data_migrator to suit MJ dataset

Browse Source
This commit is contained in:
jiangqing.vendor 2022-05-30 12:01:47 +00:00 committed by gaotongxiao
parent f1eebe9e34
commit 8ac235677e
1 changed files with 14 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
tools/data/textrecog/data_migrator.py
View File

@ -33,6 +33,10 @@ def parse_legacy_data(in_path: str,
file_path, label = line_json['filename'], line_json['text']
file_path = strip_cls(file_path)
label = strip_cls(label)
# MJ's file_path starts with './'
if file_path.startswith('./'):
file_path = file_path[2:]
file_paths.append(file_path)
labels.append(label)
return file_paths, labels
@ -42,12 +46,21 @@ def parse_legacy_data(in_path: str,
for line in f:
line = strip_cls(line)
file_path, label = line.split()[:2]
# MJ's file_path starts with './'
if file_path.startswith('./'):
file_path = file_path[2:]
file_paths.append(file_path)
labels.append(label)
elif format == 'jsonl':
for line in f:
datum = json.loads(line)
file_paths.append(datum['filename'])
file_path = datum['filename']
# MJ's file_path starts with './'
if file_path.startswith('./'):
file_path = file_path[2:]
file_paths.append(file_path)
labels.append(datum['text'])
return file_paths, labels