File tree Expand file tree Collapse file tree 2 files changed +6
-6
lines changed Expand file tree Collapse file tree 2 files changed +6
-6
lines changed Original file line number Diff line number Diff line change @@ -425,7 +425,7 @@ python3 tools/eval.py \
425
425
| 4 | PP-OCRv3中英文超轻量识别预训练模型 + fine-tune + 增加PCB图像数量 | 99.99% | +23.00% | 如果能获取更多数据量的情况,可以通过增加数据量提升效果 |
426
426
427
427
```
428
- 注:上述实验结果均是在1,500张图片(1,200张训练集 ,300张测试集)、2W张图片、添加公开通用识别数据集上训练、评估的得到,AIstudio只提供了100张数据,所以指标有所差异属于正常,只要策略有效、规律相同即可。
428
+ 注:上述实验结果均是在1500张图片(1200张训练集 ,300张测试集)、2W张图片、添加公开通用识别数据集上训练、评估的得到,AIstudio只提供了100张数据,所以指标有所差异属于正常,只要策略有效、规律相同即可。
429
429
```
430
430
431
431
# 6. 模型导出
Original file line number Diff line number Diff line change 53
53
54
54
DEFAULT_OCR_MODEL_VERSION = 'PP-OCRv3'
55
55
SUPPORT_OCR_MODEL_VERSION = ['PP-OCR' , 'PP-OCRv2' , 'PP-OCRv3' ]
56
- DEFAULT_STRUCTURE_MODEL_VERSION = 'PP-Structurev2 '
57
- SUPPORT_STRUCTURE_MODEL_VERSION = ['PP-Structure' , 'PP-Structurev2 ' ]
56
+ DEFAULT_STRUCTURE_MODEL_VERSION = 'PP-StructureV2 '
57
+ SUPPORT_STRUCTURE_MODEL_VERSION = ['PP-Structure' , 'PP-StructureV2 ' ]
58
58
MODEL_URLS = {
59
59
'OCR' : {
60
60
'PP-OCRv3' : {
272
272
}
273
273
}
274
274
},
275
- 'PP-Structurev2 ' : {
275
+ 'PP-StructureV2 ' : {
276
276
'table' : {
277
277
'en' : {
278
278
'url' :
@@ -326,10 +326,10 @@ def parse_args(mMain=True):
326
326
"--structure_version" ,
327
327
type = str ,
328
328
choices = SUPPORT_STRUCTURE_MODEL_VERSION ,
329
- default = 'PP-Structurev2 ' ,
329
+ default = 'PP-StructureV2 ' ,
330
330
help = 'Model version, the current model support list is as follows:'
331
331
' 1. PP-Structure Support en table structure model.'
332
- ' 2. PP-Structurev2 Support ch and en table structure model.' )
332
+ ' 2. PP-StructureV2 Support ch and en table structure model.' )
333
333
334
334
for action in parser ._actions :
335
335
if action .dest in [
You can’t perform that action at this time.
0 commit comments