mirrors
/
PaddleOCR
mirror of https://github.com/PaddlePaddle/PaddleOCR.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
6,451 Commits
19 Branches
16 Tags
1.1 GiB
Commit Graph

142 Commits (main)

Author SHA1 Message Date
Sunflower7788 e752487f2f
fix_ppocrv3_dy2st_config (#14721) 2025-02-19 19:28:48 +08:00
Sunflower7788 bb7e24eea3
update_det_static (#14372) 2024-12-11 20:39:21 +08:00
wangna11BD 661cda1289
fix nan in ppocrv4 for benchmark (#14072) 
* fix nan in ppocrv4 for benchmark

* fix config
 2024-10-23 11:55:43 +08:00
jzhang533 24f06d1a1b
update common pre-commit configs and commit the results of running pre-commit run -a (#12516) 2024-05-29 15:26:09 +08:00
Mattheliu 960243862f
Update ch_PP-OCRv4_det_cml.yml (#12140) 2024-05-20 10:40:03 +08:00
topduke 38c0c9ee77
openocr compti code (#12033) 
* openocr compti code

* update config and repsvtr

* svtrv2 doc
 2024-05-15 14:40:26 +08:00
sylarwcy 68b384292b
v4 det cml configs (#11258) 
* fixed several bugs

1. 修复了找不到模型的问题，原因是PPLCNetNew不在可选模型名称的列表内，将PPLCNetNew改为PPLCNetV3；
2. 修复了db_fpn.py解析in_channels的报错问题，db_fpn.py中按序列解析in_channels，而lcnetv3.py中det为false时输出是数值，为true才返回序列。在两个student模型的backbone中添加"det: true"；
3. 减少CPU占用，将cal_metric_during_train: true 改为 false；
4. 修复了训练过程中eval时的显存溢出问题。通过限制过大的测试数据可解决该问题，具体调整是，在eval→DetResizeForTest的配置中增加"limit_side_len: 960,limit_type: max"。

* 恢复cal_metric_during_train的设置
 2023-11-17 20:38:25 +08:00
Yuchen-Su bb616dbfba
add config of 'ch_PP-OCRv4_det_cml' (#10483) 
* config of ch_PP-OCRv4_det_cml

* Update ch_PP-OCRv4_det_cml .yml

* add config of ch_PP-OCRv4_det_cml

* Delete ch_PP-OCRv4_det_cml .yml

* Add files via upload
 2023-07-26 18:57:07 +08:00
Zhang Ting a288a3aac3
fix det_v3 (#10180) 2023-06-16 10:16:32 +08:00
Zhang Ting 6949448558
improve amp training (#10119) 2023-06-08 15:50:37 +08:00
huangjun12 0e9c6630ee
fix det v4 bug in dynamic ratio (#9874) 
* fix set bug

* refine 960 to 640

* fix details

* add epoch num

* add rep export

* use db head
 2023-05-24 14:59:36 +08:00
Double_V 24ff4def48
Pfhead (#9898) 
* support min_area_rect crop

* add check_install

* fix requirement.txt

* fix check_install

* add lanms-neo for drrg

* fix

* fix doc

* fix

* support set gpu_id when inference

* fix #8855

* fix #8855

* opt slim doc

* fix doc bug

* rename

* rename
 2023-05-15 10:57:30 +08:00
huangjun12 ded374035c
add v4 det pretrain link (#9829) 
* add v4 det pretrain link

* update size to 960
 2023-04-27 17:08:42 +08:00
xiaoting 26519a6d17
update PPLCNetV3 name (#9802) 2023-04-23 16:13:01 +08:00
huangjun12 ca8c8200ba
add PP-OCRv4 det code (#9766) 
* add ppocrv4 det student and teacher model

* update head and config, refine details

* refine config and head details

* refine config and head details

* refine details

* refine details

* remove application

* refine fpn

* fix bug

* update code

* fix bug

* align lcnet to rec

* align hgnet to rec

* refine make shrink

* remove theseus layer
 2023-04-21 18:10:26 +08:00
zhoujun 2e05d54af8
add d2s train for slanet and v3 (#9341) 
* add d2s train for slanet and v3

* fix bug
 2023-03-09 11:21:34 +08:00
zhoujun 623424fce0
support auto download model from bos (#9349) 2023-03-08 19:21:28 +08:00
zhoujun ce72023ce1
Merge pull request #7832 from zhiminzhang0830/drrg_branch 
add text detection algorithm DRRG
 2022-10-13 13:19:13 +08:00
Double_V 9df7730eeb
Merge pull request #7840 from LDOUBLEV/dygraph 
add polygon params
 2022-10-12 10:14:16 +08:00
zhiminzhang0830 3ee146fa99 Merge branch 'dygraph' of https://github.com/PaddlePaddle/PaddleOCR into drrg_branch 2022-10-10 15:56:08 +08:00
zhiminzhang0830 67ae525e95 删除backbone里的pretrained_model字段 2022-10-10 14:20:28 +08:00
LDOUBLEV 119196f11d add det_box_type params 2022-10-09 11:28:32 +08:00
huangqipeng 7851977157 [MLU]adapt mlu device for running dbnet network 2022-10-08 16:46:41 +08:00
zhiminzhang0830 1f9400dd73 add drrg 2022-10-08 11:56:45 +08:00
huangjun12 b5268dc3a0 add centripetal text model 2022-09-15 11:08:16 +00:00
LDOUBLEV f630700a99 fix cml 2022-08-29 14:03:05 +08:00
WenmuZhou 0a247f02d4 add exp to PP-OCRv2 det amp custom_black_list 2022-08-19 08:31:57 +00:00
WenmuZhou dd1e17fd79 add benckmark 2022-07-30 10:00:11 +08:00
Double_V 2d722451ab
Merge pull request #6870 from wangjingyeye/dyg_poly 
debug
 2022-07-13 21:05:40 +08:00
wangjingyeye 94a4416426 debug 2022-07-13 06:47:41 +00:00
LDOUBLEV 24712d933c fix 2022-07-01 18:28:02 +08:00
LDOUBLEV 7c4d0d1e91 fix kl js loss 2022-07-01 18:05:39 +08:00
wangjingyeye 04e7104194 add db++ 2022-06-28 11:45:21 +00:00
wangjingyeye 9422629cc0 add db++ 2022-06-28 04:17:14 +00:00
wangjingyeye 0ae4f66461 add db++ 2022-06-28 02:42:47 +00:00
wangjingyeye 679ec78bb2 add db++ 2022-06-27 10:01:35 +00:00
wangjingyeye 26a89db76a add resnet 2022-06-27 08:27:26 +00:00
wangjingyeye 961dca72cf add yml 2022-06-27 08:21:15 +00:00
wangjingyeye bc29ec0ba8 add yml 2022-06-27 08:17:49 +00:00
wangjingyeye 5ba74d5627 add yml 2022-06-27 08:16:10 +00:00
wangjingyeye 5f990d9cda add yml 2022-06-27 08:11:31 +00:00
Double_V 6c68808a3f
fix dml (#6067) 2022-05-01 19:54:22 +08:00
LDOUBLEV 68da674dfd fix distill 2022-04-29 10:23:51 +08:00
LDOUBLEV c9129b63ed fix 2022-04-29 09:56:48 +08:00
LDOUBLEV b077e3c559 rename 2022-04-27 19:57:48 +08:00
LDOUBLEV af4aa1aa2a fix postprocess 2022-04-27 19:06:24 +08:00
LDOUBLEV 591b92e8b5 delete teacher pretrain 2022-04-27 13:44:30 +08:00
LDOUBLEV b6cd60476c cml config 2022-04-27 10:55:21 +08:00
LDOUBLEV 62a01f7e30 fix 2022-04-26 17:29:38 +08:00
LDOUBLEV 217b0409cd fix det 2022-04-26 14:19:08 +08:00