Commit Graph

1817 Commits (496011d09b0b07a92d92af94a656eda761a17f82)

Author SHA1 Message Date
zhangyubo0722 496011d09b
fix fused_attn with fp32 (#3364) 2025-04-22 15:04:26 +08:00
gaotingquan d15bbf82cf save the inference model in json format by default 2025-04-17 16:11:08 +08:00
warrentdrew e5972f75ef pphgnet v2 support xpu3 2025-03-05 19:21:16 +08:00
gaotingquan 45b130fd6f fix ZeroDivisionError 2025-02-25 19:27:50 +08:00
drryanhuang c9ab9c7e6a Swap the order of ema and to_static 2025-02-19 19:45:34 +08:00
gaotingquan b76c7ec696 add dcu 2025-02-12 06:13:35 +00:00
hong bde3844bc4
remove max inplace grad add (#3346) 2025-01-26 17:19:10 +08:00
zhangyubo0722 4b3e6021e4
add version control for export (#3339) 2025-01-08 17:53:01 +08:00
zhangyubo0722 0e90ae5e54
modify export with pir (#3332) 2024-12-30 17:03:56 +08:00
Tingquan Gao bd23af2e7b
fix ZeroDivisionError (#3324) 2024-12-24 14:40:26 +08:00
zhangyubo0722 d821d43c55
support export with pir and no pir (#3319) 2024-12-19 19:38:24 +08:00
学卿 070716eef1
fix resnet50_face metric (#3310)
* fix resnet50_face metric

* fix face rec model to_static bug
2024-12-10 13:57:00 +08:00
EnflameGCU 1850bca5d7
[GCU] Support GCU backend (#3301) 2024-11-20 11:03:15 +08:00
zhangyubo0722 77ddc6fd7e rename_train_result 2024-11-14 15:00:52 +08:00
zhangyubo0722 4e2e0fe51f
update hpi config (#3281) 2024-11-08 17:38:13 +08:00
Sunflower7788 1cde9bce0c
rm pdmodel (#3287)
* rm pdmodel

* fix params
2024-11-05 17:19:22 +08:00
学卿 4b3503b6de
support face recgnition (#3282)
* add mobilefacenet

* exp with pplcnet on face

* support face recognition

* fix face recognition models bugs

* add face recognition README
2024-11-01 19:48:28 +08:00
zhangyubo0722 d178931fca fix gpu memory growth 2024-10-18 17:05:11 +08:00
zhangyubo0722 542b320008 fix export bug 2024-09-26 01:32:49 +08:00
zhangyubo0722 7f35c77027 fix_uniform 2024-09-25 23:30:23 +08:00
zhangyubo0722 a9a730e3b3 support uniform output 2024-09-24 23:25:41 +08:00
Sunflower7788 a7b235682c
fix flash attn mm (#3257) 2024-09-23 20:52:45 +08:00
Sunflower7788 f064291d6b
cls_config_fix (#3253)
* cls_config_fix

* fix flash attn
2024-09-23 15:00:04 +08:00
changdazhou 37c1ddb0dd
update default lable_dict for attribute (#3255) 2024-09-23 13:02:48 +08:00
cuicheng01 c736f5e0f5
update attr configs (#3254) 2024-09-22 19:36:29 +08:00
Sunflower7788 ddda8b688e
add 3 new models : MV4 FasterNet StarNet (#3251) 2024-09-21 00:26:38 +08:00
changdazhou 96e1404464
update default label list for multi-label image_classification (#3240) 2024-09-19 22:13:07 +08:00
cuicheng01 094349663a
update ShiTu Rec configs (#3246) 2024-09-13 15:50:05 +08:00
cuicheng01 716fbcc574
update config for ShiTu_rec (#3245)
* update config for ShiTu_rec

* update for ShiTu_Rec
2024-09-12 18:27:01 +08:00
cuicheng01 de0f57521d
update CLIP configs for PP-ShiTuV2-rec (#3239) 2024-09-05 10:49:04 +08:00
zhouchangda 104b378828 modify ema.pdparams to pdema 2024-09-04 15:20:28 +08:00
zhouchangda 9c37234cfb update config file for multi-label image classification 2024-08-29 17:45:23 +08:00
Tingquan Gao bc9401f861
fix: there may not be weight ratio in multilabel dataset label (#3226) 2024-08-23 11:35:56 +08:00
changdazhou 666ca9bbd1
support multi-label image classification (#3216) 2024-08-19 14:50:50 +08:00
changdazhou 617752e2b4
update dump infer config scale format (#3211) 2024-08-08 20:01:44 +08:00
zhangyubo0722 fe5700fce6
export with label (#3166) 2024-07-25 21:31:21 +08:00
Zheng-Bicheng 5dde31371c
Add support for custom scale and act (#3189) 2024-07-24 19:49:09 +08:00
Nyakku Shigure 6aa3047157
Respect `is_rec` when eval (#3187) 2024-07-10 15:45:37 +08:00
Nyakku Shigure caa6fdf57d
Fix missing label for classification model in eval mode (#3185) 2024-07-09 17:02:15 +08:00
Nyakku Shigure d6df377f44
Mark some models as rec model in yaml (#3182) 2024-07-08 17:17:53 +08:00
Tingquan Gao 587698a9df
bugfix (#3179) 2024-07-05 16:17:42 +08:00
Tingquan Gao 91e8eb3632
update to be compatible with V100 (#3178) 2024-07-05 14:52:32 +08:00
Nyakku Shigure 1dcab0a7bc
Fix wrong InputSpec setting for AST Dy2St (#3176) 2024-07-05 14:05:02 +08:00
Tingquan Gao b1ee8f911b
update to be compatible with V100 (#3177) 2024-07-04 16:56:04 +08:00
Nyakku Shigure cc24ead0ad
Use `inspect.getfullargspec` instead of deprecated `inspect.getargspec` (#3168) 2024-06-20 20:16:36 +08:00
zhuyipin cea80378ac
adapt AdaptiveAvgPool2D for npu for PPHGNet (#3162) 2024-06-17 21:07:16 +08:00
Tingquan Gao 00e018c18c
support FusedLinear (#3159) 2024-06-04 11:15:47 +08:00
gaotingquan 87ad22342e add use_fused_attn field 2024-05-29 11:24:48 +08:00
gaotingquan 46c1a0168c set use_fused_op False by default 2024-05-29 11:24:48 +08:00
gaotingquan 0bfed92cb2 perf:
1. use nn.GELU instead of QuickGELU
2. support FusedLinear
2024-05-29 11:24:48 +08:00