HydrogenSulfate
5a4874079d
fix some bugs
2022-08-23 10:59:30 +00:00
HydrogenSulfate
1b5e00e82a
add PP-ShiTuV2 code
2022-08-23 07:54:58 +00:00
Yang Nie
3248cc5819
update `DecodeImage`
2022-07-09 23:30:13 +08:00
Yang Nie
436f64ad84
Merge branch 'develop' into ConvNeXt
2022-07-09 21:50:31 +08:00
HydrogenSulfate
3b1247d069
fix bug in RandCropImageV2
2022-06-20 10:53:19 +08:00
littletomatodonkey
def286bac8
add support for no dist ( #1989 )
2022-06-07 17:25:22 +08:00
Walter
dcef409dde
Merge branch 'develop' into adaface
2022-05-27 14:19:03 +08:00
dongshuilong
759fe41ccc
modify face_dataset and ir_net name issue
2022-05-19 02:22:17 +00:00
zhiboniu
50900443f3
remove strongbaseline_attr, etc...
2022-05-18 04:54:43 +00:00
zhiboniu
aa8f4c16d2
attr ma test ok
2022-05-18 04:54:43 +00:00
dongshuilong
32c99be6aa
add adaface
2022-05-16 03:50:35 +00:00
Yang Nie
e61f392598
Merge branch 'develop' into ConvNeXt
2022-05-09 12:22:37 +08:00
flytocc
b38240b86c
add `pil` backend for DecodeImage
2022-05-09 11:51:59 +08:00
HydrogenSulfate
790815f430
refine code and docs
2022-05-05 22:14:07 +08:00
HydrogenSulfate
1c31010b14
refine code
2022-05-05 20:28:59 +08:00
HydrogenSulfate
16f910b451
add m1 and baseline config
2022-05-05 19:55:08 +08:00
HydrogenSulfate
dfd7749828
refine hard code
2022-04-21 14:37:00 +08:00
flytocc
3f84dcb44e
add random interpolation for UnifiedResize
2022-04-19 13:25:13 +08:00
dongshuilong
94cce6422f
fix augmix bug
2021-12-08 07:53:33 +00:00
cuicheng01
08952deb88
update codes
2021-11-04 10:47:13 +00:00
gaotingquan
0dccfb917d
fix: interpolation maybe be 0
2021-09-30 06:49:41 +00:00
gaotingquan
de859b4a3d
fix: compatible with opencv under version 4.4.0
2021-09-26 13:22:38 +08:00
gaotingquan
1f8cfbd69d
fix: fix augmentation
...
Fix RandomErasing, RandAugment to be consistent with Timm and compatible with earlier PaddleClas.
Add ColorJitter implemented by PaddleVision and TimmAutoAugment borrowed from Timm Lib.
2021-09-17 11:00:00 +00:00
gaotingquan
64c370008b
feat: support pil resize
...
Support PIL resizse with PIL interpolation to train transformer.
Almost all vision transformer models need using PIL.Image.BICUBIC
as interpolation in resize.
2021-09-15 06:29:22 +00:00
littletomatodonkey
9d9cd3719e
add static training ( #1037 )
...
* add static training
* fix typo
* add se fp16
* rm note
* fix loader
* fix cfg
2021-07-15 10:30:07 +08:00
dongshuilong
bba0cf8fdb
add CompCars train
2021-06-02 20:04:24 +08:00
Felix
ea08c644bd
Update operators.py
2021-05-31 15:58:46 +08:00
Felix
1d9c5710cc
Add files via upload
2021-05-31 12:57:18 +08:00