45b1296c25
Add cls_demo_person code
2022-05-14 09:31:52 +00:00
56b8710dc4
fix: fix bug get paddle version
2022-05-06 19:51:44 +08:00
bd524e8a88
Merge pull request #1819 from weisy11/reid
...
复现reid-strong-baseline
2022-05-06 12:05:07 +08:00
4bc4b7e0e2
Merge pull request #1876 from TingquanGao/dev/fix_amp_eval
...
fix: amp eval
2022-05-05 20:39:02 +08:00
1c31010b14
refine code
2022-05-05 20:28:59 +08:00
16f910b451
add m1 and baseline config
2022-05-05 19:55:08 +08:00
0b1481402b
merge devevelop
2022-05-05 14:42:06 +08:00
bb13f3c4f5
fix single card dist ( #1889 )
...
* fix single card logit
* fix distillation yaml files
2022-05-05 09:48:56 +08:00
275945dff9
fix: compatible with Paddle 2.2, 2.3, and develop.
2022-04-29 10:21:09 +00:00
24372cc6e5
update: update the default gpu num to 8 when using AdamW
2022-04-29 17:31:12 +08:00
59a3dcfc1c
fix: amp eval
2022-04-26 12:26:17 +00:00
fea9522a69
fix: dbg
2022-04-26 12:07:59 +08:00
97e8abc3db
refine code and yaml
2022-04-24 14:38:26 +08:00
4e6c36e269
Merge pull request #1833 from TingquanGao/dev/fix_dist_loss
...
fix calc metric error and calc loss error in distributed.
2022-04-22 22:41:57 +08:00
b86ddf470e
fix: when eval only, AMP can be enabled
2022-04-22 17:49:44 +08:00
b5a7eea504
fix: fix the bug that DistributedBatchSampler may sample repeatedly
2022-04-22 09:39:22 +00:00
15a0afc67c
update code
2022-04-22 15:52:16 +08:00
bc6e7f5cf5
update code
2022-04-22 15:26:05 +08:00
9dde9376fb
update code
2022-04-22 15:25:11 +08:00
15997ea9d0
update code
2022-04-22 12:03:17 +08:00
f2982e5e47
update code
2022-04-22 12:01:05 +08:00
17fd1bc2c0
refine code
2022-04-22 12:00:03 +08:00
aa26a8c1d8
fix bug for static graph
2022-04-21 16:31:28 +08:00
daf7eea23d
fix code
2022-04-21 15:43:53 +08:00
dfd7749828
refine hard code
2022-04-21 14:37:00 +08:00
41e1a86caf
add center loss
2022-04-21 00:17:54 +08:00
e6feb68bb8
Merge pull request #1824 from TingquanGao/dev/spt_amp_eval
...
fix: fp32 eval by default when enable amp
2022-04-20 14:40:46 +08:00
3a1276d315
train_loss_func only used in train mode
2022-04-19 19:54:48 +08:00
24abea151a
support for multi optimizer case
2022-04-19 14:26:42 +08:00
83ed5195c3
fix: set use_fp16_test to True when AMP O2 is enabled
2022-04-18 06:14:43 +00:00
1789da6422
fix bug
2022-04-18 11:26:32 +08:00
a35cdd2aec
uncommit: sync bn is too slow to use and convert_sync_batchnorm() is not effective for BatchNorm
2022-04-14 08:19:39 +00:00
13d5e59051
fix: convert bn to sync_bn
...
the running_mean and running_var of bn would not be synchronized in dist,
so which leads to bug that eval loss in training is inconsistent with eval only.
2022-04-14 07:36:39 +00:00
efde56ffc6
fix: only fp16 evaluation is supported when ampO2 is enabled
2022-04-13 12:14:14 +00:00
474c918b27
fix: fix bug of batch_size statistics error
2022-04-13 09:19:30 +00:00
c46189bad0
fix: fix bug about calc loss in dist
2022-04-12 06:56:44 +00:00
af90cd7c59
update center loss config and related code
2022-04-12 13:07:53 +08:00
9de22673df
dbg
2022-04-08 14:29:03 +08:00
b761325faa
fix: fp32 eval by default when enable amp
...
If you want to eval by fp16 when enable amp, please set Amp.use_fp16_test=True, False by default.
2022-04-02 19:22:10 +08:00
a944603da0
fix log twice bug
2022-03-30 08:31:35 +00:00
b62b98d79f
feat: support mlu device and amp of mlu
2022-03-14 15:48:26 +08:00
f68c098a4a
fix train acc log
2022-03-09 19:58:36 +08:00
7595ba6d70
add AFD ( #1683 )
...
* add AFD
2022-02-28 19:11:50 +08:00
dc6281a6d4
add benchmark for tipc
2022-02-10 08:25:52 +00:00
42134cd8dd
fix: raise warning when using Global.class_num
2022-01-25 15:06:36 +08:00
bb6581d21b
refactor: raise warning when gpu numbers is not 4
2022-01-25 15:06:36 +08:00
8f0bd5b582
fix: fix vdl makedir
2022-01-25 15:06:36 +08:00
10c93c55d1
fix: enable amp only in training
2022-01-25 11:58:07 +08:00
7040ce8314
refactor: change params to be consistent with amp
2022-01-25 11:58:07 +08:00
cd039a7b37
add save_dtype
2022-01-10 18:19:03 +08:00
cuicheng01
gaotingquan
Walter
Wei Shengyu
HydrogenSulfate
littletomatodonkey
cuicheng01
huangqipeng
WangChen0902
zhangbo9674