32acbbe4ad
Update mudeep.py to support testing
...
Current model fails in the case of triplet loss during testing since y,x returns a tuple where engine.py features expects a torch tensor instead to run features.cpu()
2021-04-08 11:31:27 -05:00
c2d9b0e825
update reference
2021-03-26 11:45:06 +08:00
208dc3e9b8
add self.feature_dim to osnet
2020-08-18 16:05:58 +01:00
5ce43185b1
linting code
2020-05-05 15:58:00 +01:00
188b7f321f
fixed missing import in osnet
2019-12-02 22:24:24 +00:00
08bb2cc5eb
linting and formatting code
2019-12-02 14:34:59 +00:00
6b8fe56638
linting and formatting code
2019-12-01 02:35:44 +00:00
aebb6033e0
add ibn-net
2019-11-10 21:31:28 +00:00
a136384a6a
rm unused imports/variables; fixed minor bugs
2019-10-23 17:15:37 +01:00
edf0b73e21
add OSNet-AIN code
2019-10-22 22:48:53 +01:00
894de59bff
add new ref to osnet.py
2019-10-22 22:46:30 +01:00
5365619606
add config
2019-08-26 10:34:31 +01:00
1558389f37
osnet.py: rename residual->identity; engine: add num_batches=len(dloader)
2019-08-21 23:02:05 +01:00
a20f8a66bd
output_featuremaps_only -> return_featuremaps
2019-08-05 14:01:39 +01:00
d72b85a1b8
set output_featuremaps_only=False in osnet
2019-08-03 21:52:52 +01:00
7e0249a084
add output_featuremaps_only to foward() in osnet
2019-08-03 18:25:34 +01:00
27d4f2f093
add visactmap code; upgrade visrank to plot single figure
2019-08-03 17:35:06 +01:00
60d27cf9e9
add osnet code
2019-07-03 13:37:25 +01:00
554807ad36
update mobilenetv2_1dot0 => mobilenetv2_x1_0 and the same for x1_4
2019-07-03 11:33:10 +01:00
752cb82682
change attribute name: last_linear -> classifier
2019-06-03 09:26:04 +01:00
9963cff629
update imagenet weights path
2019-05-28 15:33:04 +01:00
e3c326677a
add public keys to comments
2019-05-24 16:15:55 +01:00
fd88366d22
rm old resnext
2019-05-24 15:35:54 +01:00
97ee4b423b
merge resnet and resnext; add shufflenetv2
2019-05-24 15:35:09 +01:00
28b4580a0d
remove print in internal wrapper
2019-04-23 11:35:22 +01:00
8e5265099d
change weight path
2019-03-25 09:29:06 +00:00
4ca182dbb8
add model public keys
2019-03-24 15:52:39 +00:00
aa98ef5a7b
add docstrings to losses, metrics, models, utils, optim
2019-03-22 00:14:41 +00:00
37f92b7cc3
create engine SDK
2019-03-19 17:26:08 +00:00
03d0674dcb
comply woth PEP8
2019-03-15 23:17:38 +00:00
850da2bbed
update path to pretrained weights of resnext
2019-03-12 16:46:49 +00:00
7ff331f871
update mobilenetv2; add imagenet/reid weights
2019-03-05 10:56:39 +00:00
859b03cf19
set pretrained=True instead of pretrained="imagenet"
2019-02-19 11:10:45 +00:00
564f5a26ab
update print
2019-01-30 22:41:47 +00:00
c05f96e31b
add link to shufflenet's imagenet weights
2018-11-26 22:07:20 +00:00
47b883d0d6
bug fix: rm resnext101_32x4d
2018-11-26 22:06:33 +00:00
affeaa847d
add resnext50_32x4d's imagenet weights
2018-11-23 10:31:38 +00:00
d7755cfd16
add resnext50_32x4d_fc512
2018-11-22 12:41:22 +00:00
0545358b6f
mv resnet50mid to reid-specific model list
2018-11-09 16:17:29 +00:00
81d3cc3995
fixed division error: / -> //
2018-11-05 21:30:40 +00:00
39df4855c1
add mlfn
2018-11-04 00:18:02 +00:00
e302af017d
recover dropout in pcb
2018-11-01 17:28:15 +00:00
752a4fe498
update pcb: commented dropout
2018-10-28 22:30:19 +00:00
7588c8f581
update pcb: add support to various resnet archs
2018-10-28 11:44:09 +00:00
12cdbf96b5
update pcb: mv conv5 after eval-mode in forward
2018-10-28 11:30:35 +00:00
bfba6389d6
add pcb
2018-10-28 11:16:42 +00:00
f77b9ada10
fill space after comma
2018-10-27 23:22:23 +01:00
bc942aaf09
add variable model_url
2018-10-27 22:50:22 +01:00
3ddf9ce699
build new classifier; add squeezenet1_0_fc512
2018-10-27 22:49:53 +01:00
ee5bcee63f
add delimiter for lightweight models
2018-10-27 17:59:10 +01:00
Ramakrishnan Sundareswaran
kaiyangzhou
KaiyangZhou
KaiyangZhou