Helmut Grohne
42185c4beb
FeatureExtractor: avoid loading weights twice
...
When passing a model_path to the FeatureExtractor, weights are loaded twice.
One the pretrained once and then the given ones. Avoid the unnecessary load.
2022-01-19 09:25:20 +01:00
kaiyangzhou
6987d99074
fix missing link in model zoo
2021-08-27 09:19:08 +08:00
kaiyangzhou
ae241fd1c0
update readme
2021-08-11 17:00:01 +08:00
kaiyangzhou
f6ba78eceb
linting code
2021-08-11 16:59:54 +08:00
kaiyangzhou
b815df6ab0
add osnet_ain x0.75, x0.5 and x0.25
2021-08-11 16:51:32 +08:00
kaiyangzhou
b8b91acbd5
fpath = osp.abspath(osp.expanduser(fpath))
2021-08-11 16:50:56 +08:00
kaiyangzhou
1d143ad8a4
update readme
2021-07-19 09:08:09 +08:00
kaiyangzhou
1e26d7a6f0
update iLIDS-VID's link
2021-07-08 09:59:56 +08:00
kaiyangzhou
e580b699c3
update model zoo
2021-04-30 12:55:49 +08:00
kaiyangzhou
45cf03b65e
update print in parse_test_res.py
2021-04-28 16:54:06 +08:00
kaiyangzhou
b882bc8a09
add parse_test_res.py
2021-04-28 16:49:35 +08:00
kaiyangzhou
e108f6a3ef
update datasets instructions
2021-04-28 00:00:05 +08:00
kaiyangzhou
a379c72adc
linting code
2021-04-27 15:47:00 +08:00
kaiyangzhou
c4639e91a0
v1.4.0: added cuhksysu
2021-04-27 15:31:30 +08:00
kaiyangzhou
5bd26d8aa2
update doc
2021-04-25 15:35:51 +08:00
kaiyangzhou
ea41e69302
RandomIdentitySampler: n_pids should be larger than n_pids_to_sample
2021-04-11 16:04:06 +08:00
Kaiyang
849d08b22b
Merge pull request #430 from sramakrishnan247/patch-1
...
Update mudeep.py to support testing (return features if self.training=False)
2021-04-09 13:34:50 +08:00
Ramakrishnan Sundareswaran
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
kaiyangzhou
fac15fe098
update comments in the model zoo
2021-04-04 10:09:15 +08:00
kaiyangzhou
1da9383b43
update readme
2021-04-03 22:11:18 +08:00
kaiyangzhou
fce34a26b3
update model zoo links
2021-04-01 11:34:42 +08:00
kaiyangzhou
c2d9b0e825
update reference
2021-03-26 11:45:06 +08:00
kaiyangzhou
5d82902b3c
update readme
2021-03-05 11:38:16 +08:00
KaiyangZhou
93b8c9f3db
v1.3.6: added University-1652
2021-02-14 11:25:24 +08:00
Kaiyang
6e498f8b17
Merge pull request #417 from layumi/master
...
Add one new dataset University-1652
2021-02-14 11:17:33 +08:00
Zhedong Zheng
3923a2f259
Update README.rst
2021-02-10 15:06:56 +08:00
layumi
9b3a836308
add university
2021-02-10 17:59:05 +11:00
KaiyangZhou
08a993b123
update readme
2021-02-09 21:27:40 +08:00
Kaiyang
d064f4c966
Merge pull request #412 from lablabla/master
...
Adding support for rank cython in Windows, fixing pytorch deprecation warnings
2021-02-09 21:15:38 +08:00
Carmel Barak
c03125fb4b
Updated version
2021-02-03 15:54:07 +02:00
Carmel Barak
6379253fc8
Merge remote-tracking branch 'origin/master'
2021-02-03 15:45:55 +02:00
Carmel Barak
d78d9a86aa
Changed addmm_ to new pytorch API to remove deprecated warnings
2021-02-03 15:45:08 +02:00
Carmel Barak
7cedf08a5b
Changed cython types from long to int64_t to support Windows
2021-02-03 15:44:29 +02:00
KaiyangZhou
4df564c3df
update readme: add mixstyle
2021-01-26 11:08:16 +08:00
KaiyangZhou
a105924200
v1.3.4: features.data.cpu() => features.cpu().clone()
2021-01-22 16:21:43 +08:00
KaiyangZhou
454fe67abe
avoid huge memory consumption at eval
2021-01-22 16:20:51 +08:00
Kaiyang
28430ec309
Merge pull request #405 from Xuanmeng-Zhang/master
...
Add the code for GPU-Re-Ranking.
2021-01-16 20:07:09 +08:00
Xuanmeng-Zhang
9b73eb41d5
Add the GPU-Re-Ranking
2021-01-15 15:29:44 +08:00
KaiyangZhou
57dad66ce7
update readme
2021-01-08 10:01:16 +08:00
KaiyangZhou
626e6c3bb8
minor change in feature_extractor.py
2020-10-19 12:06:30 +01:00
KaiyangZhou
c4714fae69
add check_cfg()
2020-09-14 10:34:18 +01:00
KaiyangZhou
fefa31dc3f
1.3.3: fixed bug in visrank (forgot unpacking dsetid)
2020-08-18 22:17:35 +01:00
KaiyangZhou
208dc3e9b8
add self.feature_dim to osnet
2020-08-18 16:05:58 +01:00
KaiyangZhou
b6a3419da4
update readme
2020-08-12 13:32:18 +01:00
KaiyangZhou
ade7818402
upgrade to v1.3.2
2020-08-12 13:30:01 +01:00
KaiyangZhou
3143997135
add _junk_pids to grid & prid
2020-08-12 13:29:27 +01:00
KaiyangZhou
56f06d58b3
update docs/
2020-08-12 12:14:29 +01:00
KaiyangZhou
7c2ee1e6c6
update printing in triplet engine (taking into account weight_t & weight_x)
2020-08-12 11:03:24 +01:00
KaiyangZhou
de42729f41
v1.3.1: record results (rank1 & mAP) of each target dataset
2020-08-10 22:06:15 +01:00
KaiyangZhou
8bbe93ffb5
update version indicator: v1.3.0
2020-08-10 12:39:04 +01:00