Phil Wang
|
04f86dee3c
|
implement SOTA new self-supervised learning technique from facebook for vision transformers, Dino
0.18.1
|
2021-05-02 14:00:36 -07:00 |
|
Phil Wang
|
6549522629
|
be able to accept non-square patches, thanks to @FilipAndersson245
0.17.3
|
2021-05-01 20:04:41 -07:00 |
|
Phil Wang
|
6a80a4ef89
|
update readme
|
2021-05-01 11:51:35 -07:00 |
|
Phil Wang
|
9f05587a7d
|
0.17.2
0.17.2
|
2021-04-30 06:44:59 -07:00 |
|
Phil Wang
|
65bb350e85
|
0.17.2
|
2021-04-30 06:44:54 -07:00 |
|
Phil Wang
|
fd4a7dfcf8
|
Merge pull request #102 from jon-tow/rvt-add-use-glu-flag
Add `use_glu` flag to `RvT`
|
2021-04-30 06:44:41 -07:00 |
|
Jonathan Tow
|
6f3a5fcf0b
|
Add use_glu flag to RvT
|
2021-04-30 02:07:41 -04:00 |
|
Phil Wang
|
7807f24509
|
fix small bug
|
2021-04-29 15:39:41 -07:00 |
|
Phil Wang
|
a612327126
|
readme
|
2021-04-29 15:22:12 -07:00 |
|
Phil Wang
|
30a1335d31
|
release twins svt
|
2021-04-29 14:55:25 -07:00 |
|
Phil Wang
|
ab781f7ddb
|
add Twins SVT (small)
|
2021-04-29 14:54:06 -07:00 |
|
Phil Wang
|
4f3dbd003f
|
for PiT, project to increased dimensions on first grouped conv for depthwise-conv
0.16.13
|
2021-04-29 12:41:00 -07:00 |
|
Phil Wang
|
60b5687a79
|
cleanup rvt
|
2021-04-27 11:45:46 -07:00 |
|
Phil Wang
|
0df1505662
|
add zeroing of weight parameters of batchnorm in levit just before residual connection, noticed by @EelcoHoogendoorn
0.16.12
|
2021-04-27 08:41:16 -07:00 |
|
Phil Wang
|
3df6c31c61
|
fix norm issues in cvt
0.16.11
|
2021-04-27 08:36:17 -07:00 |
|
Phil Wang
|
54af220930
|
fix cvt
0.16.10
|
2021-04-26 20:37:51 -07:00 |
|
Phil Wang
|
bad4b94e7b
|
fix all issues with rotary vision transformer
0.16.9
|
2021-04-25 12:09:32 -07:00 |
|
Phil Wang
|
fbced01fe7
|
cite
|
2021-04-20 18:36:54 -07:00 |
|
Phil Wang
|
e42e9876bc
|
offer a way to turn off ds conv in rotary vision transformer for ablation
0.16.8
|
2021-04-20 10:12:03 -07:00 |
|
Phil Wang
|
566365978d
|
add ability to turn off rotary, for ablation
0.16.7
|
2021-04-20 09:00:27 -07:00 |
|
Phil Wang
|
34f78294d3
|
fix pooling bugs across a few new archs
0.16.6
|
2021-04-19 22:36:23 -07:00 |
|
Phil Wang
|
4c29328363
|
fix frequency in rotary vision transformer
0.16.4
|
2021-04-15 16:06:32 -07:00 |
|
Phil Wang
|
27ac10c1f1
|
0.16.3
0.16.3
|
2021-04-14 16:53:05 -07:00 |
|
Phil Wang
|
fa216c45ea
|
tweak
|
2021-04-14 16:52:53 -07:00 |
|
Phil Wang
|
1d8b7826bf
|
update personal pet vit
0.16.2
|
2021-04-14 15:56:39 -07:00 |
|
Phil Wang
|
53b3af05f6
|
use convolution on query with padding to give the network absolute spatial awareness in addition to relative encoding from rotary embeddings
|
2021-04-14 15:56:02 -07:00 |
|
Phil Wang
|
6289619e3f
|
0.16.1
0.16.1
|
2021-04-14 08:05:08 -07:00 |
|
Phil Wang
|
b42fa7862e
|
Merge pull request #91 from shabie/patch-1
Fix alpha coefficient multiplication in the loss
|
2021-04-14 08:04:50 -07:00 |
|
shabie
|
dc6622c05c
|
Fix alpha coefficient multiplication in the loss
|
2021-04-14 11:36:43 +02:00 |
|
Phil Wang
|
30b37c4028
|
add LocalViT
0.16.0
|
2021-04-12 19:17:32 -07:00 |
|
Phil Wang
|
4497f1e90f
|
add rotary vision transformer
|
2021-04-10 22:59:15 -07:00 |
|
Phil Wang
|
b50d3e1334
|
cleanup levit
0.15.2
|
2021-04-06 13:46:19 -07:00 |
|
Phil Wang
|
e075460937
|
stray print
0.15.1
|
2021-04-06 13:38:52 -07:00 |
|
Phil Wang
|
5e23e48e4d
|
Merge pull request #88 from lucidrains/levit
fix images
|
2021-04-06 13:37:46 -07:00 |
|
Phil Wang
|
db04c0f319
|
fix images
|
2021-04-06 13:37:23 -07:00 |
|
Phil Wang
|
0f31ca79e3
|
Merge pull request #87 from lucidrains/levit
levit without pos emb
0.15.0
|
2021-04-06 13:36:26 -07:00 |
|
Phil Wang
|
2cb6b35030
|
complete levit
|
2021-04-06 13:36:11 -07:00 |
|
Phil Wang
|
2ec9161a98
|
levit without pos emb
|
2021-04-06 12:58:05 -07:00 |
|
Phil Wang
|
3a3038c702
|
add layer dropout for CaiT
0.14.5
|
2021-04-01 20:30:37 -07:00 |
|
Phil Wang
|
b1f1044c8e
|
offer hard distillation as well
0.14.4
|
2021-04-01 16:56:14 -07:00 |
|
Phil Wang
|
deb96201d5
|
readme
|
2021-03-31 23:02:47 -07:00 |
|
Phil Wang
|
05b47cc070
|
make sure layerscale epsilon is a function of depth
0.14.2
|
2021-03-31 22:53:04 -07:00 |
|
Phil Wang
|
9ef8da4759
|
add CaiT, new vision transformer out of facebook AI, complete with layerscale, talking heads, and cls -> patch cross attention
0.14.1
|
2021-03-31 22:42:16 -07:00 |
|
Phil Wang
|
506fcf83a6
|
add documentation for three recent vision transformer follow-up papers
|
2021-03-31 09:22:15 -07:00 |
|
Phil Wang
|
6fb360a1ff
|
add arxiv links for now, document in readme later
|
2021-03-30 22:26:44 -07:00 |
|
Phil Wang
|
9332b9e8c9
|
cite
|
2021-03-30 22:16:14 -07:00 |
|
Phil Wang
|
da950e6d2c
|
add working PiT
0.12.0
|
2021-03-30 22:15:19 -07:00 |
|
Phil Wang
|
4b9a02d89c
|
use depthwise conv for CvT projections
0.11.1
|
2021-03-30 18:18:35 -07:00 |
|
Phil Wang
|
518924eac5
|
add CvT
0.11.0
|
2021-03-30 14:42:39 -07:00 |
|
Phil Wang
|
e712003dfb
|
add CrossViT
0.10.3
|
2021-03-30 00:53:27 -07:00 |
|