[Feature] Support ConvMixer #716
Conversation
Codecov Report
@@ Coverage Diff @@
## dev #716 +/- ##
==========================================
+ Coverage 85.04% 85.11% +0.07%
==========================================
Files 123 124 +1
Lines 7603 7666 +63
Branches 1311 1324 +13
==========================================
+ Hits 6466 6525 +59
- Misses 945 946 +1
- Partials 192 195 +3
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
yingfhu
force-pushed
the
add-convmixer
branch
2 times, most recently
from
7719006
to
a44ad12
Compare
| @@ -0,0 +1,71 @@ | |||
| _base_ = ['./pipelines/rand_aug.py'] | |||
|
|
|||
There was a problem hiding this comment.
What's the difference with imagenet_bs64_swin_224.py? If no difference, just reuse that file.
There was a problem hiding this comment.
test pipeline resize 233
| convfunc = nn.Conv2d | ||
| if digit_version(torch.__version__) < digit_version('1.9.0'): | ||
| convfunc = Conv2dAdaptivePadding |
There was a problem hiding this comment.
So the Conv2d in torch>=1.9.0 is the same as Conv2dAdaptivePadding in torch<1.9.0?
There was a problem hiding this comment.
the same as Conv2dAdaptivePadding in mmcv
|
|
* basic support for ConvMixer * simplify * add data pipeine config for timm * Add model readme and metafile * add unittest for convmixer * add copyright * modify * add tests * update model * add conv2dAdaptivePadding replacement * update model index * fix comments * Update checkpoint path Co-authored-by: mzr1996 <mzr1996@163.com>
Motivation
Support ConvMixer and add converted models from origin repo.
Checklist
Before PR:
After PR: