involution

involution copied to clipboard

Involution on sparse point clouds

2

Hi, I found your work on involution very interesting, and it relates to other ideas I am working on like deformable convolutions. So I tried reimplementing your idea for sparse...

HuguesTHOMAS

where is rednet50.pth

1

OSError: /path/to/rednet50.pth is not a checkpoint file

xiaoqianbi123

Val loss is Nan，but the train loss is normal？

1

Fateeeeee

train model

1

error: ModuleNotFoundError: No module named 'mmdet.models.dense_heads .rpn_test_mixin' I didn't see the file about rpn_test_mixin

qwertyuiop194

Why the GPU memory of rednet50 is larger than resnet50?

1

I used Tesla V100. When I train resnet50-retinanet, the gpu memory is 9.26G. But when I train rednet50-retinanet, the gpu memory is 10.98G. In your paper, the params and floats...

yushanshan05

Thanks your work, it's very useful, and I've used involution Pytorch version for a while, but recently I need to try a bigger feature than before, and my GPU memory...

hheavenknowss

Hi, which versons of mmdet,mmseg,mmcls do you used?

yanlongbinluck

Hello, thank you very much for your work.Could you tell me what version of mmcv is your code uesed?Thank you very much.

lsm-sunny

Nice work of rethinking conv modules. The question is why could involution summarize the context into a wider spatial array? In my view, only the process of changing 3x3 convolution...

ddamddi

I added some args for dilation. Therefore, the generated kernel could compute with dilated pixels. It might be helpful to expand receptive field in RedNet backbone without losing feature map...

keke1u