Deformable-Convolution-V2-PyTorch

Deformable-Convolution-V2-PyTorch copied to clipboard

RuntimeError: batch % im2col_step_ == 0 ASSERT FAILED

5

**Environment:** - python 3.6 - torch 1.0 Error occurs as follows when the **batch_size** is 128 or 192 or 256. But when `batch_size = 64` is ok. ``` File "/home/gzh/SoftWare/tf1.4/anaconda2/envs/python36/lib/python3.6/site-packages/DCN-1.0-py3.6-linux-x86_64.egg/modules/deform_conv.py",...

gzhcv

TypeError: dist must be a Distribution instance

3

torch: 1.9.0+cu111 python version:3.9 --------------------- When I run the "python build.py",the Error is "TypeError: dist must be a Distribution instance". How to solve it? I have change the version of...

Owen718

Is this version compatible with pytorch1.8

6

I try the "pytorch1.0.0" version but found that it is not compatible with PyTorch 1.8. Is there any solution for Deformable Conv in PyTorch 1.8? I want to try it...

ecnuycxie

when running make.sh ,it takes this problom.

1

RuntimeError: Error compiling objects for extension why i take this problem what's the Version of pytorch,cuda

zjm1640160263

When running make.sh, it prompts that NVCC cannot be found

1

Hello. I encountered a problem when establishing the DCN-V2 environment. After running this command in sh make.sh, it reported this error `/bin/sh: 1: :/usr/local/cuda/bin/nvcc: not found`, but I found this...

zj916716524

Questions about offsets

3

Hi ! Excellent work on making this implementation of Deformable Convolution. When I used DCN to replace some of my backbone conv to train an network for object tracking problem,...

wWHWw

error in modulated_deformable_col2im_cuda: no kernel image is available for execution on the device

3

When I run DCN on EDVR, I meet the problem as follows: 

hjynwa

Question about DCNv1 v2 on resnet

1

hi, I want to add DCNv1 and DCNv2 to resnet50, but an error occurred during the training process. Are there any examples of successfully joining dcn (for example, image classification...

Fyzjym

Is it useful to use dcn on 1x1conv?

3

hello,is it useful to use dcn on 1x1conv?

kirkzZ

QUestion: Running without Cuda

1

Hello, Is it possible to create a deformed convolutional layer that runs without Cuda ? Time computation is not an issue. Kind Regards, Rosana

rosanajurdi