Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / chengdazhi/Deformable-Convolution-V2-PyTorch issues and pull requests

#93 - Hello, do you have the frame diagram of DCNV2?

Issue - State: open - Opened by ly1035327995 4 months ago

#92 - How to change kernel size od DCN

Issue - State: open - Opened by SmallingCar over 1 year ago

#91 - Migrate THCDeviceUtils to ATen

Pull Request - State: open - Opened by llnulldisk almost 2 years ago - 3 comments

#90 - How to install DCN on google colab

Issue - State: closed - Opened by yukistavailable about 2 years ago

#89 - When running test.py, RuntimeError: Backward is not reentrant

Issue - State: open - Opened by qqpann about 2 years ago

#88 - When I run test.py, cuDNN error: CUDNN_STATUS_EXECUTION_FAILED

Issue - State: open - Opened by qqpann about 2 years ago - 1 comment

#87 - when running make.sh ,it takes this problom.

Issue - State: open - Opened by zjm1640160263 over 2 years ago - 1 comment

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

Issue - State: open - Opened by zj916716524 over 2 years ago - 1 comment

#85 - Is it useful to use dcn on 1x1conv?

Issue - State: open - Opened by kirkzZ almost 3 years ago - 3 comments

#84 - QUestion: Running without Cuda

Issue - State: open - Opened by rosanajurdi almost 3 years ago - 1 comment

#83 - problem of make.sh

Issue - State: open - Opened by b762927 almost 3 years ago

#82 - Question about DCNv1 v2 on resnet

Issue - State: closed - Opened by Fyzjym almost 3 years ago - 1 comment

#79 - Please help me! Questions about gcc or g++ in setup.py

Issue - State: closed - Opened by Fyzjym about 3 years ago

#78 - TypeError: dist must be a Distribution instance

Issue - State: open - Opened by Owen718 about 3 years ago - 3 comments

#77 - Questions about deformable roi pooling

Issue - State: open - Opened by pp00704831 over 3 years ago

#76 - using g++ not c++ to compile

Issue - State: open - Opened by Yixiang-Wu over 3 years ago

#75 - Is this version compatible with pytorch1.8

Issue - State: open - Opened by ecnuycxie over 3 years ago - 6 comments

#74 - 怎么得到 Deformable-Convolution的FLOPs呢？

Issue - State: open - Opened by AlexWangn over 3 years ago - 2 comments

#73 - [pytorch 1.0.0 branch] : no result in output tensor

Issue - State: closed - Opened by K-H-Ismail over 3 years ago

#72 - [pytorch 1.0.0 branch] : no result in output tensor

Issue - State: closed - Opened by K-H-Ismail over 3 years ago

#71 - Could you update to 3D deformable convolution?

Issue - State: open - Opened by BBQtime almost 4 years ago - 2 comments

#70 - about cuda programming tutorial

Issue - State: open - Opened by jiangxiluning almost 4 years ago

#69 - Questions about offsets

Issue - State: open - Opened by wWHWw about 4 years ago - 3 comments

#67 - Can you provide a cuda10 version?

Issue - State: closed - Opened by liuweiyy about 4 years ago

#66 - Initialization of offset layer

Issue - State: open - Opened by bluesky314 about 4 years ago

#65 - Number of output channels of conv_offset_mask

Issue - State: open - Opened by bluesky314 about 4 years ago

#64 - depriciated torch.utils.ffi used

Issue - State: open - Opened by research-boy about 4 years ago - 8 comments

#63 - Updated mmlab link for DeformConv V2 implementation in README

Pull Request - State: closed - Opened by akashpalrecha about 4 years ago - 1 comment

#61 - runs very slowly

Issue - State: open - Opened by jialeli1 about 4 years ago - 5 comments

#60 - question about backward parameter

Issue - State: open - Opened by hua-wu-que about 4 years ago - 1 comment

#59 - RuntimeError: Backward is not reentrant

Issue - State: open - Opened by FCInter over 4 years ago - 4 comments

#58 - Output channel number

Issue - State: open - Opened by lijiaying-cell over 4 years ago

#57 - Error when running python test

Issue - State: open - Opened by Susan19900316 over 4 years ago - 1 comment

#56 - Will you plan to support group convolution?

Issue - State: open - Opened by zzzmm1 over 4 years ago - 1 comment

#55 - Download model from Google Driven Drive is unavailable

Issue - State: closed - Opened by Leo63963 over 4 years ago - 1 comment

#52 - error in deformable_im2col: unknown error

Issue - State: closed - Opened by JasonLeeFdu over 4 years ago - 2 comments

#51 - Question about the gradient of Weight

Issue - State: closed - Opened by LWJ312 over 4 years ago - 1 comment

#50 - _ZN6caffe26detail37_typeMetaDataInstance_preallocated_32E

Issue - State: open - Opened by KelvinHuang666 over 4 years ago - 2 comments

#49 - How to visualize the learned position offset?

Issue - State: open - Opened by ggsDing almost 5 years ago - 2 comments

#48 - Visualization

Issue - State: open - Opened by ggsDing almost 5 years ago - 3 comments

#47 - DCN

Issue - State: open - Opened by cqray1990 almost 5 years ago - 1 comment

#46 - DCN on CPU

Issue - State: closed - Opened by abhigoku10 almost 5 years ago - 3 comments

#45 - Not supported for apex

Issue - State: open - Opened by ming71 almost 5 years ago - 2 comments

#44 - ImportError: : undefined symbol: _ZN3c105ErrorC1ENS_14SourceLocationERKSs

Issue - State: open - Opened by qianzhang2018 almost 5 years ago - 9 comments

#43 - make.sh problem

Issue - State: open - Opened by tjucwb about 5 years ago - 3 comments

#42 - pytorch 1.3 test error

Issue - State: open - Opened by a13668349935 about 5 years ago - 2 comments

#41 - The order of the offsets channels

Issue - State: open - Opened by whulc about 5 years ago - 1 comment

#40 - add missing semicolon

Pull Request - State: closed - Opened by tkkcc about 5 years ago - 2 comments

#39 - How to get pretrain weight?

Issue - State: closed - Opened by kaikaizhu about 5 years ago - 4 comments

#37 - The value corresponding to the offset position

Issue - State: open - Opened by RainHxj about 5 years ago

#36 - Value Error: Optimizing a parameter that doesn't require gradients

Issue - State: open - Opened by hufengshuo07 about 5 years ago - 1 comment

#35 - can.t compile modulated_deform_conv_cuda.cu

Issue - State: open - Opened by FourierTransform about 5 years ago - 2 comments

#34 - error in deformable_col2im_cuda: too many resources requested for launch

Issue - State: open - Opened by omrysendik over 5 years ago - 4 comments

#33 - undefined symbol: PyInt_FromLong

Issue - State: open - Opened by Jackandrome over 5 years ago

#32 - training on multiple GPU

Issue - State: open - Opened by whulc over 5 years ago - 1 comment

#30 - error in deformable_im2col: invalid argument

Issue - State: open - Opened by notagenius over 5 years ago - 3 comments

#29 - Have any documentation about this project

Issue - State: open - Opened by 18842505953 over 5 years ago

#28 - deformable_col2im invalid argument error in pytorch 1.0.0 branch

Issue - State: open - Opened by ahyunSeo over 5 years ago - 3 comments

#27 - Don't know why I get unknown error

Issue - State: open - Opened by ahyunSeo over 5 years ago

#26 - win 10

Issue - State: open - Opened by GreenTeaHua over 5 years ago - 11 comments

#25 - Update deform_conv_cuda_kernel.cu

Pull Request - State: closed - Opened by Mukosame over 5 years ago - 1 comment

#24 - How to run build.py in windows10? [cuda9、torch0.4.1、Python3.5]

Issue - State: closed - Opened by zhuwanling over 5 years ago - 3 comments

#23 - RuntimeError: batch % im2col_step_ == 0 ASSERT FAILED

Issue - State: open - Opened by gzhcv over 5 years ago - 5 comments

#22 - Solution about `ImportError: undefined symbol`

Issue - State: open - Opened by cjnjuwhy over 5 years ago - 5 comments

#20 - RuntimeError: Jacobian mismatch for output 0 with respect to input 0

Issue - State: open - Opened by hanzhangwang94 over 5 years ago - 3 comments

#19 - RuntimeError: cublas runtime error : an illegal memory access was encountered

Issue - State: open - Opened by Jiaoma over 5 years ago - 5 comments

#18 - The number of filter for generate offset is [num_deformable_groups * 2 * kH * kW]?

Issue - State: closed - Opened by gzhcv over 5 years ago - 3 comments

#17 - Add shape checking for deformable_groups?

Issue - State: closed - Opened by GengDavid over 5 years ago - 1 comment

#16 - RuntimeError: Backward is not reentrant

Issue - State: closed - Opened by leon532 over 5 years ago - 5 comments

#15 - error: Segmentation fault (core dumped), pytorch version: 1.0.0 cuda:8.0

Issue - State: open - Opened by hfutzzw over 5 years ago - 17 comments

#14 - why not use im2col_step in ModulatedDeformableConvolutionOp?

Issue - State: closed - Opened by Google1234 over 5 years ago - 1 comment

#13 - error in deformable_col2im: invalid argument

Issue - State: closed - Opened by jwwangchn over 5 years ago - 10 comments

#12 - question about "CUDA_KERNEL_LOOP"

Issue - State: open - Opened by harryhan618 over 5 years ago

#11 - DCNv2 is now supported in the official mmdetection repo

Issue - State: open - Opened by hellock almost 6 years ago - 2 comments

#9 - Need to set small learning rate in deformal conv layer?

Issue - State: closed - Opened by xyl576807077 almost 6 years ago - 3 comments

#8 - what is the meaning of num_deformable_groups， and how to use this parameter?

Issue - State: closed - Opened by miracle-fmh almost 6 years ago - 3 comments

#6 - add pytorch 1.0.0 support

Pull Request - State: closed - Opened by xvjiarui almost 6 years ago - 1 comment

#5 - how to initial the deform conv's weight?

Issue - State: closed - Opened by miracle-fmh almost 6 years ago - 1 comment

#4 - Pretrained Models

Issue - State: closed - Opened by wxzs5 almost 6 years ago - 6 comments

#3 - problem about the grad of dcn's weight

Issue - State: closed - Opened by zyxunh almost 6 years ago - 1 comment

#2 - Pytorch1.0.0 Implementation

Issue - State: closed - Opened by eugenelawrence almost 6 years ago - 2 comments

#1 - Do deformable convolution of feature_x and feature_z.

Issue - State: closed - Opened by noUmbrella almost 6 years ago - 3 comments