目前文章只是粗略的看过一遍，还不能写在博客里，要等后续再读读，恩，先欠着吧！
LZ其实是为了预研FairMOT论文，然后需要配置dcnv2的环境，期间当然不是一帆风顺的，已经习惯了。。。
其中有几个报错还是很多小伙伴会遇到的，在github上也有很多类似的issue，所以就重新开一篇，毕竟也是一篇CVPR，它值得的。
DCNv2_new/src/cuda/dcn_v2_cuda.cu(107): error: identifier "THCState_getCurrentStream" is undefined DCNv2_new/src/cuda/dcn_v2_cuda.cu(279): error: identifier "THCState_getCurrentStream" is undefined DCNv2_new/src/cuda/dcn_v2_cuda.cu(324): error: identifier "THCudaBlas_Sgemv" is undefined使用原始代码里给的dcnv2，会报如上的错误，因为原始代码用的是
conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch而LZ是安装的pytorch1.6
conda install pytorch torchvision cudatoolkit=10.2 -c pytorch非得按照自己版本来，结果就是得各种找问题
后面是在github上https://github.com/jinfagang/DCNv2_latest找到了pytorch1.6版本的DCNv2版本
后面运行make.sh应该是编译成功了的
copying build/lib.linux-x86_64-3.8/_ext.cpython-38-x86_64-linux-gnu.so -> Creating /root/anaconda3/envs/FairMOT/lib/python3.8/site-packages/DCNv2.egg-link (link to .) Adding DCNv2 0.1 to easy-install.pth file Installed /DCNv2_latest-pytorch1.6 Processing dependencies for DCNv2==0.1 Finished processing dependencies for DCNv2==0.1但是在test的时候又出现对应的问题：
python testcuda.py torch.Size([2, 64, 128, 128]) torch.Size([20, 32, 7, 7]) torch.Size([20, 32, 7, 7]) torch.Size([20, 32, 7, 7]) 0.971507, 1.943014 0.971507, 1.943014 Zero offset passed /root/anaconda3/envs/FairMOT/lib/python3.8/site-packages/torch/autograd/gradcheck.py:266: UserWarning: The {}th input requires gradient and is not a double precision floating point or complex. This check will likely fail if all the inputs are not of double precision floating point or complex. warnings.warn( check_gradient_dpooling: True Traceback (most recent call last): File "testcuda.py", line 265, in <module> check_gradient_dconv() File "testcuda.py", line 95, in check_gradient_dconv gradcheck(dcn_v2_conv, (input, offset, mask, weight, bias, File "/root/anaconda3/envs/FairMOT/lib/python3.8/site-packages/torch/autograd/gradcheck.py", line 321, in gradcheck return fail_test('Backward is not reentrant, i.e., running backward with same ' File "/root/anaconda3/envs/FairMOT/lib/python3.8/site-packages/torch/autograd/gradcheck.py", line 254, in fail_test raise RuntimeError(msg) RuntimeError: Backward is not reentrant, i.e., running backward with same input and grad_output multiple times gives different values, although analytical gradient matches numerical gradient. The tolerance for nondeterminism was 0.0.README.md中有一段
This is an adaption of the official Deformable-ConvNets.
Update: all gradient check passes with double precision.
Another issue is that it raises RuntimeError: Backward is not reentrant. However, the error is very small (<1e-7 for float <1e-15 for double), so it may not be a serious problem (?)
换句话说，这个问题可以忽略，那按照它对应的说法，这算是编译完成，可以使用了。
