To implement the faster version of RAFT, if I change the default value of --corr_implementation to reg_cuda, I get the error of "name 'corr_sampler' is not defined". Collaborator lahavlipsoncommentedMar 21, 2023 Were you able to successfully runpython setup.py installundersampler/?
这一部分比较了MXNet与Torch7、Caffe、Tensorflow在卷积网络基准测试中的性能。TensorFlow使用的是CUDA 7.0和CUDNN 2,而其他的框架使用的是CUDA 7.5和CUDNN 3。批量大小设置为32,所有的卷积神经网络在Nvidia GTX 980显卡上进行训练。图6是测试结果。可以看到,MXNet、Torch7和Caffe拥有相近的性能,这是因为大部分操作都...
Looking forward to your reply! CorrSampler is an optional faster implementation of the correlation sampler, enabled using--corr_implementation reg_cuda. If you use--corr_implementation reg, do you still get the same issue? Sorry if my question doesn't make sense, I am not so familiar with ...