Iamanorange

I'm afraid not. The OPs were written with CUDA.

https://github.com/sampepose/flownet2-tf/issues/28#issuecomment-406941839

First, you should `make` 'correlation.so' and other OPs.

Open terminal; Type `make`; Enter.

Install CUDA Toolkit.

Edit Makefile: change `GPUCC = nvcc` to `GPUCC = path/to/nvcc`. or Use your current shell to run `make`.

Paste full error log.

Can't find anything useful. More information about your environment and operation?

The error message is printed before `xx errors detected in xxxxxx`. Please paste it.

https://github.com/sampepose/flownet2-tf/issues/28#issuecomment-406941839