ubcrcl/tf-cv-itk repository overview

⁠tf-cv-itk

Ubuntu Core 16.04 + CUDA 9.0 + Python 3.5 + Tensorflow 1.9.0 + OpenCV (pip) + SimpleITK + tensorlayer

⁠Usage

Use nvidia-docker: nvidia-docker run -it ubcrcl/tf-cv-itk bash

To access port 50051, add option: -p 50051:50051

To mount your home directory inside the container, add option: --volume=$HOME:/workspace

To specify which GPU to isolate, add prefix: NV_GPU=<X> e.g. NV_GPU=1 nvidia-docker run ...