A helper for setting up up-to-date Tensorflow
building environment with Docker
. Provided with both CPU and GPU support.
- This repository was tested under:
tensorflow
version: 1.13.1python
version: 3.7.3bazel
version: 0.24.1
- To get the whole picture of building
tensorflow
, check the official site.
It's better approach to build up-to-date images by yourself, by using the Dockerfile
inside this repository, though, there is pre-built image available in the docker hub
, which may slightly shorten the build time.
Check here to learn more.
- Make folder to store built wheel
- Pull pre-built image, or build the
docker
image locally - Run the
docker
container - Build
tensorflow
inside the container - Install
tensorflow
for the host machine
Check here for detailed process!
You need to copy file from container to host, use docker cp
command:
docker cp <containerId>:/file/path/within/container /host/path/target