aleozlx/tkstack3 repository overview

⁠Tensorflow + Keras container image (Best effort documentation)

Tensorflow + Keras container image based on nvidia-docker2.

⁠What's included:

• Ubuntu 16.04
• CUDA 9.0 + cuDNN
• TensorFlow 1.11.0
• Numpy + Scipy linked against libblas + liblapack
• Isolated X11 GUI support
• Non-privileged sudoer with uid:gid matching host user
• And more python packages:
  • pandas matplotlib h5py scikit-learn scikit-image ...
  • Check out requirements.txt for complete list.

⁠Get started

Try docker.py to get a shell or jupyter.py to get a jupyter environment with current working folder mounted.
Once you get the hang of it, create your own launcher!

⁠Extra arguments for GUI application

These arguments are to be added to the docker command.

-e DISPLAY --net=host -v "/tmp/.X11-unix:/tmp/.X11-unix" -v "$HOME/.Xauthority:/home/developer/.Xauthority" aleozlx/cvstack