Docker Images

We provide docker utility scripts to help developers to setup development environment. They are also helpful run through TVM demo and tutorials. We need docker and nvidia-docker if we want to use cuda.

Get a tvm source distribution or clone the github repo to get the auxiliary scripts

git clone --recursive https://github.com/apache/tvm tvm

We can then use the following command to launch a docker image.

/path/to/tvm/docker/bash.sh <image-name>

Here the image-name can be a local docker image name, e.g. tvm.ci_cpu after you have done the local build.

This auxiliary script does the following things:

  • Mount current directory to /workspace

  • Switch user to be the same user that calls the bash.sh (so you can read/write host system)

  • Use the host-side network on Linux. Use the bridge network and expose port 8888 on macOS, because host networking driver isn’t supported. (so you can use jupyter notebook)

Then you can start a Jupyter notebook by typing

jupyter notebook

You might see an error OSError: [Errno 99] Cannot assign requested address when starting a Jupyter notebook on macOS. You can change the binding IP address by

jupyter notebook --ip=0.0.0.0

Note that on macOS, because bash.sh uses the Docker bridge network, Jupyter will be reportedly running at an URL like http://{container_hostname}:8888/?token=.... You should replace the container_hostname with localhost when pasting it into browser.

Docker Source

Check out the docker source if you are interested in building your own docker images.

Run the following command to build the docker image.

/path/to/tvm/docker/build.sh <image-name>

You can find some un-official third party pre-built images at https://hub.docker.com/r/tlcpack/. These images are used for test purposes and are NOT of the ASF release.