This Docker container runs a Jupyter notebook containing an example of the TVM custom datatypes work (also known as Bring Your Own Datatypes).
git clone <this repo>
cd <this repo>
# Build Docker image from Dockerfile
docker build -t tvm-custom-datatypes-notebook-example -f Dockerfile .
# Run Docker image (which starts a Jupyter Notebook server)
# Note that the -p option is important for connecting host port 8888 to your
# Docker container's port 8888.
docker container run -i -p 8888:8888 --rm -t \
tvm-custom-datatypes-notebook-example:latest
In the output, you should see something like:
[C 21:21:25.581 NotebookApp]
To access the notebook, open this file in a browser:
file:///root/.local/share/jupyter/runtime/nbserver-1-open.html
Or copy and paste one of these URLs:
http://b1fba94d05a3:8888/?token=bb495da54309d23324677dcaeed78e5ded9e8e34ccbbf9d3
or http://127.0.0.1:8888/?token=bb495da54309d23324677dcaeed78e5ded9e8e34ccbbf9d3
Navigate to the last address (the one using 127.0.0.1
) in your browser, and the notebook should open!