Download the Python 3 Anaconda installer for your OS and open the Anaconda prompt.
Clone this repo using git or download it manually. Then, using the Anaconda prompt, cd into the ml-tutorial directory and move on to [Starting the tutorial](#Starting the tutorial).
Clone the repo and create a virtual environment in the root folder. After activating the environment, install the required packages:
pip install -r requirements.txt
Open the release/1/index.ipynb Jupyter notebook with the following command:
jupyter notebook release/1/index.ipynb
Activate the nbgrader notebook extension as documented here.
Change the notebooks in the source/ folder, any new cells where students have to write code should be set to Autograded answer as documented here.
After saving your changes, run:
sh generate_assignment.sh
This will overwrite the existing release/ directory.
While creating the notebooks in the release/ directory, the nbgrader package will alter code segments that look like:
### BEGIN SOLUTION
foo = bar
### END SOLUTION
into:
##### Implement this part of the code #####
raise NotImplementedError("Code not implemented, follow the instructions.")
As specified in the nbgrader_config.py file.
If you want to host this workshop you can use the included Dockerfile to build a JupyterHub which you can build and launch with the following commands:
docker build . -t ml-tutorial
docker run -p 8000:80 ml-tutorial
Surf to http://localhost:8000 to start the tutorial and log in with user001 and password password.
The docker image is also available on docker hub so running it on a server should work with:
docker run -d -p 80:80 pbutenee/ml-tutorial
without the need for downloading the code or building the container. Set the port to 8000 instead of 80 to test it out on local host.
NOTE: ONLY AN OLD VERSION OF THIS CONTAINER IS AVAILABLE ON DOCKER HUB SINCE IT IS NO LONGER FREE