This course introduces graph machine learning, with a practical focus based on the StellarGraph library.
The course slides are available in the slides/ directory.
The main requirements are Python 3.6 or 3.7 and a set of python modules listed in the file requirements.txt with the correct module version numbers.
We provide installation instructions for Windows 10 and MacOS. The latter can be used as a guide for common Linux distributions such as Ubuntu. We expect that users have some experience programming in Python and using basic tools such as pip and git.
[1] Install a Python 3 (version 3.6 or 3.7) distribution. We recommend
Anaconda that can be downloaded by clicking
here. Alternatively, install
Python 3.6 from here and also
virtualenv via the command pip install virtualenv.
[2] Download and install the latest version of git from here.
[3] Create a working directory, e.g., /Users/YOUR-USER-NAME/Projects, where YOUR_USER_NAME
should be replaced with your user name and Projects is the new working directory. From this
point on we will use HOME to refer to /Users/YOUR-USER_NAME such that the working
directory will be HOME/Projects
[4] Create a custom environment using the command (if using conda)
conda create --name practical-ml python=3.6
Alternatively, if you are not using Anaconda and conda, you can create a new virtual
environment using virtualenv. Instructions on how to install and use virtualenv can be
found here.
virtualenv practical-ml
The above command will create a new Python environment at location HOME/Projects/practical-ml
[5] Create the file matplotlibrc in directory ~/.matplotlib; create the directory if it does
not exist in your system. The file contents should be,
backend:TkAgg
[6] Activate the new virtual environment created with conda by executing the command,
source activate practical-ml
or if created using virtualenv use the command,
source practical-ml/bin/activate
[7] Download or clone the course source code repository using the command (from the HOME/Projects directory),
git clone https://github.com/stellargraph/stellar-practical-ml-on-graphs.git
[8] Change to the stellar-practical-ml-on-graphs directory and install the python requirements using the command,
pip install -r requirements.txt
[9] You also need to register the practical-ml environment so that it is available in Jupyter. You
should use the following command,
python -m ipykernel install --user --name=practical-ml
[10] You can verify that you have the correct version of stellargraph installed by using the command,
python -c "import stellargraph as sg; print(sg.__version__)"
Pay attention to the double quotes and the __ is two underscores (before and after the word version.) The above command should print
0.10.0
[11] You can now run jupyter notebook using the command,
jupyter notebook
You can access the course notebooks using your web browser at localhost:8888
The installation should now be complete.
[1] Install a Python 3 (version 3.6 or 3.7) distribution. You can download the recommended version from here. Make sure to select the option to Add Python 3.6 to PATH on the Setup screen.
[2] Open a Windows Command Prompt and run the following command,
python --version
if the installation was successful then it should print,
Python 3.6.2
The default Python distribution also includes pip. On the Command Prompt type the following
command,
pip --version
if the installation was successful then it should print the pip version,
pip 19.0.2
[3] Install virtualenv using the following command,
pip install virtualenv
[4] Install Jupyter Notebook using the command,
pip install jupyter
[5] Download and install the latest version of git from here.
[6] Create a working directory, e.g., C:\users\YOUR-USER-NAME\Projects, where YOUR_USER_NAME
should be replaced with your user name and Projects is the new working directory. From this
point on we will use HOME to refer to C:\users\YOUR-USER_NAME such that the working
directory will be HOME\Projects
[7] Change to the HOME\Projects directory and create a new Python virtual environment called practical-ml
using the following command,
virtualenv practical-ml
This will create a new folder HOME\Projects\practical-ml
[8] Activate the new environment using the command,
practical-ml\Scripts\activate
[9] Download or clone the stellar-practical-ml-on-graphs repository (in HOME\Projects) using the
command,
git clone https://github.com/stellargraph/stellar-practical-ml-on-graphs.git
[10] Change to the stellar-practical-ml-on-graphs directory and install the python requirements using
the command,
pip install -r requirements.txt
[12] You also need to register the practical-ml environment so that it is available in Jupyter.
You should use the following command,
python -m ipykernel install --user --name=practical-ml
[13] You can verify that you have the correct version of stellargraph installed by using the command,
python -c “import stellargraph as sg; print(sg.__version__)”
Pay attention to the double quotes and the __ is two underscores (before and after the word version.) The above command should print
0.10.0
[14] You can now run jupyter notebook using the command,
jupyter notebook
You can access the course notebooks using your web browser at localhost:8888
The installation should now be complete.
Copyright 2010-2020 Commonwealth Scientific and Industrial Research Organisation (CSIRO).
All Rights Reserved.
NOTICE: All information contained herein remains the property of the CSIRO. The intellectual and technical concepts contained herein are proprietary to the CSIRO and are protected by copyright law. Dissemination of this information or reproduction of this material is strictly forbidden unless prior written permission is obtained from the CSIRO.