sciml-workshop

Practical notebooks for the SciML workshop

Google Colab

Directions for setup

• Go to colab https://colab.research.google.com/notebooks/intro.ipynb
• File → Open Notebook
• Click tab "Github"
• In the search box enter https://github.com/stfc-sciml/sciml-workshop
• Scroll down and choose the appropriate lesson
• Authorise yourself with Google.

Notes while running

• At the start of the session make sure you are running a GPU. NB - if you go back and change this the notebook will have to restart.
  • Runtime → Runtime Type → Hardware Accelerator = GPU
• If you wish to keep a copy with saved changes, there are two options:
  • To download the notebook to your computer: File → Download .ipynb
  • To save a copy in your Google Drive: File → Save a copy in Drive

SCD Cloud

Directions for setup

• Go to SCD Cloud's JupyterHub https://training.jupyter.stfc.ac.uk
• Click on the "sign-up"
• Enter your supplied username (sciml-workshop-XX) and a password of your choice.
• Click "Create User"
• Login with your username and password
• If prompted, on the server options page choose "SciML GPU environment"
• Click "Start"
• Wait a minute or two for your environment to load.

Credits

Keith Butler, Kuangdai Leng, Sam Jackson, Jaehoon Cha, Susmita Basak, Yeung Siu-Lun, Margaret Duff, Jeyan Thiyagalingam

Scientific Machine Learning (SciML) Group, Scientific Computing Department, Science and Technology Facilities Council

Authors of modules

• CLASSICAL: Keith Butler, Kuangdai Leng, Sam Jackson
• DNN: Keith Butler, Kuangdai Leng
• CNN: Keith Butler, Kuangdai Leng, Susmita Basak
• LSTM: Keith Butler, Kuangdai Leng
• AE: Kuangdai Leng
• VAE: Kuangdai Leng
• GAN: Kuangdai Leng
• REINFORCE: Yeung Siu-Lun
• DEBUGGING: Keith Butler
• LARGE-SCALE: Sam Jackson

Thanks to

Richard Regan, Clare Jenner, Mark Wilkinson

Distributed Research utilising Advanced Computing Community