This repository contains the content for the 2024-11-18 pyrolite workshop for the 2024 Specialist Group in Tectonics and Structural Geology (SGTSG) meeting in Armidale, NSW.
pyrolite is an open-source Python package for working with geochemical data which aims to help geoscientists new to coding get off the ground quickly. It contains a variety of functionality for the processing, transformation and visualisation of geochemical and mineralogical data (see pyrolite.rtfd.io for more information). This hands-on workshop will introduce Python and the Jupyter ecosystem, showcase some of pyrolite’s key features and demonstrate a series of geochemical data analytics workflows. No previous experience with programming or Python is necessary, but note that the workshop will be dominantly using and discussing code. Participants should bring their own laptops, but the workshop will largely be conducted via web browser, so no software installation will be necessary.
A tentative schedule and description of the content for each section of the workshop is given below. Catering will be provided beforehand.
| Time (AEDT) | Agenda |
|---|---|
| 17:30 - 18:00 | Setup, Welcome, Refreshments |
| 18:00 - 18:15 | Introduction (15 mins) |
| 18:15 - 18:45 | Using pyrolite (30 mins) |
| 18:45 - 19:25 | Geochemical Data Workflows (40 mins) |
| 19:25 - 19:30 | Wrap Up |
- Working with geochemical and petrological data
- Programmatic data processing and analysis workflows
- The open-source scientific Python ecosystem
- Introduction to Python and Jupyter
- Overview
- Key features
- Processing and transformation
- Visualisation
- Reference data
- Working with mineral compositions
- Utilities, links into other software, ML etc
- Interoperability with other components of the scientific Python ecosystem (e.g. Pandas, scikit-learn)
- Documentation and infrastructure, installation and development Roadmap
This section will feature a set of extended workflow examples, potentially including:
- A few of the more complex feaures - Calculating CIPW Norms and lambdas
- A few basic ML workflows
- Basic spatial visualisation with GeoPandas
To view the notebooks for this workshop, follow one of the links below. The first is the custom JupyterLab instance to be used for the workshop (you'll be given a unique username at the workshop, and you can create a password on first login), but the content should be available via Binder thereafter (albiet, with a short delay associated with building an environment to run it in, using free resources provided by MyBinder).
| Workshop JupyterHub | Binder |
|---|---|
 |
 |
| Facilitators | Mastodon | GitHub |
|---|---|---|
| Morgan Williams (CSIRO) | @metasomite@aus.social | @morganjwilliams |
If you want to check out some of my other open software projects, most are listed on my GitHub profile. Feel free to get in touch - send an email to morgan.williams at csiro.au.