Python Workshop Playbook

Hands-on notebooks that showcase practical pathways for moving from MATLAB-style workflows to modern Python tooling. Each folder is paced for a short workshop segment: punchy intros, focused demos, and persuasive visuals to help teams adopt the stack quickly.

Folder Index

• basic/ – Concept primers for scientific Python, including motivation to switch from MATLAB and how to layer abstraction cleanly.
• data_heavy/ – High-volume data handling with HDF5, NetCDF, Zarr, and Dask for scalable IO and lazy computation.
• geomorpho/ – Geomorphology workflows; currently a flood-routing example bridging PyTopoToolbox concepts.
• geospatial/ – Raster/vector GIS essentials with Rasterio and GeoPandas-style tooling.
• HPC_cpu_jit/ – CPU acceleration tour featuring Numba, NumExpr, JAX, F2PY, and profiling-ready scripts.
• Machine_Learning/ – Compact ML labs: titanic with CatBoost, digit recognition, anomaly detection, and number theory puzzles.
• numba/ – Fresh notebooks on Numba fundamentals, performance wins, advanced features, and a full implicit stream-power model.
• pandas_polars/ – Tabular manipulation contrasts using pandas and Polars, backed by a ready-made synthetic dataset.
• taichi_lang/ – GPU-ready numerical demos with Taichi, including heat diffusion and hydraulic erosion sims.
• visualisation/ – Plotting gallery covering Matplotlib, Seaborn, and Datashader + HoloViews.

Getting Started

1. Create or activate a conda/mamba environment with Python 3.11+.
2. Install notebook dependencies as you go (most notebooks include a !pip install ... cell for convenience).
3. Launch JupyterLab or VS Code notebooks in this repository root so relative paths resolve.

Suggested Flow

• Start with basic/ and pandas_polars/ to get comfortable with the ecosystem.
• Pick domain-specific tracks: geospatial, geomorphology, visualisation.
• Finish with acceleration stories in numba/ and taichi_lang/ to drive home performance.

Contributing

Keep notebooks concise and interactive. When adding content, include a short README entry (see folder READMEs for the tone) and prefer self-contained installs so newcomers can run cells top-to-bottom without extra setup.