Demo notebooks¶
Short, self-contained demos of popular Python libraries, grouped by topic:
| Folder | What is in it |
|---|---|
aws/ |
The AWS Common Runtime bindings. |
case_studies/ |
Longer worked examples on public data: the National Water Model, GOES-16 NetCDF files and Sentinel-2 resampling. |
data_analytics/ |
Bottleneck. |
data_processing/ |
pandas, Dask, SciPy, scikit-learn, pydantic, pandera, NLTK and more. |
data_storage/ |
Databases (PostgreSQL, MySQL, SQL Server, Oracle, SQLite and others), file systems and cloud storage, Arrow. |
data_structures/ |
jsonschema, multidict and other data structure libraries. |
data_visualization/ |
Matplotlib, Bokeh, HoloViews, GeoViews, Cartopy, Plotly, Folium, Graphviz and more. |
geospatial/ |
GeoPandas, GDAL, xarray, PySAL, NetCDF and HDF5, STAC, and Google and Azure geospatial services. |
networking/, security/, testing/, utilities/, web_servers/ |
HTTP clients, .env files, Faker and Hypothesis, small utilities, Tornado. |
Running them¶
Each notebook starts with a title cell that says what the library does and links its documentation under Resources. Install the library it demonstrates, plus JupyterLab, and run the cells from the top.
- Some notebooks, mainly those that draw plots or maps, are committed with their outputs, so the figures show on GitHub; interactive charts (Bokeh, HoloViews, Panel, Plotly, Folium) appear there as static images. The other notebooks are committed without outputs.
- Notebooks that connect to a database or a cloud service read their connection settings from environment variables. The Environment variables cell right after the title lists them; set them before you start Jupyter.
- Many demos need a running service or an account (a database server, a cloud bucket, an API key), so they cannot all run on a plain laptop.
Where the examples come from¶
Many examples follow the documentation, README, tutorials or example galleries of the library they demonstrate, and some are adapted from them. Each notebook links those sources under Resources. Material adapted from a project stays under that project's license (BSD, MIT, Apache-2.0 and others), not under this repository's BSD-3-Clause license.
Notebooks that contain adapted code say so in a Source and license entry in their first cell. THIRD_PARTY_NOTICES.md lists those projects with their copyright notices and the full license texts.
Sections¶
- AWS: 1 notebook. The AWS Common Runtime bindings.
- Case studies: 3 notebooks. Longer worked examples on public data: the National Water Model, GOES-16 NetCDF files and Sentinel-2 resampling.
- Data analytics: 1 notebook. Bottleneck.
- Data processing: 14 notebooks. pandas, Dask, SciPy, scikit-learn, pydantic, pandera, NLTK and more.
- Data storage: 47 notebooks. Databases (PostgreSQL, MySQL, SQL Server, Oracle, SQLite and others), file systems and cloud storage, Arrow.
- Data structures: 5 notebooks. jsonschema, multidict and other data structure libraries.
- Data visualization: 18 notebooks. Matplotlib, Bokeh, HoloViews, GeoViews, Cartopy, Plotly, Folium, Graphviz and more.
- Geospatial: 30 notebooks. GeoPandas, GDAL, xarray, PySAL, NetCDF and HDF5, STAC, and Google and Azure geospatial services.
- Networking: 2 notebooks
- Security: 1 notebook
- Testing: 15 notebooks
- Utilities: 5 notebooks
- Web servers: 1 notebook