Python Packages Reference

Python Packages Reference#

This page lists the core packages used in this course and common troubleshooting tips. For complete setup instructions, see the main Python Installation Guide.

Core Packages#

The course uses these packages (automatically installed via requirements.txt):

Essential Python#

  • numpy - Numerical computing and arrays

  • matplotlib - Plotting and visualization

  • scipy - Scientific computing

  • ipykernel - Jupyter kernel for VS Code

  • jupyterlab - Jupyter Lab interface

Oceanographic Data Handling#

  • xarray - N-dimensional arrays and datasets (NetCDF files)

  • netCDF4 - NetCDF file format support

  • gsw - Gibbs-SeaWater oceanographic calculations (TEOS-10)

  • seasenselib - Sea-Bird CNV file processing

Geographic Mapping#

  • cartopy - Geographic mapping and projections

  • cmocean - Oceanographic color scales

Data Access & Utilities#

  • requests - HTTP requests for data download

  • tqdm - Progress bars for downloads

  • pytest - Testing framework for assignments

Package Installation#

Packages are installed automatically when you follow the Python Installation Guide. The requirements.txt file contains all necessary packages with compatible versions.

Troubleshooting#

Cartopy/NetCDF4 Installation Issues#

These packages often need system libraries. If installation fails:

Linux:

sudo apt install libgeos-dev libproj-dev libnetcdf-dev  # Ubuntu/Debian

macOS/Windows: Use conda instead:

conda install -c conda-forge cartopy netCDF4

VS Code Doesn’t Recognize Environment#

If VS Code can’t find your virtual environment:

pip install ipykernel
python -m ipykernel install --user --name=.venv --display-name "Python (.venv)"

Then restart VS Code and select the kernel.

Exercise-Specific Packages#

Exercise 1: CTD Profile Data#

  • gsw - TEOS-10 calculations for salinity and temperature

  • seasenselib - Converting Sea-Bird CNV files to NetCDF

Exercise 2: Bathymetry Mapping#

  • cartopy - Geographic projections and mapping

  • cmocean - Ocean-appropriate color scales

  • requests & tqdm - Downloading bathymetry data with progress bars

Notes and Best Practices#

  • Always use requirements.txt: Each exercise includes a requirements.txt file with tested package versions

  • Prefer GitHub Codespaces for complex packages (cartopy, netCDF4) to avoid compilation issues

  • Use conda-forge if installing locally on macOS/Windows for packages requiring system libraries

Expert Tips#

  • Environment isolation: Each exercise should have its own virtual environment

  • Package versions matter: Stick to the versions in requirements.txt for compatibility with automated tests

  • Backup your environment: Save working configurations with pip freeze > my-requirements.txt

Contents