🌳 General

🐛 Installation

🦋 Recommended way (via conda)

EOmaps is available via the conda-forge channel and can be installed via:

conda install -c conda-forge eomaps

This should make sure all required dependencies are correctly installed.

Speed up the installation

Since EOMaps dependencies can be demanding to solve for conda it is highly recommended to have a look at mamba (a C++ re-implementation of conda that provides a remarkable speedup)

To install mamba, simply use:

conda install -c conda-forge mamba

Once mamba is installed, you just need to replace the term conda with mamba and you’re good to go!

mamba install -c conda-forge eomaps

A quick tutorial on how to get started from scratch is available here: 🚀 From 0 to EOmaps - a quickstart guide

More details on how to configure your favorite IDE to work with EOmaps can be found in the FAQ section 💻 Configuring the editor (IDE):

• 🕷 Spyder IDE

• 🗖 PyCharm IDE

• 📓 Jupyter Notebooks

🐜 optional dependencies

To use the full potential of EOmaps, some additional dependencies might be required

Note on matplotlib backends

By default EOmaps requires only a minimal version of matplotlib that does not include bindings for all available matplotlib backends.

• To install all backends, explicitly use: conda install -c conda-forge matplotlib (or mamba ...)

• For QtAgg, just install the pyqt bindings via conda install -c conda-forge pyqt

• Pandas

  • The go-to library for data-analysis.

  • If installed, pandas.DataFrames can be used as datasets in EOmaps

    • Install via conda install -c conda-forge pandas (or mamba ...)

• GeoPandas

  • An awesome extension to pandas for working with geometric (shapely) objects.

  • Required for m.add_gdf() and extends the functionalities of m.add_feature()

    • Install via conda install -c conda-forge geopandas (or mamba ...)

• Datashader

  • Provides remarkable capabilities for very fast visualization of extremely large datasets (>1M datapoints)

  • Required for the plot-shapes: m.set_shape.shade_points() and m.set_shape.shade_raster()

    • Install via conda install -c conda-forge datashader (or mamba ...)

• Xarray and RioXarray

  • Reading capabilities for NetCDF and GeoTIFF files

  • Required for the plot-shape: m.set_shape.shade_raster() and for m.read_file.GeoTIFF and m.read_file.NetCDF

    • Install via conda install -c conda-forge xarray rioxarray (or mamba ...)

• Equi7Grid

  • A library to work with data provided in Equi7Grid projections

  • Required for using the projection: Maps.CRS.Equi7Grid_projection()

    • Install via pip install equi7grid

🐞 Alternative way (via pip)

EOmaps is also available via pip and can be installed using

pip install eomaps

However, it is not guaranteed that all dependencies are correctly resolved and some manual tweaking of the environment might be required to ensure that all packages work as expected.

A list of the dependencies can be found below:

dependencies:
  - python >=3.7
  - rtree
  - numpy
  - scipy
  - matplotlib >=3.0
  - cartopy >=0.20.0
  - descartes
  - mapclassify
  - pyproj
  - pyepsg
  # -------------- for WebMaps
  - owslib
  - requests
  - xmltodict
  - cairosvg
  # -------------- optional
  - pandas
  - geopandas
  - datashader
  - xarray
  - rioxarray
  # -------------- for building the docs
  - sphinx
  - sphinx-copybutton
  - sphinx_rtd_theme
  - mock