Alternative installation methods for Conda environments
-------------------------------------------------------

In the previous section we installed our specific conda Python environment with a pre-defined environment configuration file. In this section we show an alternative variant how you can install conda environments in a more flexible way if you need to.


.. admonition:: BEWARE:

    Please DO NOT follow these steps, if you already have the environment installed via the ``environment.yml`` configuration above.

The other variant is typically more widely used in exploratory setups. It is a step-by-step procedure. Here we are making sure that Python in version 3.8 will be installed and that we want to explicitly use the additional package channel `conda-forge`.
And we give the environment a name (-n).

If you would create your environment manually, it would go like that:
Open the ``Anaconda prompt`` from the Start Menu and type the command below.

.. code::

    (C:\dev\conda3) conda create -n geopython2022alt python=3.9 -c conda-forge

Ok, now that we have installed a Python working environment with the name ``geopython2022alt`` with our desired library packages, we can check installed environments just to be sure.
In order to show all environments that have already been created you can ask conda to list these:

.. code::

    (C:\dev\conda3)  conda env list

Now we want to activate that environment, install additional packages and start working with it:

.. code::

    (C:\dev\conda3)  activate geopython2022alt

    (geopython2022alt)


Install GIS related packages with conda by running in command prompt following commands (in the same order as they are listed).
Make sure you are in the correct environment (don't install into ``base``, install new packages ideally only into your designated created environments)

.. code::

    (geopython2022alt) conda install -c conda-forge numpy pandas gdal fiona shapely geopandas

    # Install matplotlib and Jupyter Lab/Notebook
    (geopython2022alt) conda install -c conda-forge matplotlib jupyter jupyterlab

    # Install seaborn
    (geopython2022alt) conda install -c conda-forge seaborn

    # Install geoplot and cartopy
    (geopython2022alt) conda install -c conda-forge cartopy geoviews

    # Install mapclassify
    (geopython2022alt) conda install -c conda-forge mapclassify

    # Install rasterio and rasterstats
    (geopython2022alt) conda install -c conda-forge rasterio rasterstats


In the next step we will verify the installation of our conda Python environment and configure Jupyter Notebooks.


Using mamba as a faster conda version
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

**mamba** is a CLI tool to manage **conda** environments.

If you already know conda, great, you basically know mamba!


.. admonition:: BEWARE:

    The mamba command must only be installed into the "base" conda environment!!!

    AND:

    In the University computers it will not be available, and can't be installed because it requires administrator rights.

To get mamba, just install it into the base environment from the conda-forge channel:

.. code::

    conda install mamba -n base -c conda-forge


From here you can use mamba the way you use conda, especially for installing additional packages or creating environments with an environment.yml file.

For more backround, please read the manual: https://mamba.readthedocs.io/en/latest/installation.html