Installation on Mac#

This page provides instructions for installing GeoIPS using wheels from PyPI in a conda virtual environment. This installation method is recommended for users with or without administrative privleges.

Developers of core GeoIPS should follow the appropriate guide under “Contribute”.

1. Install Anaconda or Miniconda#

Use a fresh Anaconda or Miniconda environment with Python 3.11 or later.

Download the appropriate version of Conda, Miniconda, or Miniforge/Mambaforge.

# Anaconda for Intel chips
curl -L -o conda_install.sh https://repo.anaconda.com/archive/Anaconda3-2023.07-2-MacOSX-x86_64.sh
# Anaconda for ARM (M1/M2) chips
curl -L -o conda_install.sh https://repo.anaconda.com/archive/Anaconda3-2023.07-2-MacOSX-arm64.sh
# Mamba for Intel chips
curl -L -o conda_install.sh \
https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-MacOSX-x86_64.sh
# Mamba for ARM (M1/M2) chips
curl -L -o conda_install.sh \
https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-MacOSX-arm64.sh
# Miniconda for Intel chips
curl -L -o conda_install.sh https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-x86_64.sh
# Miniconda for ARM (M1/M2) chips
curl -L -o conda_install.sh https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh

Make the installer script executable and run it, following the prompts (particularly the bit about conda init / restarting terminal!):

chmod u+x conda_install.sh
./conda_install.sh
# Follow instructions regarding conda init / restarting your terminal !

# Clean up after yourself
rm -f ./Miniforge3-Linux-x86_64.sh

Create a conda environment named, for example, geoips with the appropriate dependencies:

# git required for pulling from git and for -C commands
# pyhdf and pykdtree don't have wheels for mac and don't build cleanly
#   best to install via conda
conda create -y -n geoips -c conda-forge python=3.11 cmake gfortran git libgit2 make openblas pyhdf pykdtree
conda activate geoips  # RUN EVERY TIME YOU WANT TO USE GEOIPS!

Note

You must run conda activate geoips every time you run or work with GeoIPS.

2. Install GeoIPS using pip#

Once inside the geoips environment, install GeoIPS from PyPI:

pip install geoips

3. Test Data Setup#

GeoIPS test data download requires the following environment variables to be set:

conda env config vars set GEOIPS_PACKAGES_DIR=$HOME/geoips
# Reactivate environment for variables to take effect
conda deactivate && conda activate geoips
conda env config vars set GEOIPS_TESTDATA_DIR=$GEOIPS_PACKAGES_DIR/test_data
conda env config vars set GEOIPS_OUTDIRS=$GEOIPS_PACKAGES_DIR/outdirs
conda deactivate && conda activate geoips
mkdir -p "$GEOIPS_PACKAGES_DIR" "$GEOIPS_TESTDATA_DIR" "$GEOIPS_OUTDIRS"

Note

You can customize test data location by changing the value of GEOIPS_TESTDATA_DIR.

4. Run a Sample Script#

Download the sample dataset using geoips CLI command:

geoips config install test_data_abi

Test the installation by running a sample GeoIPS processing script.

geoips run single_source $GEOIPS_TESTDATA_DIR/test_data_abi/data/goes16_20200918_1950/OR_ABI-L1b-RadF-M6C* \
--reader_name abi_netcdf \
--product_name Infrared \
--output_formatter imagery_annotated \
--sector_list conus

The end of the output from the above script should resemble the example below, showing the path to the generated output image. The final line should display Return value: 0, indicating that the script ran successfully.

# Truncated log output from the script run
.......
11_191109    log_setup.py:162  INTERACTIVE: SINGLESOURCESUCCESS ${GEOIPS_OUTDIRS}/preprocessed/annotated_imagery/NorthAmerica-UnitedStates-Continental/x-x-x/Infrared/abi/20200918.195020.goes-16.abi.Infrared.conus.97p12.noaa.3p0.png
11_191109    log_setup.py:162  INTERACTIVE: READER_NAME: abi_netcdf
11_191109    log_setup.py:162  INTERACTIVE: PRODUCT_NAME: Infrared
11_191109    log_setup.py:162  INTERACTIVE: NUM_PRODUCTS: 1
11_191109    log_setup.py:162  INTERACTIVE: NUM_DELETED_PRODUCTS: 0
11_191109    log_setup.py:162  INTERACTIVE: Completed geoips PROCFLOW single_source processing, done!
11_191109    log_setup.py:162  INTERACTIVE: Total time: 0:00:24.911853
11_191109    log_setup.py:162  INTERACTIVE: Return value: 0