Skip to content

Installing climakitae locally using conda environment

Calvin Chen edited this page Nov 14, 2024 · 6 revisions

Installing climakitae (1.0.1) locally using conda environment

climakitae is primarily designed to be run in a pangeo-notebook docker image implementation of JupyterHub. The JupyterHub allows for cloud computing and user interfaces within a Jupyter Notebook. The base climakitae package functionality is also useful outside of this environment so there is motivation to understand how to install climakitae locally.

Anaconda is the best python environment to install climakitae locally, although climakitae can be pip installed on Ubuntu (or other Unix variants) if you create a python virtual environment that uses python 3.9.13. More than likely you will have to compile python 3.9.13 from source and then use that as a basis for the virtual environment.

If you want to install the climakitaegui package that contains the GUI tools for use with climakitae do these steps: https://github.com/cal-adapt/climakitaegui/wiki/Installing-climakitaegui-locally-using-conda-environment

Ubuntu installation steps

  1. Install miniconda3: https://docs.anaconda.com/miniconda/
  2. Download the conda lockfile from the cae-environments repo: https://github.com/cal-adapt/cae-environments/blob/main/climakitae/1.0.1/conda-linux-64.lock
  3. Run: conda create -n climakitae --file conda-linux-64.lock (either in the directory with the lockfile or change the path to where it is on your system)
  4. Run: conda activate climakitae
  5. Run: pip install numba==0.53.1 h3==3.7.4 timezonefinder==6.0.2 dask-gateway==2023.1.1 dask-geopandas==0.3.1 xclim==0.42.0
  6. Run: pip install -e git+https://github.com/cal-adapt/[email protected]#egg=climakitae

Windows installation steps

  1. Install Anaconda on your system: https://www.anaconda.com/
  2. Download the conda lockfile from the cae-environments repo: https://github.com/cal-adapt/cae-environments/blob/main/climakitae/1.0.1/conda-win-64.lock
  3. Open Anaconda terminal
  4. Run: conda create -n climakitae --file conda-win-64.lock (either in the directory with the lockfile or change the path to where it is on your system)
  5. Run: conda activate climakitae
  6. Run: pip install numba==0.53.1 h3==3.7.4 timezonefinder==6.0.2 dask-gateway==2023.1.1 dask-geopandas==0.3.1 xclim==0.42.0
  7. Run: pip install -e git+https://github.com/cal-adapt/[email protected]#egg=climakitae

MacOS installation steps

  1. Install Anaconda on your system: https://www.anaconda.com/
  2. Install llvm and libomp using homebrew: brew install llvm libomp
  3. Edit .bashrc (or your preferred shell env) and add:
export CC=/opt/homebrew/opt/llvm/bin/clang
export LDFLAGS="-L/opt/homebrew/opt/libomp/lib"
export CPPFLAGS="-I/opt/homebrew/opt/libomp/include"
PATH="/opt/homebrew/opt/llvm/bin:$PATH"
  1. Run source .bashrc (or where ever your put the lines from step 3)
  2. Download the conda lockfile from the cae-environments repo: https://github.com/cal-adapt/cae-environments/blob/main/climakitae/1.0.1/conda-osx-arm64.lock
  3. Open Anaconda terminal
  4. Run: conda create -n climakitae --file conda-osx-arm64.lock (either in the directory with the lockfile or change the path to where it is on your system)
  5. Run: conda activate climakitae
  6. Run: pip install numba==0.53.1 h3==3.7.4 timezonefinder==6.0.2 dask-gateway==2023.1.1 dask-geopandas==0.3.1 xclim==0.42.0
  7. Run: pip install -e git+https://github.com/cal-adapt/[email protected]#egg=climakitae