with mamba

spudtr is packaged for installation with mamba or conda. The following examples show how to install into a fresh conda environment with some useful various options.

On computers with high performance Intel CPUs you may get better performance with MKL accelerated libraries.

mamba create --name spudtr_env -c conda-forge spudtr "blas=*=mkl*"

On AMD CPUs OpenBLAS may be faster.

mamba create --name spudtr_env -c conda-forge spudtr "blas=*=openblas*"

If you need to a specific version of Python or other packages use conda package selection syntax.

mamba create --name spudtr_env -c conda-forge spudtr "blas=*=openblas*" python=3.8 jupyter-lab

with conda

If mamba is not available, you can use conda.

conda create --name spudtr_env -c conda-forge spudtr "blas=*=mkl*"