Installation#

SCIP can be installed from the PyPi repository, or from source.

We recommend installing SCIP in a conda environment using mambaforge.

Prerequisites#

If you want to use GPU-accelerated Cellpose segmentation, make sure to install a compatible cudatoolkit. The version you need depends on which GPU model you want to use.

If you want to run SCIP on a dask-mpi cluster, an MPI implementation must be available. We recommend installing mpich.

mamba install -c conda-forge mpich-mpicc

PyPi#

SCIP can be installed from the PyPi repository. The base installation can be installed as follows:

pip install scip

There are also optional dependencies: * mpi: Add support for running SCIP using a dask-mpi cluster, * cellpose: Add support for cellpose functionality, * czi: Add support for reading Carl Zeiss Image files, and * dev: Installing extra package related to development.

These can be installed using pip’s bracket notation:

pip install scip[mpi]
pip install scip[cellpose]
pip install scip[cellpose,mpi,czi]

From source#

To install SCIP from source:

  1. Download the latest release from Github, or clone the repository.

  2. Optionally extract the release.

  3. Enter the repository directory.

  4. Run pip install . if you only need default functionality.

  5. (Optional) Run pip install .[extra] for extra functionality (refer to the list above for available extras)