Installation¶

Prerequisites¶

To access the full capabilities of CVPy, you need the following:

Python version 3 or greater (You can install the Anaconda distribution or another distribution of your choice.)
SAS Scripting Wrapper for Analytics Transfer (SWAT) package for Python
SAS Viya 3.5+ with an active Visual Data Mining and Machine Learning (VDMML) license
Mayavi for scientific data visualization in Python.

You also need valid user credentials to access the SAS Viya backend. See your system administrator for details if you do not have a SAS Viya account.

The examples in the repository are written using Jupyter notebooks. If you wish to run the notebooks in your own environment, you must also have Jupyter installed. If you do not wish to install Jupyter, you can still view the examples in your browser on the GitHub website.

Mayavi Installation¶

Mayavi is a heavy, complex package that is required by CVPy. If you do not already have Mayavi installed, follow the steps below to install it.

Install VTK, a critical dependency for Mayavi.

pip install VTK‑8.1.2‑cp37‑cp37m‑win_amd64.whl

or

pip install VTK==8.1.2

Note that a specific version of is VTK used. You can download the wheel file from here.

Install Mayavi

pip install mayavi

You may see an error message about numpy, like “Numpy is required to build Mayavi correctly, please install it first.” Ignore that message.

Install PyQt, a GUI toolkit needed to run Mayavi

pip install PyQt5

CVPy Installation¶

To install CVPy, use the following command:

pip install git+https://github.com/sassoftware/python-cvpy.git