Software Abstract
The software I chose to evaluate in my project is Python Numba. It is a compiler aiming to speed up applications. Numba is designed for Python arrays and numerical functions; specifically, it works best on code that uses loops, Numpy arrays and functions. Without having to switch languages, it increases optimization for code written in Python that is particularly math-heavy, utilizing functions with high performance. In order to call on Numba to compile functions, a selection of decorators are available to be applied. The option to parallelize functions automatically is available in the Numba decorators as well.
Installation
In order to install Python Numba on the HPCC, here are provided step-by-step instructions:
-
Log onto the HPCC
-
Log onto a development node
-
Run this command:
pip install numba -
To check if installation was successful, run these commands:
python
import numba
numba.__version__This should output the version of Numba you have installed.