Simple Python Acceleration: Optimising Loops and Functions with Numba

Speeding up Python loops and NumPy functions with minimal code changes using JIT compilation decorators.

Discover how Numba — a powerful Just-in-Time (JIT) compiler for Python — can dramatically speed up your numerical and scientific computing workflows.

Through hands-on coding notebooks, you'll Learn:

CPU Parallelisation: Leverage Numba to execute Python code at near-native speeds.

GPU Acceleration: Harness the power of CUDA-enabled GPUs for massive parallelism.

Prerequisites

1. Basic experience with Python (Programming with Python).

2. Basic experience with bash or similar Unix shells. (Software Carpentry - The Unix Shell )

3. User experience on Gadi. (Seminar on Getting Started with Gadi)

Learning Outcomes

At the completion of this training, you will be able to 

• applying Numba’s features

• optimising functions

• integrating with libraries like NumPy

• understand how to use Numba to improve code performance on both CPU and GPU