I will start off this tutorial by quickly going over the oh so important, ND Array. I will then talk about indexing and broadcasting as they are features that make Numpy unique. Once this short introduction is done, I will get into the more advanced Numpy features. I will try to cover as many of the following topics.

Ufuncs in Numpy

Stride tricks is one of my favourite Numpy features and I will show you how to do clever things with this like applying a sliding window to your array and convolution operations in a few lines of code.

Einsum is a powerful function that gives you the ability to implement Einstein summation techniques on your arrays.

I will talk about how to subclass numpy arrays, as this is useful to know if you're ever creating your own libraries that leverage Numpy capabilities.

In addition to this, I will show you cool ways to draw shapes like circles or ellipses inside a Numpy array using masks and I will introduce you to the Numpy mesh function np.ix_.

For those of you who generally work with large data sets, you'll find that loading/saving your data into files can be a performance bottleneck and sometimes infeasible. I will introduce Memory mapped files and show you how you can work with large files without compromising on performance.

I will take you through each of these functions and talk about how they can help you, and how they have helped me with some of my projects.

This will be an interactive tutorial where the focus will be on attendees getting hands on experience with these Numpy features so that they can use them in their projects.