Deep learning with fastai


Tuesday, July 6
9 am–5 pm Pacific Time

This course, suitable for people with no knowledge of machine learning, will walk you through the core concepts of neural networks and give you the tools necessary to build your own models to solve problems. We will use fastai—a deep learning library building on top of PyTorch—as it gives the option of high-level fast implementation without compromising with the lower level flexibility of PyTorch.

Instructor: Marie-Hélène Burle (WestGrid)

Prerequisites: Working knowledge of Python or attendance at the Basics of Python course.

Software: All attendees will need a remote secure shell (SSH) client installed on their computer. On Windows we recommend the free Home Edition of MobaXterm. On Mac and Linux computers SSH is usually pre-installed (try typing ssh in a terminal to make sure it is there).



Zoom   9–9:30 am Pacific Time
Introduction to the course & cluster access


On your own
AI, machine learning, deep learning
Which framework to choose?
(Optional) Local installation
Documentation
Artificial neural networks
How do neural networks learn?


Zoom   12–1:30 pm Pacific Time
Questions so far
PyTorch tensors
The MNIST


On your own
Backpropagation
A bit of calculus
Automatic differentiation


Zoom   3–5 pm Pacific Time
Our NN to classify the MNIST
Final questions


(Additional readings)
HPC with Python
A few notes on PyTorch distributed computing