Chrysolites Tech Academy

Video by freeCodeCamp.org · taught by Daniel Bourke · streamed from YouTube. All rights remain with the creator.

Lesson 1: PyTorch for Deep Learning

About this lesson

This long beginner course by Daniel Bourke teaches the foundations of machine learning and deep learning with PyTorch, the Python framework, writing code throughout rather than relying on slides. It is based on a free online book and code repository that accompany the video. The first section covers PyTorch fundamentals: what deep learning is and why it is useful, what PyTorch is used for, and how to create and manipulate tensors, the basic data structure of deep learning, including their shapes, data types and devices. The second section teaches the PyTorch workflow from end to end, preparing data, building a model, training it, evaluating it, improving it and saving and loading it, using a simple straight-line problem whose true answer is known. The third section builds neural networks for classification, starting with a toy problem of separating two groups of points and showing why non-linear activation functions are needed. The fourth turns to computer vision, using torchvision to load a dataset of clothing images and building a convolutional neural network that recognises the patterns in them. The final section covers custom datasets, showing how to load your own images into PyTorch and building the course's main project, a classifier for photos of different foods, ending with predictions on custom images. Exercises and extra-curricular material accompany every section. Afterwards, read Chapter 1 of the course book, which works through the training loop by hand.

Install Chrysolites Tech Academy

Android / Chrome / Edge: open the browser menu and choose Install app or Add to Home screen.

iPhone / iPad (Safari): tap the Share button, then Add to Home Screen.