Welcome

This website provides a comprehensive introduction to PyTorch, a popular Python library for deep learning. The content is organized into three main parts.

• Quick Reference: A concise summary of PyTorch workflows.

• Tutorials: Step-by-step explanations of deep learning techniques using PyTorch.

• Projects: Hands-on projects that apply the concepts learned in the tutorials.

The material is intended for students enrolled in a deep learning course. While no prior experience with PyTorch is needed, a basic understanding of Python and linear algebra is assumed.

Lectures

These lectures provide a high-level overview of the topics covered in the tutorials and projects.

Slides	Description
Lecture01.pdf	Introduction to supervised learning and neural networks.
Lecture02.pdf	Best practices for training and evaluating neural networks.
Lecture03.pdf	Introduction to convolutional neural networks.
Lecture04.pdf	Overview of representation learning techniques.

Downloads

The following files are available for download.

File	Description	Explained in…
train.py	Training loop	Quickref - MLP
eval.py	Evaluation loop pluggable into the training loop	Quickref - CNN
metrics.py	Evaluation metrics	Quickref
pretrained.py	MobileNetV3 with support for transfer learning	Fine Tuning
triplet.py	Triplet loss	Triplet Loss