Adding Layers and Forward Functions to Your Neural Network in PyTorch

Dive into a comprehensive 53-minute deep learning workshop focused on building neural networks with PyTorch. Learn to create network models with input, output, and hidden layers, and craft forward functions to process data effectively. Explore how to design network architectures for various data types, understand the importance of input features in model setup, and implement forward functions tailored to specific data needs. Gain hands-on experience with structured and unstructured data-based network design and coding. The workshop covers essential topics like using nn.Module, defining base neural network models, integrating models with data, and exploring complex forward functions. Practical examples include building models for heart disease prediction and MNIST digit recognition. Access the accompanying GitHub notebook for a deeper understanding of the concepts presented.

Syllabus

- Workshop #2 Introduction
- Topics in Workshop #2
- What you will learn in this workshop?
- RECAP from Workshop #1
- Workshop #2 Kickoff
- Using nn.Module in Python
- Defining base neural network model
- Design of your neural network class
- Neural network class based on data input
- Neural network shape
- Model and Data Integration
- Network with input and output layer
- Network with 1 hidden layer
- Network with multiple hidden layers
- Multilayered Network Model
- Role of Forward function in network
- CNN Explorer Intro
- Complex Forward Function
- Heart Disease Problem network model
- . MNIST digits recognition problem model
- nn.Parameters in PyTorch
- Your Homework
- Push notebook to GitHub
- RECAP
- Workshop #3 Agenda