Dive into the fundamentals of Convolutional Neural Networks (CNNs) in this comprehensive 39-minute lecture from the Full Stack Deep Learning Spring 2021 series. Begin with a review of the convolution operation, the cornerstone of CNNs, before exploring other crucial operations. Gain insights into convolutional filters, filter stacks, strides, padding, and filter math. Delve into implementation notes and advanced concepts like dilated convolutions for increasing receptive fields, and pooling and 1x1 convolutions for decreasing tensor size. Conclude by examining the classic LeNet architecture, providing a practical example of how these concepts come together in a real-world CNN design.
Convolutional Neural Networks - Full Stack Deep Learning - Spring 2021
Overview
Syllabus
- Introduction
- Convolutional Filters
- Filter Stacks and ConvNets
- Strides and Padding
- Filter Math
- Convolution Implementation Notes
- Increasing the Receptive Field with Dilated Convolutions
- Decreasing the Tensor Size with Pooling and 1x1-Convolutions
- LeNet Architecture
Taught by
The Full Stack
Related Courses
-
Neural Nets for NLP 2020 - Convolutional Neural Networks for Text
-
Introduction to Convolutional Neural Networks - Part 1 - Lecture 6
-
Neural Nets for NLP 2019 - Convolutional Neural Networks for Language
-
Convolutional Neural Networks Explained Easily
-
Convolutional Neural Networks
-
CAP5415 - Introduction to Convolutional Neural Networks Part 2 - Lecture 6
