Table of Contents

1. Convolutional Neural Network
2. Types of Neural Network
3. Computer Vision
4. Image Data
5. Image Data
6. Convolutions
7. Edge Detection
8. Parameters
9. Padding
10. Strided convolutions
11. Summary of Convolutions
12. Convolutions Over Volume
13. Your Turn: Number of Parameters in One Layer
14. Pooling Layers
15. Pooling Layers
16. Types of Layer in A Convolutional Network
17. Classic CNNs: LeNet-5
18. Classic CNNs: AlexNet
19. Classic CNNs: VGG-16
20. Classic CNNs: ResNets
21. Using Keras To Build CNN
22. Using Keras To Build CNN
23. Using Keras To Build CNN
24. Using Keras To Build CNN
25. Size of the Model
26. Different Architecture Search Algorithms:
27. Understanding Neural Networks

Convolutional Neural Network

Hui Lin @Google

Types of Neural Network

Computer Vision

Image Classification
Object Detection
Neural Style Transfer

Image Data

Image Data

Convolutions

HTML5 Icon

Edge Detection

Vertical Edge Detection

Parameters

Padding

Pad so that output size is the same as the input size.

Strided convolutions

()
()
no padding
stride

Summary of Convolutions

image; filter
padding ; stride
Output:

Convolutions Over Volume

https://www.youtube.com/watch?v=HzuFgLnhgqw

Your Turn: Number of Parameters in One Layer

Question: If you have 10 filters that are in one layer of a neural network, how many parameters does that layer have?

Pooling Layers

Pooling Layers