Scribd
Upload
77 views18 pages

Fundamentals of Deep Learning: Part 5: Pre-Trained Models

This document discusses pre-trained models and transfer learning. It reviews concepts from previous parts such as learning rates, network architecture, and data augmentation. It introduces pre-trained models available in frameworks like PyTorch Hub and how they can be used for transfer learning by taking a pre-trained model as a starting point and retraining or fine-tuning it for a new task. The document demonstrates how transfer learning can take a model trained for general image classification and customize it for a more specific task like automated dog door detection.

Uploaded by

Praveen Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
77 views18 pages

Fundamentals of Deep Learning: Part 5: Pre-Trained Models

This document discusses pre-trained models and transfer learning. It reviews concepts from previous parts such as learning rates, network architecture, and data augmentation. It introduces pre-trained models available in frameworks like PyTorch Hub and how they can be used for transfer learning by taking a pre-trained model as a starting point and retraining or fine-tuning it for a new task. The document demonstrates how transfer learning can take a model trained for general image classification and customize it for a more specific task like automated dog door detection.

Uploaded by

Praveen Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 18

FUNDAMENTALS OF

DEEP LEARNING
Part 5: Pre-trained Models

1
Part 1: An Introduction to Deep Learning

Part 2: How a Neural Network Trains

AGENDA Part 3: Convolutional Neural Networks

Part 4: Data Augmentation and Deployment

Part 5: Pre-trained Models

Part 6: Advanced Architectures


AGENDA – PART 5
• Review so far
• Pre-trained Models
• Transfer Learning
REVIEW SO FAR
4
REVIEW SO FAR

• Learning Rate
• Number of Layers
• Neurons per Layer
• Activation Functions
• Dropout
• Data

5
PRE-TRAINED MODELS
6
PRE-TRAINED MODELS

PYTORCH
HUB
7
PRE-TRAINED MODELS

IM GENET
8
THE NEXT CHALLENGE
An Automated Doggy Door

9
TRANSFER LEARNING
10
THE CHALLENGE AFTER
An Automated Presidential Doggy Door

11
TRANSFER LEARNING

12
TRANSFER LEARNING

…
…

…
(28, 28, 2) (28, 28, 2)
Stacked Images Stacked Images

(3, 3, 1, 2) (3, 3, 2, 2) (512) (512)


Kernels Kernels Dense Dense
(28, 28,1) (1568)
Image Input Flattened Image (10)
Vector Output Prediction

13
Input

More
Generalized
Convolution

Max Pooling

Convolution

Dropout

Max Pooling

Convolution

Max Pooling
TRANSFER LEARNING

Dense

Dense

Output
More
Specialized
TRANSFER LEARNING
Freezing the Model?

15
TRANSFER LEARNING

16
LET’S GET STARTED!
17
18

You might also like