Practical Transfer Learning ( Deep Learning )in Python

Don't Be Hero - Next Frontier in Deep Learning Image Classification and Object Detection Problems solution - Keras

Last updated 11/2022

English

What you'll learn

8 sections • 53 lectures • 5h 45m total length

Introduction and Course Outline3:17
Do you want to be poet?7:14
Becoming Poet Part 19:20
https://kiosk-dot-codelabs-site.appspot.com/codelabs/tensorflow-for-poets/#0
Becoming Poet Part 213:03
Becoming Poet Part 33:39
Three main motivation for Transfer Learning4:59
Success of Transfer Learning - Andrew Ng3:37
Transfer Learning vs Traditional ML and Deep Learning4:48
Transfer Learning vs Traditional ML and Deep Learning 24:25

Don't be Hero . as It is well said..

Let;s Enroll and utilize works of Hero for our problems.

Everyone can not do research like Yann Lecun or Andrew Ng. They are focused on improving machine learning algorithms for better world.

But as an individual and for industry, we are more concern with specific application and its accuracy.

Transfer Learning is the solution for many existing problems. Transfer learning uses existing knowledge of previously learned model to new frontier.

I will demonstrate code to do Transfer Learning in Image Classification.

Knowledge gain to recognize cycle and bike can be used to recognize car.

There are various ways we can achieve transfer learning. I will discuss Pre trained model, Fine tunning and feature extraction techniques.

Once again. Let's not be Hero . and enroll in this course.