Up until now you learnt about how to solve regression problems. In this video you will switch hats to solve classification problems - that is what class something belongs to. Learn how to handle data with thousands of inputs such as images, where each pixel is a feature, and instead of regression to predict a number you will solve a classification problem to predict what class an image belongs to. This video covers what classification is and how to create an image classifier for the MNIST dataset that can classify handwritten digits that works via TensorFlow.js right in the web browser introducing new concepts such as 1-hot encodings to represent your class data and categorical cross entropy.
Catch more episodes from Machine Learning for Web Developers (Web ML) → [ Ссылка ]
Check out TensorFlow on YouTube → [ Ссылка ]
Subscribe to Google Developers → [ Ссылка ]
Connect with Jason Mayes to ask questions:
LinkedIn → [ Ссылка ]
Twitter →[ Ссылка ]
Discord →[ Ссылка ]
Use #WebML to share your learnings and creations from this course to meet your peers on social media!
See what others have already made with Web ML → [ Ссылка ]
![](https://i.ytimg.com/vi/BqiOc7iCut0/maxresdefault.jpg)