TensorFlow CNN Tutorial for Handwritten Digit Recognition
Learn to build a CNN in TensorFlow to recognize handwritten digits from the MNIST dataset. Updated for compatibility. 📊

Hvass Laboratories
197.4K views • Jul 12, 2016

About this video
How to make a Convolutional Neural Network in TensorFlow for recognizing handwritten digits from the MNIST data-set.
This tutorial has been updated to work with TensorFlow 2.1 and possibly later versions using "v.1 compatibility mode".
https://github.com/Hvass-Labs/TensorFlow-Tutorials
The 2nd convolutional layer can be hard to understand because of the multiple input and output channels. Here is the math formula but it is probably even more confusing: https://www.tensorflow.org/api_docs/python/tf/nn/conv2d
This tutorial has been updated to work with TensorFlow 2.1 and possibly later versions using "v.1 compatibility mode".
https://github.com/Hvass-Labs/TensorFlow-Tutorials
The 2nd convolutional layer can be hard to understand because of the multiple input and output channels. Here is the math formula but it is probably even more confusing: https://www.tensorflow.org/api_docs/python/tf/nn/conv2d
Tags and Topics
Browse our collection to discover more content in these categories.
Video Information
Views
197.4K
Likes
2.3K
Duration
36:20
Published
Jul 12, 2016
User Reviews
4.5
(39) Related Trending Topics
LIVE TRENDSRelated trending topics. Click any trend to explore more videos.