MENU

Fun & Interesting

I Built a Neural Network in C# From Scratch. Here’s What I Learned…

Milan Jovanović 25,426 lượt xem 8 months ago
Video Not Working? Fix It Now

Get the source code for this video for FREE → https://the-dotnet-weekly.ck.page/neuralnet
Want to master Clean Architecture? Go here: https://bit.ly/3PupkOJ
Want to unlock Modular Monoliths? Go here: https://bit.ly/3SXlzSt
Join a community of 1000+ .NET developers: https://www.patreon.com/milanjovanovic

In this video, we'll cover the theoretical foundations of neural networks, explore their architecture, and then put theory into practice by building a working neural network using C#. The neural network will learn to recognize handwritten digits with more than 99%+ accuracy.

Whether you're a C# developer looking to explore AI or an AI enthusiast curious about implementation details, this video offers a practical and insightful look into the world of neural networks.

Digit Recognizer: https://github.com/m-jovanovic/digit-recognizer/

Check out my courses:
https://www.milanjovanovic.tech/courses

Read my Blog here:
https://www.milanjovanovic.tech/blog

Join my weekly .NET newsletter:
https://www.milanjovanovic.tech

Chapters
0:00 Neural Networks - Introduction
1:24 Theory behind neural networks
6:43 Gradient Descent, Backpropagation
9:28 Supervised Learning
10:42 Digit Recognizer in C#
12:54 Demo: Neural Network training
14:42 Demo: Digit Recognizer UI
16:39 Achieving 99%+ accuracy

Comment