# Neural Network

**URL:** https://forum.makecode.com/t/neural-network/45964
**Category:** Show & Tell
**Created:** [September 19, 2026, 12:07am UTC](https://forum.makecode.com/t/neural-network/45964 "2026-09-19T00:07:18Z")
**Posts on this page:** 1
**Page:** 1

<div class="post-metadata">

### Author: ![BINARIUS\_X](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.makecode.com/binarius_x/32/29055_2.png) [@BINARIUS\_X](https://forum.makecode.com/u/BINARIUS_X)
#### Post date: [September 19, 2026, 12:07am UTC](https://forum.makecode.com/t/neural-network/45964/1 "2026-09-19T00:07:18Z")

</div>

> **[Neural Network](https://makecode.microbit.org/56730-82968-12394-40565)**
>
> Erstellt mit ❤️ in Microsoft MakeCode for micro:bit.

I’ve created a small neural network. x is the input, and y is the target value the network aims to reach. The variable “Loss MSE” represents the loss value. You can only view these values ​​in debug mode. The network consists of an input layer, two hidden layers, and an output layer, with three neurons in each layer. ReLU is used for the hidden layers, and Sigmoid for the output layer. y2 represents the output value after the Sigmoid function. If you’d like to try out the network yourself, feel free to modify x and y. If you have any questions, please feel free to ask.