Neural Network Playground

Configuration

The neural network here is to solve classification problem. The dataset that is used here is logical OR with features x1 and x2. Below is truth table of OR, You can edit it all you want though. Put epoch as high as you want, overfitting here is good since ALL possible data is already defined in the dataset (that is if you haven't edited it).

Dataset
X1 X2 Y
Parameters
learning rate epoch hidden neurons hidden layers biases
1 none
Hit the start button to start training process. Please note, switching browser tab might cause a little lag.

Training

Error: N/A
Epoch: N/A
X1 X2 Y T 0 1 1 1 F 1 0 0 0 Sorry, your browser does not support inline SVG.
X1 X2 expected Y predicted Y
0 0
0 1
1 0
1 1
Matrices
input layer weight 1 hidden layer weight 2 output layer
[ ]
[ ]
[ ]
[ ]
[0 0 0]
[0 0 0]
[0 0 0]
[0 0 0]
[0 0 0]
[0 0 0]
[0 0]
[0 0]
[0 0]
[0 0]
[0 0]
[0 0]
[0 0]

Testing

This is a live testing, you can do it before or after the training process ends. Type in an input on the text input with blue underline.
Sorry, your browser does not support inline SVG.

About this playground

There are many things that I haven't implemented here like bias or adding neurons or hidden layers and many more. If you wish to contribute, please pull request this repository.