Handwriting Recognition With MNIST Data in Java

This uses my neural network Java library that can be found here. The trained weights can also be found in the GitHub repository. The test accuracy is around 95% using a very simple 3 layer (784/300/10, sigmoid/sigmoid/softmax) neural network. Some images:

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s