A simple neural network built from scratch in Java.
- Clone the repository:
$ git clone https://github.com/dannydenenberg/simple_nn_in_java.git
- Go into the source:
$ cd simple_nn_in_java/src
- Pick out the libraries you want to use and copy the file into your code base
- This is a matrix library built from scratch in Java.
- Includes all basic element wise operations as well as dot product, transpose, print functions, and more.
- A simple class to represent the shape of a matrix.
- Represents a single activation function.
- During initialization, specify the activation function you want to use, and when you call
activate(double number), it will use the specified function
- A single layer in the network.
- Has its own weights, biases, and activation functions.