NeuralNet

Creates any number of layers of feed-forward neural network (MLPs) easily.

TO GET IT WORKING

• Download or clone the project.
• Use ‘data_loader.py’ to load dataset.

• Change the values as required for loading data in ‘data_loader.py’. See the file for more info.

• Now in ‘model.py’ just specify [no. neurons, activation function to use, dropout] for each layer you create.

This will create a neural network with any no. of layers as you specify.

Change hyper-parameters as required.

Run the script; accuracy can be monitored as the training progresses.