Machine learning final project. A c++ implementation of a multi layer perceptron (MLP).
The neural network can solve classification and regression tasks.
The code implements:
- MLP training with backpropagation
- grid search for model selection
- regularization: early stopping and weight decay
In the report there is also a performance comparison with an implementation provided by KNime.
It has been applied for classification and a "blind" regression task (competition dataset).