Python implementation of feed-forward multi-layer perceptron (MLP) neural networks using numpy and scipy based on theory taught by Andrew Ng on coursera.org and adapted from the Octave code examples from this course as well as updates from the 2017 deeplearning.ai Deep Learning Specialization course..
This is still a work-in-progress. Still some bugs and fixes to do so don't recommend using this yet...
- Classification
- Prediction
- Model Predictive Control (MPC)
- Approximation of non-linear functions
MLPLayer- Multi-layer perceptron neural network layer class (used by MLPNetwork)MLPNetwork- Multi-layer perceptron neural network classMLPTrainingData- Object to store training data (inputs and outputs).
train- Used to handle training of a network.