Algorithms related to biologically inspired computing, optimisation and search implemented for educational purpose in coherence with course INF3490.
Supervised learning with a multilayer perceptron (MLP). In order to control a robotic prosthetic hand, Prosthetic hand controllers (PHCs) reads the electromyographic signals generated by contracting muscles in the under arm. The idea is that we can, through supervised learning, get a PHC to learn which hand movement the user of a robotic prosthetic hand wants toperform.
- mlp_oo_neuron ✅
- A crude and complicated implementation of an MLP as a challenge. OOP with a neuron class and no matrix functions or machine learning modules.
- mlp_2d_list 🚧
- A somewhat less complicated, but crude implementation of an MLP. 2D list and no matrix functions or machine learning modules.
- mlp_numpy ✅
- Simple two-layer MLP using numpy and matrix operations.