Machine Learning Labs

Authors: Bastian Krohg & Nicolas Siard This repo contains 3 submodules with different ML projects from INSA Toulouse

Lab 1 - Perceptron

• Develop and train a linear perceptron that can be trained using a dataset from the hydrodynamic performance of sailing yachts
• Gradient descent method to train the model by updating the weights according to losses for each iteration

Lab 2 - MLP (Multi-Layer Perceptron)

• Split data sets, train and tune a multi-layer perceptron

Lab 3 - Evaluation

• Decision Trees, Evaluation and Confusion Matrix