An academic project aimed at studying some of the simplest classification machine learning techniques by applying them on a couple of MNIST-like databases. Using an UI it allows to train and test ML models such as:
- Multilayer perceptron (tensorflow Keras)
- Multilayer perceptron (self-implemented using numpy)
- Convolutional neural network (tensorflow keras, simple)
- Convolutional neural network (tensorflow keras, complex)
- Decision tree (pytorch)
- Decision forest (pytorch)
- KNN
A precise report (in polish) available at ./Report/report.pdf