This repository contains the code and reports for the mini-projects done in McGill University's Applied Machine Learning course in Winter 2020.

The given task was to implement 2 classification models: KNN and Decision Tree with appropriate cost function in order to classify targets from 2 datasets acquired from UCI Machine Learning Repository, namely Breast Cancer dataset and Hepatitis dataset.

A report was written to indicate our results and findings.

The given task was to classify textual data by implementing Naive Bayes and K-fold cross-validation from scratch using logistic regression from scikit-learn package. The model was trained on two datasets: 20 newsgroup and IMDB Reviews. We performed hyperparameter tuning using randomized search with cross validation to achieve the best possible performance for each model.

A report was written to indicate our results and findings.

The given task was to implement a Multilayer Perceptron from scratch, including its backpropagation and a mini-batch gradient descent algorithm. Additionally, we were to develop a Convolutional Neural Network. The accuracies and losses of training and validation splits were recorded during training. We compared both models and their performance.

A report was written to indicate our results and findings.

The given task was to reproduce selected experiments from a paper.

A report was written to indicate our results and findings.