Representation-Learning

Programming assignments for the course IFT6135 Representation Learning.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Assignement 1

Problem 1: MLP

In this problem, we will build a Multilayer Perceptron (MLP) and train it on the MNIST hand- written digit dataset.

You can find the code for the first problem in the python script (mlp.py) under the folder problem 1.

python mlp.py

Problem 2: CNN

For this part of the assignment we will train a convolutional network on MNIST using Pytorch.

You can find the code for the second problem in the python script under the folder problem 2.

python cnn.py

Problem 3: Kaggle competition

(https://www.kaggle.com/c/ift6135h19) Dogs vs. Cats is an InclassKaggle challenge for image classification. We trained a CNN classifier for this task.

You can find the code for the third problem in the jupyter notebook script under the folder problem 3.

Built With

• keras - ML library
• Pytorch - ML library

Team (The Machinists)

• Fares Ben Slimane (faresbs) - check my personal webpage

Acknowledgments

• This project was made as part of the MILA course IFT6135 Representation Learning (Instructor:Prof.Aaron Courville).