Deep Learning for Image Classification
Course Code: Y1C1
ECTS Credits: 6.0
Course Description
The Deep Learning for Image Classification course provides students with the foundational knowledge and practical skills needed to build, train, and evaluate neural networks for image classification tasks. It helps students develop a clear understanding of how deep learning models function, how to structure and optimize them, and how to apply them effectively. Through a combination of theoretical insights and hands-on implementation, students learn to work with industry-standard libraries such as TensorFlow and Keras, while building a strong intuition for the key components of neural networks.
Students explore core techniques including multilayer perceptrons, convolutional neural networks, data augmentation, and transfer learning. They also learn to assess model performance, conduct error analysis, and iterate on their solutions using principles from the machine learning project lifecycle. By the end of the course, students are able to independently develop and refine deep learning models with a strong focus on experimentation, evaluation, and practical application.
Course Content
- Introduction to Deep Learning
- The Mathematical Building Blocks of Neural Networks
- Multilayer Perceptron from Scratch
- Neural Networks for Regression and Classification
- Evaluating and Improving the Performance of Neural Networks
- Introduction to TensorFlow and Keras
- Convolutional Neural Networks for Image Classification
- Data Augmentation
- Transfer Learning
- Error Analysis
- The Machine Learning Project Lifecycle
Prerequisites
- Introduction to Programming with Python (Y1A1)
- Introduction to Machine Learning (Y1B1)
Recommended Reading
- Deep Learning with Python by François Chollet