This repository contains my solutions and implementations for the Deep Learning course offered in the Fall 2025 semester at the University of Science and Technology, under the supervision of Dr. Mohammadi.
It is part of the Master’s program in Artificial Intelligence.
The course introduces the fundamental principles and practical techniques of Deep Learning, with emphasis on building, training, and evaluating neural networks using modern frameworks.
It covers both theoretical foundations and hands-on exercises involving real-world datasets.
| Week | Topic | Description |
|---|---|---|
| 1 | Introduction to Neural Networks | Perceptron, activation functions, forward & backward propagation |
| 2 | Optimization | Gradient descent, stochastic methods, momentum, Adam |
| 3 | Regularization | Overfitting, dropout, batch normalization |
| 4 | Convolutional Neural Networks (CNNs) | Image classification and feature extraction |
| 5 | Recurrent Neural Networks (RNNs) | Sequential data modeling, LSTM, GRU |
| 6 | Autoencoders | Dimensionality reduction and unsupervised feature learning |
| 7 | Generative Models | Variational Autoencoders (VAE), GANs |
| 8 | Advanced Topics | Transfer learning, attention mechanisms, and transformers |
| 9 | Final Project | Implementation of a complete deep learning pipeline |
- Python 3.10+
- NumPy, Pandas
- TensorFlow / PyTorch
- Matplotlib, Seaborn
- Jupyter Notebook
- Clone the repository:
git clone https://github.com/Rahmat-ML/IUST-DL-Fall2025.git-
Navigate to the project folder:
cd IUST-DL-Fall2025 -
Install dependencies:
pip install -r requirements.txt
-
Open Jupyter Notebook to explore the exercises:
jupyter notebook
Deep-Learning-Course-Exercises/
│
├── HW1
├── LICENCE
└── README.mdDr. Mohammadi
Faculty of Computer Engineering
Iran University of Science and Technology
Rahmat Ansari
Master’s Student in Artificial Intelligence
University of Science and Technology
Fall 2025
This repository is for educational purposes only.
All rights reserved © 2025 Rahmat.