This project, titled "Predicting Real Estate Values," served as a pivotal assignment for the Modern Analytics coursework. The primary objective was to employ Multilayer Perceptron (MLP) neural networks to predict house prices based on various features such as square footage, number of rooms, kitchen amenities, bathrooms, and location.
- Programming Language: Python
- Deep Learning Framework: PyTorch
- Neural Network Architecture: Custom MLP (Multilayer Perceptron)
The central aim of this project is to achieve accurate predictions of real estate prices through the application of advanced MLP neural networks. By leveraging key features, including square footage, room count, kitchen facilities, bathrooms, and location data, the goal is to minimize prediction errors and enhance the overall accuracy of the model.
- Deep Learning Fundamentals: Acquired a comprehensive understanding of foundational concepts in deep learning, including gradient descent, regularization methods, and dropout rates.
- Neural Network Optimization Techniques: Explored and implemented optimization techniques such as early stopping, decay values, and regularization to enhance the performance of the MLP neural network.
- Feedforward Network: Developed proficiency in building and fine-tuning feedforward neural networks for regression tasks, specifically in the context of real estate price prediction.
This project stands as a testament to the acquired knowledge and skills during the Modern Analytics coursework, emphasizing the practical application of machine learning techniques in real-world scenarios.
As the field of real estate prediction evolves, future enhancements to this project could involve incorporating additional features, experimenting with different neural network architectures, and exploring advanced optimization strategies to further improve prediction accuracy.
Contributions are welcome! If you would like to contribute to this project, please fork the repository and submit a pull request.
This project is licensed under the MIT License.