A collection of machine learning and deep learning projects, built while learning PyTorch — covering linear regression fundamentals, regression on real-world tabular datasets, and neural network classification.
| Project | Description | Key Concepts |
|---|---|---|
| Linear Regression from Scratch | Implementing y = wx + b in PyTorch from first principles | Autograd, manual training loop, SGD |
| House Price Prediction | Predicting house prices from tabular features | Data cleaning, feature normalization, one-hot encoding |
| Boston Housing Regression | Predicting housing prices using the classic Boston Housing dataset | Linear regression, EDA, PyTorch nn.Module |
| Medical Cost Prediction | Predicting individual medical charges from demographic/health data | Regression, feature visualization |
| PyTorch Classification Exercises | Binary & multi-class classification (moons, spirals) with custom neural networks | nn.Module, non-linear activations, decision boundaries |
- Python, PyTorch
- Pandas, NumPy, Scikit-learn
- Matplotlib
This repo tracks my progression from linear regression fundamentals to building and training neural networks in PyTorch, applied to real-world regression and classification problems.
Aya Dalache