AIbox 📦

AIbox is a laboratory for exploring, testing, and visualizing various AI and machine learning methods.
Main goal is to create a sandbox environment for experiments with different algorithms. Repository might be also helpful for new learners with their journey into AI/ML. Project combines ideas from statistical learning, mathematical modeling and deep learning.

Python Libraries

• [[scikit-learn]]
• [[statsmodels]]
• [[Pandas]]
• [[NumPy]]
• [[Matplotlib]]

Current Problems

• Regression: Implementing and comparing linear regression, ridge regression, and lasso regression on synthetic and real datasets.
• Classification: Exploring logistic regression, support vector machines, and decision trees for binary and multi-class classification tasks.
• SVM: Implementing support vector machines with different kernels and visualizing decision boundaries.
• Knowledge Distillation: Experimenting with teacher-student models to transfer knowledge from a larger model to a smaller one.
• Pipelines: Building end-to-end machine learning pipelines for data preprocessing, model training, and evaluation.
• Boosting: Implementing boosting algorithms like Gradient Boosting for improving model performance.

Usage

To setup the environment, follow these steps:

1. Clone the repository
2. Create a virtual environment:

uv sync

3. Experiment with different notebooks

Sources

• An Introduction to Statistical Learning