🧠 Machine Learning Demos Portfolio This repository contains a collection of demos showcasing various machine learning applicatins which aim to bring home key concepts. The notebooks are built with the latest Python libraries, support multiple GPU providers, and include references and explanations where relevant.

👉 The goal: demonstrate practical implementations of machine learning concepts — in a way that is both educational and extensible.

📂 Contents & Topics 🔹 K-Means Unsupervised learning K-Means.ipynb → There are 3 examples to demonstrate the effectiveness and implementation of the K-means algorithm.

The first two examples borrow code from a geeksforgeeks.com tutorial which computes the each part of the K-means algorithm (centroid definition, cluster assignment) separately and puts the intermediate results in a scatter plot. This allows for conceptual understanding of the algorithm and the ability to see how the algorithm converges to its result.

The second example applies feature scaling to some dummy data to show how to make the algorithm generate more reasonable clusters.

The third example leverages the scikit-learn library to identify the optimal number of clusters. It also uses the Clustering/KMeans library to run the algorithm more simply. 🚀 Tech Stack Python (latest libraries) PyTorch / HuggingFace / scikit-learn APIs: Stability.ai, OpenAI 📊 Results Sample outputs are stored in the results/ folder.

🔎 These are not required to run the demos — they simply illustrate recent experiments.

🧩 How to Use