🚀 AI Labs – M1 Student (Option IA)

Welcome to the AI Labs repository, a collection of hands-on labs and projects designed to help ENSEA 2A students explore AI concepts and practical deployment. Covering Natural Language Processing (NLP), Computer Vision, Retrieval-Augmented Generation (RAG), Recommendation Systems, and Graph Neural Networks (GNNs), this repository provides a first experience in AI deployment.

You'll train Neural Style Transfer models, build LLMs, develop recommender systems, and implement graph-based AI models—all while learning how to deploy them in real-world applications. Whether you're working on model inference, API integrations, or simple web-based AI demos, these labs offer a solid foundation for stepping into AI engineering.

---

🌟 Key Features

• Hands-On Learning – Apply AI techniques through structured, practical labs.
• Google Colab Integration – Run all labs in Google Colab with GPU acceleration.
• Diverse Topics – Covers everything from sentiment analysis to graph neural networks.
• Open Source – Licensed under MIT—feel free to use, modify, and contribute!

---

📂 Labs

🔹 TP1: Introduction to AI

Explore Deep Learning fundamentals with two techniques:

• Training an Inverting Neuron
• Neural Style Transfer

🔹 TP2: Sentiment Analysis

Perform sentiment analysis using two types of neural networks:

• Feed Forward Networks
• Convolutional Neural Networks

🔹 TP3: Language Models

Build a language model using:

• N-grams
• Feed Forward Networks

🔹 TP4: Spotify Playlist Recommendation

Develop a Deep RecSys model to recommend Spotify playlists.
📜 Slides: Recommendation Part 1

🔹 TP5: Graph Neural Networks for Recommendation

Build a deep Graph RecSys model for recommendations using the MovieLens dataset.

---

🛠️ Hands-On Sessions

🔹 TDm1: Spotify Python API Exploration

Learn to use Spotipy, the Spotify Python API, for music data exploration.

🔹 TDm2: Introduction to Recommendation Systems

Explore key techniques used in recommender systems across various datasets.

---

🛠️ Hands-On Projects

🔹 Project 1: ML Stack

Deploy a text summarization model using Gradio, Hugging Face, and Weights & Biases to streamline experimentation and monitoring.

🔹 Project 2: Retrieval-Augmented Generation (RAG)

Build a Retrieval-Augmented Generation (RAG) system from scratch—design the full pipeline, from document retrieval to response generation.

---

Let me know if you'd like any refinements! 🚀

🚀 How to Use

1. Open in Google Colab – Replace github.com with githubtocolab.com in the URL to launch notebooks.
2. Enable GPU – Accelerate training by enabling GPU (Runtime > Change runtime type > Hardware accelerator > GPU).
3. Save Your Work – Copy notebooks to your Google Drive for future reference.
4. Complete Preparations – Ensure all pre-lab exercises are done before attending lab sessions.

---

📜 License

This project is licensed under the MIT License. Feel free to use, modify, and distribute it while giving proper credit. Contributions are always welcome!

---

Happy coding! 🚀

---