AI Jupyter Notebooks

🔖 About This Project

TL;DR Learn the basics of AI on your local machine to solve real-world problems with step-by-step guides through interactive notebooks 🤖.

This repo consists of a series of Jupyter notebooks explaining how to solve problems through using various methods of AI. We use frameworks like Pytorch and scikit-learn to train and evaluate AI models on real data obtained from various sources. We also learn how to use libraries like Docling to parse through data and extract crucial information.

This project is part of my broader goal to create tutorials and resources for understanding and using AI on local machines 💻. For other in depth tutorials pertaining to building agents and the local tool servers to power them, check it out here.

Now, let's get building!

🏁 Getting Started

1. Clone the repo, head there, then create a Python environment:

   git clone https://github.com/anima-kit/ai-notebooks.git
   cd ai-notebooks
   uv venv venv

2. Activate the Python environment:

   venv/Scripts/activate

3. Install the necessary Python libraries:

   uv pip install torch torchvision --index-url https://download.pytorch.org/whl/cu126
   uv pip install -r requirements.txt

4. Setup Kaggle account and API key:

   These notebooks utilize various Kaggle datasets. To download them, you'll need a free Kaggle account and API key.

   1. Go to your Kaggle account settings. Be sure to sign in if you haven't already done so.

   2. Scroll down until you see the API settings.

   3. Click 'Create New Token' and save the kaggle.json file in the proper place:

      • For Windows: C:\Users\{username}\.kaggle\kaggle.json
      • For Linux: /home/{username}/.kaggle/kaggle.json
      • For Mac: /Users/{username}/.kaggle/kaggle.json

      where {username} is your system's user account name.

Now you can run any of the Jupyter notebooks included in the repo.

📝 Example Use Cases

These tutorials are dividing into three main aspects of AI: Machine Learning (ML), Deep Learning (DL), and Reinforcement Learning (RL) and will be released periodically.

ML

The first tutorials will be a brief introduction to Supervised Learning (SL) as well as deep dives into the one of the simplest SL algorithms - linear regression. We use a Kaggle dataset to train and evaluate a model to predict the total energy consumed given various features, and we learn techniques for improving our model which we can carry over to other learning methods.

DL

We start the DL tutorials with a dive into data extraction with Docling and the NuExtract model. We learn how to use Docling's document extraction module to get key information from unstructured formats like PDFs and images.

📚 Next Steps & Learning Resources

This project is part of a series on learning and using AI. For a deeper dive, check out my tutorials. Topics include:

• Setting up local servers to power AI agents
• Example agent workflows (simple chatbots to specialized agents)
• Implementing complex RAG techniques
• Discussing various aspects of AI beyond agents (like these notebooks)

Want to learn AI? Visit my portfolio to explore more tutorials and projects!

⚙️ Tech

• Docling: For structured data extraction
• Jupyter: For interactive notebooks
• Kaggle: For obtaining various datasets
• NuExtract: For structured data extraction
• Pytorch: For building and evaluating AI models
• scikit-learn: For building and evaluating AI models

🔗 Contributing

This repo is a work in progress. If you'd like to suggest or add improvements, fix bugs or typos etc., feel free to contribute. Check out the contributing guidelines to get started.