the-mrinal/ML-Notebook ? reverse-engineered prompt

Build me a Jupyter Notebook based ML learning repo that teaches machine learning from the ground up, starting with NumPy and plotting, then moving through regression, classification, feature scaling, PCA, NLP, and neural networks.

I want it organized into numbered lesson folders, with a lot of hands on notebooks that show both from scratch implementations and the scikit learn or TensorFlow version so learners can compare them side by side. Include practical examples like Titanic style prediction, iris classification, housing price regression, text classification, image classification with MNIST and CIFAR 10, and a few real world projects. The notebooks should feel like a guided curriculum, with clear explanations, intermediate steps, and outputs that help a beginner follow along.

Please keep everything runnable in Jupyter, use common Python data science libraries, and make sure the datasets and notebooks are easy to find. If you need to check current docs or best practices, look them up online first.