ai_projects

Why we included this project

This collection works best as the companion to the author's machine learning blog: each notebook is runnable on its own and pairs with a full article that walks through the reasoning. Engineers and data scientists will find the practical mix useful, from a gentle CNN primer on MNIST to transfer learning in PyTorch, t-SNE visualization, distributed training with DeepSpeed, and even a fraud detection service exposed through a Flask API. Some projects go further and cover the whole lifecycle, like deploying an image classifier or putting a deep autoencoder recommender into production. Since these are tutorials rather than maintained libraries, plan to read them and adapt the code into your own experiments instead of installing anything as-is.