riccardoprosdocimi/ml-predictive-maintenance ? reverse-engineered prompt

Build me a predictive maintenance project for industrial machinery that can take sensor data and learn to predict when a machine is likely to fail.

I want the flow to cover data exploration, cleaning, preprocessing, feature engineering, and then a few model experiments so I can compare a simple logistic regression baseline with a neural network model and sequence models like RNN, LSTM, and GRU. Please make it easy to run each step separately in notebooks, since I want to inspect the data and results along the way.

Include clear evaluation so I can see how well each model does, especially things like precision, recall, and AUC ROC. If you need to look up current docs online for anything, go ahead. Keep it practical and readable, with code and explanations that help me understand what each notebook is doing and how the full predictive maintenance pipeline fits together.