catlaujim/machine-learning-classification ? reverse-engineered prompt

Build me a clean Python notebook based machine learning project for classification, with two example notebooks, one for the Iris dataset and one for a marketing style classification problem.

I want it to show the full workflow from loading data and exploring it, to cleaning, training, tuning, and evaluating several models like decision trees, Naive Bayes, random forest, support vector machines, and K nearest neighbors. Please include the usual metrics people use to compare classifiers, like confusion matrix, accuracy, precision, recall, F1 score, and ROC AUC, with clear plots where they help.

Make it easy to follow for someone learning classification, with short explanations in the notebook and sensible visualizations using common Python data science libraries. If useful, add basic hyperparameter tuning and make the results easy to compare side by side. Look up current docs online if you need to.