fouadtrad/Detecting-Quishing-Attacks-with-Machine-Learning-Techniques-Through-QR-Code-Analysis ? reverse-engineered prompt

Build me a notebook that loads the QR code phishing dataset from the zip file and helps me explore it easily.

I want to open the saved pickle files, check the shapes, confirm the labels, and show a few random QR code images with their class so I can see what the data looks like. Please add simple preprocessing steps if needed, flatten or reshape the images in a way that works for machine learning, and split the data into train and test sets. Then train a few basic models for phishing detection, like logistic regression, decision tree, random forest, naive bayes, and XGBoost if it is available.

Show the main evaluation results in a clear table, including accuracy and AUC, and make a simple plot for comparing the models. If there are any missing libraries or setup steps, handle them cleanly and keep the notebook easy to run from start to finish.