VinayBR03/Math_Learning_Assessment ? reverse-engineered prompt

Build me a notebook that predicts student math performance from the MathE dataset and compares a bunch of classification models. I want the data loaded from the CSV, cleaned up with missing value handling, encoded for categorical fields, and scaled properly before training. Then reduce the feature space with PCA, balance the training data with SMOTE, and test several models like logistic regression, SVM, KNN, naive Bayes, boosting, bagging, and random forest.

Please also include a stacking model that uses random forest and KNN as the base models and logistic regression as the final model, then compare everything with accuracy and F1 score. I’d like a confusion matrix heatmap too, and a simple summary of which model performs best so it’s easy to understand the learning outcome patterns. If you need to check any current library details, look up the latest docs online.