This the final lab for the Break-Through Tech Machine Learning Fundamental course provided by Cornell Tech.

We where given 3 datasets to choose from and where asked to assign our own label and use any model of our choosing to predict that label.

I chose a United States Census Dataset and looked to predict the education of an individual comparing and contrasting multiple models to see which one would have been able to handle the limitations of the dataset and the difficulty of predicing the chosen label.

Applying techiques such as Grouping, and Hot one encoding to transform our data for better results when used with our multiple machine learning models that includes more simple models such as KNN and more complex ensemble models such as a stacking model and even attempting to use a model built specifically for the class imbalance that we have.

Overall this is an application and summarization of the 10-week course offered by cornell and allowed for me to delve into the details that interest me most about machine learning which is the optimizations and variety of models.