Welcome to week three. So far you've learned how to use the powerful aggregation framework to shape and analyze your data, the importance of schema design and how to effectively explore your data and how to migrate your data for storage and practical use. I'm very excited for the content in week three. This week we'll be putting our knowledge to practical use and using data from MongoDB to build a simple machine learning models. We'll use Python libraries including NumPy, Pandas, MLxtend and Scikit learn to work through our data, asking questions and making predictions and using Matplotlib and Seaborn for visualizations. We'll explore Pearson correlation, principal component analysis, linear aggression, market basket analysis, clustering with K-Means and decision trees. The work this week we'll use MongoDB in the role of a data store, where we'll be cleaning and transforming that data prior to feeding it into our model. MongoDB isn't a machine learning engine, but it certainly won't get in the way of our efforts and in fact aggregation can make things much easier. So the purpose of this week is to show how easy it is to work with data from MongoDB. Now we won't dive into cleaning hyperparameters or discuss the artful side of data science. Otherwise this week could easily turn into a year or more. Again, welcome to week three. We hope you have fun during this content and if you're new to data science, we hope to inspire you to dig further on your own. Best of luck.