In this lab, you will be using the Sakila database of movie rentals.

In order to optimize our inventory, we would like to know which films will be rented. We are asked to create a model to predict it. So we use the information we have from May 2005 to create the model.

1. Create a query or queries to extract the information you think may be relevant for building the prediction model. It should include some film features and some rental features (X).
2. Create a query to get the list of all unique film titles and a boolean indicating if it was rented (rental_date) in May 2005. (Create new column called - 'rented_in_may'). This will be our TARGET (y) variable.
3. Read the data into a Pandas dataframe. At this point you should have 1000 rows. Number of columns depends on the number of features you chose.
4. Analyze extracted features (X) and transform them. You may need to encode some categorical variables, or scale numerical variables.
5. Create a logistic regression model to predict 'rented_in_may' from the cleaned data.
6. Evaluate the results.