Cоntext (sаme аs the previоus questiоn) You аre given a dataset named past_leads, with 50,000 rows of data on past customer leads for a service that your company provides. makes. For each person, you have data on their gender, age, annual income, educational level, field of study, weight and occupation. This being historical data, you also have information on whether each lead finally bought your service or not, stored in a column named 'purchased'. You now have several future prospective customers for the service. You have obtained a dataset named future_leads with information on their gender, age, annual income, educational level, field of study, weight and occupation. Of course, since these are future prospects, you do not know whether they will purchase the service or not. You want to use the historical data on leads to build a model to predict for each of the rows in future_leads whether each of them will buy the service or not. Question In this scenario, is we adopt the approach of partitioning the data, we will use rows of data from past_leads to build the model