How do you optimize memory usage when loading extremely large CSV files into a Pandas DataFrame for feature extraction at Workday?
