I just realized that I never posted about my work last Monday, March 2nd. I worked on the Weekly Report for about an hour and a half.
My main goal last Monday was to create a task_zero as a precursor to the rest of the tasks currently running. I am currently trying to have data logged from all possible time intervals, but filter it before entering the report process to ensure that my systems are only calculating on relevant values (that is, when school is in session and the building climate control systems are powered on). Task 0 would allow a week of raw data to be selected (either manually by the user in an interactive front-end, or in an automated fashion each Friday), and then filtered for whether or not school is in session. I plan to use a Datetime:Boolean dictionary that includes each day of the year as the index, and the boolean value for whether it is a school day as the value. I implemented a basic version of this where I set the boolean value to (start_date.weekday() < 5), and filtered the values based on this dictionary. While it does successfully filter, it also takes around a minute to run, because reading from the incredibly large DataFrame and applying pd.to_datetime is quite time-consuming. In the future, I hope to integrate the actual school calendar into this dictionary as well as figure out ways to increase the efficiency of the program, if necessary. Comments are closed.
|
AuthorI'm a high school senior and programming enthusiast. Archives
March 2022
Categories |