Predicting Humidity with Nonlinear Methods | Dataford Interview Questions

Humidity prediction: fit a model to predict missing humidity readings; any non-linear regression or pandas interpolate passes the MSE-based tests

Asked in the HackerRank stage. Q3. Using the PostgreSQL tables provided, calculate linear interpolations for missing readings that have valid observed readings before and after them.

Output

Return one row per interpolatable missing reading with sensor_id, sensor_name, reading_time, and predicted_humidity.
Exclude missing readings without valid observations on both sides.
Order by sensor_id, then reading_time.

Schema

humidity_readings

Column	Type	Description
reading_idPK	INT	Unique humidity reading identifier
sensor_id	INT	Sensor that recorded the reading
reading_time	TIMESTAMP	Timestamp of the reading
humidity	NUMERIC(5,2)	Observed relative humidity percentage

sensors

Column	Type	Description
sensor_idPK	INT	Unique sensor identifier
sensor_name	VARCHAR(100)	Human-readable sensor name
location	VARCHAR(100)	Sensor deployment location

Your question is Predicting Humidity with Nonlinear Methods. Start with the requirements and the two tables in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem