What the Scaler Remembered

10 pts · 10 coins

Machine Learning in CodeEasy

A fitted MinMaxScaler keeps two numbers for each column, and that's all it learns.

Task: write learned_ranges(table).

table is a dict mapping column names to lists of numbers.
Fit a MinMaxScaler on the table.
Return a dict that maps each column name to [smallest, largest], using the values the scaler learned (data_min_ and data_max_), as plain floats.

table = {"weight": [110, 130, 150, 160, 170], "sweet": [2, 3, 6, 7, 8]}

returns {'weight': [110.0, 170.0], 'sweet': [2.0, 8.0]}

table = {"temp": [-5.5, 12.0, 30.25], "rain": [0, 0, 14]}

returns {'temp': [-5.5, 30.25], 'rain': [0.0, 14.0]}

solution.py

1import pandas as pd
2from sklearn.preprocessing import MinMaxScaler
3
4
5def learned_ranges(table):
6    pass
7