Exercise

Running test classes

When you ran the !pytest command in the last exercise, the test test_on_six_rows() failed. This is a test for the function split_into_training_and_testing_sets(). This means that this function is broken.

Short recap in case you forgot: this function takes a NumPy array containing housing area and prices as argument. The function randomly splits the argument array into training and testing arrays in the ratio 3:1, and returns the resulting arrays in a tuple.

A quick look revealed that during the code update, someone inadvertently changed the split from 3:1 to 9:1. This has to be changed back and the unit tests for the function, which now lives in the test class TestSplitIntoTrainingAndTestingSets, needs to be run again. Are you up to the challenge?

Instructions 1/4

undefined XP
    1
    2
    3
    4
  • Fill in with a float between 0 and 1 so that num_training is approximately \(\frac{3}{4}\) of the number of rows in data_array.