Ch 3. Line 77 in Python Code

[deulkkae]

practical-statistics-for-data-scientists/python/code/Chapter 3 - Statistial Experiments and Significance Testing.py

Line 77 in 0db4dbb

print(np.mean(perm_diffs > mean_b - mean_a))

This line brings typeerror: TypeError: '>' not supported between instances of 'list' and 'float'

It would be better to correct this line to
print(np.mean(np.array(perm_diffs) > mean_b - mean_a))

[gedeck]

This is strange. It works in the notebook:

I looked into this and it works for me because in the notebook, the type of mean_b and mean_a is numpy.float64. In this case, numpy will implicitly convert the list to np.array and make it work. Your error message suggests that the means are float on your system. The floats are from:

session_times[session_times.Page == 'Page B'].Time.mean()

What Python, numpy and pandas versions are you using? I tested now using Python 3.9.4, numpy 1.20.2, and pandas 1.2.4.

[gedeck]

In #26, the same issue was reported. In this case, pandas was version 1.1.3 and numpy 1.19.5. I assume that the issue was caused by a regression in the mean function and that this is now fixed again.

I will therefore close this issue