Will it Python? Machine Learning for Hackers, Chapter 2, Part 2: Logistic regression with statsmodels

UPDATE 1/15/2014: This blog is no longer in service.

This post is now located at: http://slendermeans.org/ml4h-ch2-p2.html

Thanks,
-c.

Advertisements
This entry was posted in Data Analysis, Will it Python and tagged , , . Bookmark the permalink.

4 Responses to Will it Python? Machine Learning for Hackers, Chapter 2, Part 2: Logistic regression with statsmodels

  1. Pingback: Will it Python? ARM Chapter 5: Logistic models of well-switching in Bangladesh | Slender Means

  2. You asked for interface to allow single decile to be passed to function without needing to put it in iterable. You can utilize variable number of arguments and let it do the building of the tuple for you:


    # Function to get arbitrary quantiles of a series.
    def my_quantiles(s, *prob):
    '''
    Calculate quantiles of a series.

    Parameters:
    -----------
    s : a pandas Series
    prob : a tuple (or other iterable) of probabilities at
    which to compute quantiles. Must be an iterable,
    even for a single probability (e.g. prob = (0.50)
    not prob = 0.50).

    Returns:
    --------
    A pandas series with the probabilities as an index.
    '''
    prob = prob or (0.0, 0.25, 0.5, 1.0)
    q = [s.quantile(p) for p in prob]
    return Series(q, index = prob)

    # With single prob argument
    print my_quantiles(heights, .10)

    • Comments are not updated, they were also in error in that singleton tuple containing 0.50 is (0.50,) Parenthesis are sometimes not needed. Looks like the code tag destroyed indent, I do not know how I should have posted it with indentation intact.

Comments are closed.