A script to calculate grade statistics for UC Berkeley Computer Science courses after removing scores of 0. This gives a more accurate representation of average score and spread.

betterstats.py

# Run like this:
# glookup -s $assignment_name -b 0.1 | python betterstats.py

from sys import stdin
import re

def getBuckets():
    buckets = []
    bucket_patt = r'\s*([0-9]+\.[0-9]+)\s+\-\s*([0-9]+\.[0-9]+)\:\s*([0-9]+).*'
    your_patt = r'Your score\:.*?([0-9]+\.[0-9]+)'

    your_score = -1
    for line in stdin.readlines():
        your_match = re.search(your_patt, line)
        if your_match:
            your_score = float(your_match.groups(1)[0])
        match = re.search(bucket_patt, line)
        if match and float(match.groups(1)[0]) > 0.0:
            buckets.append(tuple(map(float, match.groups())))
    getStats(buckets, your_score)

def getStats(buckets, your_score):
    total_scores = 1.0*sum([bucket[2] for bucket in buckets])
    total_points = 1.0*sum([bucket[0]*bucket[2] for bucket in buckets])
    mean = total_points/total_scores

    quartiles = [int(total_scores*q) for q in (.25, .5, .75)]
    quartile_scores = []

    minimum = -1
    maximum = -1
    sofar = 0
    sum_square_diffs = 0
    your_rank = total_scores


    for bucket in buckets:
        if your_score >= bucket[0]:
            your_rank -= bucket[2]
        if minimum < 0 and bucket[2] > 0:
            minimum = bucket[0]
        if bucket[2] > 0:
            maximum = bucket[0]

        sofar += bucket[2]
        if len(quartiles) > 0 and sofar >= quartiles[0]:
            quartiles.pop(0)
            quartile_scores.append(bucket[0])

        sum_square_diffs += bucket[2]*(mean - bucket[0] + .5)**2

    standard_deviation = (sum_square_diffs/total_scores)**.5

    results = """
With scores of 0 removed:
-------------------------
Your Score: %f (#%d of %d)
Mean: %f
Standard Deviation: %f
1st Quartile: %f
2nd Quartile: %f
3rd Quartile: %f
Minimum: %f
Maximum: %f
""" % (your_score, your_rank, total_scores, mean, standard_deviation,
        quartile_scores[0], quartile_scores[1], quartile_scores[2], minimum, maximum)
    print results


if __name__ == '__main__':
    getBuckets()