Need to create a median function that draws from a dictionary

Question

I need to find the median of all the integers associated with each key (AA, BB). The basic format my code leads to:

AA - 21
AA - 52
BB - 3
BB - 2

My code:

def scoreData(filename):
   d = dict() 
   fin = open(filename) 
   contents = fin.readlines()
   for line in contents:
       parts = linesplit() 
       part[i] = int(part[1]) 
       if parts[0] not in d:
           d[parts[0]] = list(parts[1])  
       else:
           d[parts[0]].append(parts[1]) 
   names = list(d.keys()) 
   names.sort() #alphabeticez the names
   print("Name\+Max\+Min\+Median")
   for name in names: #makes the table
       print (name"\+", max(d[name]),\+min(d[name]),"\+"median(d[name]))

I'm afraid following the same format as the "names" and "names.sort" will completely restructure the data. I've thought about "from statistics import median," but once again I do not know how to only select the values associated with each of the same keys.

Thanks in advance

Answer 1

You can do it easily with pandas and numpy:

import pandas
import numpy as np

and aggregating by first row:

score = pandas.read_csv(filename, delimiter=' - ', header=None)
print score.groupby(0).agg([np.median, np.min, np.max])

which returns:

         1
    median  amin  amax
0
AA    36.5    21    52
BB     2.5     2     3

Answer 2

There are many, many ways you can go about this. But here's a 'naive' implementation that will get the job done.

Assuming your data looks like:

AA  1
BB  5
AA  2
CC  7
BB  1

You can do the following:

import numpy as np
from collections import defaultdict

def find_averages(input_file)
    result_dict = defaultdict(list)
    for line in input_file.readlines()
        key, value = line.split()
        result_dict[key].append[int(value)]

    return [(key, np.mean(value)) for key,value in result_dict.iteritems()]