Adding up values in a 2D array in Python

Question

I have a numpy 2D array as follows

gona = array([['a1', 3], ['a2', 5], ['a3', 1], ['a3', 2], ['a3', 1], ['a1', 7]])

This array has 2 columns

What I want to do is create an array with 2 columns. Column 1 should have 'a1' , 'a2', 'a3' values in its' rows and column 2 should have summation of those corresponding values.

new_gona = array([['a1', 10], ['a2', 5], ['a3', 4]])

Here, corresponding values are taken as follows.

'a1' : 3 + 7 = 10
'a2' : 5 
'a3' : 1 + 2 + 1 = 4

What would be an easy method to achieve this?

I remember seeing an efficient solution with Pandas the last time this problem came up, but I don't remember what it was. Maybe someone with better search skills can find it. — user2357112
– user2357112, Commented Feb 4, 2014 at 8:53
Note: Running the code you've posted produces an array of dtype '|S2'. This means that the integers are stored as strings, instead of as int32 or some other reasonable dtype. That may be a problem. — user2357112
– user2357112, Commented Feb 4, 2014 at 8:56

eric chiang · Accepted Answer · 2014-02-04 10:13:44Z

3

Use pandas and its indexing magic:

import pandas as pd
import numpy as np

gona = np.array([['a1', 3], ['a2', 5], ['a3', 1], 
              ['a3', 2], ['a3', 1], ['a1', 7]])

# Create series where second items are data and first items are index
series = pd.Series(gona[:,1],gona[:,0],dtype=np.float)

# Compute sums across index
sums = series.sum(level=0)

# Construct new array in the format you want
new_gona = np.array(zip(sums.index,sums.values))

new_gona
# out[]:
# array([['a1', '10.0'],
#        ['a2', '5.0'],
#        ['a3', '4.0']], 
#       dtype='|S4')

It's also notable that np.arrays can only hold one datatype. So your mixing of strings and numeric types needs to be corrected for by specifying dtype=np.float. You can use np.int if you want.

answered Feb 4, 2014 at 10:13

eric chiang

2,7952 gold badges22 silver badges24 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Jaime · Accepted Answer · 2014-02-04 14:33:04Z

2

A numpy only solution:

>>> labels, indices = np.unique(gona[:, 0], return_inverse=True)
>>> sums = np.bincount(indices, weights=gona[:, 1].astype(np.float))
>>> new_gona = np.column_stack((labels, sums))
>>> new_gona
array([['a1', '10'],
       ['a2', '5.'],
       ['a3', '4.']], 
      dtype='|S2')

answered Feb 4, 2014 at 14:33

Jaime

67.7k19 gold badges128 silver badges164 bronze badges

Comments

Jakob Bowyer · Accepted Answer · 2014-02-04 08:51:00Z

1

from collections import defaultdict 
from operator import itemgetter

sums = defaultdict(int)
for key, value in gona:
    sums[key] += value

new_gona = sorted(sums.iteritems(), key=itemgetter(0))

Cheat?

answered Feb 4, 2014 at 8:51

Jakob Bowyer

34.8k8 gold badges80 silver badges92 bronze badges

Comments

ssm · Accepted Answer · 2014-02-05 01:05:01Z

0

Then, the list comprehension will do it pretty easy:

def fst(x): return x[0]
[(a, sum([int(m[1]) for m in gona if a == m[0]])) for a in set(map(fst, gona)) ]

This is basic Python. No libraries involved. The first function is defined only avoid the lambda expression in the map at the end. Both the Pandas and the NumPy solutions already mentioned seem pretty interesting though. +1 for both!

answered Feb 5, 2014 at 1:05

ssm

5,4031 gold badge26 silver badges43 bronze badges

Comments

lordkain · Accepted Answer · 2014-02-04 08:49:28Z

-2

you have to write a loop around gona and store the (a1) as a key in dictionary object. The value should be added ofcourse

answered Feb 4, 2014 at 8:49

lordkain

3,1191 gold badge16 silver badges18 bronze badges

1 Comment

lordkain Over a year ago

comments please, a solution was asked which was provided. No code but logical explanation

Collectives™ on Stack Overflow

Adding up values in a 2D array in Python

5 Answers 5

Comments

Comments

Comments

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

5 Answers 5

Comments

Comments

Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related