How do I add up all of the values of a column in a python array? Ideally I want to do this without importing any additional libraries.
input_val = [[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5]]
output_val = [3, 6, 9, 12, 15]
I know I this can be done in a nested for loop, wondering if there was a better way (like a list comprehension)?
zip and sum can get that done:
Code:
[sum(x) for x in zip(*input_val)]
zip takes the contents of the input list and transposes them so that each element of the contained lists is produced at the same time. This allows the sum to see the first elements of each contained list, then next iteration will get the second element of each list, etc...
Test Code:
input_val = [[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5]]
print([sum(x) for x in zip(*input_val)])
Results:
[3, 6, 9, 12, 15]
In case you decide to use any library, numpy easily does this:
np.sum(input_val,axis=0)
You may also use sum with zip within the map function:
# In Python 3.x
>>> list(map(sum, zip(*input_val)))
[3, 6, 9, 12, 15]
# explicitly type-cast it to list as map returns generator expression
# In Python 2.x, explicit type-casting to list is not needed as `map` returns list
>>> map(sum, zip(*input_val))
[3, 6, 9, 12, 15]
Try this:
input_val = [[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5]]
output_val = [sum([i[b] for i in input_val]) for b in range(len(input_val[0]))]
print output_val
Please construct your array using the NumPy library:
import numpy as np
create the array using the array( ) function and save it in a variable:
arr = np.array(([1, 2, 3, 4, 5],[1, 2, 3, 4, 5],[1, 2, 3, 4, 5]))
apply sum( ) function to the array specifying it for the columns by setting the axis parameter to zero:
arr.sum(axis = 0)
This should work:
[sum(i) for i in zip(*input_val)]
I guess you can use:
import numpy as np
new_list = sum(map(np.array, input_val))
I think this is the most pythonic way of doing this
map(sum, [x for x in zip(*input_val)])
One-liner using list comprehensions: for each column (length of one row), make a list of all the entries in that column, and sum that list.
output_val = [sum([input_val[i][j] for i in range(len(input_val))]) \
for j in range(len(input_val[0]))]
Try this code. This will make output_val end up as [3, 6, 9, 12, 15] given your input_val:
input_val = [[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5]]
vals_length = len(input_val[0])
output_val = [0] * vals_length # init empty output array with 0's
for i in range(vals_length): # iterate for each index in the inputs
for vals in input_val:
output_val[i] += vals[i] # add to the same index
print(output_val) # [3, 6, 9, 12, 15]
Using Numpy you can easily solve this issue in one line:
1: Input
input_val = [[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5]]
2: Numpy does the math for you
np.sum(input_val,axis=0)
3: Then finally the results
array([ 3, 6, 9, 12, 15])
output_val=input_val.sum(axis=0)
this would make the code even simpler I guess
AttributeError: 'list' object has no attribute 'sum'You can use the sum function instead of np.sum simply.
input_val = np.array([[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5],
[1, 2, 3, 4, 5]])
sum(input_val)
output: array([ 3, 6, 9, 12, 15])
