How to add arrays of tuples in Python?

Question

How to add arrays of tuples in Python?

Navigation

#1 by (1 votes)
#2 by (0 votes)

0

I want to add two arrays, follow_dismiss and display that have only one value:

follow_dismiss
[(408L,), (14L,), (59L,), (6L,), (32L,), (62L,), (59L,), (120L,), (53L,), (150L,), (3L,), (1L,), (60L,), (1L,), (126L,), (41L,), (239L,), (163L,), (11L,), (42L,), (1L,), (2L,), (5L,), (1L,), (2L,), (4L,)]
display
[(382L,), (6L,), (295L,), (1L,), (8L,), (33L,), (30L,), (24L,), (5L,), (233L,), (1L,), (65L,), (21L,), (20L,), (86L,), (171L,), (2L,), (29L,), (2L,), (3L,), (1L,)]

They come from SQL queries, for example:

cur.execute("""-- nombre de fois ou l'on propose une pub a l'utilisateur par utilisateur
SELECT COUNT (swipe.state) FROM swipe
  WHERE swipe.state= 1 OR swipe.state = 4 or swipe.state=7
    GROUP BY swipe.subscriber_id
      ORDER BY swipe.subscriber_id;""")
    display = cur.fetchall()

To get:

[790, 20, 9, ...]

Another example:

[(1,2), (3, 4)] + [(2, 3), (1, 3)] = [(3,5), (4, 7)]

Not even that works:

SUM_follow_dismiss_display_i = [x + y for x, y in zip(follow_dismiss, display)]

Because it gives:

[(408L, 381L), (14L, 6L), (59L, 294L), (6L, 1L), (32L, 8L), (62L, 33L), (59L, 30L), (120L, 24L), (53L, 5L), (150L, 233L), (3L, 1L), (1L, 65L), (60L, 21L), (1L, 20L), (126L, 86L), (41L, 171L), (239L, 2L), (163L, 29L), (11L, 2L), (42L, 3L), (1L, 1L)]

Not even that:

SUM_follow_dismiss_display_i = map(add,follow_dismiss,display)

What it gives:

SUM_follow_dismiss_display_i = map(add,follow_dismiss,display)
TypeError: can only concatenate tuple (not "NoneType") to tuple

He did it because I want to divide each square with its equivalent of the other array in a last array, result matrix. Without emargo map (truediv, arry1, arry2) it does not work because it tells me that:

print follow_dismiss
print SUM_follow_dismiss_display_i
m_i = map(truediv, SUM_follow_dismiss, SUM_follow_dismiss_display_i)

[(408L,), (14L,), (59L,), (6L,), (32L,), (62L,), (59L,), (120L,), (53L,), (150L,), (3L,), (1L,), (60L,), (1L,), (126L,), (41L,), (239L,), (163L,), (11L,), (42L,), (1L,), (2L,), (5L,), (1L,), (2L,), (4L,)]
[(408L, 381L), (14L, 6L), (59L, 294L), (6L, 1L), (32L, 8L), (62L, 33L), (59L, 30L), (120L, 24L), (53L, 5L), (150L, 233L), (3L, 1L), (1L, 65L), (60L, 21L), (1L, 20L), (126L, 86L), (41L, 171L), (239L, 2L), (163L, 29L), (11L, 2L), (42L, 3L), (1L, 1L)]
Traceback (most recent call last):
  File "./testPostreSQLPython.py", line 72, in <module>
    m_i = map(truediv, SUM_follow_dismiss, SUM_follow_dismiss_display_i)
TypeError: argument 2 to map() must support iteration

python

asked by ThePassenger 22.05.2017 в 12:31

source

2 answers

0

Having two array of values, you can help with the zip function:

uno= [1,2,3,4,5]
dos = [6,7,8,9,10]

[x + y for x, y in zip(uno, dos )]

Result:

[7, 9, 11, 13, 15]

answered by 22.05.2017 в 12:52

Double List in Pascal How to add values obtained from the same field?

score 1 · Accepted Answer

If you can use NumPy since your structure is basically an array, simply concatenate:

import numpy as np

a = np.array([(408,), (14,), (59,), (6,)])
b = np.array([(382,), (6,), (295,), (1,)])

c = a + b

The output is:

array([[790],
       [ 20],
       [354],
       [  7]])

Another option without using Numpy could be zip or itertools.izip in Python 2:

Python 2.x:

from itertools import izip

a = [(408,), (14,), (59,), (6,)]
b = [(382,), (6,), (295,), (1,)]
c = [[c+d for c, d in izip(row1, row2)] for row1, row2 in izip(a, b)]

Python 3.x:

a = [(408,), (14,), (59,), (6,)]
b = [(382,), (6,), (295,), (1,)]
c = [[c+d for c, d in zip(row1, row2)] for row1, row2 in zip(a, b)]

In Python 2 it is also valid to use zip , the difference is that zip in Python 2 returns a list and not an iterator as it does in Python 3. In any case, the use of NumPy is the best option both in efficiency as in simplicity.