matching two different arrays and making a new array in python

Question

I have two two-dimensional arrays, and I have to create a new array filtering through the 2nd array where 1st column indexes match. The arrays are of different size.

basically the idea is as follow:

file A

#x   y
1   2
3   4
2   2
5   4
6   4
7   4


file B

#x1    y1
0       1
1       1
11      1
5       1
7       1
My expected output 2D array should look like

#newx     newy
1         1
5         1
7         1

I tried it following way:

match =[]
for i in range(len(x)):
    if x[i] == x1[i]:
        new_array = x1[i]
        match.append(new_array)

print match

This does not seem to work. Please suggest a way to create the new 2D array

Tai · Accepted Answer · 2018-03-28 05:08:19Z

5

Try np.isin.

arr1 = np.array([[1,3,2,5,6,7], [2,4,2,4,4,4]])
arr2 = np.array([[0,1,11,5,7], [1,1,1,1,1]])
arr2[:,np.isin(arr2[0], arr1[0])]
array([[1, 5, 7],
       [1, 1, 1]])

np.isin(arr2[0], arr1[0]) checks whether each element of arr2[0] is in arr1[0]. Then, we use the result as the boolean index array to select elements in arr2.

edited Mar 28, 2018 at 5:08

answered Mar 28, 2018 at 5:02

Tai

8,0643 gold badges31 silver badges50 bronze badges

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

Comments

Stephen Rauch · Accepted Answer · 2018-03-28 05:34:45Z

5

If you make a set out of the first element in A, then it is fairly easy to find the elements in B to keep like:

Code:

a = ((1, 2), (3, 4), (2, 2), (5, 4), (6, 4), (7, 4))
b = ((0, 1), (1, 1), (11, 1), (5, 1), (7, 1))

in_a = {i[0] for i in a}
new_b = [i for i in b if i[0] in in_a]

print(new_b)

Results:

[(1, 1), (5, 1), (7, 1)]

Output results to file as:

with open('output.txt', 'w') as f:
    for value in new_b:
        f.write(' '.join(str(v) for v in value) + '\n')

edited Mar 28, 2018 at 5:34

answered Mar 28, 2018 at 4:41

Stephen Rauch♦

50.1k32 gold badges118 silver badges143 bronze badges

3 Comments

bhjghjh Over a year ago

could you please explain how I make pairs from my large arrays? like the way you created the a and b array?

Stephen Rauch Over a year ago

@bhjghjh, I would love to, if I understood what you were asking. Maybe you could formulate another question? If you do ask another question, feel free to leave a comment here if you want to be sure I see it. Good Luck...

bhjghjh Over a year ago

thanks a lot for helps. In my original problem, I do not have (x,y) as pairs. I just have two different .txt files from which I am reading the two columns as arrays. Could you please modify your code to output the result as two different columns in a file?

Dyno Fu · Accepted Answer · 2018-03-28 04:53:28Z

2

#!/usr/bin/env python3

from io import StringIO
import pandas as pd

fileA = """x   y
1   2
3   4
2   2
5   4
6   4
7   4
"""

fileB = """x1    y1
0       1
1       1
11      1
5       1
7       1
"""


df1 = pd.read_csv(StringIO(fileA), delim_whitespace=True, index_col="x")
df2 = pd.read_csv(StringIO(fileB), delim_whitespace=True, index_col="x1")

df = pd.merge(df1, df2, left_index=True, right_index=True)
print(df["y1"])

# 1    1
# 5    1
# 7    1

https://pandas.pydata.org/pandas-docs/stable/merging.html#database-style-dataframe-joining-merging

answered Mar 28, 2018 at 4:53

Dyno Fu

9,0764 gold badges46 silver badges73 bronze badges

Comments

Patrick Conwell · Accepted Answer · 2018-03-28 04:58:04Z

2

If you use pandas:

import pandas as pd

A = pd.DataFrame({'x': pd.Series([1,3,2,5,6,7]), 'y': pd.Series([2,4,2,4,4,4])})
B = pd.DataFrame({'x1': pd.Series([0,1,11,5,7]), 'y1': 1})

C = A.join(B.set_index('x1'), on='x')

Then if you wanted to drop the unneeded row/columns and rename the columns:

C = A.join(B.set_index('x1'), on='x')
C = C.drop(['y'], axis=1)
C.columns = ['newx', 'newy']

which gives you:

>>> C
   newx  newy
0     1   1.0
3     5   1.0
5     7   1.0

If you are going to work with arrays, dataframes, etc - pandas is definitely worth a look: https://pandas.pydata.org/pandas-docs/stable/10min.html

answered Mar 28, 2018 at 4:58

Patrick Conwell

6991 gold badge5 silver badges11 bronze badges

Comments

Vivek Pabani · Accepted Answer · 2018-03-28 04:46:55Z

1

Assuming that you have (x, y) pairs in your 2-D arrays, a simple loop may work:

arr1   = [[1, 2], [3, 4], [2, 2]]
arr2   = [[0, 1], [1, 1], [11, 1]]
result = []

for pair1 in arr1:
    for pair2 in arr2:
        if (pair1[0] == pair2[0]):
            result.append(pair2)

print(result)

answered Mar 28, 2018 at 4:46

Vivek Pabani

4503 silver badges11 bronze badges

Comments

Aritesh · Accepted Answer · 2018-03-28 04:51:12Z

1

Not the best solution for smaller arrays, but for really large arrays, works fast -

import numpy as np
import pandas as pd

n1 = np.transpose(np.array([[1,3,2,5,6,7], [2,4,2,4,4,4]]))
n2 = np.transpose(np.array([[0,1,11,5, 7], [1,1,1,1,1]]))
np.array(pd.DataFrame(n1).merge(pd.DataFrame(n2), on=0, how='inner').drop('1_x', axis=1))

answered Mar 28, 2018 at 4:51

Aritesh

2,1031 gold badge17 silver badges18 bronze badges

Collectives™ on Stack Overflow

matching two different arrays and making a new array in python

6 Answers 6

Comments

Code:

Results:

Output results to file as:

3 Comments

Comments

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

6 Answers 6

Comments

Code:

Results:

Output results to file as:

3 Comments

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related