Python array manipulation, splitting, deleting

Question

I have an array to which i would like to add 1's to the last column of the first 16 rows and 0's to the last column of rows from 17 to 27. I searched everywhere however splitting seems to work only when we want to split the array into even parts, if i could cut the array at a specific row into two smaller arrays it would be easy because i would just append 1's and 0's than bring the arrays together again.

Another option would be to hard code where the specific numbers should go by indexing but that seems a bit complicated

frankly i don't have any more ideas for it, here are the arrays for visual representation of the problem (with shortened rows):

[[0.133987 0.359897 0.000000 0.527855 0.764706 0.212121]
[0.301506 0.407752 0.013970 0.553386 0.830450 0.272727]
[0.414438 0.415456 0.023283 0.597853 0.833910 0.333333]
[0.434815 0.415844 0.046566 0.604616 0.837370 0.363636]
[0.436664 0.500333 0.093132 0.607321 0.868512 0.366667]
[0.447560 0.559649 0.098952 0.626934 0.882353 0.393939]]

I would like to have it like so:`

[[0.133987 0.359897 0.000000 0.527855 0.764706 0.212121 1]
[0.301506 0.407752 0.013970 0.553386 0.830450 0.272727 1]
[0.414438 0.415456 0.023283 0.597853 0.833910 0.333333 1]
[0.434815 0.415844 0.046566 0.604616 0.837370 0.363636 1]
[0.436664 0.500333 0.093132 0.607321 0.868512 0.366667 0]
[0.447560 0.559649 0.098952 0.626934 0.882353 0.393939 0]]

• user707650: Do you have an array, a list of lists, or a numpy array? Each has its own solutions.
• user707650: You describe your attempts, but it may be helpful to see some actual code of your (best) attempt.
• Dantuzzo: it's a numpy array, I didn't try anything because i haven't found any method for it so i don't even know where to start

Answer 1

There is probably a cleaner more efficient way to do this but I think this is what you are after.

import numpy as np
data = np.array([[0.133987, 0.359897, 0.000000, 0.527855, 0.764706, 0.212121], [0.301506, 0.407752, 0.013970, 0.553386, 0.830450, 0.272727], [0.414438, 0.415456, 0.023283, 0.597853, 0.833910, 0.333333], [0.434815, 0.415844, 0.046566, 0.604616, 0.837370, 0.363636], [0.436664, 0.500333, 0.093132, 0.607321, 0.868512, 0.366667], [0.447560, 0.559649, 0.098952, 0.626934, 0.882353, 0.393939]])
extra = []
for i in range(len(data)):
 if i < 3:
 extra.append([1])
 else:
 extra.append([0])
extra = np.array(extra)
result = np.hstack([data, extra])
print(result)

resulting in

[[ 0.133987 0.359897 0. 0.527855 0.764706 0.212121 1. ]
 [ 0.301506 0.407752 0.01397 0.553386 0.83045 0.272727 1. ]
 [ 0.414438 0.415456 0.023283 0.597853 0.83391 0.333333 1. ]
 [ 0.434815 0.415844 0.046566 0.604616 0.83737 0.363636 0. ]
 [ 0.436664 0.500333 0.093132 0.607321 0.868512 0.366667 0. ]
 [ 0.44756 0.559649 0.098952 0.626934 0.882353 0.393939 0. ]]

• Steven Summers: extra = [[int(i < len(data) // 2)] for i in range(len(data))] would be nicer. And either // or / depending on which portion of odd half should have 1 or 0.

Answer 2

Here is an abridged version which takes an input with 4 rows and adds 1 to the first 2 rows and 0 to the remaining rows. You would only have to adjust the numbers to fit your requirements.

v = [[0.133987, 0.359897, 0.000000, 0.527855, 0.764706, 0.212121],
 [0.301506, 0.407752, 0.013970, 0.553386, 0.830450, 0.272727],
 [0.414438, 0.415456, 0.023283, 0.597853, 0.833910, 0.333333],
 [0.447560, 0.559649, 0.098952, 0.626934, 0.882353, 0.393939]]
print('before **********')
for line in v:
 print(line)
for i, x in enumerate(v):
 if i < 2:
 x.append(1) # did you want ints here or floats to be consistent?
 else:
 x.append(0)
print('after **********')
for line in v:
 print(line)

output:

before **********
[0.133987, 0.359897, 0.0, 0.527855, 0.764706, 0.212121]
[0.301506, 0.407752, 0.01397, 0.553386, 0.83045, 0.272727]
[0.414438, 0.415456, 0.023283, 0.597853, 0.83391, 0.333333]
[0.44756, 0.559649, 0.098952, 0.626934, 0.882353, 0.393939]
after **********
[0.133987, 0.359897, 0.0, 0.527855, 0.764706, 0.212121, 1]
[0.301506, 0.407752, 0.01397, 0.553386, 0.83045, 0.272727, 1]
[0.414438, 0.415456, 0.023283, 0.597853, 0.83391, 0.333333, 0]
[0.44756, 0.559649, 0.098952, 0.626934, 0.882353, 0.393939, 0]

You could also shorten the append line to

x.append(1 if i < 2 else 0)

If you so wished.

Answer 3

Just make the last column and concatenate them:

arr = np.random.random(size=(27,6))
flag = [0]*16 + [1]*11
flag = np.array([flag])
arr_ = np.concatenate([arr, flag.T], axis=1)

The elements of a numpy array must have uniform dtypes, so you actually get 1. and 0. in the last column