Changing a numpy array in place based on values in a given column?

Question

I have a numpy array:

array([[ 0.68597575,  0.05544651, -1.        ],
       [ 0.33494648,  0.46368367,  1.        ],
       [ 0.42486765,  0.89427025,  1.        ],
       [ 0.62408611,  0.64633939,  1.        ],
       [ 0.37087957,  0.53077302, -1.        ],
       [ 0.21664159,  0.10786084, -1.        ],
       [ 0.13003626,  0.18425347, -1.        ]])

I want the rows having last values -1 to be multiplied by -1 and also replaced in the actual matrix.

I tried this:

def transform(data):
    for row in data:
        if row[-1] == -1:
            row = row * -1

but I know there would be something simpler than this.

Saullo G. P. Castro · Accepted Answer · 2015-09-30 23:33:31Z

2

You can avoid the for loop by doing:

data[data[:, -1] == -1] *= -1

answered Sep 30, 2015 at 23:33

Saullo G. P. Castro

59.4k28 gold badges191 silver badges244 bronze badges

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

Comments

doug · Accepted Answer · 2015-09-30 23:38:51Z

1

i prefer to do this in two steps for two reasons: (i) easier to understand the code; and (ii) often the boolean index created in the first step can be reused elsewhere in the data pipeline

create the Boolean index that selects the rows:

idx = M[:,-1] == -1

do the transform on the indexed data:

M[idx,] *= -1

edited Sep 30, 2015 at 23:38

answered Sep 30, 2015 at 23:38

doug

70.2k26 gold badges171 silver badges201 bronze badges

1 Comment

doug Over a year ago

yep--looks like you posted yours while i was still writing mine, which is consistent with the time diff (~ 3 minutes apart, and in those 3 min i split my answer into two lines and added some comments to make it easier to understand)

Collectives™ on Stack Overflow

Changing a numpy array in place based on values in a given column?

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related