Testing Numpy array to see if it is in column form

Question

In my function, sometimes I get a result that is a 1 dimensional numpy array in 2D form, so that it's shape is nx1 (n,1). Other times, I might get it in the form 1xn array.shape = (1,n)

Other times, I get just a numpy array whose shape is (n,).

when I run the following tests, I get an error on the one hand, and a false positive on the other (since the length of a shape attribute is always greater than 1, apparently):

    y_predicted = forest.predict(testX)
    if y_predicted.shape[1] != None:
        y_predicted = y_predicted.T[0]

and

    y_predicted = forest.predict(testX)
    if len(y_predicted.shape) > 1:
        y_predicted = y_predicted.T[0]

I just need to make sure the final shape of y is always in the form (n,) rather than (n,1) or (1,n)...

squeeze, ravel and flatten will all do this job; but read their docs so you understand their differences. — hpaulj
– hpaulj, Commented Feb 28, 2016 at 16:59

toasted_flakes · Accepted Answer · 2016-02-28 16:15:46Z

3

You should use numpy.squeeze:

numpy.squeeze(a) removes single-dimensional entries from the shape of an array.

Example:

>>> x = np.array([[1,2,3]])
>>> x.shape
(1, 3)
>>> np.squeeze(x)
array([1, 2, 3])
>>> np.squeeze(x).shape
(3,)

answered Feb 28, 2016 at 16:15

toasted_flakes

2,5191 gold badge24 silver badges38 bronze badges

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

1 Comment

Ami Tavory Over a year ago

Good answer. Was unaware of this method.

Collectives™ on Stack Overflow

Testing Numpy array to see if it is in column form

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related