Python: numpy.insert NaN value

Question

I'm trying to insert NaN values to specific indices of a numpy array. I keep getting this error:

TypeError: Cannot cast array data from dtype('float64') to dtype('int64') according to the rule 'safe'

When trying to do so with the following code.

x = np.array(range(1,11))
x = np.insert(x, 5, np.nan, axis=0)

However, I can append NaN values to the end of the array with no problem.

x = np.array(range(1,11))
x = np.append(x, np.nan)

Why is this and how can I insert NaN values in my array?

Pierre GM · Accepted Answer · 2013-11-28 12:10:48Z

8

With x=np.array(range(1,11)), the dtype by default is int64, which prevents you to insert a float.

The easiest is to force the dtype to float directly:

x = np.array(range(1, 11), dtype=float)

With np.insert, you're limited to the dtype of the initial array (the temporary arrays created below the hood use the dtype of the input).

With np.append, however, you're actually using np.concatenate, which creates an array with the "largest" dtype of its inputs: in your example, x is then cast to float.

Note that you could simply use the np.arange function:

x = np.arange(1, 11, dtype=float)

edited Nov 28, 2013 at 12:10

answered Nov 28, 2013 at 12:04

Pierre GM

20.5k3 gold badges58 silver badges67 bronze badges

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

Collectives™ on Stack Overflow

Python: numpy.insert NaN value

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related