Use NumPy to index one array with another

Question

Given some data

In [1]: import numpy as np
In [2]: x = np.array(['a', 'b', 'b', 'a'])

And a sorted index

In [3]: i = np.array(['a', 'b'])

I want to find the location of each data entry within the index

In [4]: # solution here
array([0, 1, 1, 0])

This is a bit like categoricals. I don't want to use Pandas here. I want to do this on fixed length strings. I need this to be somewhat efficient.

Alex Riley · Accepted Answer · 2015-03-22 21:39:03Z

5

>>> np.searchsorted(i, x)
array([0, 1, 1, 0])

The function finds out the index at which each element of x should be placed in i in order to maintain sorted order.

answered Mar 22, 2015 at 21:39

Alex Riley

178k46 gold badges274 silver badges247 bronze badges

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

1 Answer 1