Collectives™ on Stack Overflow

Find centralized, trusted content and collaborate around the technologies you use most.

Teams

Q&A for work

Connect and share knowledge within a single location that is structured and easy to search.

I am trying to get unique counts column-wise but my array has categorical variables (dtype object)

val, count = np.unique(x, axis=1, return_counts=True)
Though I am getting an error like this:
TypeError: The axis argument to unique is not supported for dtype object
How do I sove this problem?
Sample x:
array([[' Private', ' HS-grad', ' Divorced'],
       [' Private', ' 11th', ' Married-civ-spouse'],
       [' Private', ' Bachelors', ' Married-civ-spouse'],
       [' Private', ' Masters', ' Married-civ-spouse'],
       [' Private', ' 9th', ' Married-spouse-absent'],
       [' Self-emp-not-inc', ' HS-grad', ' Married-civ-spouse'],
       [' Private', ' Masters', ' Never-married'],
       [' Private', ' Bachelors', ' Married-civ-spouse'],
       [' Private', ' Some-college', ' Married-civ-spouse']], dtype=object)
Need the following counts:
for x_T in x.T:
    val, count = np.unique(x_T, return_counts=True)
    print (val,count)
[' Private' ' Self-emp-not-inc'] [8 1]
[' 11th' ' 9th' ' Bachelors' ' HS-grad' ' Masters' ' Some-college'] [1 1 2 2 2 1]
[' Divorced' ' Married-civ-spouse' ' Married-spouse-absent'
 ' Never-married'] [1 6 1 1]
                Why use axis when all you want is the count for each string?  Use axis is you want the unique rows or columns.  What's wrong with the x.T iteration?
– hpaulj
                Oct 27, 2018 at 15:59
You could use Itemfreq eventhough it the output does not look like yours it delivers the desired counts:
import numpy as np
from scipy.stats import itemfreq
x = np. array([[' Private', ' HS-grad', ' Divorced'],
       [' Private', ' 11th', ' Married-civ-spouse'],
       [' Private', ' Bachelors', ' Married-civ-spouse'],
       [' Private', ' Masters', ' Married-civ-spouse'],
       [' Private', ' 9th', ' Married-spouse-absent'],
       [' Self-emp-not-inc', ' HS-grad', ' Married-civ-spouse'],
       [' Private', ' Masters', ' Never-married'],
       [' Private', ' Bachelors', ' Married-civ-spouse'],
       [' Private', ' Some-college', ' Married-civ-spouse']], dtype=object)
itemfreq(x)
Output:
array([[' 11th', 1],
       [' 9th', 1],
       [' Bachelors', 2],
       [' Divorced', 1],
       [' HS-grad', 2],
       [' Married-civ-spouse', 6],
       [' Married-spouse-absent', 1],
       [' Masters', 2],
       [' Never-married', 1],
       [' Private', 8],
       [' Self-emp-not-inc', 1],
       [' Some-college', 1]], dtype=object)
otherwise you could try to specifiy another dtype such as:
val, count = np.unique(x.astype("<U22"), axis=1, return_counts=True)
for this however your array has to be different
        
Thanks for contributing an answer to Stack Overflow!
Please be sure to answer the question. Provide details and share your research!
But avoid …
Asking for help, clarification, or responding to other answers.
Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.