Dfyjkt

I am using pandas for extracting the categories of a dataset like in the following code:

data=pd.read_csv("iris.csv",header=None)
data[4]=data[4].astype("category")

When I print the values of data[4] I got this list:

1 Setosa
2 Setosa
3 Setosa
4 Setosa
5 Setosa
6 Setosa
7 Setosa
...
149 Virginica
150 Virginica
Name: 4, Length: 150, dtype: category
Categories (3, object): [Setosa, Versicolor, Virginica]

but I want to have the three categories in an array so to have something like:

[Setosa, Versicolor, Virginica]

I was looking around, but I could not find anything that could serve.

Any help?

data[4].cat.categories.values might be superior here, versus .unique().

Take a look at https://pandas.pydata.org/pandas-docs/stable/categorical.html, "Working with categories".

Note: The result of unique() is not always the same as
Series.cat.categories, because Series.unique() has a couple of
guarantees, namely that it returns categories in the order of
appearance, and it only includes values that are actually present.

.cat.categories.values seems much faster, because of the reasons listed above I imagine.

Example:

import pandas as pd
import numpy as np

s = pd.Series(np.random.choice(['a', 'b', 'c'], 1000), dtype = "category")

% timeit a = s.unique()

# 303 µs ± 23.2 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)

% timeit b = s.cat.categories.values

# 1.26 µs ± 27.9 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)

print(s.cat.categories.values)

# ['a' 'b' 'c']

The size of your dataset and other requirements will likely determine which is better.