How to find out number of unique values in a column along with count of the unique values in a dataframe?

Question

As per the following data set I want no get the number of unique values and count of the unique values.

My data set:

Account_Type
Gold
Gold
Platinum
Gold

Output :

no of unique values : 2
unique values : [Gold,Platinum]
Gold : 3
Platinum :1

Does this answer your question? [Counting unique values in a column in pandas dataframe like in Qlik?](https://stackoverflow.com/questions/45759966/counting-unique-values-in-a-column-in-pandas-dataframe-like-in-qlik) — jdhao, Oct 03 '22 at 06:50

score 8 · Accepted Answer · edited Jun 04 '21 at 06:33

8

Use pd.value_counts

pd.value_counts(df.Account_Type)

Gold        3
Platinum    1
Name: Account_Type, dtype: int64

Get number of unique as well

s = pd.value_counts(df.Account_Type)
s1 = pd.Series({'nunique': len(s), 'unique values': s.index.tolist()})
s.append(s1)

Gold                            3
Platinum                        1
nunique                         2
unique values    [Gold, Platinum]
dtype: object

Alternate Approach

df['col1'].value_counts(sort=True)
df['col1'].value_counts(sort=True, normalize=True) -> provides proportion

edited Jun 04 '21 at 06:33

RomanHotsiy

4,978
1
25
36

answered Mar 28 '17 at 10:01

piRSquared

285,575
57
475
624

When applying this nice code to a categorical variable of a df I receive an error: pd.value_counts(df.application_type) s = pd.value_counts(df.application_type) s1 = pd.Series({'nunique': len(s), 'unique values': s.index.tolist()}) s.append(s1) TypeError: cannot append a non-category item to a CategoricalIndex "application_type" is a category column from df. – NuValue Apr 20 '18 at 12:41

score 1 · Answer 2 · answered Mar 28 '17 at 10:01

1

You can use set() to remove duplicates and then calculate the length:

len(set(data_set))

To count the occurrence:

data_set.count(value)

answered Mar 28 '17 at 10:01

adabsurdum

91
8

score 0 · Answer 3 · answered Jul 01 '18 at 18:54

    df['Account_Type].unique()

returns unique values of the specified column (in this case 'Account_Type') as a NumPy array.

All you have to do is use the len() function to find the no of unique values in the array.

    len(df['Account_Type].unique())

To find the respective counts of unique values, you can use value_counts()

How to find out number of unique values in a column along with count of the unique values in a dataframe?

3 Answers3