We will go through how to utilize pandas groupby() and count() aggregate in this post with in depth examples. So, wait no longer but scroll down for further helpful information!

What Is Pandas Groupby?

A groupby procedure includes dividing up a lot of data into groups and doing calculations on these groups. In most cases, it involves dividing the item, using a function, and combining the outcomes.

Let’s look at the syntax on using pandas to retrieve the count of the last value in the group in this post as follows:

DataFrame.groupby(by, axis, as_index)

As such, we have:

• by (tuples, datatype-list, dict, array, series) involves list of labels, function, mapping, and label. The groups are established using the supplied function exactly as is.
• axis (default 0, datatype int): 1 divides columns, whereas 0 divides rows.
• as_index (datatype bool, default True.)  Returns an object for all aggregated output with group labels as the index.

Simple Dataframe Groupby() And Count() Examples

If you are in need of grouping by columns and retrieving the count for each group from a DataFrame quickly, have a look at the illustrations below.

# Using groupby() and count()
df2 = df.groupby(['Teachers'])['Teachers'].count()

# Using GroupBy & count() on multiple column
df2 = df.groupby(['Teachers','Prolongation'])['Fee'].count()

Let’s now construct a DataFrame with a few rows and columns by running these examples, and then check the outcomes in the end. The columns in our DataFrame will be named as Teachers, Fee, Prolongation, and Discount.

# Create a pandas DataFrame.
import pandas as pd

englishteaching   = ({
    'Teachers':["Simmon","Pinmark","Hanson","Putin","Plex","Hanson","Simmon","Putin"],
    'Fee' :[32000,45000,33000,44000,36000,45000,35000,42000],
    'Prolongation':['50days','30days','40days','35days','35days','60days','50days','55days'],
    'Discount':[2000,1300,2000,1100,2500,2300,1200,1500]
                })

df = pd.DataFrame(englishteaching, columns=['Teachers','Fee','Prolongation','Discount'])
print(df)

That way, we are achieve the below output:

Teachers    Fee Prolongation  Discount
0  Simmon  32000   50days      2000
1  Pinmark  45000   30days      1300
2  Hanson  33000   40days      2000
3  Putin  44000   35days      1100
4  Plex  36000   35days      2500
5  Hanson  45000   60days      2300
6  Simmon  35000   50days      1200
7  Putin  42000   55days      1500

Employing count() by Column Name

Pandas DataFrame.groupby() is a truly great deal in grouping the rows by column and employing the count() method to retrieve the count for each group by disregarding None and Nan values.

Better yet, it even functions with non-floating type data. The example below groups the values by the Teachers column and counts the number of times each value is present.

# Using groupby() and count()
df2 = df.groupby(['Teachers'])['Teachers'].count()

pint(df2)

The output returned will be:

Teachers
Hanson     2
Plex     1
Pinmark     1
Putin     2
Simmon      2

Name: Teachers, dtype: int64

Pandas Groupby() And Count() On List Of Columns

In order to apply groupby to several columns and determine a count over each combination group, you can also pass a list of the columns you wish to group to the groupby() method.

For instance, the function df.groupby(['Teachers',''Prolongation'])['Fee'].count() groups by the ‘Teachers’ and ‘Prolongation’ columns before determining the count.

# Using groupby() & count() on multiple column
df2 = df.groupby(['Teachers','Prolongation'])['Fee'].count()
print(df2)

The output returned will be:

Teachers  Prolongation
Hanson   40days      1
         60days      1
Plex   35days      1
Pinmark  30days      1
Putin   45days      1
         55days      1
Simmon    50days      2

Name: Fee, dtype: int64

Conclusion

Above is all the fundamental insight regarding how to operate pandas groupby() and count(). Hopefully, this article can be of great help to you somehow. See then!