4. Series Operations In Pandas

What is Series

Series is a single row or a single column in pandas.

import pandas as pd
import numpy as np

d = ['iphone', 'motorolla', 'samsung']
s = pd.Series(d, name='phones')
s

Output:

0      iphone
1   motorolla
2     samsung
Name: phones, dtype: object

How to access a Series / Columns

df = pd.read_csv('retail_sales_dataset.csv')
df['Customer ID']

Output (truncated):

0      CUST001
1      CUST002
2      CUST003
3      CUST004
4      CUST005
       ...
995    CUST996
996    CUST997
997    CUST998
998    CUST999
999    CUST1000
Name: Customer ID, Length: 1000, dtype: object

How to access multiple Columns

To access multiple columns, pass a list of columns:

cols = ['Customer ID', 'Age']
df[cols]

Output (truncated):

	Customer ID	Age
0	CUST001	34
1	CUST002	26
2	CUST003	50
3	CUST004	37
4	CUST005	30
...	...	...
995	CUST996	62
996	CUST997	52
997	CUST998	23
998	CUST999	36
999	CUST1000	47

1000 rows × 2 columns

How to create new columns

df['new col'] = 1
df

Output (truncated):

	Customer ID	Gender	Age	Product Category	Quantity	Price per Unit	new col
0	CUST001	Male	34	Beauty	3	50	1
1	CUST002	Female	26	Clothing	2	500	1
2	CUST003	Male	50	Electronics	1	30	1
3	CUST004	Male	37	Clothing	1	500	1
4	CUST005	Male	30	Beauty	2	50	1
...	...	...	...	...	...	...	...
995	CUST996	Male	62	Clothing	1	50	1
996	CUST997	Male	52	Beauty	3	30	1
997	CUST998	Female	23	Beauty	4	25	1
998	CUST999	Female	36	Electronics	3	50	1
999	CUST1000	Male	47	Electronics	4	30	1

1000 rows × 7 columns

How to create a column with Serial Numbers

df['Serial Number'] = np.arange(1, 1001)
df

Output (truncated):

	Customer ID	Gender	Age	Product Category	Quantity	Price per Unit	new col	Serial Number
0	CUST001	Male	34	Beauty	3	50	1	1
1	CUST002	Female	26	Clothing	2	500	1	2
2	CUST003	Male	50	Electronics	1	30	1	3
3	CUST004	Male	37	Clothing	1	500	1	4
4	CUST005	Male	30	Beauty	2	50	1	5
...	...	...	...	...	...	...	...	...
995	CUST996	Male	62	Clothing	1	50	1	996
996	CUST997	Male	52	Beauty	3	30	1	997
997	CUST998	Female	23	Beauty	4	25	1	998
998	CUST999	Female	36	Electronics	3	50	1	999
999	CUST1000	Male	47	Electronics	4	30	1	1000

1000 rows × 8 columns

How to create a column with random numbers

df['random column'] = np.random.randint(5,10,1000)

How to Update a Column

df['new col'] = 5
df

Output (truncated):

	Customer ID	Gender	Age	Product Category	Quantity	Price per Unit	new col	Serial Number
0	CUST001	Male	34	Beauty	3	50	5	1
1	CUST002	Female	26	Clothing	2	500	5	2
2	CUST003	Male	50	Electronics	1	30	5	3
3	CUST004	Male	37	Clothing	1	500	5	4
4	CUST005	Male	30	Beauty	2	50	5	5
...	...	...	...	...	...	...	...	...
995	CUST996	Male	62	Clothing	1	50	5	996
996	CUST997	Male	52	Beauty	3	30	5	997
997	CUST998	Female	23	Beauty	4	25	5	998
998	CUST999	Female	36	Electronics	3	50	5	999
999	CUST1000	Male	47	Electronics	4	30	5	1000

1000 rows × 8 columns

How to create a Calculative Column

df['new price'] = df['Price per Unit'] / 2
df

Output (truncated):

	Customer ID	Gender	Age	Product Category	Quantity	Price per Unit	new col	Serial Number	new price
0	CUST001	Male	34	Beauty	3	50	5	1	25
1	CUST002	Female	26	Clothing	2	500	5	2	250
2	CUST003	Male	50	Electronics	1	30	5	3	15
3	CUST004	Male	37	Clothing	1	500	5	4	250
4	CUST005	Male	30	Beauty	2	50	5	5	25
...	...	...	...	...	...	...	...	...	...
995	CUST996	Male	62	Clothing	1	50	5	996	25
996	CUST997	Male	52	Beauty	3	30	5	997	15
997	CUST998	Female	23	Beauty	4	25	5	998	12.5
998	CUST999	Female	36	Electronics	3	50	5	999	25
999	CUST1000	Male	47	Electronics	4	30	5	1000	15

1000 rows × 9 columns

Then:

df['Total Price'] = df['Price per Unit'] * df['Quantity']
df

Output (truncated):

	Customer ID	Gender	Age	Product Category	Quantity	Price per Unit	new col	Serial Number	new price	Total Price
0	CUST001	Male	34	Beauty	3	50	5	1	25	150
1	CUST002	Female	26	Clothing	2	500	5	2	250	1000
2	CUST003	Male	50	Electronics	1	30	5	3	15	30
3	CUST004	Male	37	Clothing	1	500	5	4	250	500
4	CUST005	Male	30	Beauty	2	50	5	5	25	100
...	...	...	...	...	...	...	...	...	...	...
995	CUST996	Male	62	Clothing	1	50	5	996	25	50
996	CUST997	Male	52	Beauty	3	30	5	997	15	90
997	CUST998	Female	23	Beauty	4	25	5	998	12.5	100
998	CUST999	Female	36	Electronics	3	50	5	999	25	150
999	CUST1000	Male	47	Electronics	4	30	5	1000	15	120

1000 rows × 10 columns

How to save your changes in your original file

df.to_csv('retail_sales_dataset.csv',index=False)

Assignments:

1. Load Adidas File from required files.

2. Get Retailer Column only from the dataframe.

3. Find the type of the retailer column.

4. Store top 5 Retailer , Product and Price column records in a small dataframe.

5. Create Total Price Column.

6. Create a column with serial numbers.

7. Create a discount column with random numbers ranging between 1 to 5 percent.