3.DataFrame and Series

Download Dataset

32KB

retail_sales_dataset.csv

Open

Two DataTypes in Pandas - DataFrame and Series

When working in pandas, you need to deal with only two datatypes:

• DataFrame

• Series

What is DataFrame

A DataFrame is just like a table (an Excel sheet). It has rows and columns. You can create your own DataFrame with the help of a dictionary also.

import pandas as pd

data = {
  'product': ['iphone', 'samsung', 'vivo', 'blackberry'],
  'price'  : [200, 300, 100, 50]
}
df = pd.DataFrame(data)
df

Out:

	product	price
0	iphone	200
1	samsung	300
2	vivo	100
3	blackberry	50

Whenever you load an Excel or CSV file, it gives you a result as a DataFrame. For example:

df = pd.read_csv('retail_sales_dataset.csv')
df

Out (example):

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

1000 rows × 6 columns

4 Basic Properties of DataFrame

A DataFrame has 4 properties. Each property is shown below with examples.

1

shape

Gives the number of rows and columns.

df.shape

Out:

(1000, 6)

2

index

Gives the row index.

df.index

Out:

RangeIndex(start=0, stop=1000, step=1)

3

columns

Gives all the column names.

df.columns

Out:

Index(['Customer ID', 'Gender', 'Age', 'Product Category', 'Quantity', 'Price per Unit'], dtype='object')

4

values

Gives the array of all values.

df.values

Out (example):

array([['CUST001', 'Male', 34, 'Beauty', 3, 50],
       ['CUST002', 'Female', 26, 'Clothing', 2, 500],
       ['CUST003', 'Male', 50, 'Electronics', 1, 30],
       ...,
       ['CUST998', 'Female', 23, 'Beauty', 4, 25],
       ['CUST999', 'Female', 36, 'Electronics', 3, 50],
       ['CUST1000', 'Male', 47, 'Electronics', 4, 30]], shape=(1000, 6), dtype=object)

4 Basic Functions of DataFrame

Common functions you will use with DataFrame are shown below. These are explained with examples in the stepper that follows:

1

head()

Returns the top n rows in the DataFrame. By default it returns 5 rows.

df.head()       # default 5 rows
df.head(7)      # first 7 rows

Out (example for df.head()):

	Customer ID	Gender	Age	Product Category	Quantity	Price per Unit
0	CUST001	Male	34	Beauty	3	50
1	CUST002	Female	26	Clothing	2	500
2	CUST003	Male	50	Electronics	1	30
3	CUST004	Male	37	Clothing	1	500
4	CUST005	Male	30	Beauty	2	50

2

tail()

Returns the bottom n rows in the DataFrame. By default it returns 5 rows.

df.tail()       # default 5 rows
df.tail(3)      # last 3 rows

Out (example for df.tail()):

	Customer ID	Gender	Age	Product Category	Quantity	Price per Unit
995	CUST996	Male	62	Clothing	1	50
996	CUST997	Male	52	Beauty	3	30
997	CUST998	Female	23	Beauty	4	25
998	CUST999	Female	36	Electronics	3	50
999	CUST1000	Male	47	Electronics	4	30

3

info()

Gives an overview of the DataFrame (dtypes, non-null counts, memory usage).

df.info()

Out (example):

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1000 entries, 0 to 999
Data columns (total 6 columns):
 #   Column            Non-Null Count  Dtype
---  ------            --------------  -----
 0   Customer ID       1000 non-null   object
 1   Gender            1000 non-null   object
 2   Age               1000 non-null   int64
 3   Product Category  1000 non-null   object
 4   Quantity          1000 non-null   int64
 5   Price per Unit    1000 non-null   int64
dtypes: int64(3), object(3)
memory usage: 47.0+ KB

4

describe()

Gives descriptive statistics / 5-number summary for numerical columns.

df.describe()

Out (example):

	Age	Quantity	Price per Unit
count	1000.00000	1000.00000	1000.000000
mean	41.39200	2.51400	179.890000
std	13.68143	1.13273	189.681356
min	18.00000	1.00000	25.000000
25%	29.00000	1.00000	30.000000
50%	42.00000	3.00000	50.000000
75%	53.00000	4.00000	300.000000
max	64.00000	4.00000	500.000000

Explanation for the Age column:

• count : 1000 age records

• mean : average age ≈ 41.39

• min : youngest age = 18

• max : oldest age = 64

• 25% / 50% / 75% : percentiles (part of the 5-number summary)

Note: std is the standard deviation (covered later).

5

to_excel() and to_csv()

Save a DataFrame to an Excel or CSV file. Use index=False if you don't want the index in the output file.

df.to_excel('Adidas.xlsx', index=False)
df.to_csv('Adidas.csv', index=False)

---

Assignments:

1. Load Adidas Dataset from required files.

2. Get the basic properties (shape , index , values , columns).

3. Get only number of rows in the dataset.

4. Get Top 20 rows.

5. Get Last 5 rows.

6. Get the basic information of dataframe.

7. Find the basic statistical summary of the dataframe.

8. Find the basic statistical summary of the dataframe with all columns.

9. Save the statistical summary in excel file and in csv file in your desktop.