Pandas DataFrame Attributes

DataFrame Attributes

All information related to a DataFrame is available in its attributes. Python provides a lot of dataframe attributes to access the information of a dataframe.

We can access all the information as below.

<DataFrame Object> . <Attribute Name> 

Example

import pandas as pd 

dictObj={‘EmpId’:[‘E01’,’E02’,’E03’],  

              ‘EmpName’:[‘Raj,’Ram’,’Renu’],  

              ‘Department’:[‘Accounts’,’HR’,’IT’]  

              }  

df=pd.DataFrame(dictObj) 

Get information related to Index, Columns, Axes and Data Types

Index:Returns the starting value, ending value, and the difference(step) of row index. 

>>> df.index  

 Output is RangeIndex(start=0,stop=3,step=1)  

Columns:Returns the column name of the dataframe.

>>> df.columns  

 Output : Index([‘EmpId’,’EmpName,’Department’],dtype=’object’)  

Axes:Returns a list which contains the rowindex as well as the column name of the dataframe.

>>> df.axes  

 Output : [RangeIndex(start=0,stop=3,step=1), Index([‘EmpId’,’EmpName,’Department’],dtype=’object’)]  

Dtypes:Returns datatypes of each column of a dataframe.

>>> df.dtypes  

 Output :   

 EmpId        Object  

 EmpName  Object  

 Department   Object  

 dtype : Object  

See the output of all the above attributes in running mode.

Retrieving size (no. of elements), shape, number of dimensions

 

size:Returns a total number of elements present in dataframe.

>>> df.size  

 Output 12  

shape:Returns a tuple which gives the present number of rows and number of columns of a dataframe as an element.

>>> df.shape  

Output : (3,3)  

sdim : Returns an integer value which represents the number of dimensions of a dataframe.

>>> df.ndim  

 Output : 2  

See the output of all the above attributes in running mode.

Retrieving values

 values: Returns a NumPy array which contains all rows as a value.

df.values  

empty: Returns a Boolean value which represents if the dataframe is empty or not. If it will return “True”, then dataframe is empty, otherwise, it is not empty.

>>> df.empty  

Transposing a DataFrame (T) :It transposes a dataframe, i.e., rows become columns and columns become rows.

>>> df.T  

Getting Count of non-NA values in dataframe

count(): It will return non-NA values for each COLUMNS. By default, it will take 0 as an argument.

>>> df.count()  

 Output :  

 EmpId 3  

 EmpName 3  

 Department 3   

count(1) :If we pass 1 as an argument, then instead of returning number of columns, it will return number of each rows along with index number,

>>> df.count(1)  

 Output  

 0 3  

 1 3  

 2 3  

Count with axis parameter :We can also explicitly specify an argument to count() as axis.

 

If we want to count columns value then pass argument like

>>> df.count(axis=’columns’)  

If we want to count rows value then pass argument like

>>> df.count(axix=’rows’)  

See the output of all the above count attribute in running mode,