Series
Series is a one-dimensional labeled array
We can think series as a single column or a row
Elements inside the series has a label (index) to it index starts from 0
The axis label collectively called index
Series are immutable means once a series is created we cannot change its size
Homogenous data means all the elements inside the series should be of same data type
But values are mutable
How To Create A Series Data Structure
A series can be created by using the Series() constructor
First import pandas Example 1
** as keyword is used for aliasing (temporary name )
Parameters of Series Constructor
Data - which data to use to construct a series .
Data can be a list, numpy array, tuple, dictionaries , constant value etc..
index - axis labels
Default index is np.arange(len(data))
We can give our own index label .But the index value should be unique and it should be of same length of data
dtype - data type for the data
Creating Series Data Structure By Using Above Parameters
Data
A series can be created using various data like array ,list, Dictionaries etc..
Example 2
Example 3
Example 4
Example 5
Example 6
Index
Default index will be np.arange(len(data)).But we can change the default index by using index parameter. Index should be unique
Example 7
dtype
we can use dtype parameter for changing the data type of the input data
can use numpy datatypes or python data types
Example 8
Series Attributes
axes
Returns axis label
Example 8
dtype
Returns the data type of the elements inside the series
Example 9
empty
Checks whether series is empty or not .If empty returns True or False
Example 10
size
Returns the no of elements inside the series
Example 11
values
Returns the series as ndarray (use when we need to convert series to ndarray)
Example 12
ndim
Returns rank of the array (Series will always be 1 dimensional array)
Example 13
Accessing Data (Elements) From Series
Elements in the series can be accessed by using [] (similar to accessing elements in list or ndarray)
We can also use negative indexing
Example 14
Default index will be range(len(object))
If index parameter used we can use the specified index to access the element
Example 15
We can also use slicing to access elements from the series (similar to list,ndarrays)
Note ending index not included
We can also use negative indexing
Example 16
DataFrame
DataFrame is a two-dimensional labeled array
We can think DataFrame as a table which contains rows and columns
Elements inside the DataFrame has a label (rowindex,column index) to it index starts from 0
The axis labels collectively called row index and column index
DataFrame are mutable means once a DataFrame is created we can change its shape
Homogenous data column wise means all the elements should be of same data type (columns)
Heterogenous data row wise means different elements (rows)
Values are mutable
DataFrame can be considered as a container holding series data structures (each column inside a DataFrame can be considered as a series data structure.That's why columns have homogenous data and rows have heterogenous data)
How To Create A Series Data Structure
A DataFrame can be created by using the DataFrame() constructor
Example 17
** as keyword is used for aliasing (temporary name )
Parameters of DataFrame Constructor
Data - which data to use to construct a series .
Data can be a list, numpy array, tuple, dictionaries , constant value etc..
index - axis labels (Row axis labels)
Default index is np.arange(len(data))
We can give our own index label .But the index value should be unique and it should be of same length of data
dtype - data type for the data
columns - axis labels (Column axis labels)
Default index is np.arange(len(data))
We can give our own index label .But the index value should be unique and it should be of same length of data
Creating Data Frame Data Structure By Using Above Parameters
Data
Data can be a list, numpy array, tuple, dictionaries , constant value etc..
Example 18
Example 19
Example 20
Example 21
index
As Data Frame is a 2 dimensional data structure we can provide labels for both rows and columns so index is used for row axis labels (Row axis labels)
Example 22
dtype
we can change the data type for the elements by using dtype parameters
Example 23
columns
As Data Frame is a 2 dimensional data structure we can provide labels for both rows and columns so columns is used for column axis labels (Row axis labels)
Example 24
Example 25
** when using dictionary the key will behave as column name so if we use columns argument it will check whether the defined column is there or not if there it will return only that particular column values or else it will return and empty dictionaries or elements will be filled with nan with that particular column name
