Data Analysis with Python

Read

Discuss

In this article, we will discuss how to do data analysis with Python. We will discuss all sorts of data analysis i.e. analyzing numerical data with NumPy, Tabular data with Pandas, data visualization Matplotlib, and Exploratory data analysis.

Data Analysis With Python

Data Analysis is the technique of collecting, transforming, and organizing data to make future predictions and informed data-driven decisions. It also helps to find possible solutions for a business problem. There are six steps for Data Analysis. They are:

Ask or Specify Data Requirements
Prepare or Collect Data
Clean and Process
Analyze
Share
Act or Report

Data Analysis with Python

Note: To know more about these steps refer to our Six Steps of Data Analysis Process tutorial.

Analyzing Numerical Data with NumPy

NumPy is an array processing package in Python and provides a high-performance multidimensional array object and tools for working with these arrays. It is the fundamental package for scientific computing with Python.

Arrays in NumPy

NumPy Array is a table of elements (usually numbers), all of the same types, indexed by a tuple of positive integers. In Numpy, the number of dimensions of the array is called the rank of the array. A tuple of integers giving the size of the array along each dimension is known as the shape of the array.

Creating NumPy Array

NumPy arrays can be created in multiple ways, with various ranks. It can also be created with the use of different data types like lists, tuples, etc. The type of the resultant array is deduced from the type of elements in the sequences. NumPy offers several functions to create arrays with initial placeholder content. These minimize the necessity of growing arrays, an expensive operation.

Create Array using numpy.empty(shape, dtype=float, order=’C’)

Python3

import numpy as np
  
b = np.empty(2, dtype = int)
print("Matrix b : \n", b)
  
a = np.empty([2, 2], dtype = int)
print("\nMatrix a : \n", a)
  
c = np.empty([3, 3])
print("\nMatrix c : \n", c)

Output:

Empty Matrix using pandas

Create Array using numpy.zeros(shape, dtype = None, order = ‘C’)

Python3

import numpy as np
  
b = np.zeros(2, dtype = int)
print("Matrix b : \n", b)
  
a = np.zeros([2, 2], dtype = int)
print("\nMatrix a : \n", a)
  
c = np.zeros([3, 3])
print("\nMatrix c : \n", c)

Output:

Matrix b : 
 [0 0]

Matrix a : 
 [[0 0]
 [0 0]]

Matrix c : 
 [[0. 0. 0.]
 [0. 0. 0.]
 [0. 0. 0.]]

Operations on Numpy Arrays

Arithmetic Operations

Addition:

Python3

import numpy as np
 
# Defining both the matrices
a = np.array([5, 72, 13, 100])
b = np.array([2, 5, 10, 30])
 
# Performing addition using arithmetic operator
add_ans = a+b
print(add_ans)
 
# Performing addition using numpy function
add_ans = np.add(a, b)
print(add_ans)
 
# The same functions and operations can be used for
# multiple matrices
c = np.array([1, 2, 3, 4])
add_ans = a+b+c
print(add_ans)
 
add_ans = np.add(a, b, c)
print(add_ans)

Output:

[  7  77  23 130]
[  7  77  23 130]
[  8  79  26 134]
[  7  77  23 130]

Subtraction:

Python3

import numpy as np
 
# Defining both the matrices
a = np.array([5, 72, 13, 100])
b = np.array([2, 5, 10, 30])
 
# Performing subtraction using arithmetic operator
sub_ans = a-b
print(sub_ans)
 
# Performing subtraction using numpy function
sub_ans = np.subtract(a, b)
print(sub_ans)

Output:

[ 3 67  3 70]
[ 3 67  3 70]

Multiplication:

Python3

import numpy as np
 
# Defining both the matrices
a = np.array([5, 72, 13, 100])
b = np.array([2, 5, 10, 30])
 
# Performing multiplication using arithmetic
# operator
mul_ans = a*b
print(mul_ans)
 
# Performing multiplication using numpy function
mul_ans = np.multiply(a, b)
print(mul_ans)

Output:

[  10  360  130 3000]
[  10  360  130 3000]

Division:

Python3

import numpy as np
 
# Defining both the matrices
a = np.array([5, 72, 13, 100])
b = np.array([2, 5, 10, 30])
 
# Performing division using arithmetic operators
div_ans = a/b
print(div_ans)
 
# Performing division using numpy functions
div_ans = np.divide(a, b)
print(div_ans)

Output:

[ 2.5        14.4         1.3         3.33333333]
[ 2.5        14.4         1.3         3.33333333]

For more information, refer to our NumPy – Arithmetic Operations Tutorial

NumPy Array Indexing

Indexing can be done in NumPy by using an array as an index. In the case of the slice, a view or shallow copy of the array is returned but in the index array, a copy of the original array is returned. Numpy arrays can be indexed with other arrays or any other sequence with the exception of tuples. The last element is indexed by -1 second last by -2 and so on.

Python NumPy Array Indexing

Python3

# Python program to demonstrate 
# the use of index arrays.
import numpy as np
  
# Create a sequence of integers from
# 10 to 1 with a step of -2
a = np.arange(10, 1, -2) 
print("\n A sequential array with a negative step: \n",a)
  
# Indexes are specified inside the np.array method.
newarr = a[np.array([3, 1, 2 ])]
print("\n Elements at these indices are:\n",newarr)

Output:

A sequential array with a negative step: 
 [10  8  6  4  2]

 Elements at these indices are:
 [4 8 6]

NumPy Array Slicing

Consider the syntax x[obj] where x is the array and obj is the index. The slice object is the index in the case of basic slicing. Basic slicing occurs when obj is :

a slice object that is of the form start: stop: step
an integer
or a tuple of slice objects and integers

All arrays generated by basic slicing are always the view in the original array.

Python3

# Python program for basic slicing.
import numpy as np
 
# Arrange elements from 0 to 19
a = np.arrange(20)
print("\n Array is:\n ",a)
 
# a[start:stop:step]
print("\n a[-8:17:1] = ",a[-8:17:1])
 
# The : operator means all elements till the end.
print("\n a[10:] = ",a[10:])

Output:

Array is:
[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]

a[-8:17:1]  =  [12 13 14 15 16]

a[10:] = [10 11 12 13 14 15 16 17 18 19]

Ellipsis can also be used along with basic slicing. Ellipsis (…) is the number of : objects needed to make a selection tuple of the same length as the dimensions of the array.

Python3

# Python program for indexing using basic slicing with ellipsis
import numpy as np
 
# A 3 dimensional array.
b = np.array([[[1, 2, 3],[4, 5, 6]],
            [[7, 8, 9],[10, 11, 12]]])
 
print(b[...,1]) #Equivalent to b[: ,: ,1 ]

Output:

[[ 2  5]
 [ 8 11]]

NumPy Array Broadcasting

The term broadcasting refers to how numpy treats arrays with different Dimensions during arithmetic operations which lead to certain constraints, the smaller array is broadcast across the larger array so that they have compatible shapes.

Let’s assume that we have a large data set, each datum is a list of parameters. In Numpy we have a 2-D array, where each row is a datum and the number of rows is the size of the data set. Suppose we want to apply some sort of scaling to all these data every parameter gets its own scaling factor or say Every parameter is multiplied by some factor.

Just to have a clear understanding, let’s count calories in foods using a macro-nutrient breakdown. Roughly put, the caloric parts of food are made of fats (9 calories per gram), protein (4 CPG), and carbs (4 CPG). So if we list some foods (our data), and for each food list its macro-nutrient breakdown (parameters), we can then multiply each nutrient by its caloric value (apply scaling) to compute the caloric breakdown of every food item.

NumPy Array Broadcasting

With this transformation, we can now compute all kinds of useful information. For example, what is the total number of calories present in some food or, given a breakdown of my dinner know how many calories did I get from protein and so on.

Let’s see a naive way of producing this computation with Numpy:

Python3

import numpy as np
 
macros = np.array([
[0.8, 2.9, 3.9],
[52.4, 23.6, 36.5],
[55.2, 31.7, 23.9],
[14.4, 11, 4.9]
])
 
# Create a new array filled with zeros,
# of the same shape as macros.
result = np.zeros_like(macros)
 
cal_per_macro = np.array([3, 3, 8])
 
# Now multiply each row of macros by
# cal_per_macro. In Numpy, `*` is
# element-wise multiplication between two arrays.
for i in range(macros.shape[0]):
    result[i, :] = macros[i, :] * cal_per_macro
 
result

Output:

array([[  2.4,   8.7,  31.2],
       [157.2,  70.8, 292. ],
       [165.6,  95.1, 191.2],
       [ 43.2,  33. ,  39.2]])

Broadcasting Rules: Broadcasting two arrays together follow these rules:

If the arrays don’t have the same rank then prepend the shape of the lower rank array with 1s until both shapes have the same length.
The two arrays are compatible in a dimension if they have the same size in the dimension or if one of the arrays has size 1 in that dimension.
The arrays can be broadcast together if they are compatible with all dimensions.
After broadcasting, each array behaves as if it had a shape equal to the element-wise maximum of shapes of the two input arrays.
In any dimension where one array had a size of 1 and the other array had a size greater than 1, the first array behaves as if it were copied along that dimension.

Python3

import numpy as np
 
v = np.array([12, 24, 36])
w = np.array([45, 55])
 
# To compute an outer product we first
# reshape v to a column vector of shape 3x1
# then broadcast it against w to yield an output
# of shape 3x2 which is the outer product of v and w
print(np.reshape(v, (3, 1)) * w)
 
X = np.array([[12, 22, 33], [45, 55, 66]])
 
# x has shape 2x3 and v has shape (3, )
# so they broadcast to 2x3,
print(X + v)
 
# Add a vector to each column of a matrix X has
# shape 2x3 and w has shape (2, ) If we transpose X
# then it has shape 3x2 and can be broadcast against w
# to yield a result of shape 3x2.
 
# Transposing this yields the final result
# of shape 2x3 which is the matrix.
print((X.T + w).T)
 
# Another solution is to reshape w to be a column
# vector of shape 2X1 we can then broadcast it
# directly against X to produce the same output.
print(X + np.reshape(w, (2, 1)))
 
# Multiply a matrix by a constant, X has shape 2x3.
# Numpy treats scalars as arrays of shape();
# these can be broadcast together to shape 2x3.
print(X * 2)

Output:

[[ 540  660]
 [1080 1320]
 [1620 1980]]
[[ 24  46  69]
 [ 57  79 102]]
[[ 57  67  78]
 [100 110 121]]
[[ 57  67  78]
 [100 110 121]]
[[ 24  44  66]
 [ 90 110 132]]

Note: For more information, refer to our Python NumPy Tutorial.

Analyzing Data Using Pandas

Python Pandas Is used for relational or labeled data and provides various data structures for manipulating such data and time series. This library is built on top of the NumPy library. This module is generally imported as:

import pandas as pd

Here, pd is referred to as an alias to the Pandas. However, it is not necessary to import the library using the alias, it just helps in writing less amount code every time a method or property is called. Pandas generally provide two data structures for manipulating data, They are:

Series
Dataframe

Series:

Pandas Series is a one-dimensional labeled array capable of holding data of any type (integer, string, float, python objects, etc.). The axis labels are collectively called indexes. Pandas Series is nothing but a column in an excel sheet. Labels need not be unique but must be a hashable type. The object supports both integer and label-based indexing and provides a host of methods for performing operations involving the index.

Pandas Series

It can be created using the Series() function by loading the dataset from the existing storage like SQL, Database, CSV Files, Excel Files, etc., or from data structures like lists, dictionaries, etc.

Python Pandas Creating Series

Python3

import pandas as pd
import numpy as np
 
 
# Creating empty series
ser = pd.Series()
 
print(ser)
 
# simple array
data = np.array(['g', 'e', 'e', 'k', 's'])
 
ser = pd.Series(data)
print(ser)

Output:

pnadas series

Dataframe:

Pandas DataFrame is a two-dimensional size-mutable, potentially heterogeneous tabular data structure with labeled axes (rows and columns). A Data frame is a two-dimensional data structure, i.e., data is aligned in a tabular fashion in rows and columns. Pandas DataFrame consists of three principal components, the data, rows, and columns.

Pandas Dataframe

It can be created using the Dataframe() method and just like a series, it can also be from different file types and data structures.

Python Pandas Creating Dataframe

Python3

import pandas as pd
 
# Calling DataFrame constructor
df = pd.DataFrame()
print(df)
 
# list of strings
lst = ['Geeks', 'For', 'Geeks', 'is',
            'portal', 'for', 'Geeks']
 
# Calling DataFrame constructor on list
df = pd.DataFrame(lst)
df

Output:

Creating Dataframe from python list

Creating Dataframe from CSV

We can create a dataframe from the CSV files using the read_csv() function.

Note: This dataset can be downloaded from here.

Python Pandas read CSV

Python3

import pandas as pd
 
# Reading the CSV file
df = pd.read_csv("Iris.csv")
 
# Printing top 5 rows
df.head()

Output:

head of a dataframe

Filtering DataFrame

Pandas dataframe.filter() function is used to Subset rows or columns of dataframe according to labels in the specified index. Note that this routine does not filter a dataframe on its contents. The filter is applied to the labels of the index.

Python Pandas Filter Dataframe

Python3

import pandas as pd
 
# Reading the CSV file
df = pd.read_csv("Iris.csv")
 
# applying filter function
df.filter(["Species", "SepalLengthCm", "SepalLengthCm"]).head()

Output:

Applying filter on dataset

Sorting DataFrame

In order to sort the data frame in pandas, the function sort_values() is used. Pandas sort_values() can sort the data frame in Ascending or Descending order.

Python Pandas Sorting Dataframe in Ascending Order

Output:

Sorted dataset based on a column value

Pandas GroupBy

Groupby is a pretty simple concept. We can create a grouping of categories and apply a function to the categories. In real data science projects, you’ll be dealing with large amounts of data and trying things over and over, so for efficiency, we use the Groupby concept. Groupby mainly refers to a process involving one or more of the following steps they are:

Splitting: It is a process in which we split data into group by applying some conditions on datasets.
Applying: It is a process in which we apply a function to each group independently.
Combining: It is a process in which we combine different datasets after applying groupby and results into a data structure.

The following image will help in understanding the process involve in the Groupby concept.

1. Group the unique values from the Team column

Pandas Groupby Method

2. Now there’s a bucket for each group

3. Toss the other data into the buckets

Pandas GroupBy

4. Apply a function on the weight column of each bucket.

Applying Function on the weight column of each column

Python Pandas GroupBy

Python3

# importing pandas module
import pandas as pd
 
# Define a dictionary containing employee data
data1 = {'Name': ['Jai', 'Anuj', 'Jai', 'Princi',
                  'Gaurav', 'Anuj', 'Princi', 'Abhi'],
         'Age': [27, 24, 22, 32,
                 33, 36, 27, 32],
         'Address': ['Nagpur', 'Kanpur', 'Allahabad', 'Kannuaj',
                     'Jaunpur', 'Kanpur', 'Allahabad', 'Aligarh'],
         'Qualification': ['Msc', 'MA', 'MCA', 'Phd',
                           'B.Tech', 'B.com', 'Msc', 'MA']}
 
# Convert the dictionary into DataFrame
df = pd.DataFrame(data1)
 
print("Original Dataframe")
display(df)
 
# applying groupby() function to
# group the data on Name value.
gk = df.groupby('Name')
 
# Let's print the first entries
# in all the groups formed.
print("After Creating Groups")
gk.first()

Output:

pandas groupby

Applying function to group:

After splitting a data into a group, we apply a function to each group in order to do that we perform some operations they are:

Aggregation: It is a process in which we compute a summary statistic (or statistics) about each group. For Example, Compute group sums or means
Transformation: It is a process in which we perform some group-specific computations and return a like-indexed. For Example, Filling NAs within groups with a value derived from each group
Filtration: It is a process in which we discard some groups, according to a group-wise computation that evaluates True or False. For Example, Filtering out data based on the group sum or mean

Pandas Aggregation

Aggregation is a process in which we compute a summary statistic about each group. The aggregated function returns a single aggregated value for each group. After splitting data into groups using groupby function, several aggregation operations can be performed on the grouped data.

Python Pandas Aggregation

Python3

# importing pandas module
import pandas as pd
 
# importing numpy as np
import numpy as np
 
# Define a dictionary containing employee data
data1 = {'Name': ['Jai', 'Anuj', 'Jai', 'Princi',
                  'Gaurav', 'Anuj', 'Princi', 'Abhi'],
         'Age': [27, 24, 22, 32,
                 33, 36, 27, 32],
         'Address': ['Nagpur', 'Kanpur', 'Allahabad', 'Kannuaj',
                     'Jaunpur', 'Kanpur', 'Allahabad', 'Aligarh'],
         'Qualification': ['Msc', 'MA', 'MCA', 'Phd',
                                  'B.Tech', 'B.com', 'Msc', 'MA']}
 
 
# Convert the dictionary into DataFrame
df = pd.DataFrame(data1)
 
# performing aggregation using
# aggregate method
 
grp1 = df.groupby('Name')
 
grp1.aggregate(np.sum)

Output:

Use of sum aggregate function on dataset

Concatenating DataFrame

In order to concat the dataframe, we use concat() function which helps in concatenating the dataframe. This function does all the heavy lifting of performing concatenation operations along with an axis of Pandas objects while performing optional set logic (union or intersection) of the indexes (if any) on the other axes.

Python Pandas Concatenate Dataframe

Python3

# importing pandas module
import pandas as pd 
  
# Define a dictionary containing employee data 
data1 = {'key': ['K0', 'K1', 'K2', 'K3'],
         'Name':['Jai', 'Princi', 'Gaurav', 'Anuj'], 
        'Age':[27, 24, 22, 32],} 
    
# Define a dictionary containing employee data 
data2 = {'key': ['K0', 'K1', 'K2', 'K3'],
         'Address':['Nagpur', 'Kanpur', 'Allahabad', 'Kannuaj'], 
        'Qualification':['Btech', 'B.A', 'Bcom', 'B.hons']} 
  
# Convert the dictionary into DataFrame  
df = pd.DataFrame(data1)
  
# Convert the dictionary into DataFrame  
df1 = pd.DataFrame(data2) 
   
 
display(df, df1) 
 
# combining series and dataframe
res = pd.concat([df, df1], axis=1)
  
res

Output:

Merging DataFrame

When we need to combine very large DataFrames, joins serve as a powerful way to perform these operations swiftly. Joins can only be done on two DataFrames at a time, denoted as left and right tables. The key is the common column that the two DataFrames will be joined on. It’s a good practice to use keys that have unique values throughout the column to avoid unintended duplication of row values. Pandas provide a single function, merge(), as the entry point for all standard database join operations between DataFrame objects.

There are four basic ways to handle the join (inner, left, right, and outer), depending on which rows must retain their data.

Merge Daframe Python Pandas

Python Pandas Merge Dataframe

Introduction to Data Analysis

Data Analysis Libraries

Data Visulization Libraries

Exploratory Data Analysis (EDA)

Data Preprocessing

Data Transformation

Time Series Data Analysis

Case Studies and Projects

Introduction to Data Analysis

Data Analysis Libraries

Data Visulization Libraries

Exploratory Data Analysis (EDA)

Data Preprocessing

Data Transformation

Time Series Data Analysis

Case Studies and Projects

Data Analysis with Python

Data Analysis With Python

Analyzing Numerical Data with NumPy

Arrays in NumPy

Creating NumPy Array

Python3

Python3

Operations on Numpy Arrays

Arithmetic Operations

Python3

Python3

Python3

Python3

NumPy Array Indexing

Python NumPy Array Indexing

Python3

NumPy Array Slicing

Python3

Python3

NumPy Array Broadcasting

Python3

Python3

Analyzing Data Using Pandas

Python Pandas Creating Series

Python3

Python Pandas Creating Dataframe

Python3

Creating Dataframe from CSV

Python3

Filtering DataFrame

Python3

Sorting DataFrame

Pandas GroupBy

Python3

Pandas Aggregation

Python3

Concatenating DataFrame

Python3

Merging DataFrame

Python3