UNIT 5 INTRODUCTION TO PYTHON LIBRARIES

Python Libraries- Introduction to Libraries- Creating and Exploring Packages-Numpy, SciPy,

matplotlib, Pandas, Scikit-learn- seaborn.

5.1 Python Libraries

 The Python Standard Library is a collection of exact syntax, token, and semantics of
Python.
 Python has a huge collection of libraries.
 The Python Standard Library contains hundreds of modules for performing common
tasks,
 Python’s standard library is very extensive, offering a wide range of facilities.
 A Python library is a reusable chunk of code that you may want to include in your
programs/ projects.
 Mostly it is written in C, and handles functionality like I/O and other core modules.
 Python supports very huge number of library files. Some examples are:
 Numpy
 SciPy
 Pandas
 Matplotlib
 Scikit-Learn
 Seaborn
 Keras
 PyTorch
 TensorFlow

5.2 Introduction to Libraries

 Library Definition: The library is having a collection of related functionality of codes

that allows you to perform many tasks without writing your code. It is a reusable chunk
of code that we can use by importing it in our program.
 A library means “a bundle of code.”.
 A library is a collection of pre-combined codes that can be used iteratively to reduce the
time required to code.
 They are particularly useful for accessing the pre-written frequently used codes, instead
of writing them from scratch every single time.
 The library contains built-in modules that provide access to system functionality such as
file I/O.

1
  A library is a collection of modules.

5.3 Creating and Exploring Packages

 Module Definition: The module is a simple Python file that contains collections of
functions and global variables and with having a .py extension file.
 Package Definition: The package is a simple directory having collections of modules.
This directory contains Python modules and also having __init__.py file by which the
interpreter interprets it as a Package.
 Module can contain multiple objects, such as classes, functions, etc.
 A package can contain one or more relevant modules.
 A package is actually a folder containing one or more module files.
 The package folder contains a special file called __init__.py, which stores the package's
content.

 import statement:
 A file is considered as a module in python. To use the module, you have to
import it using the import keyword.
 import modules from packages using the dot (.) operator.
 import Game.Level.start

Syntax: Example:

import module_name import math

Syntax: Example:

2
 from module_name import member_name from math import pi
Syntax: Example:

from module_name import * from math import *

Syntax: Example:

import module_name as alias_name import math as ma

 Use the following steps to create a package in Python:

1. First, we create a directory and give it a package name, preferably related to its
operation.
2. Then create more modules in a directory, we put the classes and the required
functions in a module.
3. Finally we create an __init__.py file inside the directory, to let Python know that
the directory is a package.
 Example:
1. First create a directory (package) i.e name of the directory is Arithmetic
2. Create first module with name of the module is addition.py and define one
function i.e add() inside the addition.py file
def add(a,b):
c=a+b
print(f"Addition is :{c} ")
3. Create second module with name of the module is subtraction.py and define one
function i.e sub() inside the subtraction.py file
def sub(a,b):
c=a-b
print(f"Subtraction is :{c} ")
4. Finally create the __init__.py file. This file will be placed inside Arithmetic
directory and can be left blank.
5. Call(import) Arithmetic package in a program i.e ruff.py to perform addition and
subtraction of two numbers.
import Arithmetic.addition
import Arithmetic.subtraction
Arithmetic.addition.add(10,20)
Arithmetic.subtraction.sub(50,30)

Example: Output:

1.Package name is Arithmetic Addition is :30

2.addition.py Subtraction is :20

3
 def add(a,b):
c=a+b
print(f"Addition is :{c} ")

3.subtraction.py

def add(a,b):
c=a+b
print(f"Addition is :{c} ")

4.ruff.py

import Arithmetic.addition
import Arithmetic.subtraction
Arithmetic.addition.add(10,20)
Arithmetic.subtraction.sub(50,30)

5.4 Numpy

 NumPy (Numerical Python) is the fundamental package for numerical computation in

Python
 It is the most popular machine learning library in Python.
 NumPy is a Python library used for working with arrays.
 It also has functions for working in domain of linear algebra, fourier transform, and
matrices.
 It is an open source library.
 NumPy is one of the fundamental packages for Python providing support for large
multidimensional arrays and matrices.
 NumPy relies on BLAS and LAPACK for efficient linear algebra computations.
 NumPy can also be used as an efficient multi-dimensional container of generic data.
 NumPy is one of the most used libraries for tasks involving modern scientific
computations and evolving yet powerful domains like Data Science and Machine
Learning.
 It supports for powerful N-dimensional array objects and built-in tools for performing
intensive mathematical as well as scientific calculations.
 It is used for delivering high performance, interoperability with various computing
platforms and hardware, and ease of use.

4
 NumPy is the fundamental package for scientific computing with Python, adding support
for large, multidimensional arrays and matrices, along with a large library of high-level
mathematical functions to operate on these arrays.

 Features OfNumpy
 Interactive: Numpy is very interactive and easy to use.
 Mathematics: Makes complex mathematical implementations very simple.
 Intuitive: Makes coding real easy and grasping the concepts is easy.
 Lot of Interaction: Widely used, hence a lot of open source contribution
 Applications Of Numpy
 Extensively used in data analysis
 Creates powerful N-dimensional array
 Forms the base of other libraries, such as SciPy and scikit-learn
 Replacement of MATLAB when used with SciPy and matplotlib

Example: Output:

import numpy as np [1 2 3 4 5]
a=np.array([1,2,3,4,5]) [[1 2 3]
b=np.array([[1,2,3],[4,5,6]]) [4 5 6]]
print(a) <class 'numpy.ndarray'>
print(b) 2
print(type(a)) 3
print(b.ndim) [2 3]
print(a[2]) [0 2 3 4 5]
print(a[1:3]) (2, 3)
x = a.copy()
x[0]=0
print(x)
print(b.shape)

 import numpy library.

 NumPy package can be referred to as np instead of numpy.
 The indexes in NumPy arrays start with 0.
 array() - function to create arrays
 type() - the type of the object passed to it.
 ndim attribute - that returns an integer that tells us how many dimensions the array have.
 copy() – copy the content of one array into another.
 shape attribute - It returns an array has how many dimensions, and each dimension has
how many elements.

5
5.5 SciPy

 SciPy stands for Scientific Python.

 It provides more utility functions for optimization, stats and signal processing.
 Like NumPy, SciPy is open source so we can use it freely.
 SciPy is an open-source library used for solving mathematical, scientific, engineering,
and technical problems.
 It allows users to manipulate the data and visualize the data using a wide range of high-
level Python commands.
 SciPy is built on the Python NumPy extention.
 SciPy library contains modules for optimization, linear algebra, integration, and statistics.
 SciPy is a library used by scientists, analysts, and engineers doing scientific computing
and technical computing.
 It contains modules for optimization, linear algebra, integration, interpolation, special
functions, FFT, signal and image processing, ODE solvers, and other tasks common in
science and engineering.

 Features of Scipy
 Collection of algorithms and functions built on the NumPy extension of Python
 High-level commands for data manipulation and visualization
 Multidimensional image processing with the SciPy ndimage submodule
 Includes built-in functions for solving differential equations
 Applications of Scipy
 Multidimensional image operations
 Solving differential equations and the Fourier transform
 Optimization algorithms
 Linear algebra
 Subpackages in SciPy

Package Name Description

scipy.io  File input/output
scipy.special  Special Function
scipy.linalg  Linear Algebra Operation
scipy.interpolate  Interpolation
scipy.optimize  Optimization and fit
scipy.stats  Statistics and random numbers
scipy.integrate  Numerical Integration
scipy.fftpack  Fast Fourier transforms
scipy.signal  Signal Processing

6
 scipy.ndimage  Image manipulation

 import the SciPy module using fromscipy import module statement.

 Ex : from scipy import constants - imported the constants module from SciPy

Example: Output:

from scipy import special 1000.0

a = special.exp10(3) 8.0
print(a) 1.0
0.7071067811865475
b = special.exp2(3) 1.0
print(b)

c = special.sindg(90)
print(c)

d = special.cosdg(45)
print(d)

d = special.tandg(45)
print(d)

5.6 Matplotlib

 Matplotlib is one of the most popular Python packages used for data visualization.
 It is a cross-platform library for making 2D plots from data in arrays.
 It is used to create a wide variety of visualizations, including line plots, histograms, bar
charts, pie charts, scatter plots, tables, and many other styles.
 Matplotlib is a Python 2D plotting library that produces publication-quality figures in a
variety of hard-copy formats and interactive cross-platform environments.
 This 2D plotting library of Python is very famous among data scientists for designing
varieties of figures in multiple formats which is compatible across their respected
platforms.
 Features of Matplotlib:
 Usable as a MATLAB replacement, with the advantage of being free and open
source
 Supports dozens of backends and output types, which means you can use it
regardless of which operating system you’re using or which output format you
wish to use

7
  Pandas itself can be used as wrappers around MATLAB API to drive MATLAB
like a cleaner
 Low memory consumption and better runtime behavior
 Applications of Matplotlib:
 Correlation analysis of variables
 Visualize 95 percent confidence intervals of the models
 Outlier detection using a scatter plot etc.
 Visualize the distribution of data to gain instant insights

Example: Output:

import matplotlib.pyplot as plt

import numpy as np

x = np.array([0, 6])
y = np.array([0, 250])
plt.plot(x, y)
plt.show()

x = np.array([1, 2, 6, 8])
y = np.array([3, 8, 1, 10])
plt.plot(x, y,'*r--')
plt.title("Example")
plt.xlabel("X-Axis")
plt.ylabel("Y-Axis")
plt.grid()
plt.show()

x = np.array(["A", "B", "C", "D"])

y = np.array([3, 8, 1, 10])
plt.bar(x,y,color="red")
plt.show()

y = np.array([35, 25, 25, 15])

l = ["Apples", "Bananas", "Mangoes", "Dates"]
plt.pie(y,labels=l)
plt.legend()
plt.show()

8
  Most of the Matplotlib utilities lies under the pyplot submodule, and are usually imported
under the plt alias.
 plot() function draws a line from point to point.
 keyword argument marker to emphasize each point with a specified marker
 title() function to set a title for the plot.
 xlabel() and ylabel() functions to set a label for the x- and y-axis
 grid() function to add grid lines to the plot.
 bar() function to draw bar graphs
 pie() function to draw pie charts

5.7 Pandas

 PANDAS referred as Python Data Analysis Library.

 PANDAS is another open source Python library for availing high-performance data
structures and analysis tools.
 Pandas is a machine learning library in Python that provides data structures of high-level
and a wide variety of tools for analysis
 It is an open source library.
 Ability to translate complex operations with data using one or two commands.
 Pandas have so many inbuilt methods for grouping, combining data, and filtering, as well
as time-series functionality.
 Ability to group and sort data, select best suited output for the apply method, and
provides support for performing custom types operations.
 Pandas enable the provision of easy data structure and quicker data analysis for Python
 It provides fast, expressive, and flexible data structures to easily work with structured
(tabular, multidimensional, potentially heterogeneous) and time-series data.
 It is used for solving modern Data Science and Machine Learning problems.
 Reading and writing data into multiple formats like csv, excel, sqletc
 Pandas is the “SQL of Python.”
 It is used to handle two-dimensional data tables in Python.
 Load data into data frames, select columns, filter for specific values, group by values, run
functions (sum, mean, median, min, max, etc.), merge dataframes and so on.
 It contains DataFrame as its main data structure.
 With DataFrame you can store and manage data from tables by performing manipulation
over rows and columns.
 Also create multi-dimensional data-tables.
 It was created for data analysis, data cleaning, data handling and data discovery
 Features Of Pandas
 An efficient DataFrame object for data manipulation

9
  Easy reshaping and pivoting of data sets
 Merging and joining of data sets
 Label-based data slicing, indexing, and subsetting
 working with time-series data
 Eloquent syntax and rich functionalities that gives you the freedom to deal with
missing data
 Enables you to create your own function and run it across a series of data
 High-level abstraction
 Contains high-level data structures and manipulation tools
 Manipulating data will be easier.
 Support for various operations such as Re-indexing, Iteration, Sorting,
Aggregations, Concatenations and Visualizations.
 Applications Of Pandas
 It is the best data analysis tool available for solving real-world problems
 General data wrangling and cleaning
 It is the best data analysis tool available for solving real-world problems
 ETL (extract, transform, load) jobs for data transformation and data storage, as it
has excellent support for loading CSV files into its data frame format
 Used in a variety of academic and commercial areas, including statistics, finance
and neuroscience
 Time-series-specific functionality, such as such as date range generation, moving
window, linear regression and date shifting.

Example: Output:
import pandas as pd States Capitals
df = pd.DataFrame({'States':['TN', 'AP', 'UP', 'MP'], 0 TN CHENNAI
'Capitals':['CHENNAI', 'AMARAVATHI', 1 AP AMARAVATHI
'LUCKNOW', 'BHOPAL']}) 2 UP LUCKNOW
df.to_excel('data1.xlsx') 3 MP BHOPAL
df = pd.read_excel('data1.xlsx')
print(df)

Example: Read csv file Output:

import pandas as pd Regno name science physics maths
df = pd.read_csv('stock.csv') 0 111 aaa 89 78 56
print(df.to_string()) 1 222 bbb 54 87 64
new_df = df.dropna()
print(new_df.to_string())
x = df["maths"].mean()
print(x)

10
  import pandas library alias name is pd.
 Data sets in Pandas are usually multi-dimensional tables, called DataFrames.
 loc attribute to return one or more specified row(s)
 dropna() - remove rows that contain empty cells.
 fillna() - replace empty cells with a value
 mean() - to calculate the mean value of the column
 median() - to calculate the median value of the column
 mode() - to calculate the mode value of the column
 drop_duplicates() – remove duplicates rows

5.8 scikit-Learn

 It is a Python library is associated with NumPy and SciPy.

 It is considered as one of the best libraries for working with complex data.
 It can be effectively used for a variety of applications which include classification,
regression, clustering, model selection, naive Bayes’, grade boosting, K-means, and
preprocessing.
 Scikit-Learn, which simply defines itself as “Machine Learning in Python.”
 It is designed to be interpolated into NumPy and SciPy.
 It is a simple tool for data analysis and mining-related tasks.
 It is open-source library files.
 It is being used for classification, regression and clustering to manage spam, image
recognition, drug response, stock pricing, customer segmentation etc.
 It also allows dimensionality reduction, model selection and pre-processing.
 Scikit-learn (Sklearn) is the most useful and robust library for machine learning in
Python.
 The sklearn library contains a lot of efficient tools for machine learning and statistical
modeling including classification, regression, clustering and dimensionality reduction
 Features Of Scikit-Learn
 Cross-validation: There are various methods to check the accuracy of
supervised models on unseen data.
 Supervised Learning algorithms: Almost all the popular supervised learning
algorithms, like Linear Regression, Support Vector Machine (SVM), Decision
Tree etc., are the part of scikit-learn.
 Unsupervised learning algorithms: starting from clustering, factor analysis,
principal component analysis to unsupervised neural networks.
 Feature extraction: Useful for extracting features from images and text
 Clustering: This model is used for grouping unlabeled data.

11
  Simple and efficient tools for predictive data analysis.
 It helps in all aspects and algorithms of Machine Learning, even Deep
Learning.
 It is easy to learn and use.
 Applications of Scikit-Learn
 clustering
 classification
 regression
 model selection
 dimensionality reduction
 Modelling process in Scikit-Learn
 Dataset Loading - A collection of data is called dataset.
 Preprocessing the Data - Before inputting that data to machine learning
algorithms, we need to convert it into meaningful data. This process is called
preprocessing the data.
 Splitting the dataset - To check the accuracy of our model, we can split the
dataset into two pieces-a training set and a testing set.
 Train the Model- use our dataset to train some prediction-model.

Example: Output:
import pandas as pd (1067, 13)
import matplotlib.pyplot as plt [126.18920155]
from sklearn.model_selection import Actual Predicted
train_test_split 0 356 342.055315
from sklearn.linear_model import 1 209 220.385324
LinearRegression 2 230 220.385324
3 212 232.159839
dataset=pd.read_csv("fuel.csv") 4 168 224.310162
print(dataset.head())
print(dataset.shape)
X = dataset['ENGINESIZE'].values.reshape(-1,1)
y = dataset['CO2EMISSIONS'].values.reshape(-
1,1)
X_train, X_test, y_train, y_test =
train_test_split(X, y, test_size=0.2,
random_state=0)
regr = LinearRegression()
regr.fit(X_train, y_train)
print(regr.intercept_)
y_pred = regr.predict(X_test)

12
df = pd.DataFrame({'Actual': y_test.flatten(),
'Predicted': y_pred.flatten()})
print(df)

plt.scatter(X_test, y_test, color='gray')

plt.plot(X_test, y_pred, color='red', linewidth=2)

plt.show()

5.9 seaborn

 Seaborn was designed to visualize the complex statistical models.

 It has the potential to deliver accurate graphs such as heat maps.
 Seaborn is a Python data visualization library based on matplotlib.
 It provides a high-level interface for drawing attractive and informative statistical
graphics.
 Seaborn is a data visualization library built on top of matplotlib and closely integrated
with pandas data structures in Python.
 Visualization is the central part of Seaborn which helps in exploration and understanding
of data.
 Seaborn aims to make visualization a central part of exploring and understanding data.
 Visualizing Statistical Data Using Seaborn.
 Seaborn which has powerful libraries to visualize and explore your data.
 Features Of Seaborn
 Dataset oriented API to determine the relationship between variables.
 Automatic estimation and plotting of linear regression plots.
 It supports high-level abstractions for multi-plot grids.
 Visualizing univariate and bivariate distribution.
 Provides a high-level interface to draw statistical graphics.
 Applications of Seaborn
 Distribution Plots
 Pie Chart
 Bar Chart
 Scatter Plots
 Pair Plots
 Heat maps
13
  Distplot stands for distribution plot, it takes as input an array and plots a curve
corresponding to the distribution of points in the array.
 Import the Seaborn module with another name is sns
 load_dataset()-loading dataset
 jointplot() - Bivariate Distribution is used to determine the relation between two
variables.
 catplot() - Time series can be plotted using catplot() i.e barchart

Example: Output:

import pandas as pd
import seaborn as sns
from matplotlib import pyplot as plt

dataset=pd.read_csv("fuel.csv")
sns.distplot(dataset['CO2EMISSIONS'],hist=False)
sns.jointplot(x = "ENGINESIZE",y =
'CO2EMISSIONS',data = dataset,kind='hex')
sns.catplot("CO2EMISSIONS", data=dataset,
color='green',kind='count')

14