Python Programming For Data Science And Machine Learning

#Data Types

Number: int, float, complex
Text: str
Boolean: bool
Sequence: list, tuple, range
Set: set, frozenset
Mapping: dict
Binary: bytes, memoryview

#Command to print hello world

print("!!!Hello World")

#Command to print world in next line

print("Hello\nWorld")

#Code to store string value and display it

strworld = "Dear Chetan"
print(strworld)

#Command to know type

type(strworld)

#BEDMAS

print(2 - 4 * 2 + 9 / 10)

#Arthematic operation

a = 9
b = 5
print(a + b)
print(a - b)
print(a / b)
print(a // b) #Quotient
print(a ** b) #Exponential
print(a % b) #Gives remainder

#Increment and decrement of number
inc = 8
inc += 1
print(inc)
dec = 8
dec -= 1
print(dec)

#float type checking

num = 30.3
print(num)
type(num)

#Complex number

com = 3 + 8j
print(com)
type(com)

#Logical operation

print(13<25)
print(True and False)
#It will print true if both are True
print(True or False)
#It will print true if any one is True

#Text conversion from upper case to lower or vise-versa

sentence = "i am lower case now"
print(sentence.upper())

sentence = "I AM UPPER CASE NOW"
print(sentence.lower())

#right strip sentence

strip_sentence = "There is a space at the end       "
print(strip_sentence)
print(strip_sentence.rstrip())
#rstrip will delete the space

#left strip sentence

strip_sentence = "     There is a space at the Beginning"
print(strip_sentence)
print(strip_sentence.lstrip())

#Strip all or delete all space

chetan = "     Shidling     "
print(chetan)
print(chetan.strip())

#Strip which is mentioned

percent = "555%"
print(percent.rstrip("%"))

#mixed strip

mixedstr = "*****SSSSSS######"
print(mixedstr.rstrip("#").lstrip("*"))
#rstrip means strip right side
#lstrip means strip left side

#Sentence operation

mixedstr = "20 people affected by Carona"
no_of_people = mixedstr[0:2]
print("Number of people are :", no_of_people)

#rest of string

Rest_of_string = mixedstr[2:]
print(Rest_of_string)

groupstr = "10 Apple 5 Banansa 3 Orage"
fruit = groupstr[19 :27]
print(fruit)

print(groupstr[:-2])

#Find even or odd
#even
nums_seq = "123456789"
even_nums = nums_seq[1::2]
print(even_nums)

#Find odd

odd_nums = nums_seq[0::2]
print(odd_nums)

odd_nums_again = nums_seq[ ::2]
print(odd_nums_again)

#Print first name and last name

first_name = "Chetan"
last_name = "Shidling"
name = first_name + " " + last_name
print(name)

#My age

my_age = input("Enter my age :")
my_age_sen = ("I am " + my_age +" years old")
print(my_age_sen)

#Scentence formation

A = "Data"
B = "Science"
C = "Python"
print("{0} {1} using {2}".format(A, B, C))

#Arrays

#How to create list

create_list = []
print(create_list)
type(create_list)

#Another way using constructor list()

list_1 = list()
type(list_1)

Methods performed on lists

#append() - adds an element at the end of the list
#insert() - adds an element at the specified position
#extend() - adds the elements of a list ( or any iterable), to the end of the current list
#copy() - returns a copy of the list
#count() - retuns the number of elements with the specified value
#clear() - removes all the elements from the list
#index() - returns the index of the first element with the specified value
#remove() - removes the item with the specified value
#pop() - removes the order of the list
#reverse() - reverses the order of the list
#sort() - sorts the list

#Data Structures

fruits = ['Orange', 'Apple', 'Pear', 'Banana', 'Kiwi', 'Apple', 'Banana']

#How to find count of elements inside a list

fruits.count('Apple')

#How to find index

fruits.index('Banana')
fruits.index('Banana', 4)#It will find banana index starting a position 4

#How to reverse

fruits.reverse()
fruits

#How to append

fruits.append('Grape')
fruits

#How to sort

fruits.sort()
fruits

#How to pop or delete element

fruits.pop()
fruits

#How to print particular element

print(fruits[0])
print(fruits[4])

#How to print elements from range(this to this)

print(fruits[2:5])

#How to print last element

print(fruits[-1])

#How to print elements from particular point

print(fruits[3:])

#How to print elements from particular point

print(fruits[:-3])

#How to copy all elemets and insert new element

new_fruits = fruits
new_fruits.append('gauva')
new_fruits

#How to find id

print(id(new_fruits))

#Sentence into words in sentence

a_sentence = "Corona virus is spreading all over the world"
words_in_sentence = a_sentence.split()
print(words_in_sentence)

#mutli sentence

multi_sentence = "I am Chetan. I have a pet dog. It's color is white and black, but it looks big."
sentence_inText = multi_sentence.split('.')
print(sentence_inText)

#Inser $ between all character

join_sentence = '$'.join(multi_sentence)
print(join_sentence)

#Print sentence three time

var_list = ["My name is Chetan"]
print(var_list*3)

#Cancate

another_list = ["I live in Raichur"]
print(var_list + another_list)

#Find length

print(len(another_list))

#Find length

print(len(another_list))
print(len(fruits))

#Remove element from a list using for loop

list1 = ['Apple','Banana','Orange','Orange','Banan','Banana','Apple']
count1 = list1.count('Banana')
for i in range(0,count1):
    list1.remove('Banana')
print(list1)

#Find length of number list

num_list = [2, 3, 5, 1, 9, 3, 1]
print(len(num_list))

#How to sort numbers

print(sorted(num_list))

#How to find minimum value

print(min(num_list))

#How to find maximum value

print(max(num_list))

#How to create a nested list

nested_list = [[2, 5, 1],[5, 9, 3],[9, 2, 0]]
print(nested_list[1])

#How to print any one element is nested list

print(nested_list[0][2])

#How to print any few list

print(nested_list[0:2])

#Delete particular element in the list

num_list = [-1, 3, 4, 1, 5, 6, 9, 22]
del num_list[2]
print(num_list)

#Tuple - They are immutable and usually contain a heterogeneous sequence of elements

sample_tuple = 23433, 84392, "Friends"
type(sample_tuple)

print(sample_tuple)

#Reverse packing

x,y,z = smaple_tuple
print(x)
print(y)
print(z)

#Empty tuples are contructed by an empty pair of parentheses
#A tuple with one item is constructed by following a value with a comma
#It is not sufficient to enclose a single value in parenthese. Ugly, but effective

empty_tuple = ()
single_tuple = 'Dear',
print(single_tuple)
type(single_tuple)

#Now let's see difference between list and tuple

list_1 = [12, 45, 55]
print(list_1)
print("\n")
tuple_1 = tuple(list_1)
print(tuple_1)
type(tuple_1)

#Tuple 

tuple_2 = ("Chetan","Nitin",23,"Ratan",44)
type(tuple_2)

print(tuple_2)

print(tuple_2[0])

#nested tuple

nested_tuples = tuple_1 , (1, 2, 5, 2)
print(nested_tuples)

#Nested tuples but with mutable objects

tuple_4 = ([1,2,3],[3,2,4])
print(tuple_4)
type(tuple_4)

#Set is an unordered collection with no duplicate elements
#Curly braces or the set() function can be used to create sets

#How to create a set and find type

basket = {'apple', 'orange', 'apple', 'pear', 'orange', 'banana','watermelon'}
print(basket)
type(basket)

'orange' in basket

#How to convert list into set

num_list_1 = [1,2,3,4,5,6]
num_set_1 = set(num_list_1)
print(num_set_1)
type(num_set_1)

#Let's print list

num_list_2 = [3,4,2,1,4,8]
num_set_2 = set(num_list_2)
print(num_list_2)
type(num_set_2)

num_set_1 - num_set_2
print(num_set_1)

num_set_1 | num_set_2

#It give you common element

num_set_1 & num_set_2

num_set_1 ^ num_set_2

print(num_set_1.intersection(num_set_2))

print(num_set_1.union(num_set_2))

print(num_set_1)
print(num_set_2)
print(num_set_2.difference(num_set_1))

#Dictionaries - A set of key:value pairs - with the requirement that the keys are unique (within one dictionary)
#Use {} to create dictionary
#Use 'dict()' to create dictionary
#dictionaries are indexed by keys, which can be immutable

empty_dictionary = {}
print(empty_dictionary)
type(empty_dictionary)

#Storing elements in dictionary

diction_data = {'Name':'Chetan', 'Age':21, 'Height':'5.9 ft', 'Hobby':'Blogging'}
print(diction_data)
type(diction_data)

#Print list data

list(diction_data)

diction_hobby = diction_data['Hobby']
print(diction_hobby)

diction_age = diction_data['Age']
print(diction_age)

diction_data['Age'] = 19
print(diction_data)

#Adding new key and its value

diction_data['Profession'] = 'Engineer'
print(diction_data)

#Lets check

'Profession' in diction_data

#How to get list of keys in one shot

print(list(diction_data.keys()))

#How to get list of values in one shot
print(list(diction_data.values()))

new_diction = dict([('Ratan',3000),('Nitin',3000),('Krishna',4000)])
new_diction

diction_data.update(new_diction)
print(diction_data)

#How to delete data

del diction_data['Krishna']
print(diction_data)
len(diction_data)

#Lets do something more

students_data = dict([('Kiran',22),('Akshay',24),('Raj',35)])
print(students_data)

students_data[2] = "Krishna",23
print(students_data)

#Functions

def display():
    print("Hello World")
    
display()

def square(num):
    out = num**2
    return(out)

sq_r = square(3)
print(sq_r)

#Find factorial

def factorial(n):
    if n>1:
        return n*factorial(n-1)
    else:
        return n
    
fact = factorial(6)
print(fact)

#Addition of argument

def addition(*args):
    print(args)
    return(sum(args))

print(addition(4,6,2,7,3,11))
print(addition(1,2))

#Conditional statements

score = int(input("Enter a number :"))

if score < 0:
    print("Enter positive number")
elif score == 0:
    print("You scored Zeo")
elif score == 1:
    print("You scored one")
else:
    print("It is above one")

#Multiple if statement

score = int(input("Please enter your score :"))
cutoff = 75
pass_marks = 35

if score >= 75:
    print("You got distinction")
    
if score >= 35:
    print("You passed")
    
if score < 35:
    print("You failed")

#If else

score = int(input("Enter you score"))

if score >=35:
    print("You passed")
else:
    print("You failed")

#With strings

str1 = "Chetan works at CS Electrical And Electronics"
if "Electrical" in str1:
    print("It is there")
else:
    print("Not found")
    print("\n")
    
str2 = "World is very big and lovely"
if "verymuch" in str2:
    print("It is found in str2")
else:
    print("Not found") 

if False:
    print("True for this statement")
else:
    print("False for this statement")

if True:
    print("True for this statement")
else:
    print("False for this statement")

#Loops

fruits = ['Apple','Orange','Banana']
for i in fruits:
    print(i, len(i))

#Using range function, loop through...

for w in range(6, 15):
    print(w)

#Range function with increment

for i in range(10,20,2):
    print(i)

#Looping on a string

string = "Jai Shri Ram"

for alphabet in string:
    print(alphabet)

#Triangle

str1 = ' '
for i in range(0,9):
    if i<5:
        str1 += '* '
        print(str1)
    elif i>4:
        str1 = str1[:-2]
        print(str1)

#Find count in statement

my_string = "Count the statement alpabets"

for n,alphabet in enumerate(my_string):
    print(alphabet, n)

#Find the vowels

vowels =' '
for alphabet in my_string:
    if alphabet in 'aeiou':
        vowels += ' ' + alphabet
print(vowels)

#Find the even and odd numbers

num = '0123456789'
even = ' '
odd = ' '
for number in num:
    if int(number)%2 == 0:
        even += number
    else:
        odd += number
        
print('All evens are : ' + even + ' & All odds are : '+odd)

#Sentence check

str1 = "#Chetan #Shidling is a #Entrepreneur"

for word in str1.split():
    if word.startswith('#'):
        print(word[1:])

#Dictionary

students = {1:['Chetan',21], 2:['Nitin',19], 3:['Krish', 22]}
type(students)

for key, val in students.items():
    print(key, val)

for key in students.keys():
    print(key)

for val in students.values():
    print(val)

#While loop for fibonacci series

a, b = 0, 1
n = 9
while a < n:
    print(a, end=' ')
    a, b = b, a+b

#NumPy - NUMerical PYthon

#It is used for scientific computing and data analysis

#Power of NumPy is N-dimensional array object which is in the form of rows and columns.

#Fast
#Less memory
#Convenient
#Vectorized code 

#How to import the numpy library

import numpy as np

#np is an alias, you may use any other alias, though np is standard in the industry

#Create 1-D array using list/tuple
#use np.array()

fromList_to_1dArray = np.array([10,20,30,40,50])
fromTuple_to_1dArray = np.array((1,2,3,4,5))

print(type(fromList_to_1dArray))
print(fromList_to_1dArray)
print("\n")
print(type(fromTuple_to_1dArray))
print(fromTuple_to_1dArray)

#2-D array creation using two lists

fromList_to_2dArray = np.array([[0,1,2,3,4],[5,6,7,8,9]])
print(fromList_to_2dArray)

#Let us see how Numpy works when compare to standard python

#Sum of two list elements

num_list1 = [1,2,3,4,5,6,7,8,9]
num_list2 = [10,11,12,13,14,15,16,17,18]

sum_list = list(map(lambda x, y: x+y, num_list1, num_list2))
print(sum_list)

#Use numpy arrays
#1D Array
import numpy as np
np_list1 = np.array(num_list1)
np_list2 = np.array([10,11,12,13,14,15,16,17,18])

np_sumLst = np_list1+np_list2
print(np_sumLst)
print(type(np_sumLst))

#Creation of arrays of fixed size

#from list
array_from_list = np.array([1,2,3,4,5])
array_from_list2 = np.array([1,2,3,4,5],dtype=float)
#from tuple
array_from_tuple = np.array((0,6,7,8,9))

print(array_from_list)
print(array_from_list2)
print(array_from_tuple)

#Condition statement as an argument to np.array()

array_from_list3 = array_from_list >= 2
print(array_from_list3)

#How to fetch the actual elements

array_from_list[array_from_list3]

#2D Array

array1 = np.array([[0,1,2,3],[4,5,6,7]])
print(array1)

array2 = np.array([[10,11,12,13],[14,15,16,17]])
array2

#Subtraction of two arrays

array2 - array1

###How to initialize np arrays when size is known

#Functions that are used to do so are:

#np.arange(): Array creation with defined increments
#np.zeros(): Array of 0s
#np.ones(): Array of 1s
#np.random.random(): Array of random numbers
#np.random.randint(): Random array of integers within a particular range
#np.linspace(): Array of fixed length
#np.full(): Constant array of any number 'n'
#np.eye(): Identity matrix array
#np.title(): Array is going to be created many times

#np.arange()
#range()

increment = np.arange(1,15,2)
increment

decrement = np.arange(50,10,-10)
decrement

#Let us test them and see how they work
#Zeros initialization

np.zeros(5)
np.zeros((5,5))

#Ones initialization

np.ones(4)
np.ones((4,4))
np.ones((4,4), dtype = int)

#random numbers Array


np.random.random([3,3])

#Create a 3 * 3 random array of integers ranging from 0 to 9

np.random.randint(1,10,(3,3))

#np.full() - default data type int

np.full(2,5)
np.full((2,3),5)

#Identity matrix

i_matrix = np.eye(5, dtype = int)
i_matrix

#np.linspace()
#Lenght known, step size unknown

np.linspace(1,10,5)

#Repeating array for number of times - default data type is int

array1 = ([5,4,3])
np.tile(array1,5)
np.tile(array1, (4,5))

#Generate an array (5 * 5)
import numpy as np
array1 = np.ones((5,5))
array1

#Find dtype, shape, itemsize, and ndim of array1

print(array1.dtype)
print(array1.shape)
print(array1.itemsize)
print(array1.ndim)

#Creating a 3-d array - Difficult to print it
#reshape() simply reshapes a 1-D array

array_3d = np.arange(12).reshape(2,3,2)
print(array_3d)

#Indexing, Subseting, Slicing and iterating through Arrays

#Indexing and slicing one dimensional arrays

array1d = np.arange(5)
print(array1d)

#Access thro indexing

array1d[2]
array1d[2:3]

#Read all elements

array1d[[2,3,4]]

#Subset with an increment of 2

print(array1d[0::2])

#Iterations are also similar to lists

for i in array1d:
    print(i*2)

#2 D array accessing

array2d = np.array([[1,2,3,4],[5,6,7,8],[9,10,11,12]])
print(array2d)
print(array2d.ndim)
print(array2d[1,3])

print(array2d[2,:])
#Slicing all colums
print(type(array2d[2,:]))
#Slicng all rows
print(array2d[:,1])
#Slicing all rows and the first 2 colums
print(array2d[:,:2])

#Iterating over 2D/3D arrays (row wise and columwise)

for colum in array2d:
    print(colum)

#Execution speed in NumPy and Standard Python Lists

NumPy is much faster!!!

#Import time library

import time
#create two large lists
list1 = [i for i in range(200000)]
list2 = [j**2 for j in range(200000)]
#Capture start time & end time
start_time = time.time()
perform_Ops = list(map(lambda x,y: x*y, list1, list2))
Ops_time = time.time()
diff_time = Ops_time - start_time
print(diff_time)

#numpy array
array1 = np.array([i for i in range(200000)])
array2 = np.array([j**2 for j in range(200000)])
np_startTime = time.time()
array3 = array1*array2
np_endTime = time.time()
numpy_time = np_endTime - np_startTime
print(numpy_time)

print("The ratio of the time taken {}".format(numpy_time/diff_time))

#File is a named location on a disk to store related information.
#It is used to store the data permanently in a non volatile memory
#Variables will go off

#Mantra
#Open -> R or W -> Close

#Mode

#r-Read, w-Write, a-append, r+ - read and write

#file open

#open(path,mode)
open('G:\\Chetan\\work.txt',"w")
#close()

open('G:\\Chetan\\work.txt',"r")

#Reading file and assigning file object

fo = open('G:\\Chetan\\work.txt',"r")

readdata = fo.read()
print(readdata)
fo.close()

print(type(readdata))

print(type(fo))

#Reading file and assigning file object

fo = open('G:\\Chetan\\work.txt',"r")

readdata = fo.read(6)
print(readdata)
fo.close()

#Readlines

fo = open('G:\\Chetan\\work.txt',"r")

readdata = fo.readlines()
print(readdata)
fo.close()

#One by one

fo = open('G:\\Chetan\\work.txt',"r")
linedata = fo.readlines()

for line in linedata:
    print(line)
    
fo.close()

#Writing to file

fo = open('G:\\Chetan\\work.txt',"w")

fo.write("I am in Banulure\n")
fo.write("I came here to start my own company\n")

fo.close()

#Open file you will find previous data will be vanished

#append

fo = open('G:\\Chetan\\work.txt',"a")
fo.write("\nHa ha, Deleted na?\n")

fo.close()

#Ask user

fo = open('G:\\Chetan\\work.txt',"a")
message = input("\nEnter your message :")
fo.write(message)

fo.close()

NumPy Arrays - Few Operations

Basic mathematical operations / linear algebra operations / functions

Playing with arrays using resize / reshape / stack creation

#Playing with arrays

#import the numpy library

import numpy as np

#If the arrays don't have the same dimensions
#use resize()

sample_array = np.arange(0,11)
print(sample_array)

print(sample_array.size)

print(np.resize(sample_array, (3,4)))

print(np.resize(sample_array, (6,4)))

print(sample_array.resize(3,5))

#Give a new shape to an array without changing its data

sample_array = np.arange(0,9).reshape(3,3)
sample_array

#Dimensions are automatically calculated when you use -1

smaple_array = np.arange(0, 9)
print(sample_array)
x = sample_array.reshape(3,-1)
x

#Transpose of a matrix

x.T

#Adding elements at the end of the array

y = np.append(sample_array, [10,11])
y

#Insert elements to the array at index -8

z = np.insert(sample_array,8,9)
z

#Delete element at given index array

z = np.delete(sample_array,[8])
z

Split/Join/Stack Arrays

np.r_() - Arrange arrays row-wise
np.c_() - Arrange arrays column-wise
np.concatenate() - Concatenate two arrays
np.hstack() - stack arrays horizontally
np.vstack() - stack arrays vertically

#Create 2d arrays

import numpy as np
array1 = np.arange(9).reshape(3,3)
array2 = np.arange(12).reshape(3,4)

print(array1)
print(array2)

#Concatenate

array3 = np.arange(9,18).reshape(3,3)
print(np.concatenate((array1,array3)))

#Stack the array elements horizontally

np.hstack((array1, array3))

#Stack the array elements vertically

np.vstack((array1, array3))

#Stack arrays row-wise

print(np.r_[array1,array3])

#Stack arrays column-wise

print(np.c_[array1,array3])

Basic Mathematical Operations / Linear Algebra Operations / Functions

#Applying mathematical operations

sample_array = np.arange(10,20)

#exponential, logarithms, trigometry, sine and cosine

print('\nexponential values of sample array')
print(np.exp(sample_array))
print('\nlogarithmic values of sample array')
print(np.log(sample_array))
print('\nSine values of sample array')
print(np.sin(sample_array))
print('\nCosine values of sample array')
print(np.cos(sample_array))

sample_array

#Executing a function in non-numpy way

non_numpy_fn = [(x**2+2*x+3) for x in sample_array]
print(non_numpy_fn)

#Vertorize the function and apply an numpy array

vectorised_fn = np.vectorize(lambda x: (x**2+2*x+3))

vectorised_fn (sample_array)

Linear Algebra Operations

help(np.linelg) - Provides linear algebra documentation

np.linalg - is a package

np.linalg.inv - inverse of a matrix
np.linalg.det - Determinant of a matrix
np.linalg.eig - Eigenvalues and eigenvectors of a matrix
np.dot(a,b) - dot product of two matrices

#Creating two arrays

array1 = np.arange(5,14).reshape(3,3)
array2 = np.arange(1,13).reshape(3,4)
print(array1)
print(array2)

#Determinant of 3*3 matrix vs 3*4 matrix

import numpy as np
print(np.linalg.det(array1))
#np.linalg.det(array2)
#array2 is 3*4 matrix, so it will throw error

#Eligenvalues

np.linalg.eig(array1)

#dotproduct of the matrix

np.dot(array1,array2)

Basics of Pandas library

Used for data analysis - Data manipulation, building ML models and data visualisation

Pandas data structures - Series & Dataframes

Pandas Series

01. import pandas library
02. To create series object, use pd.series()
03. Each column in a dataframe is a pandas series

#import pandas

import pandas as pd

#Creating a character series from tuple

char_series = pd.Series(('H','A','R','I'))
char_series

#Creating series from list

num_series = pd.Series([3,6,9,12,15,18,21])
print(num_series)
print(type(num_series))

#Creating a datetime series

date_series = pd.date_range(start = '01-04-2020', end = '19-04-2020')
print(date_series)
type(date_series)

#For more information try out this

help(pd.date_range(start = '01-04-2020', end = '19-04-2020'))

#Indexing

print(num_series[4])
print(num_series[1:4])
print(num_series[[1,4]])

#Setting up indexes explicity
#Use index argument while generating series

num_series = pd.Series([3,6,9,12,15], index = ['a','b','c','d','e'])
num_series

#Another way of defining index

import numpy as np
series1 = pd.Series(np.array(range(0,5))+2, index = range(0,5))
series1

The Pandas Dataframe

Dataframe - It is a table with rows and columns

Rows having an index and columns having meaningful names

Creating dataframes

Ways to create dataframes are:
    from dictionaries
    JSON objects
    Reading from CSV files, flat files or txt files, etc.

#Using dictionaries, create dataframe

basket_df = pd.DataFrame({'fruits':['Banana','Guava','Orange','Apple','Kiwi'], 'weight':[12,1,24,2,8], 'color':['Yellow','Green','Pale yellow','Red','Brown']})
basket_df

#Reading a CSV file as a dataframe

basket_df = pd.read_csv("G:/Chetan/table.csv")
basket_df

#Reading a excel file as a dataframe

basket_df = pd.read_excel("G:/Chetan/excel.xlsx")
basket_df

Play with Dataframe

#Finding the information

basket_df.info()

#Dimension of dataframe (rows * columns)

basket_df.shape

#Glance at top and bottom records of dataframe

#It will display top rows and columns
basket_df.head()

#It will display buttom 5 rows and columns

basket_df.tail()

#Summary of all the numeric columns in the dataset

basket_df.describe()

#Find columns names

basket_df.columns

#Extract the values of a dataframe as a numpy array using dataframe.values

basket_df.values

#Indexing with your own values
#Let us see how we can do it

basket_df.head()

#Set color column as index

basket_df.set_index('Name', inplace = True)
basket_df.head()