Aitt PBL
Uploaded by
RAHUL MAitt PBL
Uploaded by
RAHUL MA3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 2
Exp# 1 Week# 2 Date#
Aim :
Write a program to input graph and display adjacency nodes for a node given by user.
Libraries, Methods & Variables Used:
NONE
● adj(graph,node):returns the adjacent nodes of a particular node ariables:
V
graph(dic),
ad(list),ele(c
har)
Program:
def adj(graph,ele):
return graph[ele]
graph={
"A":["B","C"],
"B":["A","C","E","D"],
"C":["A","B","D","E"],
"D":["B","C","E","F"],
"E":["B","C","D","F"],
"F":["E","F"]
}
ad=adj(graph,"B")
print("adjacent elements of B : ",ad)
Output:
adjacent elements of B : ['A', 'C', 'E', 'D']
Observation:
G
● raph is represented using dictionary
● H ere the adjacent nodes of the graph is stored as key value pairs in a
dictionary
● When we enter A’s node then it displaying BC
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Exp# 2 Week# 2 Date#
Aim :
To demonstrate the operations of a queue data structure.
Libraries, Methods & Variables Used:
NONE
e nqueue(list,ele):Adding elements to the queue ariables:
V
dequeue():It allows both ends to remove elements fromthe queue n(int),k(input),d(i
size():return the size of the queue nt)
program:
def enqueue(queue,ele):
queue.append(ele)
def dequeue():
return queue.pop(0)
def size():
return len(queue)
queue=[]
print("Enter number of elements : ")
n=int(input())
print("Enter n elements to enqueue : ")
for i in range(0,n):
k=input()
enqueue(queue,k)
print("Queue elements : ",queue)
d=dequeue()
print("Dequeue element : ",d)
print(queue)
print("size of queue : ",size())
Output:
nter number of elements :
E
3
Enter n elements to enqueue :
2
1
A3CIL202 - AI Tools and Techniques Lab 22331A0597
4
Queue elements : ['2', '1', '4']
Dequeue element : 2
['1', '4']
size of queue : 2
Observation:
H
● ere we give input as any data type
● We can perform both insertion and deletion operations at both of its ends of the queue.
Exp# 3 Week# 2 Date#
Aim :
To demonstrate the operations of a stack data structure.
Libraries, Methods & Variables Used:
NONE
ush(list,ele):Adding elements to the stack
p ariables:
V
pop():Removes the element from one side of stack n(int),stack(list),k(
size():Return the size of stack int),p(int)
program:
def push(stack,i):
stack.append(i)
def pop():
return stack.pop()
def size():
return len(stack)
stack=[]
print("Enter no of elements to push : ")
n=int(input())
print("Enter n elements to push : ")
for i in range(0,n):
k=input()
push(stack,k)
print("stack elements : ",stack)
p=pop()
A3CIL202 - AI Tools and Techniques Lab 22331A0597
p rint("pop element : ",p)
print("Size of stack : ",size())
Output:
nter no of elements to push :
E
3
Enter n elements to push :
hi
hey
hello
stack elements : ['hi', 'hey', 'hello']
pop element : hello
Size of stack : 2
Observation:
I f we use get then it only retrieves the value
●
● Stack operations will perform at only one side
● Here we give any data type of elements as input
Exp# 4 Week# 2 Date#
Aim :
Demonstrate Breadth-First Search (BFS) AI uninformed search algorithm.
Libraries, Methods & Variables Used:
NONE
a ppend():Add element to the end of the list ariables:
V
GetNode(ele):Gives Adjacent elements of a node queue(list),vis(list),
enqueue(list,ele):Adding elements to the queue graph(dic)
Program:
q ueue =[]
vis =[]
def enqueue(val,q):
A3CIL202 - AI Tools and Techniques Lab 22331A0597
q .append(val)
def GetNode(c):
return graph[c]
def bfs(start):
enqueue(start,queue)
for i in queue:
if i not in vis:
for j in GetNode(i):
enqueue(j,queue)
vis.append(i)
graph={
"A":["B","C"],
"B":["A","C","D","E"],
"C":["A","B","D","E"],
"E":["B","C","D","F"],
"D":["B","C","E","F"],
"F":["D","E"]
}
bfs('A')
print(vis)
Output:
['A', 'B', 'C', 'D', 'E', 'F']
Observation:
● T he graph is represented as a dictionary where each key is a node and its
corresponding value is a list of neighboring nodes.
● Queue is used to represent the Breadth First Search.
● BFS explore all the neighbors in the present depth level and then moves to the next
depth level.
Exp# 5 Week# 2 Date#
Aim :
Demonstrate Depth-First Search (DFS) AI uninformed search algorithm.
Libraries, Methods & Variables Used:
A3CIL202 - AI Tools and Techniques Lab 22331A0597
NONE
a ppend():Add element to the end of the list ariables:
V
get(ele):Gives Adjacent elements of a node stack(list),vis(list),g
push(list,ele):Adding elements to the stack raph(dic),n(char),i(i
nt),j(int)
pop():Removes the element from the stack
Program:
s tack=[]
vis=[]
def push(val,s):
s.append(val)
def pop(s):
return s.pop()
def get(val):
return graph[val]
def dfs(val):
push(val,stack)
while stack:
n=pop(stack)
if n not in vis:
vis.append(n)
for j in get(n):
push(j,stack)
graph={ "A":["B","C"], "B":["A","D"], "C":["A","E"], "D":["B","E"], "E":["C","D"] }
dfs('A')
print(vis)
Output:
['A', 'C', 'E', 'D', 'B']
Observation:
S
● tack is used to implement Depth First Search
● The vis set keeps track of visited nodes to avoid revisiting them.
● The graph is represented as a dictionary where each key is a node and its corresponding
value is a list of neighboring nodes.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 3
Exp# 6 Week# 3 Date#
Aim :
Calculate distance between two nodes in a graph using euclidean distance
Libraries, Methods & Variables Used:
mathis a built-in library in Python that providesaccess to mathematical functions.
ow():It returns the power a number.
P ariables:
V
sqrt():a function from the math library thatreturnsthe square root of a(int),b(int),c(int),d
a number (int),dis(float)
Program:
import math
a=int(input())
b=int(input())
c=int(input())
d=int(input())
dis=math.sqrt(pow((d-b),2)+pow((c-a),2))
print('Distance between two points : ',dis)
Output:
2
3
4
5
Distance between two points : 2.8284271247461903
Observation:
● T
he Euclidean distance is calculated using the Pythagorean theorem, which is the
squarerootofthesumofthesquaresofthedifferencesbetweenthex-coordinatesand
the y-coordinates of the two nodes.
● In this program we use ** instead of pow().
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Exp# 7 Week# 3 Date#
Aim :
Calculate distance between two nodes in a graph using manhattan distance.
Libraries, Methods & Variables Used:
NONE
abs():Return the absolute value of a given number. ariables:
V
a(int),b(int),c(int),d
(int),mdis(int)
program:
a =int(input())
b=int(input())
c=int(input())
d=int(input())
mdis=abs(d-b)+abs(c-a)
print("manhattan distance : ",mdis)
Output:
2
3
4
5
manhattan distance : 4
Observation:
I f we give non-integer value it will raise an value error
●
● The distance is calculated by using formula abs(x2-x1)+abs(y2-y1)
● We can use fabs instead of abs, But it returns float value
Exp# 8 Week# 3 Date#
Aim :
Demonstrate Best-First Search (BFS) AI informed search algorithm.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Libraries, Methods & Variables Used:
eapq:It helps you to find the minimumormaximumelementofaPythonlist,whennew
h
elements are being added to the list periodically.
fs(list,ele):It is used for find the shortest ormost optimal path of a
b ariables:
V
graph. hq(list),vis(list),gra
ph(dic),res(list)
program:
import heapq
def bfs(graph,sc):
hq=[(0,sc)]
vis=set()
while hq:
s,d=heapq.heappop(hq)
if d not in vis:
vis.add(d)
for x,wt in graph[d]:
heapq.heappush(hq,(wt,x))
return list(vis)
graph={
'A':[('B',5),('C',7)],
'B':[('A',5),('D',4)],
'C':[('A',7),('E',6)],
'D':[('B',4),('E',2)],
'E':[('C',6),('D',2)]
}
sc='A'
res=bfs(graph,sc)
print(res)
Output:
['B', 'A', 'C', 'E', 'D']
Observation:
T
● he program can also calculate the negative costs.
● We can get an Index Error if the nodes are out of range.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 4
Exp 9 Week 4 Date
Aim :
Demonstrate various NumPy array creation, manipulation, and statistical functions.
Libraries, Methods & Variables Used:
umpy:NumPy is a Python library used for working with arrays.It also has functions for
n
working in domain of linear algebra, fourier transform, and matrices.NumPy stands for
Numerical Python.
dim:T
n he ndim property returns the number of dimensionof the DataFrame. ariables:
V
dtype:T he NumPy array object has a property called dtype that returns the arr(list),arr
data type of the array: 1(list),arr2(
reshape:Reshaping means changing the shape of an array.The shape of an list),arr3(li
array is the number of elements in each dimension. st)
Arrange():This function is used to create an arraywith a range of values.
add():this function is used when we want to compute the addition of two
arrays.
subtract(): this function is used when we want tocomputethedifferenceof
two array.
transpose():We can perform thesimplefunctionoftransposewithinoneline
by using transpose method of Numpy.
dot():D ot product of two arrays.
sin():T hismathematicalfunctionhelpsusertocalculatetrigonometricsinefor
all x
cos():This mathematical function helps usertocalculatetrigonometriccosine
for all x
tan():T hismathematicalfunctionhelpsusertocalculatetrigonometricsinefor
all x
sum():Thesum()function returns a number, the sumof all items in an iterable.
power():Arrayelementfromfirstarrayisraisedtothepowerofelementfrom
second element(all happens element-wise).
min():The min() method returns the smallest elementof an array along an axis.
max():The max() method returns the largest element.
mean():It is used to calculate the mean/average ofinput values or data set.
average():calculates the mean (average) of the givendata set.
PROGRAM:
import numpy as np
arr=np.array([1,2,3,4,5,6])
print("Given Array : ",arr)
A3CIL202 - AI Tools and Techniques Lab 22331A0597
p rint("Type : ",type(arr))
print("No of dimensions : ",arr.ndim)
print("Data type : ",arr.dtype)
rarr=arr.reshape(2,3)
print("Reshaped array : ",rarr)
a=np.random.randint(1,10)
print("Random number : ",a)
ar=np.array([1.,2.,3.,4.,0.5])
print("Reciprocal of array : ",np.reciprocal(ar))
print("Power fo array : ",np.power(arr,2))
print("Mod 2 of array : ",np.mod(arr,2))
#Arithmetic operations
arr1=np.array([1,2,3,4])
arr2=np.array([5,6,7,8])
print("Addition : ",np.add(arr1,arr2))
print("Subtraction : ",np.subtract(arr1,arr2))
print("Multiplication : ",np.multiply(arr1,arr2))
print("Division : ",np.divide(arr1,arr2))
print("Sum of array : ",np.sum(arr1))
print("dot product : ",np.dot(arr1,2))
print("Sin func : ",np.sin(arr1))
print("cos func : ",np.cos(arr1))
print("tan func : ",np.tan(arr1))
arr3=np.array([[1,2,3],[4,5,6]])
print("Transpose matrix : ",arr3.transpose())
print("Length of array : ",len(arr1))
print("Arranged values : ",np.arange(4))
print("Min of arr2 : ",np.min(arr2))
print("Max of arr2 : ",np.max(arr2))
print("mean of arr1 : ",np.mean(arr1))
print("Average of arr2 : ",np.average(arr2))
Output:
iven Array : [1 2 3 4 5 6]
G
Type : <class 'numpy.ndarray'>
No of dimensions : 1
Data type : int64
Reshaped array : [[1 2 3]
[4 5 6]]
A3CIL202 - AI Tools and Techniques Lab 22331A0597
andom number : 3
R
Reciprocal of array : [1. 0.5 0.33333333 0.25 2. ]
Power fo array : [ 1 4 9 16 25 36]
Mod 2 of array : [1 0 1 0 1 0]
Addition : [ 6 8 10 12]
Subtraction : [-4 -4 -4 -4]
Multiplication : [ 5 12 21 32]
Division : [0.2 0.33333333 0.42857143 0.5 ]
Sum of array : 10
dot product : [2 4 6 8]
Sin func : [ 0.84147098 0.90929743 0.14112001 -0.7568025 ]
cos func : [ 0.54030231 -0.41614684 -0.9899925 -0.65364362]
tan func : [ 1.55740772 -2.18503986 -0.14254654 1.15782128]
Transpose matrix : [[1 4]
[2 5]
[3 6]]
Length of array : 4
Arranged values : [0 1 2 3]
Min of arr2 : 5
Max of arr2 : 8
mean of arr1 : 2.5
Average of arr2 : 6.5
Observation:
● N umPy arrays are different from Python lists and are optimized for numerical
operations.
● import numpy as np is used to import the numpy library into a python program and
give a short alias np i.e we can use np.array instead of numpy.array to create an array.
Exp 10 Week 4 Date
Aim :
emonstrate Pandas methods that reads and analyzes a dataset, perform tasks such as data
D
loading, display, manipulation, grouping, sorting, plotting, and handling missing values.
Libraries, Methods & Variables Used:
Pandas as pd: data manipulation and analysis librarythat provides data structures like
A3CIL202 - AI Tools and Techniques Lab 22331A0597
DataFrame and series
Methods: ariables:
V
● pd.read_csv(): Reads a CSV file into a DataFrame. dt=dataofacsv
● .head():Returnsthefirstfewrows(defaultis5)oftheDataFrameto file
give a preview of the data. uni=unique
● tail():Returns the last few rows (default is 5) of the DataFrame to names
give a preview of the data. nuni=number
● .info(): Provides a summary of the DataFrame, including the of unique
number of rows, columns, column data types, and memory usage. values
● .shape: Returns the shape of the DataFrame, which is a tuple new=dataframe
containing the number of rows and columns. after dropping
● .columns: Returns the column labels of the DataFrame. a column
● .isnull(): Detects missing values in the DataFrame. sort=dataframe
● .unique(): Finds the unique values in a column. after sorting
● sort_values():Thesort_values()method sorts the DataFrameby the based on last
specified label. name
● nunique():Pandas dataframe.nunique() function returns a series with re_name=after
the specified axis’s total number of unique observations. renaming
● rename():rename columns in dataframe by using the rename() dataset
function.
PROGRAM:
import pandas as pd
df = pd.read_csv('std.csv')
print("Content of std : \n",df.to_string())
print("Last two rows : \n",df.tail(2))
print("First Two rows : \n",df.head(2))
print("Attribute data types : \n",df.dtypes)
print("Size of std : ",df.size)
print("Shape of std : ",df.shape)
print("Unique data : ",df['Name'].unique())
print("std Column names : ",df.columns)
print("Highest Ages : \n",df.nlargest(2,'Age'))
new=df.rename(columns={'Name':'Sname'})
print("Rename the Name : \n",new.head(2))
new['clg']='MVGR'
print("Adding new column : \n",new.tail(2))
Output:
Content of std :
Name Age Gender
0 Gowtham 19 M
A3CIL202 - AI Tools and Techniques Lab 22331A0597
1 Bhanu 19 M
2 Lilli 18 F
3 Bhaskar 20 M
4 Bhanu 18 M
Last two rows :
Name Age Gender
3 Bhaskar 20 M
4 Bhanu 18 M
First Two rows :
Name Age Gender
0 Gowtham 19 M
1 Bhanu 19 M
Attribute data types :
Name object
Age int64
Gender object
dtype: object
Size of std : 15
Shape of std : (5, 3)
Unique data : ['Gowtham' 'Bhanu' 'Lilli' 'Bhaskar']
std Column names : Index(['Name', 'Age', 'Gender'], dtype='object')
Highest Ages :
Name Age Gender
3 Bhaskar 20 M
0 Gowtham 19 M
Rename the Name :
Sname Age Gender
0 Gowtham 19 M
1 Bhanu 19 M
Adding new column :
Sname Age Gender clg
3 Bhaskar 20 M MVGR
4 Bhanu 18 M MVGR
Observation:
● A ttributeError if trying to use inplace with methods that do not support it.Confirm that
the method being used supports the inplace parameter. For instance, fillna() supports
inplace=True, but other methods may not.
● By using this we can get top 999 rowspd.options.display.max_rows=999;
● astype() is used to change the datatype of a column
A3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 5
Exp 11 Week 5 Date
Aim :
Demonstrate loading of remote datasets using pandas.
Libraries, Methods & Variables Used:
andas as pd:data manipulation and analysis librarythat provides data structures like
p
DataFrame and series
sklearn.datasetsimport load_iris :load_iris is afunction from sklearn . The link provides
documentation: iris in your code will be a dictionary-likeobject. X and y will be numpy
arrays, and names has the array of possible targets as text The dataset is often used in data
mining, classification and clustering examples and to test algorithms.
load_iris()=load_iris is a function from sklearn . The link provides V ariables:
documentation: iris in your code will be a dictionary-like object iris,df_iris
.map:Themap()functionisusedtoapplyagivenfunctiontoeveryitemof
an iterable, such as a list or tuple, and returns a mapobject(whichisan
iterator).
.target_names:
.data:consist of data
.feature_names:get attribute names
enumerate():Enumerate in Python is used to loop over an iterable and
automatically provide an index for each item.
PROGRAM:
# Import necessary libraries
import pandas as pd
from sklearn.datasets import load_iris # Import load_iris
iris = load_iris()
# Convert to DataFrame
df_iris = pd.DataFrame(data=iris.data, columns=iris.feature_names)
print(df_iris.columns)
#print(df_iris.to_string())
df_iris['species'] = iris.target
print(df_iris["species"])
# Map target values to species names
df_iris['species'] = df_iris['species'].map(dict(enumerate(iris.target_names)))
print(df_iris["species"])
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Output:
Index(['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)',
'petal width (cm)'],
dtype='object')
0 0
1 0
2 0
3 0
4 0
..
145 2
146 2
147 2
148 2
149 2
Name: species, Length: 150, dtype: int64
0 setosa
1 setosa
2 setosa
3 setosa
4 setosa
...
145 virginica
146 virginica
147 virginica
148 virginica
149 virginica
Name: species, Length: 150, dtype: object
Observation:
1 . Data Loaded: The Iris dataset is loaded into a DataFrame.
2. Data Structure: The data has 150 rows and 5 columns (4 features + 1 target).
3. Data Summary: The summary statistics show the central tendency and variability
of the features.
4. Data Types: The features are numeric, while the target is categorical.
5. Data Size: The dataset has 150 samples.
Exp 12 Week 5 Date
Aim :
Demonstrate loading of toy/real world datasets using sklearn.datasets
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Libraries, Methods & Variables Used:
p andas as pd
sklearn.datasets import load_iris
load_iris() ariables:
V
.data iris_df
.feature_names
.target
.target_names
Categorical.from_codes(iris.target, iris.target_names)
PROGRAM:
f rom sklearn.datasets import load_iris
import pandas as pd
# Load the Iris dataset
iris = load_iris()
# Creating a DataFrame from the dataset for easier manipulation
iris_df = pd.DataFrame(data=iris.data, columns=iris.feature_names)
iris_df['species'] = pd.Categorical.from_codes(iris.target, iris.target_names)
# Print the first few rows of the DataFrame
print(iris_df.head())
# Print a summary of the DataFrame
print(iris_df.describe())
Output:
s epal length (cm) sepal width (cm) petal length (cm) petal width (cm) \
0 5.1 3.5 1.4 0.2
1 4.9 3.0 1.4 0.2
2 4.7 3.2 1.3 0.2
3 4.6 3.1 1.5 0.2
4 5.0 3.6 1.4 0.2
species
0 setosa
1 setosa
2 setosa
3 setosa
4 setosa
sepal length (cm) sepal width (cm) petal length (cm) \
A3CIL202 - AI Tools and Techniques Lab 22331A0597
c ount 150.000000 150.000000 150.000000
mean 5.843333 3.057333 3.758000
std 0.828066 0.435866 1.765298
min 4.300000 2.000000 1.000000
25% 5.100000 2.800000 1.600000
50% 5.800000 3.000000 4.350000
75% 6.400000 3.300000 5.100000
max 7.900000 4.400000 6.900000
petal width (cm)
c ount 150.000000
mean 1.199333
std 0.762238
min 0.100000
25% 0.300000
50% 1.300000
75% 1.800000
max 2.500000
Observation:
The dataset includes four features (measurements) for each flower:
1 . epal length (cm)
S
2. Sepal width (cm)
3. Petal length (cm)
4. Petal width (cm)
Exp 13 Week 5 Date
Aim :
emonstrate identification for missing values in a dataset and apply various cleaning
D
techniques using pandas.
Libraries, Methods & Variables Used:
umPy (numpy): NumPy isafundamentalpackageforscientificcomputinginPython.It
N
provides support for arrays, matrices, and various mathematical functions.
Pandas (pandas):Pandas is a powerful data manipulation and analysis library.Itprovides
datastructureslikeDataFrames,whicharehighlyversatileforhandlingtabulardata.Rolein
Machine Learning: You canuseittoload,clean,andpreparedatasetsformodeltraining,
making it easier to manage large datasets.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Methods: ariables:
V
● pd.read_csv(): Reads a CSV file into a DataFrame. u_c=to store
● .head(): Returns the first few rows (default is 5) of the DataFrame to used cars
give a preview of the data. dataset
● tail():Returns the last few rows (default is 5) of the DataFrame to
give a preview of the data.
● .info(): Provides a summary of the DataFrame, including the
number of rows, columns, column data types, and memory usage.
● .shape: Returns the shape of the DataFrame, which is a tuple
containing the number of rows and columns.
● .columns: Returns the column labels of the DataFrame.
● .isnull(): Detects missing values in the DataFrame.
● .unique(): Finds the unique values in a column.
● sort_values():The sort_values() methodsortstheDataFramebythe
specified label.
● nunique():Pandasdataframe.nunique()functionreturnsaserieswith
the specified axis’s total number of unique observations.
● rename():rename columns in dataframe by using the rename()
function.
● sample():Pandassample()isusedtogenerateasamplerandomrow
or column from the function caller data frame.
● describe():method in Pandas is a convenient way to get a quick
overview of your data.
● tostring():The to_string() method in Pandas is used to convert a
DataFrame or Series into a string representation.
● dataframe. isnull(). sum() returns the number of missing valuesin
the dataset.
● Theduplicated()methodreturnsaSerieswithTrueandFalsevalues
that describe which rows in the DataFrame are duplicated and not.
PROGRAM:
import pandas as pd
import numpy as np
u_c=pd.read_csv("used_cars.csv")
u_c.head(5)
u_c.columns
u_c.describe()
u_c.shape
u_c.dtypes
u_c.info()
# The inplace=True argument modifies the DataFrame directly and returns None.
# Assigning it to newu_c is unnecessary and causes the error.
# Instead, use the following:
u_c.dropna(inplace=True)
A3CIL202 - AI Tools and Techniques Lab 22331A0597
# Now u_c contains the DataFrame with dropped NA values.
u_c.isnull().sum() # Check for missing values in the modified u_c
print(u_c)
#cleaning
u_c.duplicated()
u_c.drop_duplicates(inplace=True)
u_c.isnull().sum()
u_c.fuel_type=u_c.fuel_type.fillna("unknown")
u_c.isnull().sum()
#changing datatype
u_c['model_year']=u_c['model_year'].astype(int)
Output:
class 'pandas.core.frame.DataFrame'>
<
RangeIndex: 4009 entries, 0 to 4008
Data columns (total 12 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 brand 4009 non-null object
1 model 4009 non-null object
2 model_year 4009 non-null int64
3 milage 4009 non-null object
4 fuel_type 3839 non-null object
5 engine 4009 non-null object
6 transmission 4009 non-null object
7 ext_col 4009 non-null object
8 int_col 4009 non-null object
9 accident 3896 non-null object
10 clean_title 3413 non-null object
11 price 4009 non-null object
dtypes: int64(1), object(11)
Index(['brand', 'model', 'model_year', 'milage', 'fuel_type', 'engine',
'transmission', 'ext_col', 'int_col', 'accident', 'clean_title',
'price'],
dtype='object')
memory usage: 376.0+ KB
brand model model_year milage \
0 Ford Utility Police Interceptor Base 2013 51,000 mi.
1 Hyundai Palisade SEL 2021 34,742 mi.
3 INFINITI Q50 Hybrid Sport 2015 88,900 mi.
6 Audi S3 2.0T Premium Plus 2017 84,000 mi.
7 BMW 740 iL 2001 242,000 mi.
... ... ... ... ...
4003 Mercedes-Benz E-Class E 300 4MATIC 2018 53,705 mi.
4004 Bentley Continental GT Speed 2023 714 mi.
4005 Audi S4 3.0T Premium Plus 2022 10,900 mi.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
4 007 Ford F-150 Raptor 2020 33,000 mi.
4008 BMW X3 xDrive30i 2020 43,000 mi.
f uel_type engine \
0 E85 Flex Fuel 300.0HP 3.7L V6 Cylinder Engine Flex Fuel Capa...
1 Gasoline 3.8L V6 24V GDI DOHC
3 Hybrid 354.0HP 3.5L V6 Cylinder Engine Gas/Electric H...
6 Gasoline 292.0HP 2.0L 4 Cylinder Engine Gasoline Fuel
7 Gasoline 282.0HP 4.4L 8 Cylinder Engine Gasoline Fuel
... ... ...
4003 Gasoline 241.0HP 2.0L 4 Cylinder Engine Gasoline Fuel
4004 Gasoline 6.0L W12 48V PDI DOHC Twin Turbo
4005 Gasoline 349.0HP 3.0L V6 Cylinder Engine Gasoline Fuel
4007 Gasoline 450.0HP 3.5L V6 Cylinder Engine Gasoline Fuel
4008 Gasoline 248.0HP 2.0L 4 Cylinder Engine Gasoline Fuel
transmission ext_col int_col \
0 6-Speed A/T Black Black
1 8-Speed Automatic Moonlight Cloud Gray
3 7-Speed A/T Black Black
6 6-Speed A/T Blue Black
7 A/T Green Green
... ... ... ...
4003 A/T Black Black
4004 8-Speed Automatic with Auto-Shift C / C Hotspur
4005 Transmission w/Dual Shift Mode Black Black
4007 A/T Blue Black
4008 A/T Gray Brown
accident clean_title price
0 At least 1 accident or damage reported Yes 10,300
1 At least 1 accident or damage reported Yes 38,005
3 None reported Yes 15,500
6 None reported Yes 31,000
7 None reported Yes 7,300
... ... ... ...
4003 At least 1 accident or damage reported Yes 25,900
4004 None reported Yes 3,49,950
4005 None reported Yes 53,900
4007 None reported Yes 62,999
4008 At least 1 accident or damage reported Yes 40,000
brand 0
model 0
model_year 0
milage 0
fuel_type 170
A3CIL202 - AI Tools and Techniques Lab 22331A0597
e ngine 0
transmission 0
ext_col 0
int_col 0
accident 113
clean_title 596
price 0
dtype: int64
brand 0
model 0
model_year 0
milage 0
fuel_type 0
engine 0
transmission 0
ext_col 0
int_col 0
accident 0
clean_title 0
price 0
0 False
1 False
2 False
3 False
4 False
...
4004 False
4005 False
4006 False
4007 False
4008 False
Length: 4009, dtype: bool
brand 0
model 0
model_year 0
milage 0
fuel_type 0
engine 0
transmission 0
ext_col 0
int_col 0
accident 113
clean_title 596
price 0
dtype: int64
[3269 rows x 12 columns]
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Observation:
E
● ither fillna or dropna use both can’t be used
● AttributeError if trying to use inplace with methods that do not support it.Confirm that
the method being used supports the inplace parameter. For instance, fillna() supports
inplace=True, but other methods may not.
● By using this we can get top 999 rowspd.options.display.max_rows=999;
● astype() is used to change the datatype of a column
● Data Structure: The data is stored in a pandas DataFrame, which is a
2-dimensional labeled data structure.
Exp 14 Week 5 Date
Aim :
Demonstrate LogisticRegression Supervised ML algorithm with any dataset.
Libraries, Methods & Variables Used:
pandas as pd:A powerful data manipulation and analysislibrary.
●
●sklearn.datasets:A submodule of scikit-learn thatprovides access to various
datasets.
●sklearn.model_selection:A submodule of scikit-learnthat provides utilities for
modelselection and validation.
●sklearn.linear_model:A submodule of scikit-learnthat provides linear models
forclassification and regression.
●sklearn.metrics:A submodule of scikit-learn thatprovides tools for evaluating
theperformance of models.
●load_digits():A function from sklearn.datasetsthat loads the ariables:
V
digits dataset. data,X,Y,mod
●train_test_split():A function from sklearn.model_selectionthat el,accuracy,co
splits the dataset into training and testing sets. nf_matrix,X_t
●StandardScaler():A class from sklearn.preprocessingthat rain, X_test,
standardizes features by removing the mean and scaling to y_train, y_test
unitvariance.
●fit_transform():A method that fits the scaler tothe data and then
transforms the data.
●LogisticRegression():A class from sklearn.linear_modelthat
implements logistic regression.
●fit():A method that fits the model to the trainingdata.
●predict():A method that uses the trained modelto predict the
target variable.
●confusion_matrix():A function from sklearn.metricsthat
A3CIL202 - AI Tools and Techniques Lab 22331A0597
c omputes the confusion matrix to evaluate the accuracy of
aclassification.
●classification_report():A function from sklearn.metricsthat
builds a text report showing the main classification metrics.
PROGRAM:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import confusion_matrix, accuracy_score
# Create the dataset
data = {
'Study_Hours': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
'Attendance': [50, 60, 65, 70, 80, 85, 90, 95, 100, 100],
'Pass': [0, 0, 0, 0, 1, 1, 1, 1, 1, 1]
}
# Convert to DataFrame
df = pd.DataFrame(data)
# Define independent variables (features) and dependent variable (target)
X = df[['Study_Hours', 'Attendance']]
y = df['Pass']
# Split the dataset into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
print(X_train)
# Create and fit the logistic regression model
model = LogisticRegression()
model.fit(X_train, y_train)
# Predicting on the test set
y_pred = model.predict(X_test)
print(y_pred)
# Evaluate the model
accuracy = accuracy_score(y_test, y_pred)
conf_matrix = confusion_matrix(y_test, y_pred)
print('Accuracy: ',accuracy)
print("Confusion Matrix:")
print(conf_matrix)
# Visualizing the decision boundary
A3CIL202 - AI Tools and Techniques Lab 22331A0597
p lt.scatter(df['Study_Hours'], df['Attendance'], c=df['Pass'], cmap='bwr', alpha=0.7,
edgecolor='k')
plt.xlabel('Study Hours')
plt.ylabel('Attendance Percentage')
plt.title('Logistic Regression: Pass/Fail Status')
plt.show()
Output:
tudy_Hours Attendance
S
0 1 50
7 8 95
2 3 65
9 10 100
4 5 80
3 4 70
6 7 90
[1 0 1]
Accuracy: 1.0
Confusion Matrix:
[[1 0]
[0 2]]
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Observation:
1 . Data Preparation: The code loads the simple dataset, splits the data into
features and target, and splits the data into training and testing sets.
2. Model Training: A Linear Regression model is trained on the training data.
3. Model Evaluation: The model is evaluated on the test data, and the MSE and R2
score are calculated.
4. Model Performance: The MSE and R2 score indicate the model's performance,
with lower MSE and higher R2 score indicating better performance.
5. Coefficients: The coefficients of the model are printed, indicating the contribution
of each feature to the predicted target value.
Exp 15 Week 5 Date
Aim :
Demonstrate Linear Regression Supervised ML algorithm with any dataset.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Libraries, Methods & Variables Used:
numpy as np:A fundamental package for scientificcomputing with Python, used for
●
efficient array operations.
●pandas as pd:A powerful data manipulation and analysislibrary.
●sklearn.datasets:A submodule of scikit-learn thatprovides access to various datasets.
●sklearn.model_selection:A submodule of scikit-learnthat provides utilities for model
selection and validation.
●sklearn.linear_model:A submodule of scikit-learnthat provides linear models for
classification and regression.
●sklearn.metrics:A submodule of scikit-learn thatprovides tools for evaluating the
performance of models.
● L inearRegression(): Linear regression models the V ariables:
relationship between a dependentvariable(label)andone LR,df,df1,X,y,X_train,
ormoreindependentvariables(features)byfittingalinear X_test,y_train,y_test
equation to the data.
● corr():It’s useful for checking the relationships between
different features and can help identify multicollinearity.
● bar():Useful for visualizing categorical data or the
distribution of a variable.
● drop():A method in pandas usedtodroprowsorcolumns
from a DataFrame.
● fit():This methodtrainsthelinearregressionmodelonthe
training data.
● intercept_:This attribute of the trained linear regression
model represents the intercept (constant term) of the
regression line.
● Coef_:After fitting the model, you can access the
coefficients.
● predict():This method generates predictions using the
trained linear regression model.
PROGRAM:
IMPLE LINEAR REGRESSION:
S
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
df=pd.read_csv('Salary_dataset.csv')
df1=df[['YearsExperience','Salary']]
plt.scatter(df1["YearsExperience"],df1["Salary"])
df1.corr()
plt.bar(df1["YearsExperience"],df1["Salary"])
from sklearn.model_selection import train_test_split
A3CIL202 - AI Tools and Techniques Lab 22331A0597
=df1.drop('Salary',axis=1)
X
#print(X)
y=df1.Salary
X_train,X_test,y_train,y_test=train_test_split(X,y,random_state=0,test_size=0.30)
from sklearn.linear_model import LinearRegression
LR=LinearRegression()
LR.fit(X_train,y_train)
print(LR.intercept_)
print(LR.coef_)
LR.predict(X_test)
ultiple Linear regression:
M
import pandas as pd
import numpy as np
u_c=pd.read_csv("used_cars.csv")
print(u_c.head(5))
u_c.columns
u_c.describe()
print(u_c.shape)
print(u_c.dtypes)
u_c.info()
print(u_c.isnull().sum())
u_c.dropna(inplace=True)
print(u_c.isnull().sum())
#cleaning
print(u_c.duplicated())
u_c.drop_duplicates(inplace=True)
u_c['price'] = u_c['price'].str.replace(',', '', regex=True).astype(float)
from sklearn.model_selection import train_test_split
X=u_c.drop('price',axis=1)
X=pd.get_dummies(X)
y=u_c.price
x_train,x_test,y_train,y_test=train_test_split(X,y,random_state=0,test_size=0.300)
from sklearn.linear_model import LinearRegression
LR=LinearRegression()
LR.fit(x_train,y_train)
print(LR.intercept_)
print(LR.coef_)
y_pred=LR.predict(x_test)
print(y_pred)
Output:
25842.36521257826
A3CIL202 - AI Tools and Techniques Lab 22331A0597
[9360.26128619]
[ 16]:
array([ 40818.78327049, 123189.08258899, 65155.46261459, 63283.41035735,
115700.87356004, 108212.66453108, 116636.89968866, 64219.43648597,
76387.77615802])
Multiple linear regression:
brand model model_year milage \
0 Ford Utility Police Interceptor Base 2013 51,000 mi.
1 Hyundai Palisade SEL 2021 34,742 mi.
f uel_type engine \
0 E85 Flex Fuel 300.0HP 3.7L V6 Cylinder Engine Flex Fuel Capa...
1 Gasoline 3.8L V6 24V GDI DOHC
transmission ext_col int_col \
0 6-Speed A/T Black Black
1 8-Speed Automatic Moonlight Cloud Gray
a ccident clean_title price
0 At least 1 accident or damage reported Yes 10,300
1 At least 1 accident or damage reported Yes 38,005
(4009, 12)
A3CIL202 - AI Tools and Techniques Lab 22331A0597
b rand object
model object
model_year int64
milage object
fuel_type object
engine object
transmission object
ext_col object
int_col object
accident object
clean_title object
price object
dtype: object
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4009 entries, 0 to 4008
Data columns (total 12 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 brand 4009 non-null object
1 model 4009 non-null object
2 model_year 4009 non-null int64
3 milage 4009 non-null object
4 fuel_type 3839 non-null object
5 engine 4009 non-null object
6 transmission 4009 non-null object
7 ext_col 4009 non-null object
8 int_col 4009 non-null object
9 accident 3896 non-null object
10 clean_title 3413 non-null object
11 price 4009 non-null object
dtypes: int64(1), object(11)
memory usage: 376.0+ KB
fuel_type 170
accident 113
clean_title 596
dtype: int64
0 False
1 False
3 False
6 False
A3CIL202 - AI Tools and Techniques Lab 22331A0597
7 False
...
4003 False
4004 False
4005 False
4007 False
4008 False
Length: 3269, dtype: bool
-1947876.7036248783
[ 1006.12786195 -22963.95727474 -16012.05946607 ... 624.13596452
-352.07534734 0. ]
[ 3.46184624e+04 3.94432492e+04 1.37342567e+03 7.32900762e+04
5.67404888e+04 7.01105267e+04 1.47821619e+05 6.11777246e+04
-4.13817338e+02 4.37912418e+03]
Observation:
The code loads the wine dataset and splits it into training and testing sets. This
●
allows for evaluating the model's performance on unseen data.
● A LinearRegression model is trained using the training data and makes predictions
on the test data.
ValueError if train_test_split is incorrectly applied, leading to mismatched training and
testing sets.
Exp 16 Week 5 Date
Aim :
Demonstrate various plots using matplotlib.
Libraries, Methods & Variables Used:
atplotlib import pyplot as p:This will import pyplot and give it the alias plt, which is
m
commonly used in the Python community.
b ar():This is used to create bar charts in matplotlib. =1-d data
x
show():This function displays the plot y=1-d data
stem():Thisfunctioncreatesstemplots(atypeofplotthatdisplaysdataas
lines with markers on top).
p.plot():his will plot lines or markers between points.
p.title():This sets the title of the plot
p.legend():This adds a legend to the plot
hist():This creates a histogram
A3CIL202 - AI Tools and Techniques Lab 22331A0597
g roupby(dataframe).sum().plot(kind='pie',y=’dataframe'):This is a pandas
method for plotting pie charts.
PROGRAM:
# plot
from matplotlib import pyplot as pt
x = [ 1,3,5,7,9 ]
y = [ 6,9,3,15,17]
pt.plot(x, y)
pt.show()
#bar
from matplotlib import pyplot as pt
x = [ 1,3,5,7,9 ]
y = [ 6,9,3,15,17]
pt.bar(x, y)
pt.show()
#stem
from matplotlib import pyplot as pt
x = [ 1,3,5,7,9 ]
y = [ 6,9,3,15,17]
pt.stem(x, y)
pt.show()
#scatter
from matplotlib import pyplot as pt
x = [ 1,3,5,7,9 ]
y = [ 6,9,3,15,17]
pt.scatter(x, y)
pt.show()
#legend
from matplotlib import pyplot as pt
x = [ 1,3,5,7,9 ]
y = [ 6,9,3,15,17]
pt.plot(x)
pt.plot(y )
pt.legend(['Weight','Height'])
pt.show()
#pie chart
import pandas as pd;
df=pd.DataFrame({'party':['ycp','tdp','janasena','bjp'],
A3CIL202 - AI Tools and Techniques Lab 22331A0597
'result':[11,138,21,5]});
df.groupby(['party']).sum().plot(kind='pie',y='result');
Output:
A3CIL202 - AI Tools and Techniques Lab 22331A0597
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Observation:
e labeled the x-axis and y-axis and plt.show() displays the line plotting for the given
W
input on the window
Creates a scatter plot with plt.scatter(), using x and y as the coordinates for the points.
The points are marked with dots (.) and colored in blue.
● And we labeled the x-axis and y-axis.
● plt.show() displays the scatter plot,barplot,stem on the window
● Creates a bar plot with plt.bar(), using x for the bar labels and y for the bar heights.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 6
Exp 17 Week 6 Date
Aim :
Using hand written digits dataset(or any other dataset) classify using SVM classifier.
Libraries, Methods & Variables Used:
s klearn import datasets,svm,metrics:This imports thedatasets module, which provides
various datasets like the Iris dataset, digits dataset,breast cancer etc., useful for machine
learning practice and experiments.
sklearn.model_selection import train_test_split:Thisimports the train_test_split function,
which is used to split a dataset into training and testing subsets.
import matplotlib.pyplot as plt:This imports matplotlib.pyplotas plt for creating
visualizations such as line plots, histograms, and scatter plots.
load_breast_cancer():This loads the breast cancerdataset from ariables:
V
sklearn.datasets, which is commonly used for binary classification tasks. cancer,clf,pre,acc,x
feature_names:This attribute gives the names of thefeatures in the _train,x_test,y_train
dataset (e.g., mean radius, mean texture, etc.). ,y_test
target:This provides the labels (target values) ofthe dataset, where 0
indicates benign and 1 indicates malignant tumors.
Shape:You can access the shape of the data (numberof samples and
number of features) using the .shape attribute.
accuracy_score(y_test,pre):This function computesthe accuracy,
which is the ratio of correctly predicted labels to total labels.
PROGRAM:
f rom sklearn import datasets,svm,metrics
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
import matplotlib.pyplot as plt
cancer=datasets.load_breast_cancer()
data,target=cancer.data,cancer.target
print("Data shape : ",data.shape)
print("Target shape : ",target.shape)
print("Feature name : ",cancer.feature_names)
clf=svm.SVC(gamma=0.001)
A3CIL202 - AI Tools and Techniques Lab 22331A0597
x _train,x_test,y_train,y_test=train_test_split(data,target,test_size=0.5,shuffle=True,random_sta
te=42)
clf.fit(x_train,y_train)
pre=clf.predict(x_test)
print("classification report for classifier",clf)
print(metrics.classification_report(y_test,pre))
acc=accuracy_score(y_test,pre)
print("Accuracy :",f"{acc*100:.2f}")
Output:
ata shape : (569, 30)
D
Target shape : (569,)
Feature name : ['mean radius' 'mean texture' 'mean perimeter' 'mean area'
'mean smoothness' 'mean compactness' 'mean concavity'
'mean concave points' 'mean symmetry' 'mean fractal dimension'
'radius error' 'texture error' 'perimeter error' 'area error'
'smoothness error' 'compactness error' 'concavity error'
'concave points error' 'symmetry error' 'fractal dimension error'
'worst radius' 'worst texture' 'worst perimeter' 'worst area'
'worst smoothness' 'worst compactness' 'worst concavity'
'worst concave points' 'worst symmetry' 'worst fractal dimension']
classification report for classifier SVC(gamma=0.001)
precision recall f1-score support
0 0.86 0.97 0.91 98
1 0.98 0.92 0.95 187
a ccuracy 0.94 285
macro avg 0.92 0.94 0.93 285
weighted avg 0.94 0.94 0.94 285
Accuracy : 93.68
Observation:
S
● VM is best suited for classification tasks.
● The primary objective of the SVM algorithm is to identify the optimal hyperplane in
an N-dimensional space that can effectively separate data points into different classes
in the feature space.
● The algorithm ensures that the margin between the closest points of different classes,
known as support vectors, is maximized.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
● T
he dimension of thehyperplanedepends on the number of features. For instance, if
there are two input features, the hyperplane is simply a line, and if there are three input
features, the hyperplane becomes a 2-D plane.
Exp 18 Week 6 Date
Aim :
Using any dataset of your choice classify the dataset using K-MEANS classifier.
Libraries, Methods & Variables Used:
s klearn.datasets import make_blobs
sklearn.cluster import KMeans
matplotlib.pyplot as plt
ake_blobs():Generates synthetic data points groupedinto clusters, useful
m ariables:
V
for testing clustering algorithms like KMeans. n_samples,X,
KMeans():KMeans is an unsupervised machine learningalgorithm that y,random_stat
e,labels,cluste
groups data points into a predefined number of clusters based on feature
r_centers
similarity.
fit(X):Fits the KMeans model on the dataset X.
Kmeans.labels_:After fitting the KMeans model, labels_gives the cluster
assignment (or label) for each data point.
Cluster_centers_:This attribute provides the coordinatesof the cluster
centers.
Scatter:This is the scatter() function from matplotlib.pyplotused to create
scatter plots.
PROGRAM:
f rom sklearn.datasets import make_blobs
from sklearn.cluster import KMeans
import matplotlib.pyplot as plt
n_samples = 300
random_state = 170
X, y = make_blobs(n_samples=n_samples, random_state=random_state)
kmeans = KMeans(n_clusters=3, random_state=random_state, n_init="auto")
# Fit the model to the data
kmeans.fit(X)
A3CIL202 - AI Tools and Techniques Lab 22331A0597
# Get the cluster labels assigned to each data point
labels = kmeans.labels_
# Get the coordinates of the cluster centers
cluster_centers = kmeans.cluster_centers_
# Plot the data points and cluster centers
plt.scatter(X[:, 0], X[:, 1], c=labels, s=50, cmap='viridis')
plt.scatter(cluster_centers[:, 0], cluster_centers[:, 1], c='white', s=200, alpha=0.75)
plt.xlabel('Feature 1')
plt.ylabel('Feature 2')
plt.title('K-Means Clustering')
plt.show()
Output:
Observation:
● he choice of the number of clusters (K) is crucial for KMeans clustering.
T
● If K is too small, you might under-cluster your data (combine distinct clusters).
● If K is too large, you might overfit (create unnecessary clusters).
● Visualizing the results of KMeans clustering can help in understanding the cluster
structure.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 7
Exp 19 Week 7 Date
Aim :
emonstrate different image processing
D operations on colour images.
1. Reading an image and displaying the image.
2. Reading a color image and converting it into grayscale.
3. Program to Rotate an image by 90/180 degrees clockwise using OpenCV.
4. Program to write and display an image using OpenCV.
5. Program for extracting image properties using OpenCV.
6. Program for accessing and modifying pixel values.
7. Program for color mixing in OpenCV.
Libraries, Methods & Variables Used:
c v2:OpenCV (Open Source Computer Vision Library) is an open-source computer vision
and machine learning software library.Image processing,Computer vision: Real-time image
processing are tasks.
numpy as np:NumPy (Numerical Python) is a fundamental package for scientific computing
in Python, providing support for arrays and matrices.
matplotlib.pyplot as plt:
Matplotlib is a plotting library for Python and its numerical mathematics extension NumPy
c v2.imread(): Reads an image from the specified ariables:
V
file. image,color_image,gray_image,rotat
cv2.cvtColor(): Converts an image from one color ed_90,rotated_180,saved_image,hei
space to another. ght,width,channels,red_image,blue_i
cv2.rotate():Rotates an image by specified angles. mage,mixed_image
cv2.imwrite():Writes an image to a specified file.
image.shape: Returns a tuple containing the
dimensions of the image: (height, width, channels).
image.size: Returns the total number of pixels
(elements) in the image array (height × width ×
channels).
cv2.addWeighted(): Blends two images together
based on specified weights.
plt.show():Part of Matplotlib for displaying images.
PROGRAM:
1 )
import cv2
import matplotlib.pyplot as plt
A3CIL202 - AI Tools and Techniques Lab 22331A0597
# Read the image
image = cv2.imread('dog.jpg') # Replace 'image.jpg' with your image file
# Display the image
plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
plt.title('Original Image')
plt.axis('off')
plt.show()
2)
# Read the color image
color_image = cv2.imread('dog.jpg') # Replace 'image.jpg' with your image file
# Convert the image to grayscale
gray_image = cv2.cvtColor(color_image, cv2.COLOR_BGR2GRAY)
# Display the grayscale image
plt.imshow(gray_image, cmap='gray')
plt.title('Grayscale Image')
plt.axis('off')
plt.show()
3)
# Read the image
image = cv2.imread('dog.jpg') # Replace 'image.jpg' with your image file
# Rotate the image by 90 degrees clockwise
rotated_90 = cv2.rotate(image, cv2.ROTATE_90_CLOCKWISE)
# Rotate the image by 180 degrees
rotated_180 = cv2.rotate(image, cv2.ROTATE_180)
# Display the rotated images
plt.figure(figsize=(10, 5))
plt.subplot(1, 2, 1)
plt.imshow(cv2.cvtColor(rotated_90, cv2.COLOR_BGR2RGB))
plt.title('Rotated 90 Degrees Clockwise')
plt.axis('off')
plt.subplot(1, 2, 2)
plt.imshow(cv2.cvtColor(rotated_180, cv2.COLOR_BGR2RGB))
plt.title('Rotated 180 Degrees')
plt.axis('off')
plt.show()
4)
# Read the image
image = cv2.imread('dog.jpg') # Replace 'image.jpg' with your image file
# Write the image to a new file
cv2.imwrite('output_image.jpg', image)
# Display the saved image
saved_image = cv2.imread('output_image.jpg')
plt.imshow(cv2.cvtColor(saved_image, cv2.COLOR_BGR2RGB))
plt.title('Saved Image')
plt.axis('off')
plt.show()
A3CIL202 - AI Tools and Techniques Lab 22331A0597
5 )
# Read the image
image = cv2.imread('dog.jpg') # Replace 'image.jpg' with your image file
# Get image properties
height, width, channels = image.shape
size = image.size
# Print image properties
print(f"Image Height: {height}")
print(f"Image Width: {width}")
print(f"Number of Channels: {channels}")
print(f"Image Size (in bytes): {size}")
6)
# Read the image
image = cv2.imread('dog.jpg') # Replace 'image.jpg' with your image file
# Accessing a pixel value at (x=100, y=50)
pixel_value = image[50, 100] # BGR format
print(f"Original Pixel Value at (100, 50): {pixel_value}")
# Modify the pixel value to a new color (e.g., pure red)
image[50, 100] = [0, 0, 255] # BGR format for red
# Display the modified image
plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
plt.title('Modified Image')
plt.axis('off')
plt.show()
7)
import numpy as np
# Create two solid color images (Red and Blue)
red_image = np.zeros((300, 300, 3), dtype=np.uint8)
red_image[:] = [0, 0, 255] # BGR for red
blue_image = np.zeros((300, 300, 3), dtype=np.uint8)
blue_image[:] = [255, 0, 0] # BGR for blue
# Mix the images (50% each)
mixed_image = cv2.addWeighted(red_image, 0.5, blue_image, 0.5, 0)
# Display the mixed image
plt.imshow(cv2.cvtColor(mixed_image, cv2.COLOR_BGR2RGB))
plt.title('Color Mixing: Red and Blue')
plt.axis('off')
plt.show()
Output:
A3CIL202 - AI Tools and Techniques Lab 22331A0597
I mage Height: 1280
Image Width: 853
Number of Channels: 3
Image Size (in bytes): 3275520
Original Pixel Value at (100, 50): [34 28 5]
Observation:
I mage Reading and Displaying:You can load and visualizeimages using cv2.imread() and
matplotlib.pyplot.
Grayscale Conversion:Color images can be convertedto grayscale for simpler analysis.
Image Rotation:OpenCV provides straightforward functionsfor rotating images by specific
A3CIL202 - AI Tools and Techniques Lab 22331A0597
d egrees.
Image Writing:You can save images to disk using cv2.imwrite().
Image Properties:You can extract and print propertieslike dimensions and size of an image.
Pixel Access and Modification:You can directly accessand modify pixel values using
NumPy array indexing.
Color Mixing:Use cv2.addWeighted() to blend two colorsor images.
Exp 20 Week 7 Date
Aim :
Demonstrate Face and Eye detection using Haar cascade classifier in OpenCV.
Libraries, Methods & Variables Used:
c v2:OpenCV (Open Source Computer Vision Library) isan open-source computer vision
and machine learning software library.Image processing,Computer vision: Real-time image
processing are tasks.
matplotlib.pyplot as plt:Matplotlib is a plottinglibrary for Python and its numerical
mathematics extension NumPy.
c v2.CascadeClassifier:It’s mainly used for tasks suchas face detection, ariables:
V
eye detection, or other object detection (e.g., pedestrians, cars). face_cascade,
cv2.cvtColor:Converts an image from one color spaceto another. eye_cascade,i
face_cascade.detectMultiScale():Detects objects (inthis case, faces) in an mage,gray_im
image using the Haar Cascade Classifier. age,faces,roi_i
cv2.rectangle():Draws a rectangle on an image. mage,eyes
eye_cascade.detectMultiScale:Similar to
face_cascade.detectMultiScale(), but used for detecting eyes in an image
imshow():Displays an image in a window.
PROGRAM:
import cv2
import matplotlib.pyplot as plt
# Step 1: Load the Haar cascades for face and eye detection
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades +
'haarcascade_frontalface_default.xml')
eye_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_eye.xml') # Fixed:
Assigned to eye_cascade instead of face_cascade
# Step 2: Load an image
image = cv2.imread('group1.jpg')
A3CIL202 - AI Tools and Techniques Lab 22331A0597
g ray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) # Convert to grayscale
# Step 3: Detect faces in the image
faces = face_cascade.detectMultiScale(gray_image, scaleFactor=1.3, minNeighbors=5)
# Step 4: Draw rectangles around faces and detect eyes within each face
for (x, y, w, h) in faces:
cv2.rectangle(image, (x, y), (x + w, y + h), (255, 0, 0), 2) # Blue rectangle for face
# Define the region of interest (ROI) in the grayscale image for the face
roi_gray = gray_image[y:y + h, x:x + w]
roi_color = image[y:y + h, x:x + w]
# Step 5: Detect eyes within the face ROI
eyes = eye_cascade.detectMultiScale(roi_gray) # Now eye_cascade is defined and can be used
for (ex, ey, ew, eh) in eyes:
cv2.rectangle(roi_color, (ex, ey), (ex + ew, ey + eh), (0, 255, 0), 2) # Green rectangles for eyes
# Step 6: Display the result using matplotlib
plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
plt.title('Face and Eye Detection')
plt.axis('off')
plt.show()
Output:
Observation:
aar Cascade Efficiency:Haar cascades are fast andefficient for detecting objects,
H
especially faces. This method can detect faces in real-time if applied to video streams.
Detection Parameters:The choice of scaleFactor, minNeighbors,and minSize parameters
affects the performance and accuracy of the detection.
Single Face Detection:This code is suitable for detectinga single face in a well-lit
environment.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 8
Exp 21 Week 8 Date
Aim :
perform corner detection algorithms harris corner and shi thomshi corner detection algorithms.
Libraries, Methods & Variables Used:
c v2:OpenCV (Open Source Computer Vision Library)is an open-source computer vision
and machine learning software library.
numpy as np:NumPy (Numerical Python) is a fundamentalpackage for scientific
computing in Python, providing support for arrays and matrices.
matplotlib.pyplot as plt:
Matplotlib is a plotting library for Python and its numerical mathematics extension NumPy.
cv2.imread():Reads an image from a file and returnsit as a NumPy array. ariables:
V
cv2.cvtColor:Converts an image from one color spaceto another. image,gray_i
cv2.cornerHarris():Detects corners in an image usingthe Harris corner mage,dst,thres
detection algorithm. hold,corners,r
Cv2.dilate:Applies a dilation operation to an image,which is typically esult,image,x,
used to expand the white regions of an image or enhance features like y
edges or corners.
dst.max():This is not an OpenCV method but a NumPymethod. If dst is a
NumPy array, dst.max() returns the maximum value in the array.
cv2.addWeighted():Blends two images by adding them together with
specific weights.
imshow():Displays an image in a window.
PROGRAM:
import cv2
import numpy as np
import matplotlib.pyplot as plt
# Step 1: Load the image
image = cv2.imread('geeks.jpg')
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# Step 2: Convert to float32
gray_image = np.float32(gray_image
# Step 3: Harris corner detection
dst = cv2.cornerHarris(gray_image, blockSize=2, ksize=3, k=0.04)
# Step 4: Dilate the result to mark the corners
dst = cv2.dilate(dst, None)
# Step 5: Thresholding to identify strong corners
threshold = 0.01 * dst.max()
corners = np.zeros_like(image)
A3CIL202 - AI Tools and Techniques Lab 22331A0597
c orners[dst > threshold] = [0, 0, 255] # Mark corners in red
# Step 6: Display the original image with corners marked
result = cv2.addWeighted(image, 0.7, corners, 0.3, 0)
plt.imshow(cv2.cvtColor(result, cv2.COLOR_BGR2RGB))
plt.title('Harris Corners')
plt.axis('off')
plt.show()
Shi thomasi
import cv2
import numpy as np
import matplotlib.pyplot as plt
# Step 1: Load the image
image = cv2.imread('geeks.jpg')
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# Step 2: Shi-Tomasi corner detection
corners = cv2.goodFeaturesToTrack(gray_image, maxCorners=100, qualityLevel=0.01,
minDistance=10)
# Step 3: Draw the corners on the image
for corner in corners:
x, y = corner.ravel()
# Convert x and y to integers before passing to cv2.circle
x = int(x)
y = int(y)
cv2.circle(image, (x, y), 3, (0, 255, 0), -1)
# Step 4: Display the image with corners marked
plt.imshow(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
plt.title('Shi-Tomasi Corners')
plt.axis('off')
plt.show()
Output:
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Observation:
threshold is applied to retain only the strongest corners. Points with values in the corner
A
response image (dst) greater than 1% of the maximum response (0.01 * dst.max()) are
considered corners.
These corners are marked in red ([0, 0, 255]).
Harris Corner Detectionis sensitive to changes inthe image, making it effective in detecting
sharp edges or corner-like features.
Shi-Tomasi Corner Detection:
● Selects only the strongest corners, which results in fewer, more distinct corners.
● More computationally efficient than Harris and performs better in noisy environments.
Exp 22 Week 8 Date
Aim :
redict whether two images are similar using SIFT for feature detection and BF matcher
P
algorithms.
Libraries, Methods & Variables Used:
A3CIL202 - AI Tools and Techniques Lab 22331A0597
cv2:OpenCV (Open Source Computer Vision Library) is an open-source computer vision
and machine learning software library.
atplotlib.pyplot as plt :Matplotlib is a plotting library for Python and its numerical
m
mathematics extension NumPy.
r ead_image(path1, path2):Function to read the images by taking their ariables:
V
paths img1,img2,sift,kp1,
convert_to_grayscale(pic1, pic2):Function to convert images from kp2,des1,des2,bf,m
RGB to grayscale atches,matched_im
cv2.cvtColor():convert color from rgb to grayscale and vice versa g,flags,num_matche
detectAndCompute() is used in OpenCV to detect keypoints and s,img1_path,img2_
compute descriptors in a single step. path
BF_FeatureMatcher(des1, des2): Function to find best detected
features using brute force matcher
cv2.BFMatcher: Brute-Force Matcher (BFMatcher) is a simple, yet
powerful algorithm to match descriptors between two images.
title():Adds a title to the plot.
axis():Turns the axis on or off for the plot.
show():Displays the current plot.
PROGRAM:
import cv2
import matplotlib.pyplot as plt
def detect_and_match(img1_path, img2_path):
# Read images
img1 = cv2.imread(img1_path, cv2.IMREAD_GRAYSCALE)
img2 = cv2.imread(img2_path, cv2.IMREAD_GRAYSCALE)
# Initialize SIFT detector
sift = cv2.SIFT_create()
# Detect keypoints and descriptors with SIFT
kp1, des1 = sift.detectAndCompute(img1, None)
kp2, des2 = sift.detectAndCompute(img2, None)
# BFMatcher with default params (L2 norm)
bf = cv2.BFMatcher(cv2.NORM_L2, crossCheck=True)
# Match descriptors
matches = bf.match(des1, des2)
# Sort them in the order of their distances
matches = sorted(matches, key=lambda x: x.distance)
# Draw the matches
matched_img = cv2.drawMatches(img1, kp1, img2, kp2, matches[:50], None,
flags=cv2.DrawMatchesFlags_NOT_DRAW_SINGLE_POINTS)
# Display the image with matches
plt.figure(figsize=(12, 6))
plt.imshow(matched_img, cmap='gray')
plt.title(f'Number of Matches: {len(matches)}')
A3CIL202 - AI Tools and Techniques Lab 22331A0597
p lt.axis('off')
plt.show()
# Calculate the number of good matches
num_matches = len(matches)
# Define a threshold for determining similarity
similarity_threshold = 10 # Adjust this based on experimentation
if num_matches > similarity_threshold:
return f"Images are similar with {num_matches} good matches."
else:
return f"Images are not similar with only {num_matches} matches."
# Example usage
img1_path = 'dog.jpg'
img2_path = 'dog1.jpg'
result = detect_and_match(img1_path, img2_path)
print(result)
Output:
Images are similar with 243 good matches.
Observation:
● T he Brute-Force Matcher (BFMatcher) in OpenCV is a straightforward and effective
algorithm for matching feature descriptors between images
● BFMatcher compares each descriptor from the first set to every descriptor in the
second set. It computes the distance between each pair of descriptors and identifies the
best matches.
A3CIL202 - AI Tools and Techniques Lab 22331A0597
WEEK - 9
Exp 23 Week 9 Date
Aim :
Demonstrate implementation of a chatbot in botpress.
Description:
chatbotis an artificial intelligence (AI) softwarethat can simulate a conversation with a
A
user in natural language through messaging applications, websites, mobile apps, or via the
telephone. It is designed to automate interactions and provide instant responses to user
inquiries, helping to streamline customer service, enhance user engagement, and improve
productivity.
Output:
A3CIL202 - AI Tools and Techniques Lab 22331A0597
Observation:
The distinction between rule-based and AI-powered chatbots helps readers understand the
●
different levels of chatbot capabilities, from handling simple to complex queries.
● They enhance user experience through personalised interactions and continuous learning
while being flexible and easy to manage.
● The platform's visual tools and modular design make it accessible for users with varying
technical skills.
You might also like
- Google Coding Round Test Top 10 Questions and SolutionNo ratings yetGoogle Coding Round Test Top 10 Questions and Solution9 pages
- AIML Lab - SRMTRPEC - Observation - 20 Feb 24No ratings yetAIML Lab - SRMTRPEC - Observation - 20 Feb 2473 pages
- Data Structures and Algorithms - Unit 3 & 4 SolutionsNo ratings yetData Structures and Algorithms - Unit 3 & 4 Solutions31 pages
- CD3291 Data Structurres and Algorithm Lab ManualNo ratings yetCD3291 Data Structurres and Algorithm Lab Manual84 pages
- ADA Practical-9: 1. Depth First Search:-CODENo ratings yetADA Practical-9: 1. Depth First Search:-CODE3 pages
- Artificial Intelligence and Machine Learning FundamentalsNo ratings yetArtificial Intelligence and Machine Learning Fundamentals23 pages
- COMSATS University Islamabad, Lahore Campus: Lab Assignment 1 Statement - FALL 2021No ratings yetCOMSATS University Islamabad, Lahore Campus: Lab Assignment 1 Statement - FALL 20213 pages
- Download Full Point Process Calculus in Time and Space 1st Edition Pierre Brémaud PDF All Chapters100% (1)Download Full Point Process Calculus in Time and Space 1st Edition Pierre Brémaud PDF All Chapters55 pages
- 2k2901k90rjg920jr90jf90j093j0e9jgr0j0j40j09j0jr01dj09k09kt03kd09k094k02k0kg0i!o$ "!) ? "!i$?i? ) KC? "!?K?" DK? !KD? K ) K) KF ? K K DK FK"K K FK "K KNo ratings yet2k2901k90rjg920jr90jf90j093j0e9jgr0j0j40j09j0jr01dj09k09kt03kd09k094k02k0kg0i!o$ "!) ? "!i$?i? ) KC? "!?K?" DK? !KD? K ) K) KF ? K K DK FK"K K FK "K K114 pages
- Date Time Code No. Subject of Examinations: 9.30 AM TO 12.30 PMNo ratings yetDate Time Code No. Subject of Examinations: 9.30 AM TO 12.30 PM22 pages
- 6440 Board Diploma Examination, (C-16) : WWW - Manaresults.co - inNo ratings yet6440 Board Diploma Examination, (C-16) : WWW - Manaresults.co - in3 pages
- VCE Corporate Skill Improvement Program: Name Email-ID College Name100% (1)VCE Corporate Skill Improvement Program: Name Email-ID College Name5 pages
- CENG250 Slides Chapter 3 Part 2 - Implicants and Simplification in K-Maps - With NarrationNo ratings yetCENG250 Slides Chapter 3 Part 2 - Implicants and Simplification in K-Maps - With Narration16 pages
- Introduction - 2017 - Rhythmic Stimulation Procedures in Neuromodulation PDFNo ratings yetIntroduction - 2017 - Rhythmic Stimulation Procedures in Neuromodulation PDF6 pages
- Principles and Methods of Food Processing & Preservation100% (1)Principles and Methods of Food Processing & Preservation34 pages
