Ass3.py

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from PIL import Image
import time

# Part 1: Pandas

# Question 1: Filter rows with CPI > 60 and calculate statistics
# Create the DataFrame
data = {'Std Name': ['Alice', 'Bob', 'Charlie', 'David'],
        'Roll no': [1, 2, 3, 4],
        'CPI': [75.5, 45.3, 68.7, 82.1]}
df = pd.DataFrame(data)

# Filter rows where CPI > 60
filtered_df = df[df['CPI'] > 60]

# Calculate statistics
mean_cpi = df['CPI'].mean()
median_cpi = df['CPI'].median()
std_cpi = df['CPI'].std()

print("Filtered DataFrame:")
print(filtered_df)
print(f"Mean: {mean_cpi}, Median: {median_cpi}, Std Dev: {std_cpi}")

# Question 2: Total books per subject
# Create the DataFrame
data = {'Subjects': ['Mathematics', 'Physics', 'Chemistry', 'Mathematics', 'Physics'],
        'Authors': ['R.D.Sharma', 'H.C.Verma', 'O.P.Tandon', 'S.K.Goyal', 'Resnick'],
        'No of Books': [15, 12, 8, 10, 5]}
df_books = pd.DataFrame(data)

# Calculate total books per subject
total_books = df_books.groupby('Subjects')['No of Books'].sum()
print("\nTotal Books per Subject:")
print(total_books)

# Question 3: Image pixel data to CSV
# Read the image and convert it to a NumPy array
# Uncomment the following lines to test with an actual image file
# image = Image.open('example.jpg').convert('L')  # Convert to grayscale
# image_data = np.array(image)

# Save to CSV
# pd.DataFrame(image_data).to_csv('image_data.csv', index=False)

# Load the CSV back, excluding the last row and last column
# df_image = pd.read_csv('image_data.csv')
# df_image = df_image.iloc[:-1, :-1]
# print("\nImage Data from CSV:")
# print(df_image)

# Question 4: Clean DataFrame
dirty_data = pd.DataFrame({'Name': ['A', 'B', 'C', 'D'],
                           'Age': [16, 25, 17, 30],
                           'Salary': [50000, -20000, 40000, 60000]})

# Remove rows where Age < 18 or Salary < 0
cleaned_data = dirty_data[(dirty_data['Age'] >= 18) & (dirty_data['Salary'] >= 0)]
print("\nCleaned DataFrame:")
print(cleaned_data)

# Question 5: Extract hour from timestamp
time_data = pd.DataFrame({'TimeStamp': pd.date_range('2025-01-01', periods=5, freq='H')})
time_data['Hour'] = time_data['TimeStamp'].dt.hour
print("\nTime Data with Hour Column:")
print(time_data)

# Part 2: NumPy Application

# Question 1: Matrix multiplication and speedup
# Create random matrices
P = np.random.rand(10, 10)
Q = np.random.rand(10, 10)

# Using nested loops
start = time.time()
result = np.zeros((P.shape[0], Q.shape[1]))
for i in range(P.shape[0]):
    for j in range(Q.shape[1]):
        result[i, j] = np.dot(P[i, :], Q[:, j])
t1 = time.time() - start

# Using NumPy vectorized operations
start = time.time()
result_np = P @ Q.T
t2 = time.time() - start

# Speedup
speedup = t1 / t2
print(f"\nt1: {t1:.4f}, t2: {t2:.4f}, Speedup: {speedup:.2f}")

# Question 2: Extract second column and last row
arr = np.random.rand(5, 5)
second_column = arr[:, 1]
last_row = arr[-1, :]
print("\nSecond Column:", second_column)
print("Last Row:", last_row)

# Question 3: Frequency of unique numbers
array = np.random.randint(1, 10, size=(5, 6))
frequency = np.unique(array, return_counts=True)
print("\nFrequency of Unique Numbers:")
print(dict(zip(frequency[0], frequency[1])))

# Question 4: Solve linear equations
n = 5
A = np.random.rand(n, n)
b = np.random.rand(n)
x = np.linalg.solve(A, b)
print("\nSolution to Linear Equations:")
print(x)

# Question 5: Frequency of score 85
scores = np.array([85, 92, 75, 85, 90, 92, 85, 75, 85, 92, 75, 85, 90, 92, 85, 75, 85, 92])
frequency_85 = np.sum(scores == 85)
print(f"\nFrequency of Score 85: {frequency_85}")

# Question 6: Eigenvalues and eigenvectors
A = np.array([[2, -1, 0],
              [-1, 2, -1],
              [0, -1, 2]])
eigenvalues, eigenvectors = np.linalg.eig(A)
print("\nEigenvalues:", eigenvalues)
print("Eigenvectors:\n", eigenvectors)

# Question 7: Standardize features
X = np.array([[1, 2, 3],
              [4, 5, 6],
              [7, 8, 9]])
mean = X.mean(axis=0)
std = X.std(axis=0)
X_standardized = (X - mean) / std
print("\nStandardized Features:\n", X_standardized)

# Question 8: Singular Value Decomposition
M = np.array([[1, 2],
              [3, 4],
              [5, 6]])
U, S, Vt = np.linalg.svd(M)
print("\nSVD Results:")
print("U:\n", U)
print("S:", S)
print("Vt:\n", Vt)

# Part 3: Data Visualization

# Question 1: Bar chart for total books per subject
total_books.plot(kind='bar', color=['blue', 'orange', 'green'])
plt.title("Total Books per Subject")
plt.xlabel("Subject")
plt.ylabel("Number of Books")
plt.show()

# Question 2: Histogram for salary distribution
salaries = [55, 62, 65, 58, 70, 55, 60, 68, 63, 55, 62, 150]
plt.hist(salaries, bins=6, color='purple', edgecolor='black')
plt.title("Distribution of Employee Salaries")
plt.xlabel("Salary (in thousands of dollars)")
plt.ylabel("Frequency")
plt.show()