Muhammad al-xozazmiy nomidagi toshkent axborot texnologiyalari universiteti mashinali o’qitish fanidan Amaliy ish Bajardi: Boymurodov Zuhriddin Tekshirdi

OʻZBEKISTON RESPUBLIKASI AXBOROT TEXNOLOGIYALARI VA KOMMUNIKATSIYALARINI RIVOJLANTIRISH VAZIRLIGI
MUHAMMAD AL-XOZAZMIY NOMIDAGI TOSHKENT AXBOROT TEXNOLOGIYALARI UNIVERSITETI

Mashinali o’qitish fanidan
Amaliy ish

Bajardi: Boymurodov Zuhriddin
Tekshirdi: Qo’chqorov Muslimjon


TOSHKENT – 2023
import random
import matplotlib.pyplot as plt

def mean(data):

return sum(data) / len(data)

def gradient(x, y, w):

grad = 0
for i in range(len(x)):
grad += (y[i] - w * x[i]) * x[i]
grad = -2 * grad / len(x)
return grad

def loss(x, y, w):

l = 0
for i in range(len(x)):
y_pred = w * x[i]
l += (y[i] - y_pred) ** 2
l = l / len(x)
return l

# Generate random data points

random.seed(42)
x = [random.random() for _ in range(100)]
y = [3 * x[i] + 2 + 0.1 * random.random() for i in range(100)]

# Initialize weight and learning rate

w = 0
lr = 0.01

# Perform gradient descent

n_iterations = 100
losses, weights = [], []
for i in range(n_iterations):
grad = gradient(x, y, w)
w -= lr * grad
l = loss(x, y, w)
losses.append(l)
weights.append(w)

print(f"Loss qiymatlari: {losses}")

# Output final weight
print(f"Weight qiymatlari: {weights}")

# Plot loss and weight values

plt.subplot(2, 1, 1)
plt.plot(losses)
plt.xlabel("Iteration")
plt.ylabel("Loss")
plt.subplot(1, 1, 1)
plt.plot(weights)
plt.xlabel("Iteration")
plt.ylabel("Weight")
plt.show()

import random

import matplotlib.pyplot as plt

# Generate random data points

import numpy as np

random.seed(42)

X = [[random.random()] for _ in range(10)]
y = [[2*x[0]**2 + 3*x[0] + random.gauss(0, 1)] for x in X]

# Initialize coefficients and learning rate

w1 = random.random()
w2 = random.random()
lr = 0.1

# Perform gradient descent

n_iterations = 100
losses = []
W1 = []
W2 = []
for i in range(n_iterations):
y_pred = [[w1*x[0]**2 + w2*x[0]] for x in X]
loss = sum([(y[i][0] - y_pred[i][0])**2 for i in range(len(X))]) / len(X)
gradient_w1 = -2*sum([(y[i][0] - y_pred[i][0])*X[i][0]**2 for i in range(len(X))]) / len(X)
gradient_w2 = -2*sum([(y[i][0] - y_pred[i][0])*X[i][0] for i in range(len(X))]) / len(X)
w1 -= lr*gradient_w1
w2 -= lr*gradient_w2
W1.append(w1)
W2.append(w2)
losses.append(loss)

# Print coefficients

print(f"Loss qiymatlari {losses}")
print(f" w1 qiymatlari {W1}")
print(f" w2 qiymatlari {W2}")
# Plot loss graph
plt.plot(range(n_iterations), losses)
plt.title("Loss")
plt.xlabel("Iteration")
plt.ylabel("Loss")
plt.show()