|
Guruh: 214-21. Bajardi: Yuldoshov Dilshodbek
|
| Sana | 22.04.2024 | | Hajmi | 41.91 Kb. | | #204529 |
Bog'liq Mashinali o\'qitishga kirish amaliy Yuzi, Документ Microsoft Word, Arduino led1
MUHAMMAD AL-XORAZMIY NOMIDAGI
TOSHKENT AXBOROT TEXNOLOGIYALARI
UNIVERSITETI
1-Topshiriq
Guruh: 214-21.
Bajardi: Yuldoshov Dilshodbek
Tekshirdi: Abdug‘aniyev Muxriddin
Toshkent 2024
Chiziqli regressiya java kodi:
import java.util.ArrayList;
import java.util.List;
import java.util.Scanner;
public class Main {
public static void main(String[] args) {
Scanner scanner = new Scanner(System.in);
List xValues = new ArrayList<>();
List yValues = new ArrayList<>();
char continueInput = 'y';
while (continueInput == 'y' || continueInput == 'Y') {
System.out.print("X qiymatini kiriting: ");
double xInput = scanner.nextDouble();
System.out.print("Y qiymatini kiriting: ");
double yInput = scanner.nextDouble();
xValues.add(xInput);
yValues.add(yInput);
System.out.print("Yana ma'lumot qo'shishni xohlaysizmi? (y/n): ");
continueInput = scanner.next().charAt(0);
}
double[] regressionResult = linearRegression(xValues, yValues);
double slope = regressionResult[0];
double intercept = regressionResult[1];
System.out.println("Chiziqli regressiya: y = " + slope + "x + " + intercept);
}
public static double[] linearRegression(List x, List y) {
int n = x.size();
double sumX = 0, sumY = 0, sumXY = 0, sumXX = 0;
for (int i = 0; i < n; ++i) {
sumX += x.get(i);
sumY += y.get(i);
sumXY += x.get(i) * y.get(i);
sumXX += x.get(i) * x.get(i);
}
double xMean = sumX / n;
double yMean = sumY / n;
double slope = (sumXY - n * xMean * yMean) / (sumXX - n * xMean * xMean);
double intercept = yMean - slope * xMean;
return new double[]{slope, intercept};
}
}
import numpy as np
import matplotlib.pyplot as plt
x = np.array([11,7,23,9,2,17,11,9,4,11,14,9,6])
y = np.array([99,86,89,88,101,86,103,87,95,78,77,85,86])
def linear_regression(X, y):
X = np.column_stack((np.ones_like(X), X))
w = np.linalg.inv(X.T @ X) @ X.T @ y
return w
def predict(X, w):
X = np.column_stack((np.ones_like(X), X))
y_pred = X @ w
return y_pred
def cost_function(X, y, w):
m = len(y)
J = np.sum((X @ w - y) ** 2) / (2 * m)
return J
w = linear_regression(x, y)
w1_values = np.linspace(w[1] - 10, w[1] + 10, 100)
J_values = [cost_function(np.column_stack((np.ones_like(x), x)), y, np.array([w[0], w1])) for w1 in w1_values]
# plt.scatter(w1_values, J_values, color='blue')
plt.plot(w1_values, J_values, color='red')
plt.xlabel('w1')
plt.ylabel('J(w)')
plt.title('Cost Function')
plt.grid(True)
plt.show()
print("Parameters of the linear regression model:")
print("w0:", w[0])
print("w1:", w[1])
y_pred = predict(x, w)
J = cost_function(np.column_stack((np.ones_like(x), x)), y, w)
print("Cost (Mean Squared Error):", J)
plt.scatter(x, y, color='blue', label='Data points')
plt.plot(x, y_pred, color='red', label='Linear Regression Line')
plt.xlabel('x')
plt.ylabel('y')
plt.title('Linear Regression')
plt.legend()
plt.grid(True)
plt.show()
|
| |
