Implement Linear Regression with Gradient Descent from Scratch

Problem: Implement Univariate Linear Regression Training with Gradient Descent Given data points (x_i, y_i), train a univariate linear regression model using batch gradient descent: [ \hat{y} = w x + b ] Use mean squared error (MSE): [ J(w,b) = \frac{1}{n}\sum_{i=1}^{n}(wx_i + b - y_i)^2 ] Input Line 1: n lr steps Next n lines: x_i y_i as floats Output Print the final w and b separated by a space, to 6 decimal places. Constraints 1 <= n <= 1e5 0 < lr <= 1 1 <= steps <= 1e5 Initialize w = 0, b = 0 Example Input: 3 0.1 1000 1 2 2 4 3 6 Output (with tolerance): 2.000000 0.000000 Example Input 3 0.1 1000 1 2 2 4 3 6 Output 2.000000 0.000000