import pickle class LinearRegression: def __init__(self, learning_rate=0.01, epochs=1000): self.learning_rate = learning_rate self.epochs = epochs self.w1 = 0.0 self.w2 = 0.0 self.b = 0.0 def predict(self, x1, x2): return self.w1 * x1 + self.w2 * x2 + self.b def compute_gradients(self, X1, X2, y): n = len(y) dw1, dw2, db = 0.0, 0.0, 0.0 for i in range(n): y_pred = self.predict(X1[i], X2[i]) error = y[i] - y_pred dw1 += -2 * X1[i] * error dw2 += -2 * X2[i] * error db += -2 * error return dw1 / n, dw2 / n, db / n def update(self, X1, X2, y): dw1, dw2, db = self.compute_gradients(X1, X2, y) self.w1 -= self.learning_rate * dw1 self.w2 -= self.learning_rate * dw2 self.b -= self.learning_rate * db def mean_squared_error(self, y, y_pred): n = len(y) return sum((y[i] - y_pred[i]) ** 2 for i in range(n)) / n def fit(self, X1, X2, y): for epoch in range(self.epochs): self.update(X1, X2, y) if epoch % 200 == 0: y_pred = [self.predict(X1[i], X2[i]) for i in range(len(y))] mse = self.mean_squared_error(y, y_pred) print(f"Epoch {epoch}: MSE = {mse:.4f}") # ------------------------------- # TRAINING # ------------------------------- X1 = [2, 4, 6, 8] X2 = [2, 4, 6, 8] y = [4, 8, 12, 16] model = LinearRegression(learning_rate=0.01, epochs=1000) model.fit(X1, X2, y) print("Prediction before saving:", model.predict(2, 2)) # ------------------------------- # SAVE MODEL # ------------------------------- with open("linear_regression_model.pkl", "wb") as f: pickle.dump(model, f) print("Model saved!") # ------------------------------- # LOAD MODEL # ------------------------------- with open("linear_regression_model.pkl", "rb") as f: loaded_model = pickle.load(f) print("Prediction after loading:", loaded_model.predict(2, 2))