add: R2 Score to get confidence

confidence.py

@@ -0,0 +1,44 @@
+import pandas as pd
+'''
+Class Confidence:
+    Calculate the confidence using the R-Square score
+    https://en.wikipedia.org/wiki/Coefficient_of_determination
+'''
+
+class Confidence:
+    def __init__(self, thetas_path="thetas.csv", data_path="datasets/data.csv"):
+        self.theta_path = thetas_path
+        self.data_path = data_path
+        self.data = []
+        self.theta0 = 0
+        self.theta1 = 0
+        self.get_thetas()
+        self.get_data()
+    def get_thetas(self):
+        try:
+            with open(self.theta_path, 'r') as file:
+                data = pd.read_csv(file)
+                self.theta0 = data["theta0"].iloc[0]
+                self.theta1 = data["theta1"].iloc[0]
+        except:
+            print("! Warning, no trained model has been found")
+
+    def get_data(self):
+        try:
+            with open(self.data_path, 'r') as file:
+                self.data = pd.read_csv(file)
+        except:
+            print("! Warning, no data has been found")
+
+    def estimate_price(self, mileage):
+        return self.theta0 + (self.theta1 * mileage)
+
+    def get_confidence(self):
+        predicted_price = []
+        for data in self.data["km"]:
+            predicted_price.append(self.estimate_price(data))
+        avg_price = sum(self.data["price"]) / len(self.data["price"])
+        ss_tot = sum((y - avg_price) ** 2 for y in self.data["price"])
+        ss_res = sum((y - y_hat) ** 2 for y, y_hat in zip(self.data["price"], predicted_price))
+        r2 = 1 - (ss_res / ss_tot)
+        return r2

model.py

@@ -3,8 +3,12 @@ import numpy as np
 import matplotlib.pyplot as plt
 import csv
 
+from confidence import Confidence
+
+data_path="datasets/data.csv"
+
 class Model:
-    def __init__(self, data_name="data.csv", learning_rate=0.01, epochs=2000):
+    def __init__(self, data_name=data_path, learning_rate=0.01, epochs=2000):
         with open(data_name, 'r') as file:
             self.data = pd.read_csv(file)
             self.km = [self.data["km"][i] for i in range(len(self.data))]
@@ -61,6 +65,8 @@ class Model:
 
             writer.writerows(data)
     def visualize(self):
+        confidence = Confidence(data_path=data_path)
+
         km_range = np.linspace(min(self.data["km"]), max(self.data["km"]), 100)
         predicted_prices = self.theta0 + (self.theta1 * (km_range - self.km_min) / (self.km_max - self.km_min))
         fig, (regression, cost) = plt.subplots(nrows=1, ncols=2, figsize=(12, 5))
@@ -73,11 +79,11 @@ class Model:
         regression.set_title("Linear Regression")
         regression.legend()
         regression.grid(True)
-
+        regression.text(0.05, 0.95, f"Confidence: {confidence.get_confidence()}", transform=regression.transAxes, fontsize=8, verticalalignment='top', horizontalalignment='left', color="black")
         cost.plot(range(1, len(self.cost_history) + 1), self.cost_history, color='green')
-        cost.set_title("Coût vs Itérations")
-        cost.set_xlabel("Itérations")
-        cost.set_ylabel("Coût")
+        cost.set_title("Cost vs Epochs")
+        cost.set_xlabel("Epochs")
+        cost.set_ylabel("Cost")
         cost.grid(True)
         plt.show()