import numpy as np
from sklearn.preprocessing import PolynomialFeatures
from sklearn.linear_model import LinearRegression

def findRegressionOutput(x_test:float):
    data = []
    # Ensure trainingdata.txt is in the same directory
    with open('trainingdata.txt','r') as file:
        for line in file:
            charging_time, hour = map(float, line.strip().split(','))
            data.append((charging_time, hour))
        
        data = np.array(data)
        x_raw = data[:, 0]
        y_raw = data[:, 1]
        
        # 1. Identify the threshold where battery is fully charged
        threshold = min(x_raw[y_raw == 8])
        
        # 2. Filter correctly using a mask to keep shapes aligned
        mask = x_raw < threshold
        x = x_raw[mask].reshape(-1, 1)
        y = y_raw[mask]
        
        # 3. Model training
        features = PolynomialFeatures(degree=2)
        x_poly = features.fit_transform(x)
        
        model = LinearRegression()
        model.fit(x_poly, y)
        
        # 4. Prediction Logic
        if x_test >= threshold:
            print(8.00)
        else:
            # Fixed np.array syntax and 2D shape
            x_test_feature = features.transform(np.array([[x_test]]))
            y_pred = model.predict(x_test_feature)
            print(round(float(y_pred[0]), 2))

if __name__ == '__main__':
    try:
        timeCharged = float(input().strip())
        findRegressionOutput(timeCharged)
    except EOFError:
        pass

print([8 if timeCharged >= 4 else timeCharged * 2][0])