Не знаете, как это сделать?

---

implement the algorithms for the bisection and false-point variant discussed from the 04-Lecture-part-a notebook to approximate the  𝑟  that makes the centered-difference approximation of  𝑑𝑆/𝑑𝑟=0  (use  ℎ=0.001  when calling this finite-difference function). Use an interval of  (0,40]  and a tolerance of  0.01  in both algorithms. Comment on the number of times  𝑆  must be evaluated to achieve the desired tolerance for each algorithm (hint: look at the total number of iterations and check how many times  𝑆  must be evaluated by the finite-different function). Compare this to the brute-force approach.

Что я уже пробовал:

def for_diff(f,x=0,h=.1):
    '''
    Computes a forward difference approximation to the derivative of f at
    x using a step-size of h.
    
    >>> for_diff(lambda x : 5)
    0.0
    
    >>> for_diff(lambda x : 5*x)
    5
    '''
    deriv=(f(x+h)-f(x))/h
    return deriv

def back_diff(f,x=0,h=.1):
    deriv=(f(x)-f(x-h))/h
    return deriv

def cent_diff(f,x=0,h=.1):
    deriv=(f(x+h)-f(x-h))/(2*h)
    return deriv

---