TheAlgorithms/Python · error · ValueError

At least one simulation is necessary to estimate PI.

Error message

At least one simulation is necessary to estimate PI.

What it means

estimate_pi() in maths/pi_monte_carlo_estimation.py estimates pi as 4 * (points in unit circle / total points) over number_of_simulations draws. If number_of_simulations < 1 it raises ValueError('At least one simulation is necessary to estimate PI.') because the estimator's ratio m/n is undefined for n = 0 and meaningless for negative n. This is a statistical pre-condition, not a performance knob: even 1 is statistically worthless but technically allowed.

Source

Thrown at maths/pi_monte_carlo_estimation.py:47

    The estimate is generated by Monte Carlo simulations. Let U be uniformly drawn from
    the unit square [0, 1) x [0, 1). The probability that U lies in the unit circle is:

        P[U in unit circle] = 1/4 PI

    and therefore

        PI = 4 * P[U in unit circle]

    We can get an estimate of the probability P[U in unit circle].
    See https://en.wikipedia.org/wiki/Empirical_probability by:

        1. Draw a point uniformly from the unit square.
        2. Repeat the first step n times and count the number of points in the unit
            circle, which is called m.
        3. An estimate of P[U in unit circle] is m/n
    """
    if number_of_simulations < 1:
        raise ValueError("At least one simulation is necessary to estimate PI.")

    number_in_unit_circle = 0
    for _ in range(number_of_simulations):
        random_point = Point.random_unit_square()

        if random_point.is_in_unit_circle():
            number_in_unit_circle += 1

    return 4 * number_in_unit_circle / number_of_simulations


if __name__ == "__main__":
    # import doctest

    # doctest.testmod()
    from math import pi

    prompt = "Please enter the desired number of Monte Carlo simulations: "

View on GitHub (pinned to f5988cc097)

Solutions

1. Pass at least 1; realistically pass a large count (e.g. 100_000+) since accuracy grows with sqrt(n).
2. If the count is user/config supplied, clamp or validate it (max(1, n) or explicit error) before calling.
3. Guard callers that compute n dynamically so n = 0 fails loudly upstream with a clearer message.

Example fix

# before
estimate_pi(num_points)  # ValueError when num_points == 0

# after
if num_points < 1:
    raise ValueError(f"need >= 1 simulation, got {num_points}")
estimate_pi(num_points)

if number_of_simulations < 1:
    raise ValueError('simulation count must be >= 1')
estimate_pi(number_of_simulations)

def is_valid_simulation_count(v) -> bool:
    return isinstance(v, int) and v >= 1

When it happens

Trigger: Calling estimate_pi(0), estimate_pi(-100), or passing a computed count (e.g. int(request.args['n']) defaulting to 0, or a variable that underflowed to 0) as number_of_simulations.

Common situations: Config/default values left at 0; a loop or formula producing 0 simulations for tiny inputs; CLI flag parsing that yields 0 when the flag is omitted.

Related errors

• number must be an integer
• multiplicative_persistence() only accepts integral values
• multiplicative_persistence() does not accept negative values
• additive_persistence() only accepts integral values
• additive_persistence() does not accept negative values

AI-assisted analysis of TheAlgorithms/Python@f5988cc097 (2026-08-14). Data as JSON: /api/errors/c0ea10903ada95ce. Report an issue: GitHub.