SteppableOptimizer¶

class SteppableOptimizer(maxiter=100)[source]¶

Bases: Optimizer

Base class for a steppable optimizer.

This family of optimizers uses the ask and tell interface. When using this interface the user has to call ask() to get information about how to evaluate the function (we are asking the optimizer about how to do the evaluation). This information is typically the next points at which the function is evaluated, but depending on the optimizer it can also determine whether to evaluate the function or its gradient. Once the function has been evaluated, the user calls the method tell() to tell the optimizer what the result of the function evaluation(s) is. The optimizer then updates its state accordingly and the user can decide whether to stop the optimization process or to repeat a step.

This interface is more customizable, and allows the user to have full control over the evaluation of the function.

Examples

An example where the evaluation of the function has a chance of failing. The user, with specific knowledge about his function can catch this errors and handle them before passing the result to the optimizer.

import random
import numpy as np
from qiskit_machine_learning.optimizers import GradientDescent

def objective(x):
    if random.choice([True, False]):
        return None
    else:
        return (np.linalg.norm(x) - 1) ** 2

def grad(x):
    if random.choice([True, False]):
        return None
    else:
        return 2 * (np.linalg.norm(x) - 1) * x / np.linalg.norm(x)


initial_point = np.random.normal(0, 1, size=(100,))

optimizer = GradientDescent(maxiter=20)
optimizer.start(x0=initial_point, fun=objective, jac=grad)

while optimizer.continue_condition():
    ask_data = optimizer.ask()
    evaluated_gradient = None

    while evaluated_gradient is None:
        evaluated_gradient = grad(ask_data.x_center)
        optimizer.state.njev += 1

    optimizer.state.nit += 1

     cf  = TellData(eval_jac=evaluated_gradient)
    optimizer.tell(ask_data=ask_data, tell_data=tell_data)

result = optimizer.create_result()

Users that aren’t dealing with complicated functions and who are more familiar with step by step optimization algorithms can use the step() method which wraps the ask() and tell() methods. In the same spirit the method minimize() will optimize the function and return the result.

To see other libraries that use this interface one can visit: https://optuna.readthedocs.io/en/stable/tutorial/20_recipes/009_ask_and_tell.html

Parameters:: maxiter (int) – Number of steps in the optimization process before ending the loop.

Attributes

bounds_support_level¶: Returns bounds support level

gradient_support_level¶: Returns gradient support level

initial_point_support_level¶: Returns initial point support level

is_bounds_ignored¶: Returns is bounds ignored

is_bounds_required¶: Returns is bounds required

is_bounds_supported¶: Returns is bounds supported

is_gradient_ignored¶: Returns is gradient ignored

is_gradient_required¶: Returns is gradient required

is_gradient_supported¶: Returns is gradient supported

is_initial_point_ignored¶: Returns is initial point ignored

is_initial_point_required¶: Returns is initial point required

is_initial_point_supported¶: Returns is initial point supported

setting¶: Return setting

settings¶

The optimizer settings in a dictionary format.

The settings can for instance be used for JSON-serialization (if all settings are serializable, which e.g. doesn’t hold per default for callables), such that the optimizer object can be reconstructed as

settings = optimizer.settings
# JSON serialize and send to another server
optimizer = OptimizerClass(**settings)

state¶: Return the current state of the optimizer.

Methods

ask()[source]¶

Ask the optimizer for a set of points to evaluate.

This method asks the optimizer which are the next points to evaluate. These points can, e.g., correspond to function values and/or its derivative. It may also correspond to variables that let the user infer which points to evaluate. It is the first method inside of a step() in the optimization process.

Returns:: An object containing the data needed to make the function evaluation to advance the optimization process.
Return type:: AskData

continue_condition()[source]¶

Condition that indicates the optimization process should continue.

Returns:: True if the optimization process should continue, False otherwise.
Return type:: bool

abstract create_result()[source]¶

Returns the result of the optimization.

All the information needed to create such a result should be stored in the optimizer state and will typically contain the best point found, the function value and gradient at that point, the number of function and gradient evaluation and the number of iterations in the optimization.

Returns:: The result of the optimization process.
Return type:: OptimizerResult

abstract evaluate(ask_data)[source]¶

Evaluates the function according to the instructions contained in ask_data.

If the user decides to use step() instead of ask() and tell() this function will contain the logic on how to evaluate the function.

Parameters:: ask_data (AskData) – Contains the information on how to do the evaluation.
Returns:: Data of all relevant information about the function evaluation.
Return type:: TellData

abstract get_support_level()¶: Return support level dictionary

static gradient_num_diff(x_center, f, epsilon, max_evals_grouped=None)¶

We compute the gradient with the numeric differentiation in the parallel way, around the point x_center.

Parameters:

x_center (ndarray) – point around which we compute the gradient
f (func) – the function of which the gradient is to be computed.
epsilon (float) – the epsilon used in the numeric differentiation.
max_evals_grouped (int) – max evals grouped, defaults to 1 (i.e. no batching).

Returns:

the gradient computed

Return type:

grad

minimize(fun, x0, jac=None, bounds=None)[source]¶

Minimizes the function.

For well behaved functions the user can call this method to minimize a function. If the user wants more control on how to evaluate the function a custom loop can be created using ask() and tell() and evaluating the function manually.

Parameters:

fun (Callable[[float | ndarray], float]) – Function to minimize.
x0 (float | ndarray) – Initial point.
jac (Callable[[float | ndarray], float | ndarray] | None) – Function to compute the gradient.
bounds (list[tuple[float, float]] | None) – Bounds of the search space.

Returns:

Object containing the result of the optimization.

Return type:

OptimizerResult

print_options()¶: Print algorithm-specific options.

set_max_evals_grouped(limit)¶: Set max evals grouped

set_options(**kwargs)¶

Sets or updates values in the options dictionary.

The options dictionary may be used internally by a given optimizer to pass additional optional values for the underlying optimizer/optimization function used. The options dictionary may be initially populated with a set of key/values when the given optimizer is constructed.

Parameters:: kwargs (dict) – options, given as name=value.

abstract start(fun, x0, jac=None, bounds=None)[source]¶

Populates the state of the optimizer with the data provided and sets all the counters to 0.

Parameters:

fun (Callable[[float | ndarray], float]) – Function to minimize.
x0 (float | ndarray) – Initial point.
jac (Callable[[float | ndarray], float | ndarray] | None) – Function to compute the gradient.
bounds (list[tuple[float, float]] | None) – Bounds of the search space.

step()[source]¶

Performs one step in the optimization process.

This method composes ask(), evaluate(), and tell() to make a “step” in the optimization process.

tell(ask_data, tell_data)[source]¶

Updates the optimization state using the results of the function evaluation.

A canonical optimization example using ask() and tell() can be seen in step().

Parameters:

ask_data (AskData) – Contains the information on how the evaluation was done.
tell_data (TellData) – Contains all relevant information about the evaluation of the objective function.

static wrap_function(function, args)¶

Wrap the function to implicitly inject the args at the call of the function.

Parameters:

function (func) – the target function
args (tuple) – the args to be injected

Returns:

wrapper

Return type:

function_wrapper