An example to import existing experimental results into Optuna's storage.

quadratic_trial_import.py

"""
Optuna example that optimizes a simple quadratic function.

In this example, we demonstrate how to import existing experimental results 
and continue the optimization.

We have the following two ways to execute this example:

(1) Execute this code directly.
    $ python quadratic_trial_import.py


(2) Execute through CLI.
    $ STUDY_NAME=`optuna create-study --storage sqlite:///example.db`
    $ optuna study optimize quadratic_trial_import.py objective --n-trials=100 \
      --study $STUDY_NAME --storage sqlite:///example.db

"""


import optuna
from optuna.trial import InjectedTrial


def suggest_all(trial):
    x = trial.suggest_uniform('x', -100, 100)
    y = trial.suggest_categorical('y', [-1, 0, 1])
    return x, y


# Define a simple 2-dimensional quadratic function whose minimum value is -1 when (x, y) = (0, -1).
def quadratic(x, y):
    return x ** 2 + y


def objective(trial):
    x, y = suggest_all(trial)
    return quadratic(x, y)


def import_trials(study, param_value_pairs):
    for pv in param_value_pairs:
        trial = InjectedTrial(study, pv[0])
        # We can use InjectedTrial.suggest_xx to save the parameter values along with their ranges.
        suggest_all(trial)
        # Save the objective value and change trial state from RUNNING to COMPLETE.
        trial.report(pv[1])
        study.storage.set_trial_state(trial.trial_id, optuna.structs.TrialState.COMPLETE)


if __name__ == '__main__':
    # Let us minimize the objective function above.
    study = optuna.create_study()

    # 1. Run grid search.
    param_value_pairs = []
    step = 40
    for x in range(-100, 100 + step, step):
        for y in [-1, 0, 1]:
            value = quadratic(x, y)
            param_value_pairs.append(({'x': x, 'y': y}, value))

    # 2. Import grid-search results.
    import_trials(study, param_value_pairs)

    print(study.trials_dataframe())
    print('Best value: {} (params: {})\n'.format(study.best_value, study.best_params))

    # 3. Continue optimization with Optuna.
    study.optimize(objective, n_trials=10)

    print(study.trials_dataframe())
    print('Best value: {} (params: {})\n'.format(study.best_value, study.best_params))