Source code for ax.exceptions.generation_strategy

#!/usr/bin/env python3
# Copyright (c) Meta Platforms, Inc. and affiliates.
#
# This source code is licensed under the MIT license found in the
# LICENSE file in the root directory of this source tree.

# pyre-strict

import typing  # noqa F401, this is to enable type-checking
from typing import Optional

from ax.exceptions.core import AxError, OptimizationComplete


[docs]class AxGenerationException(AxError):
    """Raised when there is an issue with the generation strategy."""

    pass


[docs]class MaxParallelismReachedException(AxGenerationException):
    """Special exception indicating that maximum number of trials running in
    parallel set on a given step (as `GenerationStep.max_parallelism`) has been
    reached. Upon getting this exception, users should wait until more trials
    are completed with data, to generate new trials.
    """

    def __init__(
        self,
        model_name: str,
        num_running: int,
        step_index: Optional[int] = None,
        node_name: Optional[str] = None,
    ) -> None:
        if node_name is not None:
            msg_start = (
                f"Maximum parallelism for generation node #{node_name} ({model_name})"
            )
        else:
            msg_start = (
                f"Maximum parallelism for generation step #{step_index} ({model_name})"
            )
        super().__init__(
            msg_start
            + f" has been reached: {num_running} trials are currently 'running'. Some "
            "trials need to be completed before more trials can be generated. See "
            "https://ax.dev/docs/bayesopt.html to understand why limited parallelism "
            "improves performance of Bayesian optimization."
        )


[docs]class GenerationStrategyCompleted(OptimizationComplete):
    """Special exception indicating that the generation strategy has been
    completed.
    """

    pass


[docs]class GenerationStrategyRepeatedPoints(GenerationStrategyCompleted):
    """Special exception indicating that the generation strategy is repeatedly
    suggesting previously sampled points.
    """

    pass


[docs]class GenerationStrategyMisconfiguredException(AxGenerationException):
    """Special exception indicating that the generation strategy is misconfigured."""

    def __init__(self, error_info: Optional[str]) -> None:
        super().__init__(
            "This GenerationStrategy was unable to be initialized properly. Please "
            + "check the documentation, and adjust the configuration accordingly. "
            + f"{error_info}"
        )