_mn_sopt_baseline.py

import os
from machinable import Interface, get


class _MnSoptBaseline(Interface):

    class Config:
        ranges: bool = False

    def launch(self):
        from models.ops import import_initial_samples

        c = ""
        if self.config.ranges:
            c = "'benchmarks/motoneuron_modeling/config/wide_ranges.yaml'"

        for trial in range(3):
            with get("machinable.scope", {"trial": trial}):
                initial = get(
                    "interface.sopt_modeling",
                    [
                        f"""~from_protocol({c})""",
                        {
                            "dopt_params.surrogate_method_name": "gpr",
                            "dopt_params.n_epochs": 0,
                        },
                    ],
                ).launch()
                initial.save_attribute("preflight", True)
                if initial.cached():
                    with get("machinable.scope", {"parent": initial.hash}):
                        for version in [
                            {},
                            {
                                "dopt_params.surrogate_method_name": "gpr",
                            },
                            {
                                "dopt_params.surrogate_method_name": "megp",
                            },
                            "~joint_model(mode='o', backbone='resnet')",
                            "~joint_model(mode='c+o', backbone='resnet')",
                            "~joint_model(mode='o', backbone='fttransformer')",
                            "~joint_model(mode='c+o', backbone='fttransformer')",
                        ]:
                            e = get(
                                "interface.sopt_modeling",
                                [
                                    f"""~from_protocol({c})""",
                                    {
                                        "dopt_params.n_epochs": 25,
                                    },
                                ]
                                + [version],
                            ).launch()

                            if os.environ.get("LAUNCH", 0) and not os.path.isfile(
                                e.output_filepath
                            ):
                                e.commit()
                                import_initial_samples(
                                    file_path=e.output_filepath,
                                    source=initial.output_filepath,
                                    num=e.num_initial_samples,
                                    opt_id=e.config.dopt_params.opt_id,
                                    feature_dtypes=e.feature_dtypes,
                                    param_names=e.parameter_names,
                                )