def from_inputs()

in src/gluonts/model/seq2seq/_mq_dnn_estimator.py [0:0]

• 39 lines of code
• 11 McCabe index (conditional complexity)

    def from_inputs(cls, train_iter, **params):
        logger = logging.getLogger(__name__)
        logger.info(
            f"gluonts[from_inputs]: User supplied params set to {params}"
        )
        # auto_params usually include `use_feat_dynamic_real`, `use_past_feat_dynamic_real`,
        # `use_feat_static_cat` and `cardinality`
        auto_params = cls.derive_auto_fields(train_iter)

        fields = [
            "use_feat_dynamic_real",
            "use_past_feat_dynamic_real",
            "use_feat_static_cat",
        ]
        # user defined arguments become implications
        for field in fields:
            if field in params.keys():
                is_params_field = (
                    params[field]
                    if type(params[field]) == bool
                    else strtobool(params[field])
                )
                if is_params_field and not auto_params[field]:
                    logger.warning(
                        f"gluonts[from_inputs]: {field} set to False since it is not present in the data."
                    )
                    params[field] = False
                    if field == "use_feat_static_cat":
                        params["cardinality"] = None
                elif (
                    field == "use_feat_static_cat"
                    and not is_params_field
                    and auto_params[field]
                ):
                    params["cardinality"] = None

        # user specified 'params' will take precedence:
        params = {**auto_params, **params}
        logger.info(
            f"gluonts[from_inputs]: use_past_feat_dynamic_real set to "
            f"'{params['use_past_feat_dynamic_real']}', use_feat_dynamic_real set to "
            f"'{params['use_feat_dynamic_real']}', and use_feat_static_cat set to "
            f"'{params['use_feat_static_cat']}' with cardinality of '{params['cardinality']}'"
        )
        return cls.from_hyperparameters(**params)