mergeron 2024.738953.1__py3-none-any.whl → 2025.739265.0__py3-none-any.whl

mergeron/gen/upp_tests.py

@@ -1,248 +1,157 @@
 """
-Routines to estimate intrinsic clearnace rates and intrinsic enforcement rates
+Methods to compute intrinsic clearance rates and intrinsic enforcement rates
 from generated market data.
 
 """
 
 from collections.abc import Sequence
 from contextlib import suppress
-from dataclasses import fields
-from importlib.metadata import version
 from pathlib import Path
-from typing import Literal, TypeAlias, TypedDict
+from typing import Any, Literal, TypedDict
 
 import numpy as np
 import tables as ptb  # type: ignore
-from attrs import evolve
-from attrs import fields as attrs_fields
-from joblib import Parallel, cpu_count, delayed  # type: ignore
 from numpy.random import SeedSequence
 from numpy.typing import NDArray
 
-from mergeron.core.pseudorandom_numbers import TF, TI
-
-from .. import _PKG_NAME, RECConstants, UPPAggrSelector  # noqa: TID252
+from .. import (  # noqa
+    VERSION,
+    ArrayBIGINT,
+    ArrayBoolean,
+    ArrayDouble,
+    ArrayFloat,
+    ArrayINT,
+    HMGPubYear,
+    UPPAggrSelector,
+)
 from ..core import guidelines_boundaries as gbl  # noqa: TID252
 from . import (
-    EMPTY_ARRAY_DEFAULT,
+    DEFAULT_EMPTY_ARRAY,
     DataclassInstance,
     INVResolution,
     MarketDataSample,
-    MarketSampleSpec,
     UPPTestRegime,
     UPPTestsCounts,
     UPPTestsRaw,
 )
-from . import data_generation as dgl
-from . import investigations_stats as isl
-
-__version__ = version(_PKG_NAME)
+from . import enforcement_stats as esl
 
+__version__ = VERSION
 
-ptb.parameters.MAX_NUMEXPR_THREADS = 8
-ptb.parameters.MAX_BLOSC_THREADS = 4
+type SaveData = Literal[False] | tuple[Literal[True], ptb.File, ptb.Group]
 
-SaveData: TypeAlias = Literal[False] | tuple[Literal[True], ptb.File, ptb.Group]
 
+class INVRESCntsArgs(TypedDict, total=False):
+    "Keyword arguments of function, :code:`sim_enf_cnts`"
 
-class IVNRESCntsArgs(TypedDict, total=False):
-    sim_test_regime: UPPTestRegime
-    saved_array_name_suffix: str
-    save_data_to_file: SaveData
-    seed_seq_list: list[SeedSequence]
+    sample_size: int
+    seed_seq_list: Sequence[SeedSequence] | None
     nthreads: int
+    save_data_to_file: SaveData
+    saved_array_name_suffix: str
 
 
-def sim_invres_cnts_ll(
-    _invres_parm_vec: gbl.HMGThresholds,
-    _mkt_sample_spec: MarketSampleSpec,
-    _sim_invres_cnts_kwargs: IVNRESCntsArgs,
+def compute_upp_test_counts(
+    _market_data_sample: MarketDataSample,
+    _upp_test_parms: gbl.HMGThresholds,
+    _upp_test_regime: UPPTestRegime,
     /,
 ) -> UPPTestsCounts:
-    """
-    A function to parallelize simulations
-
-    The parameters _sim_invres_cnts_kwargs is passed unaltered to
-    the parent function, sim_invres_cnts(), except that, if provided,
-    "seed_seq_list" is used to spawn a seed sequence for each thread,
-    to assure independent samples in each thread. The number of draws
-    in each thread may be tuned, by trial and error, to the amount of
-    memory (RAM) available.
-
-    """
-
-    _sample_sz = _mkt_sample_spec.sample_size
-    _subsample_sz = 10**6
-    _iter_count = int(_sample_sz / _subsample_sz) if _subsample_sz < _sample_sz else 1
-    _thread_count = cpu_count()
+    """Estimate enforcement and clearance counts from market data sample
 
-    # Crate a copy, to avoid side effects in the outer scope
-    _mkt_sample_spec_here = evolve(_mkt_sample_spec, sample_size=_subsample_sz)
+    Parameters
+    ----------
+    _market_data_sample
+        Market data sample
 
-    if (
-        _mkt_sample_spec.recapture_rate is None
-        and _mkt_sample_spec.share_spec.recapture_spec != RECConstants.OUTIN
-    ):
-        _mkt_sample_spec_here = evolve(
-            _mkt_sample_spec_here, recapture_rate=_invres_parm_vec.rec
-        )
-    elif _mkt_sample_spec.recapture_rate != _invres_parm_vec.rec:
-        raise ValueError(
-            "{} {} {} {}".format(
-                f"Value, {_mkt_sample_spec.recapture_rate}",
-                "of recapture rate in the second positional argument",
-                f"must equal its value, {_invres_parm_vec.rec}",
-                "in the first positional argument.",
-            )
-        )
+    _upp_test_parms
+        Threshold values for various Guidelines criteria
 
-    _rng_seed_seq_list = [None] * _iter_count
-    if _sim_invres_cnts_kwargs:
-        if _sseql := _sim_invres_cnts_kwargs.get("seed_seq_list", None):
-            _rng_seed_seq_list = list(
-                zip(*[g.spawn(_iter_count) for g in _sseql], strict=True)  # type: ignore
-            )
+    _upp_test_regime
+        Specifies whether to analyze enforcement, clearance, or both
+        and the GUPPI and diversion ratio aggregators employed, with
+        default being to analyze enforcement based on the maximum
+        merging-firm GUPPI and maximum diversion ratio between the
+        merging firms
 
-        _sim_invres_cnts_ll_kwargs: IVNRESCntsArgs = {  # type: ignore
-            _k: _v
-            for _k, _v in _sim_invres_cnts_kwargs.items()
-            if _k != "seed_seq_list"
-        }
-    else:
-        _sim_invres_cnts_ll_kwargs = {}
-
-    _res_list = Parallel(n_jobs=_thread_count, prefer="threads")(
-        delayed(sim_invres_cnts)(
-            _invres_parm_vec,
-            _mkt_sample_spec_here,
-            **_sim_invres_cnts_ll_kwargs,
-            saved_array_name_suffix=f"{_iter_id:0{2 + int(np.ceil(np.log10(_iter_count)))}d}",
-            seed_seq_list=_rng_seed_seq_list_ch,
-        )
-        for _iter_id, _rng_seed_seq_list_ch in enumerate(_rng_seed_seq_list)
-    )
-
-    _res_list_stacks = UPPTestsCounts(*[
-        np.stack([getattr(_j, _k) for _j in _res_list])
-        for _k in ("by_firm_count", "by_delta", "by_conczone")
-    ])
-    upp_test_results = UPPTestsCounts(*[
-        np.column_stack((
-            (_gv := getattr(_res_list_stacks, _g.name))[0, :, :_h],
-            np.einsum("ijk->jk", np.int64(1) * _gv[:, :, _h:]),
-        ))
-        for _g, _h in zip(fields(_res_list_stacks), [1, 1, 3], strict=True)
-    ])
-    del _res_list, _res_list_stacks
+    Returns
+    -------
+    UPPTestsCounts
+        Enforced and cleared counts
 
-    return upp_test_results
-
-
-def sim_invres_cnts(
-    _upp_test_parms: gbl.HMGThresholds,
-    _mkt_sample_spec: MarketSampleSpec,
-    /,
-    *,
-    sim_test_regime: UPPTestRegime,
-    saved_array_name_suffix: str = "",
-    save_data_to_file: SaveData = False,
-    seed_seq_list: list[SeedSequence] | None = None,
-    nthreads: int = 16,
-) -> UPPTestsCounts:
-    # Generate market data
-    _market_data = dgl.gen_market_sample(
-        _mkt_sample_spec, seed_seq_list=seed_seq_list, nthreads=nthreads
-    )
-
-    _invalid_array_names = (
-        ("fcounts", "choice_prob_outgd", "nth_firm_share", "hhi_post")
-        if _mkt_sample_spec.share_spec.dist_type == "Uniform"
-        else ()
-    )
-
-    save_data_to_hdf5(
-        _market_data,
-        saved_array_name_suffix,
-        _invalid_array_names,
-        save_data_to_file=save_data_to_file,
-    )
+    """
 
-    _upp_tests_data = gen_upp_arrays(
-        _upp_test_parms,
-        _market_data,
-        sim_test_regime,
-        saved_array_name_suffix=saved_array_name_suffix,
-        save_data_to_file=save_data_to_file,
+    _enf_cnts_sim_array = -1 * np.ones((6, 2), np.int64)
+    _upp_test_arrays = compute_upp_test_arrays(
+        _market_data_sample, _upp_test_parms, _upp_test_regime
     )
 
     _fcounts, _hhi_delta, _hhi_post = (
-        getattr(_market_data, _g) for _g in ["fcounts", "hhi_delta", "hhi_post"]
+        getattr(_market_data_sample, _g) for _g in ("fcounts", "hhi_delta", "hhi_post")
     )
-    del _market_data
 
     _stats_rowlen = 6
     # Clearance/enforcement counts --- by firm count
-    _firm_counts_weights = _mkt_sample_spec.share_spec.firm_counts_weights
-    if _firm_counts_weights is not None and np.all(_firm_counts_weights >= 0):
-        _max_firm_count = len(_firm_counts_weights)
+    _firmcounts_list = np.unique(_fcounts)
+    if _firmcounts_list is not None and np.all(_firmcounts_list >= 0):
+        _max_firmcount = max(_firmcounts_list)
 
-        _invres_cnts_sim_byfirmcount_array = -1 * np.ones(_stats_rowlen, np.int64)
-        for _firm_cnt in 2 + np.arange(_max_firm_count):
-            _firm_count_test = _fcounts == _firm_cnt
+        _enf_cnts_sim_byfirmcount_array = -1 * np.ones(_stats_rowlen, np.int64)
+        for _firmcount in np.arange(2, _max_firmcount + 1):
+            _firmcount_test = _fcounts == _firmcount
 
-            _invres_cnts_sim_byfirmcount_array = np.row_stack((
-                _invres_cnts_sim_byfirmcount_array,
+            _enf_cnts_sim_byfirmcount_array = np.vstack((
+                _enf_cnts_sim_byfirmcount_array,
                 np.array([
-                    _firm_cnt,
-                    np.einsum("ij->", 1 * _firm_count_test),
+                    _firmcount,
+                    np.einsum("ij->", 1 * _firmcount_test),
                     *[
                         np.einsum(
                             "ij->",
-                            1 * (_firm_count_test & getattr(_upp_tests_data, _f.name)),
+                            1 * (_firmcount_test & getattr(_upp_test_arrays, _f)),
                         )
-                        for _f in fields(_upp_tests_data)
+                        for _f in _upp_test_arrays.__dataclass_fields__
                     ],
                 ]),
             ))
-        _invres_cnts_sim_byfirmcount_array = _invres_cnts_sim_byfirmcount_array[1:]
+        _enf_cnts_sim_byfirmcount_array = _enf_cnts_sim_byfirmcount_array[1:]
     else:
-        _invres_cnts_sim_byfirmcount_array = np.array(
+        _enf_cnts_sim_byfirmcount_array = np.array(
             np.nan * np.empty((1, _stats_rowlen)), np.int64
         )
-        _invres_cnts_sim_byfirmcount_array[0] = 2
+        _enf_cnts_sim_byfirmcount_array[0] = 2
 
-    # Clearance/enfrocement counts --- by delta
-    _hhi_delta_ranged = isl.hhi_delta_ranger(_hhi_delta)
-    _invres_cnts_sim_bydelta_array = -1 * np.ones(_stats_rowlen, np.int64)
-    for _hhi_delta_lim in isl.HHI_DELTA_KNOTS[:-1]:
+    # Clearance/enforcement counts --- by delta
+    _hhi_delta_ranged = esl.hhi_delta_ranger(_hhi_delta)
+    _enf_cnts_sim_bydelta_array = -1 * np.ones(_stats_rowlen, np.int64)
+    for _hhi_delta_lim in esl.HHI_DELTA_KNOTS[:-1]:
         _hhi_delta_test = _hhi_delta_ranged == _hhi_delta_lim
 
-        _invres_cnts_sim_bydelta_array = np.row_stack((
-            _invres_cnts_sim_bydelta_array,
+        _enf_cnts_sim_bydelta_array = np.vstack((
+            _enf_cnts_sim_bydelta_array,
             np.array([
                 _hhi_delta_lim,
                 np.einsum("ij->", 1 * _hhi_delta_test),
                 *[
                     np.einsum(
-                        "ij->",
-                        1 * (_hhi_delta_test & getattr(_upp_tests_data, _f.name)),
+                        "ij->", 1 * (_hhi_delta_test & getattr(_upp_test_arrays, _f))
                     )
-                    for _f in fields(_upp_tests_data)
+                    for _f in _upp_test_arrays.__dataclass_fields__
                 ],
             ]),
         ))
 
-    _invres_cnts_sim_bydelta_array = _invres_cnts_sim_bydelta_array[1:]
+    _enf_cnts_sim_bydelta_array = _enf_cnts_sim_bydelta_array[1:]
 
-    # Clearance/enfrocement counts --- by zone
+    # Clearance/enforcement counts --- by zone
     try:
-        _hhi_zone_post_ranged = isl.hhi_zone_post_ranger(_hhi_post)
+        _hhi_zone_post_ranged = esl.hhi_zone_post_ranger(_hhi_post)
     except ValueError as _err:
         print(_hhi_post)
         raise _err
 
     _stats_byconczone_sim = -1 * np.ones(_stats_rowlen + 1, np.int64)
-    for _hhi_zone_post_knot in isl.HHI_POST_ZONE_KNOTS[:-1]:
+    for _hhi_zone_post_knot in esl.HHI_POST_ZONE_KNOTS[:-1]:
         _level_test = _hhi_zone_post_ranged == _hhi_zone_post_knot
 
         for _hhi_zone_delta_knot in [0, 100, 200]:
@@ -254,7 +163,7 @@ def sim_invres_cnts(
 
             _conc_test = _level_test & _delta_test
 
-            _stats_byconczone_sim = np.row_stack((
+            _stats_byconczone_sim = np.vstack((
                 _stats_byconczone_sim,
                 np.array([
                     _hhi_zone_post_knot,
@@ -262,45 +171,53 @@ def sim_invres_cnts(
                     np.einsum("ij->", 1 * _conc_test),
                     *[
                         np.einsum(
-                            "ij->", 1 * (_conc_test & getattr(_upp_tests_data, _f.name))
+                            "ij->", 1 * (_conc_test & getattr(_upp_test_arrays, _f))
                         )
-                        for _f in fields(_upp_tests_data)
+                        for _f in _upp_test_arrays.__dataclass_fields__
                     ],
                 ]),
             ))
 
-    _invres_cnts_sim_byconczone_array = isl.invres_cnts_byconczone(
-        _stats_byconczone_sim[1:]
-    )
+    _enf_cnts_sim_byconczone_array = esl.enf_cnts_byconczone(_stats_byconczone_sim[1:])
     del _stats_byconczone_sim
     del _hhi_delta, _hhi_post, _fcounts
 
     return UPPTestsCounts(
-        _invres_cnts_sim_byfirmcount_array,
-        _invres_cnts_sim_bydelta_array,
-        _invres_cnts_sim_byconczone_array,
+        _enf_cnts_sim_byfirmcount_array,
+        _enf_cnts_sim_bydelta_array,
+        _enf_cnts_sim_byconczone_array,
     )
 
 
-def gen_upp_arrays(
-    _upp_test_parms: gbl.HMGThresholds,
+def compute_upp_test_arrays(
     _market_data: MarketDataSample,
+    _upp_test_parms: gbl.HMGThresholds,
     _sim_test_regime: UPPTestRegime,
     /,
-    *,
-    saved_array_name_suffix: str = "",
-    save_data_to_file: SaveData = False,
 ) -> UPPTestsRaw:
+    """
+    Generate UPP tests arrays for given configuration and market sample
+
+    Given a standards vector, market
+
+    Parameters
+    ----------
+    _market_data
+        market data sample
+    _upp_test_parms
+        guidelines thresholds for testing UPP and related statistics
+    _sim_test_regime
+        configuration to use for generating UPP tests
+
+    """
     _g_bar, _divr_bar, _cmcr_bar, _ipr_bar = (
         getattr(_upp_test_parms, _f) for _f in ("guppi", "divr", "cmcr", "ipr")
     )
 
-    _invres_resolution, _guppi_aggregator, _divr_aggregator = (
-        getattr(_sim_test_regime, _f)
-        for _f in ("resolution", "guppi_aggregator", "divr_aggregator")
+    _guppi_array, _ipr_array, _cmcr_array = (
+        np.empty_like(_market_data.price_array) for _ in range(3)
     )
 
-    _guppi_array = np.empty_like(_market_data.divr_array)
     np.einsum(
         "ij,ij,ij->ij",
         _market_data.divr_array,
@@ -309,157 +226,146 @@ def gen_upp_arrays(
         out=_guppi_array,
     )
 
-    _cmcr_array = np.empty_like(_market_data.divr_array)
-    np.divide(
-        np.einsum("ij,ij->ij", _market_data.pcm_array, _market_data.divr_array),
-        np.einsum("ij,ij->ij", 1 - _market_data.pcm_array, 1 - _market_data.divr_array),
-        out=_cmcr_array,
-    )
-
-    _ipr_array = np.empty_like(_market_data.divr_array)
     np.divide(
         np.einsum("ij,ij->ij", _market_data.pcm_array, _market_data.divr_array),
         1 - _market_data.divr_array,
         out=_ipr_array,
     )
 
-    # This one needs further testing:
-    # _ipr_array_alt = np.empty_like(_market_data.divr_array)
-    # np.divide(_guppi_array, (1 - _market_data.divr_array[:, ::-1]), out=_ipr_array_alt)
+    np.divide(_ipr_array, 1 - _market_data.pcm_array, out=_cmcr_array)
+
+    (_divr_test_vector,) = _compute_test_array_seq(
+        (_market_data.divr_array,),
+        _market_data.frmshr_array,
+        _sim_test_regime.divr_aggregator,
+    )
+
+    (_guppi_test_vector, _cmcr_test_vector, _ipr_test_vector) = _compute_test_array_seq(
+        (_guppi_array, _cmcr_array, _ipr_array),
+        _market_data.frmshr_array,
+        _sim_test_regime.guppi_aggregator,
+    )
+    del _cmcr_array, _ipr_array, _guppi_array
+
+    if _sim_test_regime.resolution == INVResolution.ENFT:
+        _upp_test_arrays = UPPTestsRaw(
+            _guppi_test_vector >= _g_bar,
+            (_guppi_test_vector >= _g_bar) | (_divr_test_vector >= _divr_bar),
+            _cmcr_test_vector >= _cmcr_bar,
+            _ipr_test_vector >= _ipr_bar,
+        )
+    else:
+        _upp_test_arrays = UPPTestsRaw(
+            _guppi_test_vector < _g_bar,
+            (_guppi_test_vector < _g_bar) & (_divr_test_vector < _divr_bar),
+            _cmcr_test_vector < _cmcr_bar,
+            _ipr_test_vector < _ipr_bar,
+        )
 
-    _test_measure_seq = (_market_data.divr_array, _guppi_array, _cmcr_array, _ipr_array)
+    return _upp_test_arrays
 
+
+def _compute_test_array_seq(
+    _test_measure_seq: tuple[ArrayDouble, ...],
+    _wt_array: ArrayDouble,
+    _aggregator: UPPAggrSelector,
+) -> tuple[ArrayDouble, ...]:
     _wt_array = (
-        _market_data.frmshr_array
-        / np.einsum("ij->i", _market_data.frmshr_array)[:, None]
-        if _guppi_aggregator
+        _wt_array / np.einsum("ij->i", _wt_array)[:, None]
+        if _aggregator
         in (
             UPPAggrSelector.CPA,
             UPPAggrSelector.CPD,
             UPPAggrSelector.OSA,
             UPPAggrSelector.OSD,
         )
-        else EMPTY_ARRAY_DEFAULT
+        else DEFAULT_EMPTY_ARRAY
     )
 
-    match _guppi_aggregator:
+    match _aggregator:
         case UPPAggrSelector.AVG:
-            _test_value_seq = (
+            _test_array_seq = (
                 1 / 2 * np.einsum("ij->i", _g)[:, None] for _g in _test_measure_seq
             )
         case UPPAggrSelector.CPA:
-            _test_value_seq = (
+            _test_array_seq = (
                 np.einsum("ij,ij->i", _wt_array[:, ::-1], _g)[:, None]
                 for _g in _test_measure_seq
             )
         case UPPAggrSelector.CPD:
-            _test_value_seq = (
+            _test_array_seq = (
                 np.sqrt(np.einsum("ij,ij,ij->i", _wt_array[:, ::-1], _g, _g))[:, None]
                 for _g in _test_measure_seq
             )
         case UPPAggrSelector.DIS:
-            _test_value_seq = (
+            _test_array_seq = (
                 np.sqrt(1 / 2 * np.einsum("ij,ij->i", _g, _g))[:, None]
                 for _g in _test_measure_seq
             )
         case UPPAggrSelector.MAX:
-            _test_value_seq = (
+            _test_array_seq = (
                 _g.max(axis=1, keepdims=True) for _g in _test_measure_seq
             )
         case UPPAggrSelector.MIN:
-            _test_value_seq = (
+            _test_array_seq = (
                 _g.min(axis=1, keepdims=True) for _g in _test_measure_seq
             )
         case UPPAggrSelector.OSA:
-            _test_value_seq = (
+            _test_array_seq = (
                 np.einsum("ij,ij->i", _wt_array, _g)[:, None]
                 for _g in _test_measure_seq
             )
         case UPPAggrSelector.OSD:
-            _test_value_seq = (
+            _test_array_seq = (
                 np.sqrt(np.einsum("ij,ij,ij->i", _wt_array, _g, _g))[:, None]
                 for _g in _test_measure_seq
             )
         case _:
             raise ValueError("GUPPI/diversion ratio aggregation method is invalid.")
-    del _cmcr_array, _guppi_array
-    (_divr_test_vector, _guppi_test_vector, _cmcr_test_vector, _ipr_test_vector) = (
-        _test_value_seq
-    )
-
-    if _divr_aggregator == UPPAggrSelector.MAX:
-        _divr_test_vector = _market_data.divr_array.max(axis=1, keepdims=True)
-
-    if _invres_resolution == INVResolution.ENFT:
-        _upp_tests_data = UPPTestsRaw(
-            _guppi_test_vector >= _g_bar,
-            (_guppi_test_vector >= _g_bar) | (_divr_test_vector >= _divr_bar),
-            _cmcr_test_vector >= _cmcr_bar,
-            _ipr_test_vector >= _ipr_bar,
-        )
-    else:
-        _upp_tests_data = UPPTestsRaw(
-            _guppi_test_vector < _g_bar,
-            (_guppi_test_vector < _g_bar) & (_divr_test_vector < _divr_bar),
-            _cmcr_test_vector < _cmcr_bar,
-            _ipr_test_vector < _ipr_bar,
-        )
-    del _guppi_test_vector, _divr_test_vector, _cmcr_test_vector, _ipr_test_vector
-
-    save_data_to_hdf5(
-        _upp_tests_data,
-        saved_array_name_suffix,
-        (),
-        save_data_to_file=save_data_to_file,
-    )
-
-    return _upp_tests_data
+    return tuple(_test_array_seq)
 
 
 def initialize_hd5(
-    _h5_path: Path, _hmg_pub_year: gbl.HMGPubYear, _test_regime: UPPTestRegime, /
+    _h5_path: Path, _hmg_pub_year: HMGPubYear, _test_regime: UPPTestRegime, /
 ) -> tuple[SaveData, str]:
     _h5_title = f"HMG version: {_hmg_pub_year}; Test regime: {_test_regime}"
     if _h5_path.is_file():
         _h5_path.unlink()
-    _h5_file = ptb.open_file(_h5_path, mode="w", title=_h5_title)
-    _save_data_to_file: tuple[Literal[True], ptb.File, str] = (True, _h5_file, "/")
-    _next_subgroup_name = "invres_{}_{}_{}_{}".format(
+    _h5_file = ptb.open_file(_h5_path, mode="w", title=_h5_title)  # pyright: ignore
+    _save_data_to_file: SaveData = (True, _h5_file, _h5_file.root)
+    _next_subgroup_name_root = "enf_{}_{}_{}_{}".format(
         _hmg_pub_year,
-        *(
-            getattr(_test_regime, _f.name).name
-            for _f in attrs_fields(type(_test_regime))
-        ),
+        *(getattr(_test_regime, _f.name).name for _f in _test_regime.__attrs_attrs__),  # pyright: ignore
     )
-    return _save_data_to_file, _next_subgroup_name
+    return _save_data_to_file, _next_subgroup_name_root
 
 
 def save_data_to_hdf5(
     _dclass: DataclassInstance,
-    _saved_array_name_suffix: str = "",
-    _excl_attrs: Sequence[str] = (),
     /,
     *,
+    saved_array_name_suffix: str | None = "",
+    excluded_attrs: Sequence[str] | None = (),
     save_data_to_file: SaveData = False,
 ) -> None:
     if save_data_to_file:
         _, _h5_file, _h5_group = save_data_to_file
         # Save market data arrays
-        for _array_field in fields(_dclass):
-            _array_name = _array_field.name
-            if _array_name in _excl_attrs:
+        excluded_attrs = excluded_attrs or ()
+        for _array_name in _dclass.__dataclass_fields__:
+            if _array_name in excluded_attrs:
                 continue
             save_array_to_hdf5(
                 getattr(_dclass, _array_name),
                 _array_name,
                 _h5_group,
                 _h5_file,
-                saved_array_name_suffix=_saved_array_name_suffix,
+                saved_array_name_suffix=saved_array_name_suffix,
             )
 
 
 def save_array_to_hdf5(
-    _array_obj: NDArray[np.floating[TF] | np.integer[TI] | np.bool_],
+    _array_obj: NDArray[Any],
     _array_name: str,
     _h5_group: ptb.Group,
     _h5_file: ptb.File,
@@ -467,7 +373,7 @@ def save_array_to_hdf5(
     *,
     saved_array_name_suffix: str | None = None,
 ) -> None:
-    _h5_array_name = f"{_array_name}_{saved_array_name_suffix or ""}".rstrip("_")
+    _h5_array_name = f"{_array_name}_{saved_array_name_suffix or ''}".rstrip("_")
 
     with suppress(ptb.NoSuchNodeError):
         _h5_file.remove_node(_h5_group, name=_array_name)
@@ -477,6 +383,12 @@ def save_array_to_hdf5(
         _h5_array_name,
         atom=ptb.Atom.from_dtype(_array_obj.dtype),
         shape=_array_obj.shape,
-        filters=ptb.Filters(complevel=3, complib="blosc:lz4hc", fletcher32=True),
+        filters=ptb.Filters(complevel=3, complib="blosc:lz4hc", fletcher32=True),  # pyright: ignore
     )
     _h5_array[:] = _array_obj
+
+
+if __name__ == "__main__":
+    print(
+        "This module defines classes with methods for generating UPP test arrays and UPP test-counts arrays on given data."
+    )