iframe-proxy | Sunbelt Computer Software

557 lines (475 loc) · 22.7 KB
"""End-to-end tests for ``diff_diff.local_linear.mse_optimal_bandwidth``.
Parity against nprobust 0.5.0 (SHA 36e4e53) is the primary test. Golden
values live at ``benchmarks/data/nprobust_mse_dpi_golden.json``; see
``benchmarks/R/generate_nprobust_golden.R`` for the generator.
Behavioural tests cover input validation, kernel dispatch, and
``return_diagnostics`` shape.
from __future__ import annotations
import json
from pathlib import Path
import numpy as np
import pytest
from diff_diff import BandwidthResult, local_linear_fit, mse_optimal_bandwidth
# =============================================================================
# Golden-value parity: per-stage at 1% relative tolerance (DGP 1 / 2 / 3)
# =============================================================================
GOLDEN_PATH = (
    Path(__file__).resolve().parents[1] / "benchmarks" / "data" / "nprobust_mse_dpi_golden.json"
_PARITY_TOL = 0.01  # 1% relative error; commit criterion #4
@pytest.fixture(scope="module")
def golden():
    """Load R-generated nprobust golden values. Skip if the file is
    absent (e.g. in CI's isolated-install path where benchmarks/ is
    not copied alongside tests/)."""
    if not GOLDEN_PATH.exists():
        pytest.skip(
            "Golden values file not found; "
            "run: Rscript benchmarks/R/generate_nprobust_golden.R"
    with GOLDEN_PATH.open() as f:
        return json.load(f)
@pytest.fixture(scope="module", params=["dgp1", "dgp2", "dgp3"])
def dgp_case(request, golden):
    name = request.param
    d = np.array(golden[name]["d"], dtype=np.float64)
    y = np.array(golden[name]["y"], dtype=np.float64)
    return name, d, y, golden[name]
class TestNprobustParity:
    """1% per-stage parity against R nprobust::lpbwselect(bwselect="mse-dpi")."""
    def test_c_bw_parity(self, dgp_case):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        assert br.c_bw == pytest.approx(g["c_bw"], rel=_PARITY_TOL), name
    def test_bw_mp2_parity(self, dgp_case):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        assert br.bw_mp2 == pytest.approx(g["bw_mp2"], rel=_PARITY_TOL), name
    def test_bw_mp3_parity(self, dgp_case):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        assert br.bw_mp3 == pytest.approx(g["bw_mp3"], rel=_PARITY_TOL), name
    def test_b_mse_parity(self, dgp_case):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        assert br.b_mse == pytest.approx(g["b_mse_dpi"], rel=_PARITY_TOL), name
    def test_h_mse_parity(self, dgp_case):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        assert br.h_mse == pytest.approx(g["h_mse_dpi"], rel=_PARITY_TOL), name
class TestStageDiagnosticsParity:
    """Per-stage (V, B1, B2, R) parity at 1% each. Catches formula
    divergences that might cancel out in the final bandwidth."""
    @pytest.mark.parametrize(
        "stage",
        ["stage_d1", "stage_d2", "stage_b", "stage_h"],
    def test_V_parity(self, dgp_case, stage):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        actual = getattr(br, f"{stage}_V")
        expected = g[stage]["V"]
        assert actual == pytest.approx(expected, rel=_PARITY_TOL), f"{name} {stage}"
    @pytest.mark.parametrize(
        "stage",
        ["stage_d1", "stage_d2", "stage_b", "stage_h"],
    def test_B1_parity(self, dgp_case, stage):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        actual = getattr(br, f"{stage}_B1")
        expected = g[stage]["B1"]
        if expected == 0:
            assert actual == pytest.approx(0, abs=1e-10), f"{name} {stage}"
        else:
            assert actual == pytest.approx(expected, rel=_PARITY_TOL), f"{name} {stage}"
    @pytest.mark.parametrize(
        "stage",
        ["stage_d1", "stage_d2", "stage_b", "stage_h"],
    def test_B2_parity(self, dgp_case, stage):
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        actual = getattr(br, f"{stage}_B2")
        expected = g[stage]["B2"]
        if expected == 0:
            assert actual == pytest.approx(0, abs=1e-10), f"{name} {stage}"
        else:
            assert actual == pytest.approx(expected, rel=_PARITY_TOL), f"{name} {stage}"
    @pytest.mark.parametrize(
        "stage",
        ["stage_d1", "stage_d2", "stage_b", "stage_h"],
    def test_R_parity(self, dgp_case, stage):
        """R (BWreg) parity. stage_d1 / stage_d2 use scale=0 so R=0;
        stage_b / stage_h use scale=bwregul=1 so R is non-trivial and
        must match nprobust."""
        name, d, y, g = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        actual = getattr(br, f"{stage}_R")
        expected = g[stage]["R"]
        if expected == 0:
            assert actual == pytest.approx(0, abs=1e-10), f"{name} {stage}"
        else:
            assert actual == pytest.approx(expected, rel=_PARITY_TOL), (
                f"{name} {stage}: py={actual!r} R={expected!r}"
# =============================================================================
# Behavioral tests
# =============================================================================
class TestReturnShape:
    def test_returns_float_by_default(self, dgp_case):
        _, d, y, _ = dgp_case
        h = mse_optimal_bandwidth(d, y)
        assert isinstance(h, float)
        assert h > 0.0
    def test_return_diagnostics_true_returns_dataclass(self, dgp_case):
        _, d, y, _ = dgp_case
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        assert isinstance(br, BandwidthResult)
        assert br.h_mse > 0.0
        assert br.c_bw > 0.0
        assert br.b_mse > 0.0
    def test_float_return_matches_h_mse(self, dgp_case):
        _, d, y, _ = dgp_case
        h = mse_optimal_bandwidth(d, y)
        br = mse_optimal_bandwidth(d, y, return_diagnostics=True)
        # Different code paths but same computation; should be bit-exact.
        assert h == br.h_mse
class TestInputValidation:
    def test_mismatched_shapes_raise(self):
        rng = np.random.default_rng(0)
        d = rng.uniform(size=100)
        y = rng.normal(size=50)
        with pytest.raises(ValueError, match="same shape"):
            mse_optimal_bandwidth(d, y)
    def test_non_finite_d_raises(self):
        d = np.array([0.1, np.nan, 0.3, 0.4])
        y = np.array([1.0, 2.0, 3.0, 4.0])
        with pytest.raises(ValueError, match="d contains non-finite"):
            mse_optimal_bandwidth(d, y)
    def test_non_finite_y_raises(self):
        d = np.array([0.1, 0.2, 0.3, 0.4])
        y = np.array([1.0, 2.0, np.inf, 4.0])
        with pytest.raises(ValueError, match="y contains non-finite"):
            mse_optimal_bandwidth(d, y)
    def test_non_finite_boundary_raises(self):
        d = np.array([0.1, 0.2, 0.3, 0.4])
        y = np.array([1.0, 2.0, 3.0, 4.0])
        with pytest.raises(ValueError, match="boundary"):
            mse_optimal_bandwidth(d, y, boundary=np.nan)
    def test_unknown_kernel_raises(self):
        rng = np.random.default_rng(0)
        d = rng.uniform(size=100)
        y = rng.normal(size=100)
        with pytest.raises(ValueError, match="Unknown kernel"):
            mse_optimal_bandwidth(d, y, kernel="gaussian")
    def test_weights_raises_not_implemented(self):
        rng = np.random.default_rng(0)
        d = rng.uniform(size=100)
        y = rng.normal(size=100)
        w = np.ones_like(d)
        with pytest.raises(NotImplementedError, match="weights"):
            mse_optimal_bandwidth(d, y, weights=w)
    def test_bwcheck_exceeds_sample_size_raises(self):
        """bwcheck > N would cause IndexError inside the selector; guard it."""
        from diff_diff._nprobust_port import lpbwselect_mse_dpi
        d = np.array([0.1, 0.2, 0.3, 0.4, 0.5])
        y = np.array([1.0, 2.0, 3.0, 4.0, 5.0])
        with pytest.raises(ValueError, match="bwcheck"):
            # Default bwcheck=21 exceeds N=5.
            lpbwselect_mse_dpi(y, d, eval_point=0.0, bwcheck=21)
    def test_bwcheck_zero_raises(self):
        from diff_diff._nprobust_port import lpbwselect_mse_dpi
        rng = np.random.default_rng(0)
        d = rng.uniform(size=100)
        y = rng.normal(size=100)
        with pytest.raises(ValueError, match="bwcheck"):
            lpbwselect_mse_dpi(y, d, eval_point=0.0, bwcheck=0)
    def test_bwcheck_negative_raises(self):
        from diff_diff._nprobust_port import lpbwselect_mse_dpi
        rng = np.random.default_rng(0)
        d = rng.uniform(size=100)
        y = rng.normal(size=100)
        with pytest.raises(ValueError, match="bwcheck"):
            lpbwselect_mse_dpi(y, d, eval_point=0.0, bwcheck=-1)
    def test_public_wrapper_fixes_vce_nn_nnmatch_3(self):
        """Pin the public API scope restriction documented in
        REGISTRY.md and the mse_optimal_bandwidth docstring.
        The wrapper hard-codes vce='nn', nnmatch=3 for Phase 1b; users
        needing other variance modes must go through the private port.
        This test ensures the behavior is frozen: changing it would be
        a scope expansion that should update REGISTRY.md and the
        docstring in lockstep.
        """
        from diff_diff._nprobust_port import lpbwselect_mse_dpi
        rng = np.random.default_rng(20260419)
        d = rng.uniform(0, 1, size=G)
        y = d + d**2 + rng.normal(0, 0.5, size=G)
        via_wrapper = mse_optimal_bandwidth(
            d, y, kernel="epanechnikov", return_diagnostics=True
        via_port_nn = lpbwselect_mse_dpi(
            eval_point=0.0,
            deriv=0,
            kernel="epa",
            bwcheck=21,
            bwregul=1.0,
            vce="nn",
            nnmatch=3,
            interior=False,
        # vce/nnmatch/interior/p/deriv chosen by the wrapper must match
        # what the port call explicitly sets.
        assert via_wrapper.h_mse == via_port_nn.h_mse_dpi
        assert via_wrapper.b_mse == via_port_nn.b_mse_dpi
        assert via_wrapper.c_bw == via_port_nn.c_bw
    def test_bwcheck_none_on_tiny_sample_raises_valueerror(self):
        """bwcheck=None on a tiny sample must raise a clear ValueError
        from the per-stage support/rank guard in lprobust_bw, NOT an
        opaque LinAlgError or IndexError."""
        from diff_diff._nprobust_port import lpbwselect_mse_dpi
        d = np.array([0.1, 0.2, 0.3, 0.4, 0.5])
        y = np.array([1.0, 2.0, 3.0, 4.0, 5.0])
        with pytest.raises(ValueError, match="lprobust_bw"):
            lpbwselect_mse_dpi(y, d, eval_point=0.0, bwcheck=None)
    def test_interior_boundary_rejected(self):
        """boundary strictly inside the support must be rejected by
        the Phase 1b wrapper. Running the boundary selector at an
        interior point would silently use a symmetric kernel and
        produce a bandwidth incompatible with the one-sided fitter."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(0.0, 1.0, size=500)
        y = d + d**2 + rng.normal(0, 0.5, size=500)
        with pytest.raises(ValueError, match="boundary"):
            mse_optimal_bandwidth(d, y, boundary=0.5)
    def test_upper_boundary_rejected(self):
        """boundary at d.max() (upper support edge) must be rejected."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(0.0, 1.0, size=500)
        y = d + d**2 + rng.normal(0, 0.5, size=500)
        with pytest.raises(ValueError, match="boundary"):
            mse_optimal_bandwidth(d, y, boundary=float(d.max()))
    def test_boundary_equal_to_min_d_accepted(self):
        """Design 1 continuous-near-d_lower uses boundary = min(d)
        exactly; this must pass the applicability check."""
        rng = np.random.default_rng(20260419)
        d = rng.uniform(1.0, 2.0, size=1500)
        y = 3.0 + 0.5 * (d - 1.0) ** 2 + rng.normal(0, 0.3, size=1500)
        h = mse_optimal_bandwidth(d, y, boundary=float(d.min()))
        assert np.isfinite(h)
        assert h > 0.0
    def test_boundary_zero_design_1_prime_accepted(self):
        """Design 1' with support at 0: boundary=0 passes."""
        rng = np.random.default_rng(20260419)
        d = rng.uniform(0.0, 1.0, size=3000)
        y = d + d**2 + rng.normal(0, 0.5, size=3000)
        h = mse_optimal_bandwidth(d, y, boundary=0.0)
        assert np.isfinite(h)
        assert h > 0.0
    def test_boundary_zero_thin_boundary_density_accepted(self):
        """Beta(2,2) Design 1' case: boundary density vanishes at 0
        but the estimand is well-defined. Must not be mistakenly
        rejected by any design heuristic."""
        rng = np.random.default_rng(20260419)
        d = rng.beta(2.0, 2.0, size=2000)  # f_D(0) = 0
        y = d + d**2 + rng.normal(0, 0.5, size=2000)
        h = mse_optimal_bandwidth(d, y, boundary=0.0)
        assert np.isfinite(h)
        assert h > 0.0
    def test_boundary_zero_with_data_far_from_zero_rejected(self):
        """boundary=0 with d.min() substantially positive fails the
        Design 1' support check (d.min() > 5% of median(|d|)). The
        caller must either pass boundary=float(d.min()) for the
        continuous-near-d_lower path or confirm Design 1' applicability."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(0.5, 1.0, size=1500)  # d.min() ~ 0.5 >> 5% of median
        y = d + rng.normal(0, 0.3, size=1500)
        with pytest.raises(ValueError, match="Design 1'"):
            mse_optimal_bandwidth(d, y, boundary=0.0)
    def test_empty_input_rejected(self):
        """Empty d/y must raise a targeted ValueError up front, not
        leak a NumPy reduction error from d.min()."""
        d = np.array([], dtype=np.float64)
        y = np.array([], dtype=np.float64)
        with pytest.raises(ValueError, match="non-empty"):
            mse_optimal_bandwidth(d, y)
    def test_boundary_zero_with_d_min_mass_point_rejected(self):
        """boundary=0 with d.min() > 0 AND mass at d.min() is a
        Design 1 mass-point design and must be redirected to 2SLS."""
        rng = np.random.default_rng(2026)
        n_mass = 300  # 15% at 0.1
        n_cont = 1700
        d_mass = np.full(n_mass, 0.1)
        d_cont = rng.uniform(0.1, 1.0, size=n_cont)
        d = np.concatenate([d_mass, d_cont])
        y = d + rng.normal(0, 0.5, size=d.size)
        with pytest.raises(NotImplementedError, match="mass-point"):
            mse_optimal_bandwidth(d, y, boundary=0.0)
    def test_off_support_boundary_rejected(self):
        """boundary must equal 0 or d.min() within tolerance; any
        other lower off-support value must be rejected."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(1.0, 2.0, size=1500)  # d.min() ~ 1.0
        y = d + rng.normal(0, 0.3, size=1500)
        # boundary = 0.5 is between 0 and d.min(); neither documented
        # estimand.
        with pytest.raises(ValueError, match="not at a supported HAD estimand"):
            mse_optimal_bandwidth(d, y, boundary=0.5)
    def test_negative_boundary_rejected(self):
        """boundary < 0 is off-support and rejected."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(0.0, 1.0, size=1500)
        y = d + rng.normal(0, 0.3, size=1500)
        with pytest.raises(ValueError, match="not at a supported HAD estimand"):
            mse_optimal_bandwidth(d, y, boundary=-0.1)
    def test_negative_dose_rejected_boundary_zero(self):
        """HAD requires D_{g,2} >= 0. Negative-dose samples must be
        rejected at the wrapper level, even under boundary=0."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(-0.5, 0.5, size=1500)  # mix of negative and positive
        y = d + rng.normal(0, 0.3, size=1500)
        with pytest.raises(ValueError, match="Negative dose values"):
            mse_optimal_bandwidth(d, y, boundary=0.0)
    def test_negative_dose_rejected_boundary_at_d_min(self):
        """Negative-dose samples must also be rejected when the caller
        tries to pass boundary=d.min() < 0."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(-1.0, -0.1, size=1500)  # entirely negative
        y = d + rng.normal(0, 0.3, size=1500)
        with pytest.raises(ValueError, match="Negative dose values"):
            mse_optimal_bandwidth(d, y, boundary=float(d.min()))
    def test_mass_point_design_rejected(self):
        """Design 1 mass-point case (boundary > 0, modal fraction > 2%)
        must be rejected with NotImplementedError pointing to 2SLS."""
        rng = np.random.default_rng(2026)
        n_mass = 200  # 10% mass at d_lower
        n_cont = 1800
        d_mass = np.full(n_mass, 0.1)
        d_cont = rng.uniform(0.1, 1.0, size=n_cont)
        d = np.concatenate([d_mass, d_cont])
        y = d + rng.normal(0, 0.5, size=d.size)
        with pytest.raises(NotImplementedError, match="mass-point"):
            mse_optimal_bandwidth(d, y, boundary=float(d.min()))
    def test_continuous_near_d_lower_accepted(self):
        """Design 1 continuous-near-d_lower (boundary > 0, modal
        fraction <= 2%) must pass through to nonparametric."""
        rng = np.random.default_rng(20260419)
        d = rng.uniform(0.1, 1.0, size=1500)  # no mass point
        y = d + rng.normal(0, 0.3, size=1500)
        h = mse_optimal_bandwidth(d, y, boundary=float(d.min()))
        assert np.isfinite(h)
        assert h > 0.0
    def test_untreated_at_zero_accepted(self):
        """Paper Section 3.1.5 / Garrett et al. application: untreated
        units at d=0 are OK for Design 1'. boundary=0 with mass at 0
        must NOT trigger the mass-point rejection."""
        rng = np.random.default_rng(2026)
        # ~15% at d=0 (genuinely untreated), rest continuous on (0, 1).
        d_zero = np.zeros(300)
        d_pos = rng.uniform(0.01, 1.0, size=1700)
        d = np.concatenate([d_zero, d_pos])
        y = d + rng.normal(0, 0.5, size=d.size)
        h = mse_optimal_bandwidth(d, y, boundary=0.0)
        assert np.isfinite(h)
        assert h > 0.0
    def test_rank_deficient_design_raises_valueerror(self):
        """Duplicate-support windows must fail with a clear ValueError
        from qrXXinv's Cholesky guard, not an opaque LinAlgError."""
        from diff_diff._nprobust_port import qrXXinv
        # Rank-1 X: all rows identical -> X.T @ X is rank-1.
        X = np.tile([[1.0, 2.0, 3.0]], (10, 1))
        with pytest.raises(ValueError, match="qrXXinv"):
            qrXXinv(X)
    def test_wrapper_rank_deficient_raises_valueerror(self):
        """Public-wrapper regression: a continuous-near-d_lower sample
        whose kernel window contains too few DISTINCT d values drives
        a rank-deficient X'X in one of the DPI stages. The wrapper
        must surface a clear ValueError from qrXXinv's Cholesky guard,
        not an opaque LinAlgError.
        Construction: d.min() is unique (modal_fraction = 1/G < 2% so
        mass-point check passes), but the remaining data concentrates
        on a single value so the kernel window has only 2 distinct d
        values and the design-matrix columns become linearly
        dependent at higher polynomial orders.
        """
        rng = np.random.default_rng(2026)
        # G = 151: d.min=0.1 unique, 50 obs each at 0.15 / 0.3 / 0.4.
        # Modal fraction = 1/151 < 2% passes mass-point check.
        # The B1 / B2 auxiliary fits at stage d1 use h_B1 = h_B2 =
        # range = 0.3, which retains all 4 distinct values (0.1, 0.15,
        # 0.3, 0.4). The B1 design matrix has o_B+1 = 5 columns but
        # only 4 independent rows -> rank-deficient X'X -> Cholesky
        # fails in qrXXinv.
        d = np.concatenate(
            [[0.1], np.full(50, 0.15), np.full(50, 0.3), np.full(50, 0.4)]
        y = d + rng.normal(0, 0.01, size=d.size)
        with pytest.raises(ValueError, match="qrXXinv"):
            mse_optimal_bandwidth(d, y, boundary=float(d.min()))
class TestKernelDispatch:
    """Different kernels produce different bandwidths."""
    def test_kernel_epa_vs_uni_differ(self):
        rng = np.random.default_rng(42)
        d = rng.uniform(0, 1, size=G)
        y = d + d**2 + rng.normal(0, 0.5, size=G)
        h_epa = mse_optimal_bandwidth(d, y, kernel="epanechnikov")
        h_uni = mse_optimal_bandwidth(d, y, kernel="uniform")
        # Different C_rot constants (2.34 vs 1.843) -> different c_bw -> cascade.
        assert h_epa != h_uni
        assert h_epa > 0
        assert h_uni > 0
    def test_kernel_epa_vs_tri_differ(self):
        rng = np.random.default_rng(42)
        d = rng.uniform(0, 1, size=1000)
        y = d + d**2 + rng.normal(0, 0.5, size=1000)
        h_epa = mse_optimal_bandwidth(d, y, kernel="epanechnikov")
        h_tri = mse_optimal_bandwidth(d, y, kernel="triangular")
        assert h_epa != h_tri
class TestBoundary:
    def test_nonzero_boundary(self):
        """Design 1 continuous-near-d_lower case: boundary = d.min()
        (not the theoretical infimum of the support). Under the
        strict boundary-applicability check, the user must pass the
        sample minimum, not a known theoretical lower bound like 1.0
        on U(1, 2) data."""
        rng = np.random.default_rng(2026)
        d = rng.uniform(1.0, 2.0, size=1500)
        y = 3.0 + 0.5 * (d - 1.0) ** 2 + rng.normal(0, 0.3, size=1500)
        h = mse_optimal_bandwidth(d, y, boundary=float(d.min()))
        assert np.isfinite(h)
        assert h > 0.0
class TestDownstreamIntegration:
    """End-to-end: bandwidth feeds into local_linear_fit without error."""
    def test_selector_feeds_local_linear_fit(self, dgp_case):
        name, d, y, _ = dgp_case
        h = mse_optimal_bandwidth(d, y)
        fit = local_linear_fit(d, y, bandwidth=h, boundary=0.0, kernel="epanechnikov")
        # Basic sanity -- fit returned finite intercept and slope.
        assert np.isfinite(fit.intercept), name
        assert np.isfinite(fit.slope), name
        assert fit.n_effective > 0, name
class TestRateScaling:
    """h* scales as G^{-1/5} in Monte Carlo (15% MC tolerance)."""
    def test_rate_scaling_across_g(self):
        rng = np.random.default_rng(20260419)
        ratios = []
        for G in [500, 2000, 8000]:
            d = rng.uniform(0, 1, size=G)
            y = d + d**2 + rng.normal(0, 0.5, size=G)
            h = mse_optimal_bandwidth(d, y)
            ratios.append((G, h))
        # h(G_2) / h(G_1) should be approx (G_1 / G_2)^{1/5}.
        g1, h1 = ratios[0]
        g2, h2 = ratios[-1]
        expected_ratio = (g1 / g2) ** (1.0 / 5.0)
        actual_ratio = h2 / h1
        # 15% MC tolerance; rate check, not parity.
        assert actual_ratio == pytest.approx(expected_ratio, rel=0.15)
Sunbelt Computer Software

PL/B Language Development and Support

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Sunbelt Computer Software

PL/B Language Development and Support

Uh oh!

FilesExpand file tree

test_bandwidth_selector.py

Latest commit

History

test_bandwidth_selector.py

File metadata and controls
