How to use assert_between method in Testify

Best Python code snippet using Testify_python

test_stacking_models.py

Source:test_stacking_models.py

...89        fit_2gaussian(vels, twogauss_spectrum)90    actual_params = [0.5, 0.0, 0.05, 0.5, 0.3]91    for fit, act, err in zip(parvals, actual_params, parerrs):92        npt.assert_allclose(act, fit, atol=2 * err)93def assert_between(act, val, low_err, up_err):94    assert (act >= val - low_err) & (act <= val + up_err)95def test_hwhm_model():96    vels = np.linspace(-1, 1, 100)97    model, gauss, lor = \98        gauss_with_wings_noise(vels, 1., 0.0, 0.1, 0.0, return_models=True)99    noisy_model = \100        gauss_with_wings_noise(vels, 1., 0.0, 0.1, 0.01)101    parvals, parerrs, param_names, hwhm_gauss = \102        fit_hwhm(vels, noisy_model, sigma_noise=0.01,103                 niters=100)104    parvals_clean, parerrs_clean = \105        fit_hwhm(vels, model, sigma_noise=0.01,106                 niters=100)[:2]107    f_wings = np.sum((model - gauss)) / np.sum(model)108    sigma_wings = np.sqrt(np.sum([(m - g) * v**2 for m, g, v in109                                  zip(model, gauss, vels)]) /110                          np.sum((model - gauss)))111    asymm = 0.0112    kappa = 0.0113    actual_params = [0.1, 0.0, f_wings, sigma_wings, asymm, kappa]114    # Compare the clean parameter estimates115    for act, val, low, up in zip(actual_params, parvals_clean, *parerrs_clean):116        assert_between(act, val, low, up)117    # Compare the noisy parameter estimates118    for act, val, low, up in zip(actual_params, parvals, *parerrs):119        assert_between(act, val, low, up)120def test_hwhm_model_wgauss():121    vels = np.linspace(-1, 1, 100)122    model, gauss = \123        gauss_with_noise(vels, 1., 0.0, 0.1, 0.0, return_models=True)124    noisy_model = \125        gauss_with_noise(vels, 1., 0.0, 0.1, 0.01)126    parvals, parerrs, param_names, hwhm_gauss = \127        fit_hwhm(vels, noisy_model, sigma_noise=0.01,128                 niters=100)129    parvals_clean, parerrs_clean = \130        fit_hwhm(vels, model, sigma_noise=0.01,131                 niters=100)[:2]132    f_wings = 0.0133    sigma_wings = 0.0134    asymm = 0.0135    kappa = 0.0136    actual_params = [0.1, 0.0, f_wings, sigma_wings, asymm, kappa]137    names = ['sigma', 'vcen', 'fwing', 'sig_wing', 'asymm', 'kappa']138    # Compare the clean parameter estimates139    for name, act, val, low, up in zip(names, actual_params, parvals_clean,140                                       *parerrs_clean):141        if name == 'sig_wing':142            continue143        assert_between(act, val, low, up)144    # Compare the noisy parameter estimates145    for name, act, val, low, up in zip(names, actual_params, parvals,146                                       *parerrs):147        if name == 'sig_wing':148            continue149        assert_between(act, val, low, up)150def test_hwhm_model_asymmwings():151    vels = np.linspace(-1, 1, 100)152    model, gauss, lor = \153        gauss_with_wings_noise(vels, 1., 0.0, 0.1, 0.0, return_models=True,154                               asymm=-0.85)155    noisy_model = \156        gauss_with_wings_noise(vels, 1., 0.0, 0.1, 0.01,157                               asymm=-0.85)158    parvals, parerrs, param_names, hwhm_gauss = \159        fit_hwhm(vels, noisy_model, sigma_noise=0.01,160                 niters=100)161    parvals_clean, parerrs_clean = \162        fit_hwhm(vels, model, sigma_noise=0.01,163                 niters=100)[:2]164    high_mask = vels > 0.0 + 0.1 * np.sqrt(2 * np.log(2))165    low_mask = vels < 0.0 - 0.1 * np.sqrt(2 * np.log(2))166    # peak_high_mask = np.logical_and(vels > 0.,167    #                                 vels < 0.0 + 0.1 * np.sqrt(2 * np.log(2)))168    # peak_low_mask = np.logical_and(vels < 0.,169    #                                vels > 0.0 - 0.1 * np.sqrt(2 * np.log(2)))170    f_wings = (np.sum(model[low_mask] - gauss[low_mask]) +171               np.sum(model[high_mask] - gauss[high_mask])) / np.sum(model)172    sigma_wings = np.NaN173    pos_mask = vels > 0174    neg_mask = vels < 0175    asymm = np.sum(model[neg_mask] - model[pos_mask]) / np.sum(model)176    kappa = 0.0177    actual_params = [0.1, 0.0, f_wings, sigma_wings, asymm, kappa]178    names = ['sigma', 'vcen', 'fwing', 'sig_wing', 'asymm', 'kappa']179    # Not comparing the sigma_wings b/c we aren't using it in any analysis180    # and the definition doesn't seem well-suited to substantial asymmetry181    # Compare the clean parameter estimates182    for name, act, val, low, up in zip(names, actual_params, parvals_clean,183                                       *parerrs_clean):184        if name == 'sig_wing':185            continue186        assert_between(act, val, low, up)187    # Compare the noisy parameter estimates188    for name, act, val, low, up in zip(names, actual_params, parvals,189                                       *parerrs):190        if name == 'sig_wing':191            continue192        assert_between(act, val, low, up)193def test_symm_asymm_fwing():194    '''195    Use above asymmetric setup.196    '''197    vels = np.linspace(-1, 1, 100)198    model, gauss, lor = \199        gauss_with_wings_noise(vels, 1., 0.0, 0.1, 0.0, return_models=True,200                               asymm=-0.85)201    noisy_model = \202        gauss_with_wings_noise(vels, 1., 0.0, 0.1, 0.01,203                               asymm=-0.85)204    s_model = model205    n_model = model[::-1]206    tot_model = n_model + s_model207    params_clean = find_linewing_asymm(vels, n_model, s_model)208    s_model_noisy = noisy_model209    n_model_noisy = noisy_model[::-1]210    params, low_lim, up_lim = \211        find_linewing_asymm(vels, n_model_noisy, s_model_noisy,212                            niters=100,213                            sigma_noise_n=0.01,214                            sigma_noise_s=0.01)215    high_mask = vels > 0.0 + 0.1 * np.sqrt(2 * np.log(2))216    low_mask = vels < 0.0 - 0.1 * np.sqrt(2 * np.log(2))217    # We added the model twice, so multiple gauss by 2218    f_wings = (np.sum(tot_model[low_mask] - 2 * gauss[low_mask]) +219               np.sum(tot_model[high_mask] - 2 * gauss[high_mask])) / \220        np.sum(tot_model)221    blue_excess = np.sum(s_model[low_mask] - n_model[low_mask])222    red_excess = np.sum(n_model[high_mask] - s_model[high_mask])223    f_asymm = (blue_excess + red_excess) / np.sum(tot_model)224    f_symm = f_wings - f_asymm225    npt.assert_allclose(f_wings, params_clean[0], atol=0.005)226    assert_between(f_wings, params[0], low_lim[0], up_lim[0])227    npt.assert_allclose(f_symm, params_clean[1], atol=0.005)228    assert_between(f_symm, params[1], low_lim[1], up_lim[1])229    npt.assert_allclose(f_asymm, params_clean[2], atol=0.005)230    # Errors seem to be a bit more asymmetric for f_asymm. Allow231    # a larger range232    avg_sig = 0.5 * ((params[2] - low_lim[2]) + (up_lim[2] - params[2]))233    assert_between(f_asymm, params[2], low_lim[2] - 0.5 * avg_sig,...

test_pca.py

Source:test_pca.py

...26    slice_used = slice(0, tester.n_eigs)27    npt.assert_allclose(tester.eigvals[slice_used],28                        computed_data['pca_val'][slice_used])29    fit_values = computed_data["pca_fit_vals"].reshape(-1)[0]30    assert_between(fit_values["index"], tester.index_error_range[0],31                   tester.index_error_range[1])32    assert_between(fit_values["gamma"], tester.gamma_error_range[0],33                   tester.gamma_error_range[1])34    assert_between(fit_values["intercept"].value,35                   tester.intercept_error_range[0].value,36                   tester.intercept_error_range[1].value)37    assert_between(fit_values["sonic_length"].value,38                   tester.sonic_length()[1][0].value,39                   tester.sonic_length()[1][1].value)40@pytest.mark.skipif("not EMCEE_INSTALLED")41def test_PCA_method_w_bayes():42    tester = PCA(dataset1["cube"])43    tester.run(mean_sub=True, n_eigs=50,44               spatial_method='contour',45               spectral_method='walk-down',46               fit_method='bayes', brunt_beamcorrect=False)47    slice_used = slice(0, tester.n_eigs)48    npt.assert_allclose(tester.eigvals[slice_used],49                        computed_data['pca_val'][slice_used])50    fit_values = computed_data["pca_fit_vals"].reshape(-1)[0]51    assert_between(fit_values["index_bayes"], tester.index_error_range[0],52                   tester.index_error_range[1])53    assert_between(fit_values["gamma_bayes"], tester.gamma_error_range[0],54                   tester.gamma_error_range[1])55    assert_between(fit_values["intercept_bayes"].value,56                   tester.intercept_error_range[0].value,57                   tester.intercept_error_range[1].value)58    assert_between(fit_values["sonic_length_bayes"].value,59                   tester.sonic_length()[1][0].value,60                   tester.sonic_length()[1][1].value)61@pytest.mark.parametrize(("method", "min_eigval"),62                         [("proportion", 0.99), ("value", 0.001)])63def test_PCA_auto_n_eigs(method, min_eigval):64    tester = PCA(dataset1["cube"])65    tester.run(mean_sub=True, n_eigs='auto', min_eigval=min_eigval,66               eigen_cut_method=method, decomp_only=True)67    fit_values = computed_data["pca_fit_vals"].reshape(-1)[0]68    assert tester.n_eigs == fit_values["n_eigs_" + method]69def test_PCA_distance():70    tester_dist = \71        PCA_Distance(dataset1["cube"],72                     dataset2["cube"]).distance_metric()...

Automation Testing Tutorials

Learn to execute automation testing from scratch with LambdaTest Learning Hub. Right from setting up the prerequisites to run your first automation test, to following best practices and diving deeper into advanced test scenarios. LambdaTest Learning Hubs compile a list of step-by-step guides to help you be proficient with different test automation frameworks i.e. Selenium, Cypress, TestNG etc.