Python sklearn.linear_model.LassoLars() Examples

The following are 24 code examples of sklearn.linear_model.LassoLars(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module sklearn.linear_model , or try the search function .
Example #1
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 6 votes vote down vote up
def test_rank_deficient_design():
    # consistency test that checks that LARS Lasso is handling rank
    # deficient input data (with n_features < rank) in the same way
    # as coordinate descent Lasso
    y = [5, 0, 5]
    for X in (
              [[5, 0],
               [0, 5],
               [10, 10]],
              [[10, 10, 0],
               [1e-32, 0, 0],
               [0, 0, 1]]
             ):
        # To be able to use the coefs to compute the objective function,
        # we need to turn off normalization
        lars = linear_model.LassoLars(.1, normalize=False)
        coef_lars_ = lars.fit(X, y).coef_
        obj_lars = (1. / (2. * 3.)
                    * linalg.norm(y - np.dot(X, coef_lars_)) ** 2
                    + .1 * linalg.norm(coef_lars_, 1))
        coord_descent = linear_model.Lasso(.1, tol=1e-6, normalize=False)
        coef_cd_ = coord_descent.fit(X, y).coef_
        obj_cd = ((1. / (2. * 3.)) * linalg.norm(y - np.dot(X, coef_cd_)) ** 2
                  + .1 * linalg.norm(coef_cd_, 1))
        assert_less(obj_lars, obj_cd * (1. + 1e-8)) 
Example #2
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 6 votes vote down vote up
def test_lasso_lars_vs_lasso_cd_early_stopping():
    # Test that LassoLars and Lasso using coordinate descent give the
    # same results when early stopping is used.
    # (test : before, in the middle, and in the last part of the path)
    alphas_min = [10, 0.9, 1e-4]

    for alpha_min in alphas_min:
        alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                       alpha_min=alpha_min)
        lasso_cd = linear_model.Lasso(fit_intercept=False, tol=1e-8)
        lasso_cd.alpha = alphas[-1]
        lasso_cd.fit(X, y)
        error = linalg.norm(lasso_path[:, -1] - lasso_cd.coef_)
        assert_less(error, 0.01)

    # same test, with normalization
    for alpha_min in alphas_min:
        alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                       alpha_min=alpha_min)
        lasso_cd = linear_model.Lasso(fit_intercept=True, normalize=True,
                                      tol=1e-8)
        lasso_cd.alpha = alphas[-1]
        lasso_cd.fit(X, y)
        error = linalg.norm(lasso_path[:, -1] - lasso_cd.coef_)
        assert_less(error, 0.01) 
Example #3
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 6 votes vote down vote up
def test_multitarget():
    # Assure that estimators receiving multidimensional y do the right thing
    Y = np.vstack([y, y ** 2]).T
    n_targets = Y.shape[1]
    estimators = [
        linear_model.LassoLars(),
        linear_model.Lars(),
        # regression test for gh-1615
        linear_model.LassoLars(fit_intercept=False),
        linear_model.Lars(fit_intercept=False),
    ]

    for estimator in estimators:
        estimator.fit(X, Y)
        Y_pred = estimator.predict(X)
        alphas, active, coef, path = (estimator.alphas_, estimator.active_,
                                      estimator.coef_, estimator.coef_path_)
        for k in range(n_targets):
            estimator.fit(X, Y[:, k])
            y_pred = estimator.predict(X)
            assert_array_almost_equal(alphas[k], estimator.alphas_)
            assert_array_almost_equal(active[k], estimator.active_)
            assert_array_almost_equal(coef[k], estimator.coef_)
            assert_array_almost_equal(path[k], estimator.coef_path_)
            assert_array_almost_equal(Y_pred[:, k], y_pred) 
Example #4
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 6 votes vote down vote up
def test_estimatorclasses_positive_constraint():
    # testing the transmissibility for the positive option of all estimator
    # classes in this same function here
    default_parameter = {'fit_intercept': False}

    estimator_parameter_map = {'LassoLars': {'alpha': 0.1},
                               'LassoLarsCV': {},
                               'LassoLarsIC': {}}
    for estname in estimator_parameter_map:
        params = default_parameter.copy()
        params.update(estimator_parameter_map[estname])
        estimator = getattr(linear_model, estname)(positive=False, **params)
        estimator.fit(X, y)
        assert estimator.coef_.min() < 0
        estimator = getattr(linear_model, estname)(positive=True, **params)
        estimator.fit(X, y)
        assert min(estimator.coef_) >= 0 
Example #5
Source File: test_least_angle.py    From twitter-stock-recommendation with MIT License 6 votes vote down vote up
def test_lasso_lars_vs_lasso_cd_early_stopping(verbose=False):
    # Test that LassoLars and Lasso using coordinate descent give the
    # same results when early stopping is used.
    # (test : before, in the middle, and in the last part of the path)
    alphas_min = [10, 0.9, 1e-4]

    for alpha_min in alphas_min:
        alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                       alpha_min=alpha_min)
        lasso_cd = linear_model.Lasso(fit_intercept=False, tol=1e-8)
        lasso_cd.alpha = alphas[-1]
        lasso_cd.fit(X, y)
        error = linalg.norm(lasso_path[:, -1] - lasso_cd.coef_)
        assert_less(error, 0.01)

    # same test, with normalization
    for alpha_min in alphas_min:
        alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                       alpha_min=alpha_min)
        lasso_cd = linear_model.Lasso(fit_intercept=True, normalize=True,
                                      tol=1e-8)
        lasso_cd.alpha = alphas[-1]
        lasso_cd.fit(X, y)
        error = linalg.norm(lasso_path[:, -1] - lasso_cd.coef_)
        assert_less(error, 0.01) 
Example #6
Source File: scikitlearn.py    From sia-cog with MIT License 6 votes vote down vote up
def getModels():
    result = []
    result.append("LinearRegression")
    result.append("BayesianRidge")
    result.append("ARDRegression")
    result.append("ElasticNet")
    result.append("HuberRegressor")
    result.append("Lasso")
    result.append("LassoLars")
    result.append("Rigid")
    result.append("SGDRegressor")
    result.append("SVR")
    result.append("MLPClassifier")
    result.append("KNeighborsClassifier")
    result.append("SVC")
    result.append("GaussianProcessClassifier")
    result.append("DecisionTreeClassifier")
    result.append("RandomForestClassifier")
    result.append("AdaBoostClassifier")
    result.append("GaussianNB")
    result.append("LogisticRegression")
    result.append("QuadraticDiscriminantAnalysis")
    return result 
Example #7
Source File: test_least_angle.py    From twitter-stock-recommendation with MIT License 6 votes vote down vote up
def test_rank_deficient_design():
    # consistency test that checks that LARS Lasso is handling rank
    # deficient input data (with n_features < rank) in the same way
    # as coordinate descent Lasso
    y = [5, 0, 5]
    for X in ([[5, 0],
               [0, 5],
               [10, 10]],

              [[10, 10, 0],
               [1e-32, 0, 0],
               [0, 0, 1]],
              ):
        # To be able to use the coefs to compute the objective function,
        # we need to turn off normalization
        lars = linear_model.LassoLars(.1, normalize=False)
        coef_lars_ = lars.fit(X, y).coef_
        obj_lars = (1. / (2. * 3.)
                    * linalg.norm(y - np.dot(X, coef_lars_)) ** 2
                    + .1 * linalg.norm(coef_lars_, 1))
        coord_descent = linear_model.Lasso(.1, tol=1e-6, normalize=False)
        coef_cd_ = coord_descent.fit(X, y).coef_
        obj_cd = ((1. / (2. * 3.)) * linalg.norm(y - np.dot(X, coef_cd_)) ** 2
                  + .1 * linalg.norm(coef_cd_, 1))
        assert_less(obj_lars, obj_cd * (1. + 1e-8)) 
Example #8
Source File: test_sklearn_grid_search_cv_converter.py    From sklearn-onnx with MIT License 6 votes vote down vote up
def test_grid_search_regressor_float(self):
        tuned_parameters = [{'alpha': np.logspace(-4, -0.5, 4)}]
        clf = GridSearchCV(LassoLars(max_iter=100),
                           tuned_parameters, cv=5)
        model, X = fit_regression_model(clf)
        model_onnx = convert_sklearn(
            model, "GridSearchCV",
            [("input", FloatTensorType([None, X.shape[1]]))])
        self.assertIsNotNone(model_onnx)
        dump_data_and_model(
            X,
            model,
            model_onnx,
            basename="SklearnGridSearchRegressionFloat-OneOffArray-Dec4",
            allow_failure="StrictVersion("
            "onnxruntime.__version__) "
            "<= StrictVersion('0.2.1') or "
            "StrictVersion(onnx.__version__) "
            "== StrictVersion('1.4.1')",
        ) 
Example #9
Source File: test_sklearn_glm_regressor_converter.py    From sklearn-onnx with MIT License 6 votes vote down vote up
def test_model_lasso_lars_bool(self):
        model, X = fit_regression_model(
            linear_model.LassoLars(), is_bool=True)
        model_onnx = convert_sklearn(
            model, "lasso lars",
            [("input", BooleanTensorType([None, X.shape[1]]))])
        self.assertIsNotNone(model_onnx)
        dump_data_and_model(
            X,
            model,
            model_onnx,
            basename="SklearnLassoLarsBool",
            allow_failure="StrictVersion("
            "onnxruntime.__version__)"
            "<= StrictVersion('0.2.1')",
        ) 
Example #10
Source File: test_sklearn_glm_regressor_converter.py    From sklearn-onnx with MIT License 5 votes vote down vote up
def test_model_lasso_lars(self):
        model, X = fit_regression_model(linear_model.LassoLars(alpha=0.01))
        model_onnx = convert_sklearn(
            model, "lasso lars",
            [("input", FloatTensorType([None, X.shape[1]]))])
        self.assertIsNotNone(model_onnx)
        dump_data_and_model(
            X,
            model,
            model_onnx,
            basename="SklearnLassoLars-Dec4",
            allow_failure="StrictVersion("
            "onnxruntime.__version__)"
            "<= StrictVersion('0.2.1')",
        ) 
Example #11
Source File: test_least_angle.py    From twitter-stock-recommendation with MIT License 5 votes vote down vote up
def test_multitarget():
    # Assure that estimators receiving multidimensional y do the right thing
    X = diabetes.data
    Y = np.vstack([diabetes.target, diabetes.target ** 2]).T
    n_targets = Y.shape[1]
    estimators = [
        linear_model.LassoLars(),
        linear_model.Lars(),
        # regression test for gh-1615
        linear_model.LassoLars(fit_intercept=False),
        linear_model.Lars(fit_intercept=False),
    ]

    for estimator in estimators:
        estimator.fit(X, Y)
        Y_pred = estimator.predict(X)
        alphas, active, coef, path = (estimator.alphas_, estimator.active_,
                                      estimator.coef_, estimator.coef_path_)
        for k in range(n_targets):
            estimator.fit(X, Y[:, k])
            y_pred = estimator.predict(X)
            assert_array_almost_equal(alphas[k], estimator.alphas_)
            assert_array_almost_equal(active[k], estimator.active_)
            assert_array_almost_equal(coef[k], estimator.coef_)
            assert_array_almost_equal(path[k], estimator.coef_path_)
            assert_array_almost_equal(Y_pred[:, k], y_pred) 
Example #12
Source File: test_least_angle.py    From twitter-stock-recommendation with MIT License 5 votes vote down vote up
def test_lasso_lars_path_length():
    # Test that the path length of the LassoLars is right
    lasso = linear_model.LassoLars()
    lasso.fit(X, y)
    lasso2 = linear_model.LassoLars(alpha=lasso.alphas_[2])
    lasso2.fit(X, y)
    assert_array_almost_equal(lasso.alphas_[:3], lasso2.alphas_)
    # Also check that the sequence of alphas is always decreasing
    assert_true(np.all(np.diff(lasso.alphas_) < 0)) 
Example #13
Source File: test_least_angle.py    From twitter-stock-recommendation with MIT License 5 votes vote down vote up
def test_lasso_lars_vs_lasso_cd(verbose=False):
    # Test that LassoLars and Lasso using coordinate descent give the
    # same results.
    X = 3 * diabetes.data

    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso')
    lasso_cd = linear_model.Lasso(fit_intercept=False, tol=1e-8)
    for c, a in zip(lasso_path.T, alphas):
        if a == 0:
            continue
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01)

    # similar test, with the classifiers
    for alpha in np.linspace(1e-2, 1 - 1e-2, 20):
        clf1 = linear_model.LassoLars(alpha=alpha, normalize=False).fit(X, y)
        clf2 = linear_model.Lasso(alpha=alpha, tol=1e-8,
                                  normalize=False).fit(X, y)
        err = linalg.norm(clf1.coef_ - clf2.coef_)
        assert_less(err, 1e-3)

    # same test, with normalized data
    X = diabetes.data
    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso')
    lasso_cd = linear_model.Lasso(fit_intercept=False, normalize=True,
                                  tol=1e-8)
    for c, a in zip(lasso_path.T, alphas):
        if a == 0:
            continue
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01) 
Example #14
Source File: test_least_angle.py    From twitter-stock-recommendation with MIT License 5 votes vote down vote up
def test_lars_lstsq():
    # Test that Lars gives least square solution at the end
    # of the path
    X1 = 3 * diabetes.data  # use un-normalized dataset
    clf = linear_model.LassoLars(alpha=0.)
    clf.fit(X1, y)
    coef_lstsq = np.linalg.lstsq(X1, y)[0]
    assert_array_almost_equal(clf.coef_, coef_lstsq) 
Example #15
Source File: test_linear_model.py    From pandas-ml with BSD 3-Clause "New" or "Revised" License 5 votes vote down vote up
def test_objectmapper(self):
        df = pdml.ModelFrame([])
        self.assertIs(df.linear_model.ARDRegression, lm.ARDRegression)
        self.assertIs(df.linear_model.BayesianRidge, lm.BayesianRidge)
        self.assertIs(df.linear_model.ElasticNet, lm.ElasticNet)
        self.assertIs(df.linear_model.ElasticNetCV, lm.ElasticNetCV)

        self.assertIs(df.linear_model.HuberRegressor, lm.HuberRegressor)

        self.assertIs(df.linear_model.Lars, lm.Lars)
        self.assertIs(df.linear_model.LarsCV, lm.LarsCV)
        self.assertIs(df.linear_model.Lasso, lm.Lasso)
        self.assertIs(df.linear_model.LassoCV, lm.LassoCV)
        self.assertIs(df.linear_model.LassoLars, lm.LassoLars)
        self.assertIs(df.linear_model.LassoLarsCV, lm.LassoLarsCV)
        self.assertIs(df.linear_model.LassoLarsIC, lm.LassoLarsIC)

        self.assertIs(df.linear_model.LinearRegression, lm.LinearRegression)
        self.assertIs(df.linear_model.LogisticRegression, lm.LogisticRegression)
        self.assertIs(df.linear_model.LogisticRegressionCV, lm.LogisticRegressionCV)
        self.assertIs(df.linear_model.MultiTaskLasso, lm.MultiTaskLasso)
        self.assertIs(df.linear_model.MultiTaskElasticNet, lm.MultiTaskElasticNet)
        self.assertIs(df.linear_model.MultiTaskLassoCV, lm.MultiTaskLassoCV)
        self.assertIs(df.linear_model.MultiTaskElasticNetCV, lm.MultiTaskElasticNetCV)

        self.assertIs(df.linear_model.OrthogonalMatchingPursuit, lm.OrthogonalMatchingPursuit)
        self.assertIs(df.linear_model.OrthogonalMatchingPursuitCV, lm.OrthogonalMatchingPursuitCV)
        self.assertIs(df.linear_model.PassiveAggressiveClassifier, lm.PassiveAggressiveClassifier)
        self.assertIs(df.linear_model.PassiveAggressiveRegressor, lm.PassiveAggressiveRegressor)

        self.assertIs(df.linear_model.Perceptron, lm.Perceptron)
        self.assertIs(df.linear_model.RandomizedLasso, lm.RandomizedLasso)
        self.assertIs(df.linear_model.RandomizedLogisticRegression, lm.RandomizedLogisticRegression)
        self.assertIs(df.linear_model.RANSACRegressor, lm.RANSACRegressor)
        self.assertIs(df.linear_model.Ridge, lm.Ridge)
        self.assertIs(df.linear_model.RidgeClassifier, lm.RidgeClassifier)
        self.assertIs(df.linear_model.RidgeClassifierCV, lm.RidgeClassifierCV)
        self.assertIs(df.linear_model.RidgeCV, lm.RidgeCV)
        self.assertIs(df.linear_model.SGDClassifier, lm.SGDClassifier)
        self.assertIs(df.linear_model.SGDRegressor, lm.SGDRegressor)
        self.assertIs(df.linear_model.TheilSenRegressor, lm.TheilSenRegressor) 
Example #16
Source File: test_sklearn_glm_regressor_converter.py    From sklearn-onnx with MIT License 5 votes vote down vote up
def test_model_lasso_lars_int(self):
        model, X = fit_regression_model(linear_model.LassoLars(), is_int=True)
        model_onnx = convert_sklearn(
            model, "lasso lars",
            [("input", Int64TensorType([None, X.shape[1]]))])
        self.assertIsNotNone(model_onnx)
        dump_data_and_model(
            X,
            model,
            model_onnx,
            basename="SklearnLassoLarsInt-Dec4",
            allow_failure="StrictVersion("
            "onnxruntime.__version__)"
            "<= StrictVersion('0.2.1')",
        ) 
Example #17
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 5 votes vote down vote up
def test_lars_lstsq():
    # Test that Lars gives least square solution at the end
    # of the path
    X1 = 3 * X  # use un-normalized dataset
    clf = linear_model.LassoLars(alpha=0.)
    clf.fit(X1, y)
    # Avoid FutureWarning about default value change when numpy >= 1.14
    rcond = None if LooseVersion(np.__version__) >= '1.14' else -1
    coef_lstsq = np.linalg.lstsq(X1, y, rcond=rcond)[0]
    assert_array_almost_equal(clf.coef_, coef_lstsq) 
Example #18
Source File: cnmf.py    From minian with GNU General Public License v3.0 5 votes vote down vote up
def update_spatial_perpx(y, alpha, sub, C):
    res = np.zeros_like(sub, dtype=y.dtype)
    if np.sum(sub) > 0:
        C = C[:, sub]
        clf = LassoLars(alpha=alpha, positive=True)
        coef = clf.fit(C, y).coef_
        res[np.where(sub)[0]] = coef
    return res 
Example #19
Source File: MCFS.py    From fsfc with MIT License 5 votes vote down vote up
def _create_regressor(self):
        if self.mode == 'default':
            return Lars()
        if self.mode == 'lasso':
            return LassoLars(alpha=self.alpha)
        raise ValueError('Unexpected mode ' + self.mode + '. Expected "default" or "lasso"') 
Example #20
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 5 votes vote down vote up
def test_lasso_lars_path_length():
    # Test that the path length of the LassoLars is right
    lasso = linear_model.LassoLars()
    lasso.fit(X, y)
    lasso2 = linear_model.LassoLars(alpha=lasso.alphas_[2])
    lasso2.fit(X, y)
    assert_array_almost_equal(lasso.alphas_[:3], lasso2.alphas_)
    # Also check that the sequence of alphas is always decreasing
    assert np.all(np.diff(lasso.alphas_) < 0) 
Example #21
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 5 votes vote down vote up
def test_lasso_lars_vs_lasso_cd():
    # Test that LassoLars and Lasso using coordinate descent give the
    # same results.
    X = 3 * diabetes.data

    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso')
    lasso_cd = linear_model.Lasso(fit_intercept=False, tol=1e-8)
    for c, a in zip(lasso_path.T, alphas):
        if a == 0:
            continue
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01)

    # similar test, with the classifiers
    for alpha in np.linspace(1e-2, 1 - 1e-2, 20):
        clf1 = linear_model.LassoLars(alpha=alpha, normalize=False).fit(X, y)
        clf2 = linear_model.Lasso(alpha=alpha, tol=1e-8,
                                  normalize=False).fit(X, y)
        err = linalg.norm(clf1.coef_ - clf2.coef_)
        assert_less(err, 1e-3)

    # same test, with normalized data
    X = diabetes.data
    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso')
    lasso_cd = linear_model.Lasso(fit_intercept=False, normalize=True,
                                  tol=1e-8)
    for c, a in zip(lasso_path.T, alphas):
        if a == 0:
            continue
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01) 
Example #22
Source File: scikitlearn.py    From sia-cog with MIT License 4 votes vote down vote up
def getSKLearnModel(modelName):
    if modelName == 'LinearRegression':
        model = linear_model.LinearRegression()
    elif modelName == 'BayesianRidge':
        model = linear_model.BayesianRidge()
    elif modelName == 'ARDRegression':
        model = linear_model.ARDRegression()
    elif modelName == 'ElasticNet':
        model = linear_model.ElasticNet()
    elif modelName == 'HuberRegressor':
        model = linear_model.HuberRegressor()
    elif modelName == 'Lasso':
        model = linear_model.Lasso()
    elif modelName == 'LassoLars':
        model = linear_model.LassoLars()
    elif modelName == 'Rigid':
        model = linear_model.Ridge()
    elif modelName == 'SGDRegressor':
        model = linear_model.SGDRegressor()
    elif modelName == 'SVR':
        model = SVR()
    elif modelName=='MLPClassifier':
        model = MLPClassifier()
    elif modelName=='KNeighborsClassifier':
        model = KNeighborsClassifier()
    elif modelName=='SVC':
        model = SVC()
    elif modelName=='GaussianProcessClassifier':
        model = GaussianProcessClassifier()
    elif modelName=='DecisionTreeClassifier':
        model = DecisionTreeClassifier()
    elif modelName=='RandomForestClassifier':
        model = RandomForestClassifier()
    elif modelName=='AdaBoostClassifier':
        model = AdaBoostClassifier()
    elif modelName=='GaussianNB':
        model = GaussianNB()
    elif modelName=='LogisticRegression':
        model = linear_model.LogisticRegression()
    elif modelName=='QuadraticDiscriminantAnalysis':
        model = QuadraticDiscriminantAnalysis()

    return model 
Example #23
Source File: test_least_angle.py    From Mastering-Elasticsearch-7.0 with MIT License 4 votes vote down vote up
def test_lasso_lars_vs_lasso_cd_positive():
    # Test that LassoLars and Lasso using coordinate descent give the
    # same results when using the positive option

    # This test is basically a copy of the above with additional positive
    # option. However for the middle part, the comparison of coefficient values
    # for a range of alphas, we had to make an adaptations. See below.

    # not normalized data
    X = 3 * diabetes.data

    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                   positive=True)
    lasso_cd = linear_model.Lasso(fit_intercept=False, tol=1e-8, positive=True)
    for c, a in zip(lasso_path.T, alphas):
        if a == 0:
            continue
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01)

    # The range of alphas chosen for coefficient comparison here is restricted
    # as compared with the above test without the positive option. This is due
    # to the circumstance that the Lars-Lasso algorithm does not converge to
    # the least-squares-solution for small alphas, see 'Least Angle Regression'
    # by Efron et al 2004. The coefficients are typically in congruence up to
    # the smallest alpha reached by the Lars-Lasso algorithm and start to
    # diverge thereafter.  See
    # https://gist.github.com/michigraber/7e7d7c75eca694c7a6ff

    for alpha in np.linspace(6e-1, 1 - 1e-2, 20):
        clf1 = linear_model.LassoLars(fit_intercept=False, alpha=alpha,
                                      normalize=False, positive=True).fit(X, y)
        clf2 = linear_model.Lasso(fit_intercept=False, alpha=alpha, tol=1e-8,
                                  normalize=False, positive=True).fit(X, y)
        err = linalg.norm(clf1.coef_ - clf2.coef_)
        assert_less(err, 1e-3)

    # normalized data
    X = diabetes.data
    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                   positive=True)
    lasso_cd = linear_model.Lasso(fit_intercept=False, normalize=True,
                                  tol=1e-8, positive=True)
    for c, a in zip(lasso_path.T[:-1], alphas[:-1]):  # don't include alpha=0
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01) 
Example #24
Source File: test_least_angle.py    From twitter-stock-recommendation with MIT License 4 votes vote down vote up
def test_lasso_lars_vs_lasso_cd_positive(verbose=False):
    # Test that LassoLars and Lasso using coordinate descent give the
    # same results when using the positive option

    # This test is basically a copy of the above with additional positive
    # option. However for the middle part, the comparison of coefficient values
    # for a range of alphas, we had to make an adaptations. See below.

    # not normalized data
    X = 3 * diabetes.data

    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                   positive=True)
    lasso_cd = linear_model.Lasso(fit_intercept=False, tol=1e-8, positive=True)
    for c, a in zip(lasso_path.T, alphas):
        if a == 0:
            continue
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01)

    # The range of alphas chosen for coefficient comparison here is restricted
    # as compared with the above test without the positive option. This is due
    # to the circumstance that the Lars-Lasso algorithm does not converge to
    # the least-squares-solution for small alphas, see 'Least Angle Regression'
    # by Efron et al 2004. The coefficients are typically in congruence up to
    # the smallest alpha reached by the Lars-Lasso algorithm and start to
    # diverge thereafter.  See
    # https://gist.github.com/michigraber/7e7d7c75eca694c7a6ff

    for alpha in np.linspace(6e-1, 1 - 1e-2, 20):
        clf1 = linear_model.LassoLars(fit_intercept=False, alpha=alpha,
                                      normalize=False, positive=True).fit(X, y)
        clf2 = linear_model.Lasso(fit_intercept=False, alpha=alpha, tol=1e-8,
                                  normalize=False, positive=True).fit(X, y)
        err = linalg.norm(clf1.coef_ - clf2.coef_)
        assert_less(err, 1e-3)

    # normalized data
    X = diabetes.data
    alphas, _, lasso_path = linear_model.lars_path(X, y, method='lasso',
                                                   positive=True)
    lasso_cd = linear_model.Lasso(fit_intercept=False, normalize=True,
                                  tol=1e-8, positive=True)
    for c, a in zip(lasso_path.T[:-1], alphas[:-1]):  # don't include alpha=0
        lasso_cd.alpha = a
        lasso_cd.fit(X, y)
        error = linalg.norm(c - lasso_cd.coef_)
        assert_less(error, 0.01)