diff --git a/GPy/testing/inference_tests.py b/GPy/testing/inference_tests.py index 58ee0288..7a091589 100644 --- a/GPy/testing/inference_tests.py +++ b/GPy/testing/inference_tests.py @@ -28,7 +28,10 @@ class InferenceXTestCase(unittest.TestCase): def test_inferenceX_BGPLVM_RBF(self): Ys = self.genData() m = GPy.models.BayesianGPLVM(Ys,3,kernel=GPy.kern.RBF(3,ARD=True)) - m.optimize() + import warnings + with warnings.catch_warnings(): + warnings.simplefilter("ignore") + m.optimize() x, mi = m.infer_newX(m.Y, optimize=True) np.testing.assert_array_almost_equal(m.X.mean, mi.X.mean, decimal=2) np.testing.assert_array_almost_equal(m.X.variance, mi.X.variance, decimal=2) diff --git a/GPy/testing/plotting_tests.py b/GPy/testing/plotting_tests.py index a680fa6d..179bf2e0 100644 --- a/GPy/testing/plotting_tests.py +++ b/GPy/testing/plotting_tests.py @@ -101,40 +101,42 @@ def test_figure(): matplotlib.rcParams.update(matplotlib.rcParamsDefault) matplotlib.rcParams[u'figure.figsize'] = (4,3) matplotlib.rcParams[u'text.usetex'] = False + import warnings + with warnings.catch_warnings(): + warnings.simplefilter("ignore") - ax, _ = pl().new_canvas(num=1) - def test_func(x): - return x[:, 0].reshape(3,3) - pl().imshow_interact(ax, test_func, extent=(-1,1,-1,1), resolution=3) + ax, _ = pl().new_canvas(num=1) + def test_func(x): + return x[:, 0].reshape(3,3) + pl().imshow_interact(ax, test_func, extent=(-1,1,-1,1), resolution=3) - ax, _ = pl().new_canvas() - def test_func_2(x): - y = x[:, 0].reshape(3,3) - anno = np.argmax(x, axis=1).reshape(3,3) - return y, anno - pl().annotation_heatmap_interact(ax, test_func_2, extent=(-1,1,-1,1), resolution=3) - pl().annotation_heatmap_interact(ax, test_func_2, extent=(-1,1,-1,1), resolution=3, imshow_kwargs=dict(interpolation='nearest')) + ax, _ = pl().new_canvas() + def test_func_2(x): + y = x[:, 0].reshape(3,3) + anno = np.argmax(x, axis=1).reshape(3,3) + return y, anno + + pl().annotation_heatmap_interact(ax, test_func_2, extent=(-1,1,-1,1), resolution=3) + pl().annotation_heatmap_interact(ax, test_func_2, extent=(-1,1,-1,1), resolution=3, imshow_kwargs=dict(interpolation='nearest')) - ax, _ = pl().new_canvas(figsize=(4,3)) - x = np.linspace(0,1,100) - y = [0,1,2] - array = np.array([.4,.5]) - cmap = matplotlib.colors.LinearSegmentedColormap.from_list('WhToColor', ('r', 'b'), N=array.size) - pl().fill_gradient(ax, x, y, facecolors=['r', 'g'], array=array, cmap=cmap) - try: - pl().show_canvas(ax, tight_layout=True) - except: - # macosx tight layout not stable - pl().show_canvas(ax, tight_layout=False) + ax, _ = pl().new_canvas(figsize=(4,3)) + x = np.linspace(0,1,100) + y = [0,1,2] + array = np.array([.4,.5]) + cmap = matplotlib.colors.LinearSegmentedColormap.from_list('WhToColor', ('r', 'b'), N=array.size) + + pl().fill_gradient(ax, x, y, facecolors=['r', 'g'], array=array, cmap=cmap) - ax, _ = pl().new_canvas(num=4, figsize=(4,3), projection='3d', xlabel='x', ylabel='y', zlabel='z', title='awsome title', xlim=(-1,1), ylim=(-1,1), zlim=(-3,3)) - z = 2-np.abs(np.linspace(-2,2,(100)))+1 - x, y = z*np.sin(np.linspace(-2*np.pi,2*np.pi,(100))), z*np.cos(np.linspace(-np.pi,np.pi,(100))) - pl().plot(ax, x, y, z, linewidth=2) - for do_test in _image_comparison( - baseline_images=['coverage_{}'.format(sub) for sub in ["imshow_interact",'annotation_interact','gradient','3d_plot',]], - extensions=extensions): - yield (do_test, ) + ax, _ = pl().new_canvas(num=4, figsize=(4,3), projection='3d', xlabel='x', ylabel='y', zlabel='z', title='awsome title', xlim=(-1,1), ylim=(-1,1), zlim=(-3,3)) + z = 2-np.abs(np.linspace(-2,2,(100)))+1 + x, y = z*np.sin(np.linspace(-2*np.pi,2*np.pi,(100))), z*np.cos(np.linspace(-np.pi,np.pi,(100))) + + pl().plot(ax, x, y, z, linewidth=2) + + for do_test in _image_comparison( + baseline_images=['coverage_{}'.format(sub) for sub in ["imshow_interact",'annotation_interact','gradient','3d_plot',]], + extensions=extensions): + yield (do_test, ) def test_kernel(): @@ -143,19 +145,22 @@ def test_kernel(): matplotlib.rcParams.update(matplotlib.rcParamsDefault) matplotlib.rcParams[u'figure.figsize'] = (4,3) matplotlib.rcParams[u'text.usetex'] = False - k = GPy.kern.RBF(5, ARD=True) * GPy.kern.Linear(3, active_dims=[0,2,4], ARD=True) + GPy.kern.Bias(2) - k.randomize() - k2 = GPy.kern.RBF(5, ARD=True) * GPy.kern.Linear(3, active_dims=[0,2,4], ARD=True) + GPy.kern.Bias(2) + GPy.kern.White(4) - k2[:-1] = k[:] - k2.plot_ARD(['rbf', 'linear', 'bias'], legend=True) - k2.plot_covariance(visible_dims=[0, 3], plot_limits=(-1,3)) - k2.plot_covariance(visible_dims=[2], plot_limits=(-1, 3)) - k2.plot_covariance(visible_dims=[2, 4], plot_limits=((-1, 0), (5, 3)), projection='3d') - k2.plot_covariance(visible_dims=[1, 4]) - for do_test in _image_comparison( - baseline_images=['kern_{}'.format(sub) for sub in ["ARD", 'cov_2d', 'cov_1d', 'cov_3d', 'cov_no_lim']], - extensions=extensions): - yield (do_test, ) + import warnings + with warnings.catch_warnings(): + warnings.simplefilter("ignore") + k = GPy.kern.RBF(5, ARD=True) * GPy.kern.Linear(3, active_dims=[0,2,4], ARD=True) + GPy.kern.Bias(2) + k.randomize() + k2 = GPy.kern.RBF(5, ARD=True) * GPy.kern.Linear(3, active_dims=[0,2,4], ARD=True) + GPy.kern.Bias(2) + GPy.kern.White(4) + k2[:-1] = k[:] + k2.plot_ARD(['rbf', 'linear', 'bias'], legend=True) + k2.plot_covariance(visible_dims=[0, 3], plot_limits=(-1,3)) + k2.plot_covariance(visible_dims=[2], plot_limits=(-1, 3)) + k2.plot_covariance(visible_dims=[2, 4], plot_limits=((-1, 0), (5, 3)), projection='3d') + k2.plot_covariance(visible_dims=[1, 4]) + for do_test in _image_comparison( + baseline_images=['kern_{}'.format(sub) for sub in ["ARD", 'cov_2d', 'cov_1d', 'cov_3d', 'cov_no_lim']], + extensions=extensions): + yield (do_test, ) def test_plot(): np.random.seed(111) @@ -163,18 +168,21 @@ def test_plot(): matplotlib.rcParams.update(matplotlib.rcParamsDefault) matplotlib.rcParams[u'figure.figsize'] = (4,3) matplotlib.rcParams[u'text.usetex'] = False - X = np.random.uniform(-2, 2, (40, 1)) - f = .2 * np.sin(1.3*X) + 1.3*np.cos(2*X) - Y = f+np.random.normal(0, .1, f.shape) - m = GPy.models.SparseGPRegression(X, Y, X_variance=np.ones_like(X)*[0.06]) - #m.optimize() - m.plot_data() - m.plot_mean() - m.plot_confidence() - m.plot_density() - m.plot_errorbars_trainset() - m.plot_samples() - m.plot_data_error() + import warnings + with warnings.catch_warnings(): + warnings.simplefilter("ignore") + X = np.random.uniform(-2, 2, (40, 1)) + f = .2 * np.sin(1.3*X) + 1.3*np.cos(2*X) + Y = f+np.random.normal(0, .1, f.shape) + m = GPy.models.SparseGPRegression(X, Y, X_variance=np.ones_like(X)*[0.06]) + #m.optimize() + m.plot_data() + m.plot_mean() + m.plot_confidence() + m.plot_density() + m.plot_errorbars_trainset() + m.plot_samples() + m.plot_data_error() for do_test in _image_comparison(baseline_images=['gp_{}'.format(sub) for sub in ["data", "mean", 'conf', 'density', 'out_error', diff --git a/GPy/testing/plotting_tests/baseline/coverage_gradient.png b/GPy/testing/plotting_tests/baseline/coverage_gradient.png index de5fb4f3..60bd7fb9 100644 Binary files a/GPy/testing/plotting_tests/baseline/coverage_gradient.png and b/GPy/testing/plotting_tests/baseline/coverage_gradient.png differ diff --git a/travis_tests.py b/travis_tests.py index d034fcfd..8252ec36 100644 --- a/travis_tests.py +++ b/travis_tests.py @@ -35,5 +35,5 @@ import matplotlib matplotlib.use('agg') import nose -nose.main('GPy', defaultTest='GPy/testing/') +nose.main('GPy', defaultTest='GPy/testing/', verbose=10)