apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-09 20:12:38 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

some cleaning on WarpedGP code

Browse source
This commit is contained in:
beckdaniel 2015-08-05 19:18:25 +01:00
parent 96441382b9
commit 20a1e8d3c7
2 changed files with 6 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

10
GPy/models/warped_gp.py
View file

@ -16,7 +16,7 @@ class WarpedGP(GP):
if warping_function == None: if warping_function == None:
self.warping_function = TanhWarpingFunction_d(warping_terms) self.warping_function = TanhWarpingFunction_d(warping_terms)
self.warping_params = (np.random.randn(self.warping_function.n_terms * 3 + 1,) * 1) self.warping_params = (np.random.randn(self.warping_function.n_terms * 3 + 1) * 1)
else: else:
self.warping_function = warping_function self.warping_function = warping_function
@ -56,7 +56,6 @@ class WarpedGP(GP):
self.warping_function.psi.gradient[:] = warping_grads[:, :-1] self.warping_function.psi.gradient[:] = warping_grads[:, :-1]
self.warping_function.d.gradient[:] = warping_grads[0, -1] self.warping_function.d.gradient[:] = warping_grads[0, -1]
def transform_data(self): def transform_data(self):
Y = self.warping_function.f(self.Y_untransformed.copy()).copy() Y = self.warping_function.f(self.Y_untransformed.copy()).copy()
return Y return Y
@ -84,6 +83,9 @@ class WarpedGP(GP):
return gh_weights.dot(self._get_warped_term(mean, std, gh_samples)) / np.sqrt(np.pi) return gh_weights.dot(self._get_warped_term(mean, std, gh_samples)) / np.sqrt(np.pi)
def _get_warped_variance(self, mean, std, pred_init=None, deg_gauss_hermite=100): def _get_warped_variance(self, mean, std, pred_init=None, deg_gauss_hermite=100):
"""
Calculate the warped variance by using Gauss-Hermite quadrature.
"""
gh_samples, gh_weights = np.polynomial.hermite.hermgauss(deg_gauss_hermite) gh_samples, gh_weights = np.polynomial.hermite.hermgauss(deg_gauss_hermite)
gh_samples = gh_samples[:,None] gh_samples = gh_samples[:,None]
gh_weights = gh_weights[None,:] gh_weights = gh_weights[None,:]
@ -105,13 +107,10 @@ class WarpedGP(GP):
if self.predict_in_warped_space: if self.predict_in_warped_space:
std = np.sqrt(var) std = np.sqrt(var)
if median: if median:
#print 'MEDIAN!'
wmean = self.warping_function.f_inv(mean, y=pred_init) wmean = self.warping_function.f_inv(mean, y=pred_init)
else: else:
#print 'MEAN!'
wmean = self._get_warped_mean(mean, std, pred_init=pred_init, wmean = self._get_warped_mean(mean, std, pred_init=pred_init,
deg_gauss_hermite=deg_gauss_hermite).T deg_gauss_hermite=deg_gauss_hermite).T
#var = self.warping_function.f_inv(var)
wvar = self._get_warped_variance(mean, std, pred_init=pred_init, wvar = self._get_warped_variance(mean, std, pred_init=pred_init,
deg_gauss_hermite=deg_gauss_hermite).T deg_gauss_hermite=deg_gauss_hermite).T
else: else:
@ -147,6 +146,7 @@ class WarpedGP(GP):
return [new_a, new_b] return [new_a, new_b]
#return self.likelihood.predictive_quantiles(m, v, quantiles, Y_metadata) #return self.likelihood.predictive_quantiles(m, v, quantiles, Y_metadata)
if __name__ == '__main__': if __name__ == '__main__':
X = np.random.randn(100, 1) X = np.random.randn(100, 1)
Y = np.sin(X) + np.random.randn(100, 1)*0.05 Y = np.sin(X) + np.random.randn(100, 1)*0.05

2
GPy/testing/model_tests.py
View file

@ -221,7 +221,7 @@ class MiscTests(unittest.TestCase):
np.testing.assert_almost_equal(preds, warp_preds) np.testing.assert_almost_equal(preds, warp_preds)
@unittest.skip('Comment this to plot the modified sine function') #@unittest.skip('Comment this to plot the modified sine function')
def test_warped_gp_sine(self): def test_warped_gp_sine(self):
""" """
A test replicating the sine regression problem from A test replicating the sine regression problem from