apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-27 14:25:16 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

migrate pep_tests to pytest

Browse source
This commit is contained in:
Martin Bubel 2023-10-10 18:56:16 +02:00
parent 1df86e2216
commit b81fe8aaeb
1 changed files with 12 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

17
GPy/testing/pep_tests.py
View file

@ -1,13 +1,12 @@
# Copyright (c) 2014, James Hensman, 2016, Thang Bui # Copyright (c) 2014, James Hensman, 2016, Thang Bui
# Licensed under the BSD 3-clause license (see LICENSE.txt) # Licensed under the BSD 3-clause license (see LICENSE.txt)
import unittest
import numpy as np import numpy as np
import GPy import GPy
class PEPgradienttest(unittest.TestCase): class TestPEPgradient:
def setUp(self): def setup(self):
###################################### ######################################
# # 1 dimensional example # # 1 dimensional example
np.random.seed(10) np.random.seed(10)
@ -39,6 +38,7 @@ class PEPgradienttest(unittest.TestCase):
self.lik_noise_var = 0.01 self.lik_noise_var = 0.01
def test_pep_1d_gradients(self): def test_pep_1d_gradients(self):
self.setup()
m = GPy.models.SparseGPRegression(self.X1D, self.Y1D) m = GPy.models.SparseGPRegression(self.X1D, self.Y1D)
m.inference_method = GPy.inference.latent_function_inference.PEP( m.inference_method = GPy.inference.latent_function_inference.PEP(
alpha=np.random.rand() alpha=np.random.rand()
@ -46,6 +46,7 @@ class PEPgradienttest(unittest.TestCase):
assert m.checkgrad() assert m.checkgrad()
def test_pep_2d_gradients(self): def test_pep_2d_gradients(self):
self.setup()
m = GPy.models.SparseGPRegression(self.X2D, self.Y2D) m = GPy.models.SparseGPRegression(self.X2D, self.Y2D)
m.inference_method = GPy.inference.latent_function_inference.PEP( m.inference_method = GPy.inference.latent_function_inference.PEP(
alpha=np.random.rand() alpha=np.random.rand()
@ -53,6 +54,7 @@ class PEPgradienttest(unittest.TestCase):
assert m.checkgrad() assert m.checkgrad()
def test_pep_vfe_consistency(self): def test_pep_vfe_consistency(self):
self.setup()
vfe_model = GPy.models.SparseGPRegression( vfe_model = GPy.models.SparseGPRegression(
self.X1, self.Y1, kernel=self.kernel, Z=self.Z self.X1, self.Y1, kernel=self.kernel, Z=self.Z
) )
@ -69,9 +71,12 @@ class PEPgradienttest(unittest.TestCase):
pep_model.Gaussian_noise.variance = self.lik_noise_var pep_model.Gaussian_noise.variance = self.lik_noise_var
pep_lml = pep_model.log_likelihood() pep_lml = pep_model.log_likelihood()
self.assertAlmostEqual(vfe_lml[0, 0], pep_lml[0], delta=abs(0.01 * pep_lml[0])) np.testing.assert_almost_equal(
vfe_lml[0, 0], pep_lml[0], decimal=abs(0.01 * pep_lml[0])
)
def test_pep_fitc_consistency(self): def test_pep_fitc_consistency(self):
self.setup()
fitc_model = GPy.models.SparseGPRegression( fitc_model = GPy.models.SparseGPRegression(
self.X1D, self.Y1D, kernel=self.kernel, Z=self.Z self.X1D, self.Y1D, kernel=self.kernel, Z=self.Z
) )
@ -88,4 +93,6 @@ class PEPgradienttest(unittest.TestCase):
pep_model.Gaussian_noise.variance = self.lik_noise_var pep_model.Gaussian_noise.variance = self.lik_noise_var
pep_lml = pep_model.log_likelihood() pep_lml = pep_model.log_likelihood()
self.assertAlmostEqual(fitc_lml, pep_lml[0], delta=abs(0.001 * pep_lml[0])) np.testing.assert_almost_equal(
fitc_lml, pep_lml[0], decimal=abs(0.001 * pep_lml[0])
)