apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-06-05 14:55:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

psistattests update

Browse source
This commit is contained in:
Max Zwiessele 2013-11-20 11:58:30 +00:00
parent 4948fb1345
commit 76bfbee545
3 changed files with 51 additions and 32 deletions
Download patch file Download diff file Expand all files Collapse all files

3
GPy/kern/kern.py
View file

@ -412,6 +412,9 @@ class kern(Parameterized):
[p.dpsi0_dtheta(dL_dpsi0, Z[:, i_s], mu[:, i_s], S[:, i_s], target[ps]) for p, ps, i_s in zip(self.parts, self.param_slices, self.input_slices)] [p.dpsi0_dtheta(dL_dpsi0, Z[:, i_s], mu[:, i_s], S[:, i_s], target[ps]) for p, ps, i_s in zip(self.parts, self.param_slices, self.input_slices)]
return self._transform_gradients(target) return self._transform_gradients(target)
def dpsi0_dZ(self, dL_dpsi0, Z, mu, S):
return np.zeros_like(Z)
def dpsi0_dmuS(self, dL_dpsi0, Z, mu, S): def dpsi0_dmuS(self, dL_dpsi0, Z, mu, S):
target_mu, target_S = np.zeros_like(mu), np.zeros_like(S) target_mu, target_S = np.zeros_like(mu), np.zeros_like(S)
[p.dpsi0_dmuS(dL_dpsi0, Z[:, i_s], mu[:, i_s], S[:, i_s], target_mu[:, i_s], target_S[:, i_s]) for p, i_s in zip(self.parts, self.input_slices)] [p.dpsi0_dmuS(dL_dpsi0, Z[:, i_s], mu[:, i_s], S[:, i_s], target_mu[:, i_s], target_S[:, i_s]) for p, i_s in zip(self.parts, self.input_slices)]

40
GPy/testing/psi_stat_expectation_tests.py
View file

@ -28,8 +28,8 @@ def ard(p):
class Test(unittest.TestCase): class Test(unittest.TestCase):
input_dim = 9 input_dim = 9
num_inducing = 13 num_inducing = 13
N = 30 N = 300
Nsamples = 9e6 Nsamples = 1e6
def setUp(self): def setUp(self):
i_s_dim_list = [2,4,3] i_s_dim_list = [2,4,3]
@ -50,20 +50,20 @@ class Test(unittest.TestCase):
# GPy.kern.linear(self.input_dim, ARD=True) + # GPy.kern.linear(self.input_dim, ARD=True) +
# GPy.kern.bias(self.input_dim) + # GPy.kern.bias(self.input_dim) +
# GPy.kern.white(self.input_dim)), # GPy.kern.white(self.input_dim)),
# (GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True) + (#GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True)
(GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True) GPy.kern.linear(self.input_dim, np.random.rand(self.input_dim), ARD=True)
+GPy.kern.linear(self.input_dim, np.random.rand(self.input_dim), ARD=True)
# GPy.kern.bias(self.input_dim) +
# GPy.kern.white(self.input_dim)),
),
(GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True)
+GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True) +GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True)
#+GPy.kern.bias(self.input_dim, np.random.rand()) # +GPy.kern.bias(self.input_dim)
#+GPy.kern.white(self.input_dim, np.random.rand())), # +GPy.kern.white(self.input_dim)),
), ),
(GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True) + # (GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True) +
GPy.kern.bias(self.input_dim, np.random.rand()) + # GPy.kern.bias(self.input_dim, np.random.rand())),
GPy.kern.white(self.input_dim, np.random.rand())), # (GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True)
# +GPy.kern.rbf(self.input_dim, np.random.rand(), np.random.rand(self.input_dim), ARD=True)
# #+GPy.kern.bias(self.input_dim, np.random.rand())
# #+GPy.kern.white(self.input_dim, np.random.rand())),
# ),
# GPy.kern.white(self.input_dim, np.random.rand())),
# GPy.kern.rbf(self.input_dim), GPy.kern.rbf(self.input_dim, ARD=True), # GPy.kern.rbf(self.input_dim), GPy.kern.rbf(self.input_dim, ARD=True),
# GPy.kern.linear(self.input_dim, ARD=False), GPy.kern.linear(self.input_dim, ARD=True), # GPy.kern.linear(self.input_dim, ARD=False), GPy.kern.linear(self.input_dim, ARD=True),
# GPy.kern.linear(self.input_dim) + GPy.kern.bias(self.input_dim), # GPy.kern.linear(self.input_dim) + GPy.kern.bias(self.input_dim),
@ -120,25 +120,25 @@ class Test(unittest.TestCase):
diffs = [] diffs = []
for i, q_x_sample_stripe in enumerate(np.array_split(self.q_x_samples, self.Nsamples / Nsamples)): for i, q_x_sample_stripe in enumerate(np.array_split(self.q_x_samples, self.Nsamples / Nsamples)):
K = kern.K(q_x_sample_stripe, self.Z) K = kern.K(q_x_sample_stripe, self.Z)
K = (K[:, :, None] * K[:, None, :]).mean(0) K = (K[:, :, None] * K[:, None, :])
K_ += K K_ += K.sum(0) / self.Nsamples
diffs.append(((psi2 - (K_ / (i + 1)))**2).mean()) diffs.append(((psi2 - (K_*self.Nsamples/((i+1)*Nsamples)))**2).mean())
K_ /= self.Nsamples / Nsamples #K_ /= self.Nsamples / Nsamples
msg = "psi2: {}".format("+".join([p.name + ard(p) for p in kern.parts])) msg = "psi2: {}".format("+".join([p.name + ard(p) for p in kern.parts]))
try: try:
import pylab import pylab
pylab.figure(msg) pylab.figure(msg)
pylab.plot(diffs, marker='x', mew=1.3) pylab.plot(diffs, marker='x', mew=.2)
# print msg, np.allclose(psi2.squeeze(), K_, rtol=1e-1, atol=.1) # print msg, np.allclose(psi2.squeeze(), K_, rtol=1e-1, atol=.1)
self.assertTrue(np.allclose(psi2.squeeze(), K_), self.assertTrue(np.allclose(psi2.squeeze(), K_),
#rtol=1e-1, atol=.1), #rtol=1e-1, atol=.1),
msg=msg + ": not matching") msg=msg + ": not matching")
# sys.stdout.write(".") # sys.stdout.write(".")
except: except:
# import ipdb;ipdb.set_trace()
# kern.psi2(self.Z, self.q_x_mean, self.q_x_variance) # kern.psi2(self.Z, self.q_x_mean, self.q_x_variance)
# sys.stdout.write("E") # sys.stdout.write("E")
print msg + ": not matching" print msg + ": not matching"
import ipdb;ipdb.set_trace()
pass pass
if __name__ == "__main__": if __name__ == "__main__":

36
GPy/testing/psi_stat_gradient_tests.py
View file

@ -40,10 +40,9 @@ class PsiStatModel(Model):
return self.kern.__getattribute__(self.which)(self.Z, self.X, self.X_variance).sum() return self.kern.__getattribute__(self.which)(self.Z, self.X, self.X_variance).sum()
def _log_likelihood_gradients(self): def _log_likelihood_gradients(self):
psimu, psiS = self.kern.__getattribute__("d" + self.which + "_dmuS")(numpy.ones_like(self.psi_), self.Z, self.X, self.X_variance) psimu, psiS = self.kern.__getattribute__("d" + self.which + "_dmuS")(numpy.ones_like(self.psi_), self.Z, self.X, self.X_variance)
try: #psimu, psiS = numpy.ones(self.N * self.input_dim), numpy.ones(self.N * self.input_dim)
psiZ = self.kern.__getattribute__("d" + self.which + "_dZ")(numpy.ones_like(self.psi_), self.Z, self.X, self.X_variance) psiZ = self.kern.__getattribute__("d" + self.which + "_dZ")(numpy.ones_like(self.psi_), self.Z, self.X, self.X_variance)
except AttributeError: #psiZ = numpy.ones(self.num_inducing * self.input_dim)
psiZ = numpy.zeros(self.num_inducing * self.input_dim)
thetagrad = self.kern.__getattribute__("d" + self.which + "_dtheta")(numpy.ones_like(self.psi_), self.Z, self.X, self.X_variance).flatten() thetagrad = self.kern.__getattribute__("d" + self.which + "_dtheta")(numpy.ones_like(self.psi_), self.Z, self.X, self.X_variance).flatten()
return numpy.hstack((psimu.flatten(), psiS.flatten(), psiZ.flatten(), thetagrad)) return numpy.hstack((psimu.flatten(), psiS.flatten(), psiZ.flatten(), thetagrad))
@ -116,9 +115,9 @@ if __name__ == "__main__":
# m.randomize() # m.randomize()
# # self.assertTrue(m.checkgrad()) # # self.assertTrue(m.checkgrad())
numpy.random.seed(0) numpy.random.seed(0)
input_dim = 5 input_dim = 3
N = 50 N = 3
num_inducing = 10 num_inducing = 2
D = 15 D = 15
X = numpy.random.randn(N, input_dim) X = numpy.random.randn(N, input_dim)
X_var = .5 * numpy.ones_like(X) + .1 * numpy.clip(numpy.random.randn(*X.shape), 0, 1) X_var = .5 * numpy.ones_like(X) + .1 * numpy.clip(numpy.random.randn(*X.shape), 0, 1)
@ -143,10 +142,27 @@ if __name__ == "__main__":
# num_inducing=num_inducing, kernel=kernel) # num_inducing=num_inducing, kernel=kernel)
# m2 = PsiStatModel('psi2', X=X, X_variance=X_var, Z=Z, # m2 = PsiStatModel('psi2', X=X, X_variance=X_var, Z=Z,
# num_inducing=num_inducing, kernel=GPy.kern.rbf(input_dim)) # num_inducing=num_inducing, kernel=GPy.kern.rbf(input_dim))
m3 = PsiStatModel('psi2', X=X, X_variance=X_var, Z=Z, # m3 = PsiStatModel('psi2', X=X, X_variance=X_var, Z=Z,
num_inducing=num_inducing, kernel=GPy.kern.linear(input_dim, ARD=True, variances=numpy.random.rand(input_dim))) # num_inducing=num_inducing, kernel=GPy.kern.linear(input_dim, ARD=True, variances=numpy.random.rand(input_dim)))
# + GPy.kern.bias(input_dim)) # + GPy.kern.bias(input_dim))
# m4 = PsiStatModel('psi2', X=X, X_variance=X_var, Z=Z, # m = PsiStatModel('psi2', X=X, X_variance=X_var, Z=Z,
# num_inducing=num_inducing, kernel=GPy.kern.rbf(input_dim) + GPy.kern.bias(input_dim)) # num_inducing=num_inducing,
# kernel=(
# GPy.kern.rbf(input_dim, ARD=1)
# +GPy.kern.linear(input_dim, ARD=1)
# +GPy.kern.bias(input_dim))
# )
# m.ensure_default_constraints()
m2 = PsiStatModel('psi2', X=X, X_variance=X_var, Z=Z,
num_inducing=num_inducing, kernel=(
GPy.kern.rbf(input_dim, numpy.random.rand(), numpy.random.rand(input_dim), ARD=1)
#+GPy.kern.linear(input_dim, numpy.random.rand(input_dim), ARD=1)
#+GPy.kern.rbf(input_dim, numpy.random.rand(), numpy.random.rand(input_dim), ARD=1)
#+GPy.kern.rbf(input_dim, numpy.random.rand(), numpy.random.rand(), ARD=0)
+GPy.kern.bias(input_dim)
+GPy.kern.white(input_dim)
)
)
m2.ensure_default_constraints()
else: else:
unittest.main() unittest.main()