def test_se(self,thr=1e-6):
""" Test squared exponential covariance.
"""
k = cov.se(ell=np.random.rand())
self.run_verifications(k,thr)
k = cov.se(ell=np.random.rand(self.D))
self.run_verifications(k,thr)
def test_se(self, thr=1e-6):
""" Test squared exponential covariance.
"""
k = cov.se(ell=np.random.rand())
self.run_verifications(k, thr)
k = cov.se(ell=np.random.rand(self.D))
self.run_verifications(k, thr)
def test_prodsum(self,thr=3e-6):
""" Test the product of sums of several covariance functions.
"""
k1 = cov.se(ell=np.random.rand())
k2 = cov.se(ell=np.random.rand(self.D))
k3 = cov.se(ell=np.random.rand())
k4 = cov.se(ell=np.random.rand(self.D))
self.run_verifications((2.3+k1)*(k2+1.2)*(k3+k4),thr)
def test_sumprod(self,thr=1e-6):
""" Test the sum of products of several covariance functions.
"""
k1 = cov.se(ell=np.random.rand())
k2 = cov.se(ell=np.random.rand(self.D))
k3 = cov.se(ell=np.random.rand())
k4 = cov.se(ell=np.random.rand(self.D))
self.run_verifications(2.3*k1+k2*1.2+k3*k4,thr)
def test_prodsum(self, thr=3e-6):
""" Test the product of sums of several covariance functions.
"""
k1 = cov.se(ell=np.random.rand())
k2 = cov.se(ell=np.random.rand(self.D))
k3 = cov.se(ell=np.random.rand())
k4 = cov.se(ell=np.random.rand(self.D))
self.run_verifications((2.3 + k1) * (k2 + 1.2) * (k3 + k4), thr)
def test_sumprod(self, thr=1e-6):
""" Test the sum of products of several covariance functions.
"""
k1 = cov.se(ell=np.random.rand())
k2 = cov.se(ell=np.random.rand(self.D))
k3 = cov.se(ell=np.random.rand())
k4 = cov.se(ell=np.random.rand(self.D))
self.run_verifications(2.3 * k1 + k2 * 1.2 + k3 * k4, thr)
def test_prod(self,thr=1e-6):
""" Test the product of several covariance functions.
"""
n,m,D,X,Z = self.n,self.m,self.D,self.X,self.Z
k1 = cov.se(ell=np.random.rand())
k2 = cov.se(ell=np.random.rand(self.D))
k = k1*k2*1.2
d = np.linalg.norm( k1(X)*k2(X)*1.2 - k(X) )
self.assertLessEqual(d,thr)
self.run_verifications(k,thr)
def test_prod(self, thr=1e-6):
""" Test the product of several covariance functions.
"""
n, m, D, X, Z = self.n, self.m, self.D, self.X, self.Z
k1 = cov.se(ell=np.random.rand())
k2 = cov.se(ell=np.random.rand(self.D))
k = k1 * k2 * 1.2
d = np.linalg.norm(k1(X) * k2(X) * 1.2 - k(X))
self.assertLessEqual(d, thr)
self.run_verifications(k, thr)
def test_ep(self,thr=1e-4):
""" Test expectation propagation approximate inference.
"""
im = inf.ep
cf,mf,lf = 1.1*cov.se(ell=0.8),mean.zero(),lik.erf()
self.run_verifications(im,cf,mf,lf,thr)
lf = lik.gauss(sn=0.2)
self.run_verifications(im,cf,mf,lf,thr)
mf = 2.3*mean.one()
self.run_verifications(im,cf,mf,lf,thr)
cf = 1.3*cov.se(ell=[0.8,0.7,0.3])
self.run_verifications(im,cf,mf,lf,thr)
mf = 1.3*mean.linear(a=[0.8,0.7,0.3])
self.run_verifications(im,cf,mf,lf,thr)
def test_exact(self,thr=1e-4):
""" Test exact inference.
"""
im = inf.exact
cf,mf,lf = 1.1*cov.se(ell=0.8),mean.zero(),lik.gauss(sn=1e-5)
self.run_verifications(im,cf,mf,lf,thr)
lf = lik.gauss(sn=0.2)
self.run_verifications(im,cf,mf,lf,thr)
mf = 2.3*mean.one()
self.run_verifications(im,cf,mf,lf,thr)
cf = 1.3*cov.se(ell=[0.8,0.7,0.3])
self.run_verifications(im,cf,mf,lf,thr)
mf = 1.3*mean.linear(a=[0.8,0.7,0.3])
self.run_verifications(im,cf,mf,lf,thr)
def __init__(self,inf=exact,cov=1.0*se(ell=1.0),
mean=zero(),lik=gauss(sn=1.0),X=None,y=None):
""" Set up a gp, perform inference if X and y are provided.
"""
self.inf = inf # inference method
self.cov,self.mean,self.lik = cov,mean,lik # covariance,mean,likelihood
self.post = None # init posterior approximation
if X!=None and y!=None: self.post = self.inference(X,y) # do inference
def __init__(self,
inf=exact,
cov=1.0 * se(ell=1.0),
mean=zero(),
lik=gauss(sn=1.0),
X=None,
y=None):
""" Set up a gp, perform inference if X and y are provided.
"""
self.inf = inf # inference method
self.cov, self.mean, self.lik = cov, mean, lik # covariance,mean,likelihood
self.post = None # init posterior approximation
if X != None and y != None:
self.post = self.inference(X, y) # do inference