def test_gp_gradient2():
random = RandomState(94584)
N = 10
X = random.randn(N, 2)
K0 = random.randn(N, N + 1)
K0 = K0 @ K0.T + eye(N) * 1e-5
K1 = random.randn(N, N + 1)
K1 = K1 @ K1.T + eye(N) * 1e-5
mean = LinearMean(X)
mean.effsizes = [-1.2, +0.9]
cov0 = GivenCov(K0)
cov1 = GivenCov(K1)
cov2 = EyeCov(N)
cov = SumCov([cov0, cov1, cov2])
cov0.scale = 0.1
cov1.scale = 2.3
cov2.scale = 0.3
y = random.randn(N)
gp = GP(y, mean, cov)
assert_allclose(gp._check_grad(), 0, atol=1e-5)
def _get_data():
random = RandomState(0)
N = 10
X = random.randn(N, N + 1)
X -= X.mean(0)
X /= X.std(0)
X /= sqrt(X.shape[1])
offset = 1.0
mean = OffsetMean(N)
mean.offset = offset
cov_left = LinearCov(X)
cov_left.scale = 1.5
cov_right = EyeCov(N)
cov_right.scale = 1.5
cov = SumCov([cov_left, cov_right])
lik = BernoulliProdLik(LogitLink())
y = GGPSampler(lik, mean, cov).sample(random)
return dict(mean=mean,
cov=cov,
lik=lik,
y=y,
cov_left=cov_left,
cov_right=cov_right)
def append_iid_noise(self):
from glimix_core.cov import EyeCov
cov = EyeCov()
cov.set_data((self._sample_idx, self._sample_idx))
self._covariance_matrices["impl"].append(cov)
self._covariance_matrices["user"].append(user_cov.EyeCov(cov))
self._covariance_matrices["user"][-1].name = "residual"
self._cov = None
def _outcome_sample(random, offset, X):
n = X.shape[0]
mean = OffsetMean(n)
mean.offset = offset
cov_left = LinearCov(X)
cov_left.scale = 1.5
cov_right = EyeCov(n)
cov_right.scale = 1.5
cov = SumCov([cov_left, cov_right])
lik = DeltaProdLik()
return GGPSampler(lik, mean, cov).sample(random)
def linear_eye_cov():
from numpy import sqrt
from numpy.random import RandomState
from glimix_core.cov import EyeCov, LinearCov, SumCov
random = RandomState(458)
X = random.randn(_nsamples(), _nsamples() + 1)
X -= X.mean(0)
X /= X.std(0)
X /= sqrt(X.shape[1])
cov_left = LinearCov(X)
cov_left.scale = 1.0
cov_right = EyeCov(_nsamples())
cov_right.scale = 1.0
return SumCov([cov_left, cov_right])
def append_iid(self, name="residual"):
from glimix_core.cov import EyeCov
if self._multi_trait():
cov = MTEyeCov(self._y.shape[1])
else:
cov = EyeCov(self._y.shape[0])
cov.name = name
self._covariance.append(cov)
def test_lmm_predict():
random = RandomState(9458)
n = 30
X = random.randn(n, n + 1)
X -= X.mean(0)
X /= X.std(0)
X /= sqrt(X.shape[1])
offset = 1.0
mean = OffsetMean(n)
mean.offset = offset
cov_left = LinearCov(X)
cov_left.scale = 1.5
cov_right = EyeCov(n)
cov_right.scale = 1.5
cov = SumCov([cov_left, cov_right])
lik = DeltaProdLik()
y = GGPSampler(lik, mean, cov).sample(random)
QS = economic_qs_linear(X)
lmm = LMM(y, ones((n, 1)), QS)
lmm.fit(verbose=False)
plmm = LMMPredict(y, lmm.beta, lmm.v0, lmm.v1, lmm.mean(),
lmm.covariance())
K = dot(X, X.T)
pm = plmm.predictive_mean(ones((n, 1)), K, K.diagonal())
assert_allclose(corrcoef(y, pm)[0, 1], 0.8358820971891354)
def test_ggp_expfam_tobi():
random = RandomState(2)
n = 30
ntrials = random.randint(30, size=n)
K = random.randn(n, n)
K = matmul(K, K.T)
lik = BinomialProdLik(ntrials=ntrials, link=LogitLink())
mean = OffsetMean(n)
cov2 = EyeCov(n)
y = GGPSampler(lik, mean, cov2).sample(random)
ggp = ExpFamGP(y, ("binomial", ntrials), mean, cov2)
assert_allclose(ggp.lml(), -67.84095700542488)
ggp.fit(verbose=False)
assert_allclose(ggp.lml(), -64.26701904994792)
def test_eyecov():
cov = EyeCov(2)
cov.scale = 1.5
assert_allclose(cov.value(), 1.5 * eye(2))
assert_allclose(cov._check_grad(), 0, atol=1e-5)
def append_iid(self, name="residual"):
from glimix_core.cov import EyeCov
c = EyeCov(self._y.shape[0])
c.name = name
self._covariance.append(c)