def __init__(self, layer_lower,
dim_down, dim_up,
likelihood,
X=None, X_variance=None, init='PCA',
Z=None, num_inducing=10, kernel=None,
inference_method=None, uncertain_inputs=True,
mpi_comm=None, mpi_root=0, back_constraint=True,
encoder=None, auto_update=True, name='layer'):
self.uncertain_inputs = uncertain_inputs
self.layer_lower = layer_lower
Y = self.Y if self.layer_lower is None else self.layer_lower.X
self.back_constraint = back_constraint
from deepgp.util.util import initialize_latent
if X is None: X, _ = initialize_latent(init, Y.shape[0], dim_up, Y.mean.values if isinstance(Y, VariationalPosterior) else Y)
if X_variance is None: X_variance = 0.01*np.ones(X.shape) + 0.01*np.random.rand(*X.shape)
if Z is None:
if self.back_constraint: Z = np.random.rand(num_inducing, dim_up)*2-1.
else:
if num_inducing<=X.shape[0]:
Z = X[np.random.permutation(X.shape[0])[:num_inducing]].copy()
else:
Z_more = np.random.rand(num_inducing-X.shape[0],X.shape[1])*(X.max(0)-X.min(0))+X.min(0)
Z = np.vstack([X.copy(),Z_more])
assert Z.shape[1] == X.shape[1]
if mpi_comm is not None:
from ..util.parallel import broadcastArrays
broadcastArrays([Z], mpi_comm, mpi_root)
if uncertain_inputs: X = NormalPosterior(X, X_variance)
if kernel is None: kernel = kern.RBF(dim_up, ARD = True)
assert kernel.input_dim==X.shape[1], "The dimensionality of input has to be equal to the input dimensionality of the kernel!"
self.Kuu_sigma = Param('Kuu_var', np.zeros(num_inducing)+1e-3, Logexp())
super(Layer, self).__init__(X, Y, Z, kernel,
likelihood, inference_method=inference_method,
mpi_comm=mpi_comm, mpi_root=mpi_root,
auto_update=auto_update, name=name)
self.link_parameter(self.Kuu_sigma)
if back_constraint: self.encoder = encoder
if self.uncertain_inputs and not self.back_constraint:
self.link_parameter(self.X)
def _init_Xs(self, Ya, Xa):
if isinstance(Ya, list):
return [None for d in self.nDims[1:]]
else:
nDims, N = self.nDims, Ya.shape[0]
if self.back_constraint:
if self.X_observed:
Xs = [np.random.rand(N, d) for d in nDims[1:-1]] + [Xa]
else:
Xs = [np.random.rand(N, d) for d in nDims[1:]]
elif self.X_observed and len(nDims) == 3:
X = Xa.mean.values if isinstance(Xa, VariationalPosterior) else Xa
from deepgp.util.util import initialize_latent
X_mid = initialize_latent('PCA', N, nDims[1], X)[0]
if X.shape[1] < self.input_dim:
tmp = np.random.randn(*X.shape)
tmp[:, :X_mid.shape[1]] = X_mid
X_mid = tmp
if self.repeatX:
if isinstance(X, VariationalPosterior):
#-- Haven't tested this case
# ...
raise NotImplementedError()
else:
X_mid2 = np.hstack((X_mid, X))
# Should we re-normalize everything????
Xmean, Xstd = X_mid2.mean(0), X_mid2.std(0) + 1e-20
X_mid2 -= Xmean[np.newaxis, :]
X_mid2 /= Xstd[np.newaxis, :]
self.repeatX_Xmean = Xmean.copy()[X_mid.shape[1]:]
self.repeatX_Xstd = Xstd.copy()[X_mid.shape[1]:]
self.nDimsOrig = nDims[:]
nDims[1] = X_mid2.shape[1]
X_mid = X_mid2
Xs = [X_mid, X]
elif self.X_observed:
X = Xa.mean.values if isinstance(Xa, VariationalPosterior) else Xa
Xs = [None for d in nDims[1:-1]] + [X]
else:
Xs = [None for d in nDims[1:]]
return Xs
def __init__(self, layer_lower, dim_down, dim_up, likelihood, X=None, X_variance=None, init='PCA', Z=None, num_inducing=10, kernel=None, inference_method=None, uncertain_inputs=True,mpi_comm=None, mpi_root=0, back_constraint=True, encoder=None, auto_update=True, name='layer'):
self.uncertain_inputs = uncertain_inputs
self.layer_lower = layer_lower
Y = self.Y if self.layer_lower is None else self.layer_lower.X
self.back_constraint = back_constraint
from deepgp.util.util import initialize_latent
if X is None: X, _ = initialize_latent(init, Y.shape[0], dim_up, Y.mean.values if isinstance(Y, VariationalPosterior) else Y)
if X_variance is None: X_variance = 0.01*np.ones(X.shape) + 0.01*np.random.rand(*X.shape)
if Z is None:
if self.back_constraint: Z = np.random.rand(num_inducing, dim_up)*2-1.
else:
if num_inducing<=X.shape[0]:
Z = X[np.random.permutation(X.shape[0])[:num_inducing]].copy()
else:
Z_more = np.random.rand(num_inducing-X.shape[0],X.shape[1])*(X.max(0)-X.min(0))+X.min(0)
Z = np.vstack([X.copy(),Z_more])
assert Z.shape[1] == X.shape[1]
if mpi_comm is not None:
from ..util.parallel import broadcastArrays
broadcastArrays([Z], mpi_comm, mpi_root)
if uncertain_inputs: X = NormalPosterior(X, X_variance)
if kernel is None: kernel = kern.RBF(dim_up, ARD = True)
assert kernel.input_dim==X.shape[1], "The dimensionality of input has to be equal to the input dimensionality of kernel!"
self.Kuu_sigma = Param('Kuu_var', np.zeros(num_inducing)+1e-3, Logexp())
super(Layer, self).__init__(X, Y, Z, kernel, likelihood, inference_method=inference_method, mpi_comm=mpi_comm, mpi_root=mpi_root, auto_update=auto_update, name=name)
self.link_parameter(self.Kuu_sigma)
if back_constraint: self.encoder = encoder
if self.uncertain_inputs and not self.back_constraint:
self.link_parameter(self.X)