def compute_phi_from_parents(u_parents):
# Compute weighted average of the parameters
# Cluster parameters
Phi = distribution.compute_phi_from_parents(u_parents[1:])
# Contributions/weights/probabilities
P = u_parents[0][0]
phi = list()
for ind in range(len(Phi)):
# Compute element-wise product and then sum over K clusters.
# Note that the dimensions aren't perfectly aligned because
# the cluster dimension (K) may be arbitrary for phi, and phi
# also has dimensions (Dd,..,D0) of the parameters.
# Shape(phi) = [Nn,..,K,..,N0,Dd,..,D0]
# Shape(p) = [Nn,..,N0,K]
# Shape(result) = [Nn,..,N0,Dd,..,D0]
# General broadcasting rules apply for Nn,..,N0, that is,
# preceding dimensions may be missing or dimension may be
# equal to one. Probably, shape(phi) has lots of missing
# dimensions and/or dimensions that are one.
if cluster_plate < 0:
cluster_axis = cluster_plate - distribution.ndims[ind]
#else:
# cluster_axis = cluster_plate
# Move cluster axis to the last:
# Shape(phi) = [Nn,..,N0,Dd,..,D0,K]
phi.append(utils.moveaxis(Phi[ind], cluster_axis, -1))
# Add axes to p:
# Shape(p) = [Nn,..,N0,K,1,..,1]
p = utils.add_trailing_axes(P, distribution.ndims[ind])
# Move cluster axis to the last:
# Shape(p) = [Nn,..,N0,1,..,1,K]
p = utils.moveaxis(p, -(distribution.ndims[ind] + 1), -1)
#print('Mixture.compute_phi, p:', np.sum(p, axis=-1))
#print('mixture.compute_phi shapes:')
#print(np.shape(p))
#print(np.shape(phi[ind]))
# Now the shapes broadcast perfectly and we can sum
# p*phi over the last axis:
# Shape(result) = [Nn,..,N0,Dd,..,D0]
phi[ind] = utils.sum_product(p, phi[ind], axes_to_sum=-1)
return phi
def compute_phi_from_parents(u_parents):
# Compute weighted average of the parameters
# Cluster parameters
Phi = distribution.compute_phi_from_parents(u_parents[1:])
# Contributions/weights/probabilities
P = u_parents[0][0]
phi = list()
for ind in range(len(Phi)):
# Compute element-wise product and then sum over K clusters.
# Note that the dimensions aren't perfectly aligned because
# the cluster dimension (K) may be arbitrary for phi, and phi
# also has dimensions (Dd,..,D0) of the parameters.
# Shape(phi) = [Nn,..,K,..,N0,Dd,..,D0]
# Shape(p) = [Nn,..,N0,K]
# Shape(result) = [Nn,..,N0,Dd,..,D0]
# General broadcasting rules apply for Nn,..,N0, that is,
# preceding dimensions may be missing or dimension may be
# equal to one. Probably, shape(phi) has lots of missing
# dimensions and/or dimensions that are one.
if cluster_plate < 0:
cluster_axis = cluster_plate - distribution.ndims[ind]
#else:
# cluster_axis = cluster_plate
# Move cluster axis to the last:
# Shape(phi) = [Nn,..,N0,Dd,..,D0,K]
phi.append(utils.moveaxis(Phi[ind], cluster_axis, -1))
# Add axes to p:
# Shape(p) = [Nn,..,N0,K,1,..,1]
p = utils.add_trailing_axes(P, distribution.ndims[ind])
# Move cluster axis to the last:
# Shape(p) = [Nn,..,N0,1,..,1,K]
p = utils.moveaxis(p, -(distribution.ndims[ind]+1), -1)
#print('Mixture.compute_phi, p:', np.sum(p, axis=-1))
#print('mixture.compute_phi shapes:')
#print(np.shape(p))
#print(np.shape(phi[ind]))
# Now the shapes broadcast perfectly and we can sum
# p*phi over the last axis:
# Shape(result) = [Nn,..,N0,Dd,..,D0]
phi[ind] = utils.sum_product(p, phi[ind], axes_to_sum=-1)
return phi
def compute_message(index, u, u_parents):
""" . """
#print('Mixture.compute_message:')
if index == 0:
# Shape(phi) = [Nn,..,K,..,N0,Dd,..,D0]
# Shape(L) = [Nn,..,K,..,N0]
# Shape(u) = [Nn,..,N0,Dd,..,D0]
# Shape(result) = [Nn,..,N0,K]
# Compute g:
# Shape(g) = [Nn,..,K,..,N0]
g = distribution.compute_g_from_parents(u_parents[1:])
# Reshape(g):
# Shape(g) = [Nn,..,N0,K]
g = utils.moveaxis(g, cluster_plate, -1)
# Compute phi:
# Shape(phi) = [Nn,..,K,..,N0,Dd,..,D0]
phi = distribution.compute_phi_from_parents(u_parents[1:])
# Reshape phi:
# Shape(phi) = [Nn,..,N0,K,Dd,..,D0]
for ind in range(len(phi)):
phi[ind] = utils.moveaxis(
phi[ind], cluster_plate - distribution.ndims[ind],
-1 - distribution.ndims[ind])
# Reshape u:
# Shape(u) = [Nn,..,N0,1,Dd,..,D0]
u_self = list()
for ind in range(len(u)):
u_self.append(
np.expand_dims(u[ind],
axis=(-1 - distribution.ndims[ind])))
# Compute logpdf:
# Shape(L) = [Nn,..,N0,K]
L = distribution.compute_logpdf(u_self, phi, g, 0)
# Sum over other than the cluster dimensions? No!
# Hmm.. I think the message passing method will do
# that automatically
## print(np.shape(phi[0]))
## print(np.shape(u_self[0]))
## print(np.shape(g))
## print(np.shape(L))
return [L]
elif index >= 1:
# Parent index for the distribution used for the
# mixture.
index = index - 1
# Reshape u:
# Shape(u) = [Nn,..1,..,N0,Dd,..,D0]
u_self = list()
for ind in range(len(u)):
if cluster_plate < 0:
cluster_axis = cluster_plate - distribution.ndims[ind]
else:
cluster_axis = cluster_plate
u_self.append(np.expand_dims(u[ind], axis=cluster_axis))
# Message from the mixed distribution
m = distribution.compute_message(index, u_self, u_parents[1:])
# Weigh the messages with the responsibilities
for i in range(len(m)):
# Shape(m) = [Nn,..,K,..,N0,Dd,..,D0]
# Shape(p) = [Nn,..,N0,K]
# Shape(result) = [Nn,..,K,..,N0,Dd,..,D0]
# Number of axes for the variable dimensions for
# the parent message.
D = distribution.ndims_parents[index][i]
# Responsibilities for clusters are the first
# parent's first moment:
# Shape(p) = [Nn,..,N0,K]
p = u_parents[0][0]
# Move the cluster axis to the proper place:
# Shape(p) = [Nn,..,K,..,N0]
p = utils.moveaxis(p, -1, cluster_plate)
# Add axes for variable dimensions to the contributions
# Shape(p) = [Nn,..,K,..,N0,1,..,1]
p = utils.add_trailing_axes(p, D)
if cluster_plate < 0:
# Add the variable dimensions
cluster_axis = cluster_plate - D
# Add axis for clusters:
# Shape(m) = [Nn,..,1,..,N0,Dd,..,D0]
#m[i] = np.expand_dims(m[i], axis=cluster_axis)
#
# TODO: You could do summing here already so that
# you wouldn't compute huge matrices as
# intermediate result. Use einsum.
## print(np.shape(m[i]))
## print(np.shape(p))
# Compute the message contributions for each
# cluster:
# Shape(result) = [Nn,..,K,..,N0,Dd,..,D0]
m[i] = m[i] * p
#print(np.shape(m[i]))
return m
def compute_message(index, u, u_parents):
""" . """
#print('Mixture.compute_message:')
if index == 0:
# Shape(phi) = [Nn,..,K,..,N0,Dd,..,D0]
# Shape(L) = [Nn,..,K,..,N0]
# Shape(u) = [Nn,..,N0,Dd,..,D0]
# Shape(result) = [Nn,..,N0,K]
# Compute g:
# Shape(g) = [Nn,..,K,..,N0]
g = distribution.compute_g_from_parents(u_parents[1:])
# Reshape(g):
# Shape(g) = [Nn,..,N0,K]
g = utils.moveaxis(g, cluster_plate, -1)
# Compute phi:
# Shape(phi) = [Nn,..,K,..,N0,Dd,..,D0]
phi = distribution.compute_phi_from_parents(u_parents[1:])
# Reshape phi:
# Shape(phi) = [Nn,..,N0,K,Dd,..,D0]
for ind in range(len(phi)):
phi[ind] = utils.moveaxis(phi[ind],
cluster_plate-distribution.ndims[ind],
-1-distribution.ndims[ind])
# Reshape u:
# Shape(u) = [Nn,..,N0,1,Dd,..,D0]
u_self = list()
for ind in range(len(u)):
u_self.append(np.expand_dims(u[ind],
axis=(-1-distribution.ndims[ind])))
# Compute logpdf:
# Shape(L) = [Nn,..,N0,K]
L = distribution.compute_logpdf(u_self, phi, g, 0)
# Sum over other than the cluster dimensions? No!
# Hmm.. I think the message passing method will do
# that automatically
## print(np.shape(phi[0]))
## print(np.shape(u_self[0]))
## print(np.shape(g))
## print(np.shape(L))
return [L]
elif index >= 1:
# Parent index for the distribution used for the
# mixture.
index = index - 1
# Reshape u:
# Shape(u) = [Nn,..1,..,N0,Dd,..,D0]
u_self = list()
for ind in range(len(u)):
if cluster_plate < 0:
cluster_axis = cluster_plate - distribution.ndims[ind]
else:
cluster_axis = cluster_plate
u_self.append(np.expand_dims(u[ind], axis=cluster_axis))
# Message from the mixed distribution
m = distribution.compute_message(index, u_self, u_parents[1:])
# Weigh the messages with the responsibilities
for i in range(len(m)):
# Shape(m) = [Nn,..,K,..,N0,Dd,..,D0]
# Shape(p) = [Nn,..,N0,K]
# Shape(result) = [Nn,..,K,..,N0,Dd,..,D0]
# Number of axes for the variable dimensions for
# the parent message.
D = distribution.ndims_parents[index][i]
# Responsibilities for clusters are the first
# parent's first moment:
# Shape(p) = [Nn,..,N0,K]
p = u_parents[0][0]
# Move the cluster axis to the proper place:
# Shape(p) = [Nn,..,K,..,N0]
p = utils.moveaxis(p, -1, cluster_plate)
# Add axes for variable dimensions to the contributions
# Shape(p) = [Nn,..,K,..,N0,1,..,1]
p = utils.add_trailing_axes(p, D)
if cluster_plate < 0:
# Add the variable dimensions
cluster_axis = cluster_plate - D
# Add axis for clusters:
# Shape(m) = [Nn,..,1,..,N0,Dd,..,D0]
#m[i] = np.expand_dims(m[i], axis=cluster_axis)
#
# TODO: You could do summing here already so that
# you wouldn't compute huge matrices as
# intermediate result. Use einsum.
## print(np.shape(m[i]))
## print(np.shape(p))
# Compute the message contributions for each
# cluster:
# Shape(result) = [Nn,..,K,..,N0,Dd,..,D0]
m[i] = m[i] * p
#print(np.shape(m[i]))
return m
