def __init__(self, model, batch_size):
# Call constructor of superclass
super(PCD, self).__init__(model=model, batch_size=batch_size)
self.sampler = RBM_SAMPLER.PersistentGibbsSampler(
self.rbm, self.batch_size)
def test_Persistent_Gibbs_sampler(self):
sys.stdout.write('RBM Sampler -> Performing PersistentGibbsSampler test ... ')
sys.stdout.flush()
numx.random.seed(42)
sampler = Sampler.PersistentGibbsSampler(self.bbrbm, 1)
probCD1, probCD2, probCS1, probCS2, probCS3, probCS4, sumProbs = self.execute_sampler(sampler, self.num_samples)
assert numx.all(numx.abs(1.0 / 4.0 - probCD1) < self.epsilon)
assert numx.all(numx.abs(1.0 / 4.0 - probCD2) < self.epsilon)
assert numx.all(numx.abs(1.0 / 8.0 - probCS1) < self.epsilon)
assert numx.all(numx.abs(1.0 / 8.0 - probCS2) < self.epsilon)
assert numx.all(numx.abs(1.0 / 8.0 - probCS3) < self.epsilon)
assert numx.all(numx.abs(1.0 / 8.0 - probCS4) < self.epsilon)
assert numx.all(numx.abs(1.0 - sumProbs) < self.epsilon)
print('successfully passed!')
sys.stdout.flush()
def __init__(self, model, num_chains, data=None):
""" The constructor initializes the PCD trainer with a given models and data.
:param model: The models to sample from.
:type model: Valid models class.
:param num_chains: The number of chains that should be used.
.. Note:: You should use the data's batch size!
:type num_chains: int
:param data: Data for initialization, only has effect if the centered gradient is used.
:type data: numpy array [num. samples x input dim]
"""
# Call super constructor of CD
super(PCD, self).__init__(model, data)
self.sampler = sampler.PersistentGibbsSampler(model, num_chains)