def __init__(self, n_visible, n_hidden):
super(BinaryBinaryRBM, self).__init__()
# data shape
self.n_visible = n_visible
self.n_hidden = n_hidden
# units
self.v = units.BinaryUnits(self, name='v') # visibles
self.h = units.BinaryUnits(self, name='h') # hiddens
# parameters
self.W = parameters.ProdParameters(self, [self.v, self.h],
theano.shared(
value=self._initial_W(),
name='W'),
name='W') # weights
self.bv = parameters.BiasParameters(self,
self.v,
theano.shared(
value=self._initial_bv(),
name='bv'),
name='bv') # visible bias
self.bh = parameters.BiasParameters(self,
self.h,
theano.shared(
value=self._initial_bh(),
name='bh'),
name='bh') # hidden bias
def __init__(self, n_visible, n_hidden_mean, n_hidden_precision):
super(LearntPrecisionSeparateGaussianBinaryRBM, self).__init__()
# data shape
self.n_visible = n_visible
self.n_hidden_mean = n_hidden_mean
self.n_hidden_precision = n_hidden_precision
# units
self.v = units.LearntPrecisionGaussianUnits(self, name='v') # visibles
self.hm = units.BinaryUnits(self, name='hm') # hiddens for mean
self.hp = units.BinaryUnits(self, name='hp') # hiddens for precision
# parameters
self.Wm = parameters.ProdParameters(
self, [self.v, self.hm],
theano.shared(value=self._initial_W(self.n_visible,
self.n_hidden_mean),
name='Wm'),
name='Wm') # weights
self.Wp = parameters.ProdParameters(
self, [self.v.precision_units, self.hp],
theano.shared(value=-np.abs(
self._initial_W(self.n_visible, self.n_hidden_precision)) /
1000,
name='Wp'),
name='Wp') # weights
self.bvm = parameters.BiasParameters(
self,
self.v,
theano.shared(value=self._initial_bias(self.n_visible),
name='bvm'),
name='bvm') # visible bias
self.bvp = parameters.BiasParameters(
self,
self.v.precision_units,
theano.shared(value=self._initial_bias(self.n_visible),
name='bvp'),
name='bvp') # precision bias
self.bhm = parameters.BiasParameters(
self,
self.hm,
theano.shared(value=self._initial_bias(self.n_hidden_mean),
name='bhm'),
name='bhm') # hidden bias for mean
self.bhp = parameters.BiasParameters(
self,
self.hp,
theano.shared(value=self._initial_bias(self.n_hidden_precision) +
1.0,
name='bhp'),
name='bhp') # hidden bias for precision
def __init__(self, n_visible, n_hidden):
super(LearntPrecisionGaussianBinaryRBM, self).__init__()
# data shape
self.n_visible = n_visible
self.n_hidden = n_hidden
# units
self.v = units.LearntPrecisionGaussianUnits(self, name='v') # visibles
self.h = units.BinaryUnits(self, name='h') # hiddens
# parameters
self.Wm = parameters.ProdParameters(self, [self.v, self.h],
theano.shared(
value=self._initial_W(),
name='Wm'),
name='Wm') # weights
self.Wp = parameters.ProdParameters(
self, [self.v.precision_units, self.h],
theano.shared(value=-np.abs(self._initial_W()) / 1000, name='Wp'),
name='Wp') # weights
self.bvm = parameters.BiasParameters(
self,
self.v,
theano.shared(value=self._initial_bias(self.n_visible),
name='bvm'),
name='bvm') # visible bias
self.bvp = parameters.BiasParameters(
self,
self.v.precision_units,
theano.shared(value=self._initial_bias(self.n_visible),
name='bvp'),
name='bvp') # precision bias
self.bh = parameters.BiasParameters(
self,
self.h,
theano.shared(value=self._initial_bias(self.n_hidden), name='bh'),
name='bh') # hidden bias
def morbrun1(f1=1, f2=1, v1=1, v2=1, kern = 1):
test_set_x = np.array(eval_print1).flatten(order='C')
valid_set_x = np.array(eval_print3).flatten(order='C')
train_set_x = np.array(eval_print2).flatten(order='C')
train_set_x = train_set_x.reshape(np.array(eval_print2).shape[0]*batchm,kern,v1,v2)
valid_set_x = valid_set_x.reshape(np.array(eval_print3).shape[0]*batchm,kern,v1,v2)
test_set_x = test_set_x.reshape(np.array(eval_print1).shape[0]*batchm,kern,v1,v2)
visible_maps = kern
hidden_maps = neuron
filter_height = f1
filter_width = f2
mb_size = batchm # 1 minibatch
print(">> Constructing RBM...")
fan_in = visible_maps * filter_height * filter_width
"""
initial_W = numpy.asarray(
self.numpy_rng.uniform(
low = - numpy.sqrt(3./fan_in),
high = numpy.sqrt(3./fan_in),
size = self.filter_shape
), dtype=theano.config.floatX)
"""
numpy_rng = np.random.RandomState(123)
initial_W = np.asarray(
numpy_rng.normal(
0, 0.5 / np.sqrt(fan_in),
size = (hidden_maps, visible_maps, filter_height, filter_width)
), dtype=theano.config.floatX)
initial_bv = np.zeros(visible_maps, dtype = theano.config.floatX)
initial_bh = np.zeros(hidden_maps, dtype = theano.config.floatX)
shape_info = {
'hidden_maps': hidden_maps,
'visible_maps': visible_maps,
'filter_height': filter_height,
'filter_width': filter_width,
'visible_height': v1, #45+8,
'visible_width': v2, #30,
'mb_size': mb_size
}
# rbms.SigmoidBinaryRBM(n_visible, n_hidden)
rbm = base.RBM()
rbm.v = units.BinaryUnits(rbm, name='v') # visibles
rbm.h = units.BinaryUnits(rbm, name='h') # hiddens
rbm.W = parameters.Convolutional2DParameters(rbm, [rbm.v, rbm.h], theano.shared(value=initial_W, name='W'), name='W', shape_info=shape_info)
# one bias per map (so shared across width and height):
rbm.bv = parameters.SharedBiasParameters(rbm, rbm.v, 3, 2, theano.shared(value=initial_bv, name='bv'), name='bv')
rbm.bh = parameters.SharedBiasParameters(rbm, rbm.h, 3, 2, theano.shared(value=initial_bh, name='bh'), name='bh')
initial_vmap = { rbm.v: T.tensor4('v') }
# try to calculate weight updates using CD-1 stats
print(">> Constructing contrastive divergence updaters...")
s = stats.cd_stats(rbm, initial_vmap, visible_units=[rbm.v], hidden_units=[rbm.h], k=5, mean_field_for_stats=[rbm.v], mean_field_for_gibbs=[rbm.v])
lr_cd = 0.001
if indk == -1:
lr_cd = 0
umap = {}
for var in rbm.variables:
pu = var + lr_cd * updaters.CDUpdater(rbm, var, s)
umap[var] = pu
print(">> Compiling functions...")
t = trainers.MinibatchTrainer(rbm, umap)
m = monitors.reconstruction_mse(s, rbm.v)
e_data = rbm.energy(s['data']).mean()
e_model = rbm.energy(s['model']).mean()
# train = t.compile_function(initial_vmap, mb_size=32, monitors=[m], name='train', mode=mode)
train = t.compile_function(initial_vmap, mb_size=mb_size, monitors=[m, e_data, e_model], name='train', mode=mode)
# TRAINING
epochs = epoch_cd
print(">> Training for %d epochs..." % epochs)
for epoch in range(epochs):
monitoring_data_train = [(cost, energy_data, energy_model) for cost, energy_data, energy_model in train({ rbm.v: train_set_x })]
mses_train, edata_train_list, emodel_train_list = zip(*monitoring_data_train)
lay1w = rbm.W.var.get_value()
Wl = theano.shared(lay1w)
lay1bh = rbm.bh.var.get_value()
bhl = theano.shared(lay1bh)
return [Wl, bhl]