def __init__(self,
degree,
depth,
mark=None,
max_volterra_order=3,
**kwargs):
# Check parameters
if degree < 1: raise ValueError('!! Degree must be a positive integer')
if depth < 0: raise ValueError('!! Depth must be a positive integer')
# Call parent's constructor
Model.__init__(self, mark)
# Initialize fields
self.degree = degree
self.depth = depth
self._max_volterra_order = min(max_volterra_order, degree)
self.T = {}
self._input = Input([depth], name='input')
self._output = None
self._target = None
self._alpha = 1.1
self._outputs = {}
# Initialize operators in each degree
orders = kwargs.get('orders', None)
if orders is None: orders = list(range(1, self.degree + 1))
self.orders = orders
self._init_T()
def __init__(self, mark=None):
Model.__init__(self, mark)
RNet.__init__(self, 'RecurrentNet')
self.superior = self
self._default_net = self
# Attributes
self._state = NestedTensorSlot(self, 'State')
# mascot will be initiated as a placeholder with no shape specified
# .. and will be put into initializer argument of tf.scan
self._mascot = None
def advanced_one_step(self, *args, lr_list=None, **kwargs):
if lr_list is None: lr_list = [0.000088] * self.branches_num
for i in range(self.branches_num):
self.set_branch_index(i)
if i > 0:
FLAGS.overwrite = False
FLAGS.save_best = True
self._optimizer_lr_modify(lr_list[i])
Model.train(self, *args, **kwargs)
lr_list = [0.000088, 0.00088, 0.000088]
self.train(branch_index=self.branches_num, lr_list=lr_list, **kwargs)
def build(self,
optimizer=None,
loss='euclid',
metric=None,
metric_is_like_loss=True,
metric_name='Metric',
**kwargs):
Model.build(self,
optimizer=optimizer,
loss=loss,
metric=metric,
metric_name=metric_name,
metric_is_like_loss=metric_is_like_loss)
def __init__(self, mark=None, **kwargs):
# Call parent's initializer
Model.__init__(self, mark)
Net.__init__(self, 'Bamboo_Broad_Net', inter_type=pedia.fork)
assert self._inter_type == pedia.fork
self.outputs = None
# Private fields
self._losses = []
self._metrics = []
self._train_ops = []
self._var_list = []
self._output_list = []
self._branch_index = 0
self._identity_initial = kwargs.get('ientity', False)
def train(self, *args, branch_index=0, lr_list=None, **kwargs):
if lr_list is None: lr_list = [0.000088] * self.branches_num
self.set_branch_index(branch_index)
freeze = kwargs.get('freeze', True)
if not freeze:
train_step = []
for i in range(branch_index + 1):
self._optimizer_lr_modify(lr_list[i])
train_step.append(
self._optimizer.minimize(loss=self._loss,
var_list=self._var_list[i]))
self._train_step = train_step
Model.train(self, *args, **kwargs)
def __init__(self,
z_dim=None,
sample_shape=None,
output_shape=None,
mark=None,
classes=0):
# Call parent's constructor
Model.__init__(self, mark)
self._targets = None
self._conditional = classes > 0
self._classes = classes
if self._conditional:
with self._graph.as_default():
self._targets = tf.placeholder(dtype=tf.float32,
shape=[None, classes],
name='one_hot_labels')
# Define generator and discriminator
self.Generator = Net(pedia.Generator)
self.Discriminator = Net(pedia.Discriminator)
# Alias
self.G = self.Generator
self.D = self.Discriminator
# If z_dim/sample_shape is provided, define the input for
# generator/discriminator accordingly
if z_dim is not None:
self.G.add(Input(sample_shape=[None, z_dim], name='z'))
if sample_shape is not None:
if (not isinstance(sample_shape, list)
and not isinstance(sample_shape, tuple)):
raise TypeError('sample shape must be a list or a tuple')
self.D.add(
Input(sample_shape=[None] + list(sample_shape),
name='samples'))
self._z_dim = z_dim
self._sample_shape = sample_shape
self._output_shape = output_shape
self._sample_num = None
# Private tensors and ops
self._G, self._outputs = None, None
self._Dr, self._Df = None, None
self._logits_Dr, self._logits_Df = None, None
self._loss_G, self._loss_D = None, None
self._loss_Dr, self._loss_Df = None, None
self._train_step_G, self._train_step_D = None, None
self._merged_summary_G, self._merged_summary_D = None, None
def build(self,
optimizer=None,
loss='euclid',
metric=None,
batch_metric=None,
eval_metric=None,
**kwargs):
context.metric_name = 'unknown' # TODO: to be deprecated
Model.build(self,
optimizer=optimizer,
loss=loss,
metric=metric,
batch_metric=batch_metric,
eval_metric=eval_metric,
**kwargs)
def __init__(self, mark=None):
Model.__init__(self, mark)
RNet.__init__(self, 'RecurrentNet')
self.superior = self
self._default_net = self
# Attributes
self._state_slot = NestedTensorSlot(self, 'State')
# mascot will be initiated as a placeholder with no shape specified
# .. and will be put into initializer argument of tf.scan
self._mascot = None
self._while_loop_free_output = None
# TODO: BETA
self.last_scan_output = None
self.grad_delta_slot = NestedTensorSlot(self, 'GradDelta')
self._grad_buffer_slot = NestedTensorSlot(self, 'GradBuffer')
def __init__(self,
z_dim=None,
sample_shape=None,
output_shape=None,
mark=None,
classes=0):
# Call parent's constructor
Model.__init__(self, mark)
# Fields
self._output_shape = output_shape
# Define encoder and decoder
self.Encoder = Net(pedia.Encoder)
self.Decoder = Net(pedia.Decoder)
self.Q = self.Encoder
self.P = self.Decoder
# If z_dim/sample_shape is provided, define the input for
# decoder/encoder accordingly
if z_dim is not None:
self.P.add(Input(sample_shape=[None, z_dim], name='z'))
if sample_shape is not None:
if (not isinstance(sample_shape, list)
and not isinstance(sample_shape, tuple)):
raise TypeError('sample shape must be a list or a tuple')
self.Q.add(
Input(sample_shape=[None] + list(sample_shape),
name='samples'))
# Placeholders
self._sample_num = None
self._classes = classes
self._conditional = classes > 0
if self._conditional:
self._targets = tf.placeholder(dtype=tf.float32,
shape=[None, classes],
name='one_hot_labels')
self._P, self._outputs = None, None
# ...
pass
def __init__(self, mark=None):
Model.__init__(self, mark)
Net.__init__(self, 'FeedforwardNet')
self.superior = self
self._default_net = self
def __init__(self, mark=None):
Model.__init__(self, mark)
Net.__init__(self, 'FeedforwardNet')
self.outputs = None
