def __init__(self, axis=0, **kwargs):
super().__init__(**kwargs)
self.axis = axis
self.input_spec = [
InputSpec(dtype='int32'),
InputSpec(dtype=K.floatx())
]
def __init__(self, upsampling, name='', **kwargs):
self.name = name
self.upsampling = upsampling
self.channels = None
self.input_spec = [InputSpec(ndim=4)]
super(BilinearUpsampling, self).__init__(**kwargs)
def __init__(self,
downsampling_factor=10,
init='glorot_uniform',
activation='linear',
weights=None,
W_regularizer=None,
activity_regularizer=None,
W_constraint=None,
input_dim=None,
**kwargs):
self.downsampling_factor = downsampling_factor
self.init = initializations.get(init)
self.activation = activations.get(activation)
self.W_regularizer = regularizers.get(W_regularizer)
self.activity_regularizer = regularizers.get(activity_regularizer)
self.W_constraint = constraints.get(W_constraint)
self.initial_weights = weights
self.input_dim = input_dim
if self.input_dim:
kwargs['input_shape'] = (self.input_dim, )
self.input_spec = [InputSpec(ndim=4)]
super(EltWiseProduct, self).__init__(**kwargs)
def build(self, input_shape):
self.input_spec = [
InputSpec(dtype=K.floatx(), shape=(None, ) + input_shape[1:])
]
self.activity_in_bounds.set_layer(self)
self.regularizers = [self.activity_in_bounds]
super().build(input_shape)
def __init__(self,
padding=1,
mode='CONSTANT',
constant_values=0,
**kwargs):
super(type(self), self).__init__(**kwargs)
self.padding = normalize_tuple(padding, 2, 'padding')
self.mode = mode
self.constant_values = constant_values
self.input_spec = InputSpec(ndim=3)
def __init__(self, upsampling, input_dim=None, name='', **kwargs):
self.name = name
self.upsampling = upsampling
self.channels = None
self.input_dim = input_dim
if self.input_dim:
kwargs['input_shape'] = (self.input_dim, )
self.input_spec = [InputSpec(ndim=4)]
super(BilinearUpsampling, self).__init__(**kwargs)
def __init__(self, upsampling=(2, 2), output_size=None, data_format=None, **kwargs):
super(BilinearUpsampling, self).__init__(**kwargs)
self.data_format = normalize_data_format(data_format)
self.input_spec = InputSpec(ndim=4)
if output_size:
self.output_size = conv_utils.normalize_tuple(
output_size, 2, 'output_size')
self.upsampling = None
else:
self.output_size = None
self.upsampling = conv_utils.normalize_tuple(
upsampling, 2, 'upsampling')
def build(self, input_shape):
self.input_spec = [InputSpec(ndim=4), InputSpec(ndim=2)]