def get_config(self): config = { 'units': self.units, 'kl_weight': self.kl_weight, 'kl_use_exact': self.kl_use_exact, 'activation': tf.keras.activations.serialize(self.activation), 'use_bias': self.use_bias, 'activity_regularizer': tf.keras.regularizers.serialize(self.activity_regularizer), } function_keys = [ 'make_posterior_fn', 'make_prior_fn', ] for function_key in function_keys: function = getattr(self, function_key) if function is None: function_name = None function_type = None else: function_name, function_type = tfp_layers_util.serialize_function( function) config[function_key] = function_name config[function_key + '_type'] = function_type base_config = super(DenseVariational, self).get_config() return dict(list(base_config.items()) + list(config.items()))
def get_config(self): """Returns the config of the layer. A layer config is a Python dictionary (serializable) containing the configuration of a layer. The same layer can be reinstantiated later (without its trained weights) from this configuration. Returns: config: A Python dictionary of class keyword arguments and their serialized values. """ config = { "filters": self.filters, "kernel_size": self.kernel_size, "strides": self.strides, "padding": self.padding, "data_format": self.data_format, "dilation_rate": self.dilation_rate, "activation": ( tf.keras.activations.serialize(self.activation) if self.activation else None ), "activity_regularizer": tf.keras.initializers.serialize( self.activity_regularizer ), } function_keys = [ "kernel_posterior_fn", "kernel_posterior_tensor_fn", "kernel_prior_fn", "kernel_divergence_fn", "bias_posterior_fn", "bias_posterior_tensor_fn", "bias_prior_fn", "bias_divergence_fn", ] for function_key in function_keys: function = getattr(self, function_key) if function is None: function_name = None function_type = None else: function_name, function_type = tfp_layers_util.serialize_function( function ) config[function_key] = function_name config[function_key + "_type"] = function_type base_config = super(_ConvVariational, self).get_config() return dict(list(base_config.items()) + list(config.items()))
def get_config(self): """Returns the config of the layer. A layer config is a Python dictionary (serializable) containing the configuration of a layer. The same layer can be reinstantiated later (without its trained weights) from this configuration. Returns: config: A Python dictionary of class keyword arguments and their serialized values. """ config = { 'units': self.units, 'activation': (tf.keras.activations.serialize(self.activation) if self.activation else None), 'activity_regularizer': tf.keras.initializers.serialize(self.activity_regularizer), } function_keys = [ 'kernel_posterior_fn', 'kernel_posterior_tensor_fn', 'kernel_prior_fn', 'kernel_divergence_fn', 'bias_posterior_fn', 'bias_posterior_tensor_fn', 'bias_prior_fn', 'bias_divergence_fn', ] for function_key in function_keys: function = getattr(self, function_key) if function is None: function_name = None function_type = None else: function_name, function_type = tfp_layers_util.serialize_function( function) config[function_key] = function_name config[function_key + '_type'] = function_type base_config = super(_DenseVariational, self).get_config() return dict(list(base_config.items()) + list(config.items()))