Type BahdanauAttentionV2
Namespace tensorflow.contrib.seq2seq
Parent _BaseAttentionMechanismV2
Interfaces IBahdanauAttentionV2
Methods
Properties
- activity_regularizer
- activity_regularizer_dyn
- alignments_size
- alignments_size_dyn
- attention_b
- attention_g
- attention_v
- batch_size
- built
- dtype
- dtype_dyn
- dynamic
- dynamic_dyn
- inbound_nodes
- inbound_nodes_dyn
- input
- input_dyn
- input_mask
- input_mask_dyn
- input_shape
- input_shape_dyn
- input_spec
- input_spec_dyn
- kernel_initializer
- keys
- losses
- losses_dyn
- memory_layer
- metrics
- metrics_dyn
- name
- name_dyn
- name_scope
- name_scope_dyn
- non_trainable_variables
- non_trainable_variables_dyn
- non_trainable_weights
- non_trainable_weights_dyn
- normalize
- outbound_nodes
- outbound_nodes_dyn
- output
- output_dyn
- output_mask
- output_mask_dyn
- output_shape
- output_shape_dyn
- probability_fn
- probability_fn_name
- PythonObject
- query_layer
- score_mask_value
- state_size
- state_size_dyn
- stateful
- submodules
- submodules_dyn
- supports_masking
- trainable
- trainable_dyn
- trainable_variables
- trainable_variables_dyn
- trainable_weights
- trainable_weights_dyn
- units
- updates
- updates_dyn
- values
- variables
- variables_dyn
- weights
- weights_dyn