Fix namedtuple generated documentation (#2130)

* Fix namedtuple generated documentation

Use "Attributes" block to document attributes.

* Format with black 19.10b0

Author: guillaumekln

tensorflow_addons/seq2seq/attention_wrapper.py

@@ -1366,18 +1366,17 @@ class AttentionWrapperState(
 ):
     """`namedtuple` storing the state of a `AttentionWrapper`.
 
-    Contains:
-
-      - `cell_state`: The state of the wrapped RNN cell at the previous time
+    Attributes:
+      cell_state: The state of the wrapped RNN cell at the previous time
         step.
-      - `attention`: The attention emitted at the previous time step.
-      - `alignments`: A single or tuple of `Tensor`(s) containing the
+      attention: The attention emitted at the previous time step.
+      alignments: A single or tuple of `Tensor`(s) containing the
          alignments emitted at the previous time step for each attention
          mechanism.
-      - `alignment_history`: (if enabled) a single or tuple of `TensorArray`(s)
+      alignment_history: (if enabled) a single or tuple of `TensorArray`(s)
          containing alignment matrices from all time steps for each attention
          mechanism. Call `stack()` on each to convert to a `Tensor`.
-      - `attention_state`: A single or tuple of nested objects
+      attention_state: A single or tuple of nested objects
          containing attention mechanism state for each attention mechanism.
          The objects may contain Tensors or TensorArrays.
     """

tensorflow_addons/seq2seq/basic_decoder.py

@@ -31,12 +31,11 @@ class BasicDecoderOutput(
 ):
     """Outputs of a `BasicDecoder` step.
 
-    Contains:
-
-      - `rnn_output`: The output for this step. If the `output_layer` argument
+    Attributes:
+      rnn_output: The output for this step. If the `output_layer` argument
          of `BasicDecoder` was set, it is the output of this layer, otherwise it
          is the output of the RNN cell.
-      - `sample_id`: The token IDs sampled for this step, as returned by the
+      sample_id: The token IDs sampled for this step, as returned by the
         `sampler` instance passed to `BasicDecoder`.
     """
 

tensorflow_addons/seq2seq/beam_search_decoder.py

@@ -46,16 +46,15 @@ class BeamSearchDecoderState(
 ):
     """State of a `BeamSearchDecoder`.
 
-    Contains:
-
-      - `cell_state`: The cell state returned at the previous time step.
-      - `log_probs`: The accumulated log probabilities of each beam.
+    Attributes:
+      cell_state: The cell state returned at the previous time step.
+      log_probs: The accumulated log probabilities of each beam.
         A `float32` `Tensor` of shape `[batch_size, beam_width]`.
-      - `finished`: The finished status of each beam.
+      finished: The finished status of each beam.
         A `bool` `Tensor` of shape `[batch_size, beam_width]`.
-      - `lengths`: The accumulated length of each beam.
+      lengths: The accumulated length of each beam.
         An `int64` `Tensor` of shape `[batch_size, beam_width]`.
-      - `accumulated_attention_prob`: Accumulation of the attention
+      accumulated_attention_prob: Accumulation of the attention
         probabilities (used to compute the coverage penalty)
     """
 
@@ -69,19 +68,18 @@ class BeamSearchDecoderOutput(
 ):
     """Outputs of a `BeamSearchDecoder` step.
 
-    Contains:
-
-      - `scores`: The scores this step, which are the log
+    Attributes:
+      scores: The scores this step, which are the log
         probabilities over the output vocabulary, possibly penalized by length
         and attention coverage. When `BeamSearchDecoder` is created with
         `output_all_scores=False` (default), this will be a `float32` `Tensor`
         of shape `[batch_size, beam_width]` containing the top scores
         corresponding to the predicted IDs. When `output_all_scores=True`,
         this contains the scores for all token IDs and has shape
         `[batch_size, beam_width, vocab_size]`.
-      - `predicted_ids`: The token IDs predicted for this step.
+      predicted_ids: The token IDs predicted for this step.
         A `int32` `Tensor` of shape `[batch_size, beam_width]`.
-      - `parent_ids`: The indices of the parent beam of each beam.
+      parent_ids: The indices of the parent beam of each beam.
         A `int32` `Tensor` of shape `[batch_size, beam_width]`.
     """
 
@@ -93,10 +91,9 @@ class FinalBeamSearchDecoderOutput(
         "FinalBeamDecoderOutput", ["predicted_ids", "beam_search_decoder_output"]
     )
 ):
-    """Final outputs returned by the beam search after all decoding is
-    finished.
+    """Final outputs returned by the beam search after all decoding is finished.
 
-    Args:
+    Attributes:
       predicted_ids: The final prediction. A tensor of shape
         `[batch_size, T, beam_width]` (or `[T, batch_size, beam_width]` if
         `output_time_major` is True). Beams are ordered from best to worst.