Skip to content

Refine BeamSearchDecoder doc #48882

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Dec 8, 2022
Merged

Refine BeamSearchDecoder doc #48882

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 8 additions & 9 deletions python/paddle/nn/decode.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,8 +26,6 @@

class Decoder:
"""
:api_attr: Static Graph

Decoder is the base class for any decoder instance used in `dynamic_decode`.
It provides interface for output generation for one time step, which can be
used to generate sequences.
Expand Down Expand Up @@ -146,13 +144,14 @@ class BeamSearchDecoder(Decoder):
Please refer to `Beam search <https://en.wikipedia.org/wiki/Beam_search>`_
for more details.

**NOTE** When decoding with beam search, the `inputs` and `states` of cell
would be tiled to `beam_size` (unsqueeze and tile), resulting to shapes like
`[batch_size * beam_size, ...]` , which is built into `BeamSearchDecoder` and
done automatically. Thus any other tensor with shape `[batch_size, ...]` used
in `cell.call` needs to be tiled manually first, which can be completed by using
:code:`BeamSearchDecoder.tile_beam_merge_with_batch` . The most common case
for this is the encoder output in attention mechanism.
Note:
When decoding with beam search, the `inputs` and `states` of cell
would be tiled to `beam_size` (unsqueeze and tile), resulting to shapes like
`[batch_size * beam_size, ...]` , which is built into `BeamSearchDecoder` and
done automatically. Thus any other tensor with shape `[batch_size, ...]` used
in `cell.call` needs to be tiled manually first, which can be completed by using
:code:`BeamSearchDecoder.tile_beam_merge_with_batch` . The most common case
for this is the encoder output in attention mechanism.

Returns:
BeamSearchDecoder: An instance of decoder which can be used in \
Expand Down