Skip to content

Add return_attention_scores support to CachedMultiHeadAttention #2213

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

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Add return_attention_scores support to CachedMultiHeadAttention #2213

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
b40bb86
Add return_attention_scores support to CachedMultiHeadAttention
DakshBegani Apr 13, 2025
54b4789
Fix: properly passing return_attention_scores to super().__init__ and…
DakshBegani Apr 15, 2025
a4aa8d3
Updated return logic
DakshBegani May 4, 2025
6d8441e
Flatteing if else condition and removing unused __int__
DakshBegani May 7, 2025
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
35 changes: 17 additions & 18 deletions keras_hub/src/layers/modeling/cached_multi_head_attention.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,11 +14,11 @@ class CachedMultiHeadAttention(keras.layers.MultiHeadAttention):

- No cache, same as regular multi-head attention.
- Static cache (`cache_update_index` is None). In this case, the
cached key/value projections will be used and the input values will
be ignored.
cached key/value projections will be used and the input values will
be ignored.
- Updated cache (`cache_update_index` is not None). In this case, new
key/value projections are computed using the input, and spliced into
the cache at the specified index.
key/value projections are computed using the input, and spliced into
the cache at the specified index.

Note that caching is useful only during inference and should not be used
during training.
Expand Down Expand Up @@ -56,12 +56,11 @@ class CachedMultiHeadAttention(keras.layers.MultiHeadAttention):
training mode or in inference mode.

Returns:
An `(attention_output, cache)` tuple. `attention_output` is the result
of the computation, of shape `(B, T, dim)`, where `T` is for target
sequence shapes and `dim` is the query input last dimension if
`output_shape` is `None`. Otherwise, the multi-head outputs are
projected to the shape specified by `output_shape`. `cache` is the
updated cache.
Depending on the `return_attention_scores` and `cache` arguments, this method returns one of:
- `attention_output`
- `(attention_output, attention_scores)`
- `(attention_output, cache)`
- `(attention_output, attention_scores, cache)`
"""

def call(
Expand All @@ -72,19 +71,14 @@ def call(
attention_mask=None,
cache=None,
cache_update_index=None,
return_attention_scores=False,
training=None,
):
if key is None:
key = value

query = self._query_dense(query)

# If cache is not `None`, we will use the cache to compute the final key
# and value tensors. If `cache_update_index` is not None, we will first
# update the cache before use. To do this, we first call the
# `_key_dense` and `_value_dense` layers, and copy the outputs into the
# cache at the specified index. `cache = None` handles the training
# case, where we don't use the cache at all.
if cache is not None:
key_cache = cache[:, 0, ...]
value_cache = cache[:, 1, ...]
Expand Down Expand Up @@ -118,6 +112,11 @@ def call(

attention_output = self._output_dense(attention_output)

if cache is not None:
if return_attention_scores and cache is not None:
return attention_output, attention_scores, cache
elif return_attention_scores:
return attention_output, attention_scores
elif cache is not None:
return attention_output, cache
return attention_output
else:
return attention_output
Loading