Skip to content

Commit

Permalink
internal changes
Browse files Browse the repository at this point in the history
PiperOrigin-RevId: 592031242
  • Loading branch information
SeqIO Team authored and SeqIO committed Dec 18, 2023
1 parent d38e77a commit ad99fd3
Show file tree
Hide file tree
Showing 2 changed files with 3 additions and 0 deletions.
2 changes: 2 additions & 0 deletions seqio/dataset_providers.py
Original file line number Diff line number Diff line change
Expand Up @@ -1665,6 +1665,8 @@ def get_dataset(
# Shuffle before mixing since preprocessor can output multiple
# (correlated) examples per input.
ds = ds.shuffle(shuffle_buffer_size, seed=seed)


return ds.prefetch(tf.data.experimental.AUTOTUNE)

def _get_cached_source(
Expand Down
1 change: 1 addition & 0 deletions seqio/dataset_providers_test.py
Original file line number Diff line number Diff line change
Expand Up @@ -945,6 +945,7 @@ def test_plaintext_to_pretokenized_rename(self):
),
)


def test_list_shards(self):
def _get_formatted_shards_list(task_name, split):
shards = dataset_providers.get_mixture_or_task(
Expand Down

0 comments on commit ad99fd3

Please sign in to comment.