Skip to content
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

Made trainer more efficient. Loading full files instead of using TextLineDataset. #280

Merged
merged 3 commits into from
Jul 27, 2022
+7 −1
Merged

Made trainer more efficient. Loading full files instead of using TextLineDataset. #280

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5ad9d58
Made trainer more efficient. Loading full files instead of using .
jessechancy Jul 27, 2022
1f0d8b3
removed optional param
jessechancy Jul 27, 2022
f046698
added comments
jessechancy Jul 27, 2022
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
8 changes: 7 additions & 1 deletion keras_nlp/tokenizers/word_piece_tokenizer_trainer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -118,14 +118,20 @@ def compute_word_piece_vocabulary(
f"Received: {type(data)}."
)
if isinstance(data, list):
# Processing list of file paths.
if not split:
raise ValueError(
"When learning a vocab from files, `split` must be `True`. "
"To compute a vocabulary with custom split rules, load your "
"data as a dataset, split it, and pass it to "
"`compute_word_piece_vocabulary()` with split=False."
)
data = tf.data.TextLineDataset(data)
path_ds = tf.data.Dataset.from_tensor_slices(data)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should we rename data to filepaths?

# Uses map to read filepaths.
data = path_ds.map(
lambda path: tf.io.read_file(path),
num_parallel_calls=tf.data.AUTOTUNE,
)

words_data = data.map(
lambda text: pretokenize(text, lowercase, strip_accents, split),
Expand Down