Skip to content

Handle OOV token in XLMRoBERTaTokenizer's token_to_id function #968

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 5 commits into from
Apr 12, 2023
Merged

Handle OOV token in XLMRoBERTaTokenizer's token_to_id function #968

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
4b739c6
Handle OOV token in XLMRoBERTaTokenizer's token_to_id function
abheesht17 Apr 6, 2023
a95b8ac
Add special token check in UT
abheesht17 Apr 6, 2023
3eabd3c
Optimise
abheesht17 Apr 7, 2023
8fb746a
Format code
abheesht17 Apr 7, 2023
978cf09
Simplify
abheesht17 Apr 7, 2023
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
10 changes: 9 additions & 1 deletion keras_nlp/models/xlm_roberta/xlm_roberta_tokenizer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -123,10 +123,18 @@ def id_to_token(self, id):

def token_to_id(self, token):
"""Convert a string token to an integer id."""

if token in self._vocabulary_prefix:
return self._vocabulary_prefix.index(token)

return int(self._sentence_piece.string_to_id(token).numpy()) + 1
spm_token_id = self._sentence_piece.string_to_id(token)

# OOV token
spm_unk_token_id = self._sentence_piece.string_to_id("<unk>")
if spm_token_id == spm_unk_token_id:
return self.unk_token_id

return int(spm_token_id.numpy()) + 1

def tokenize(self, inputs):
tokens = super().tokenize(inputs)
Expand Down
4 changes: 4 additions & 0 deletions keras_nlp/models/xlm_roberta/xlm_roberta_tokenizer_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,6 +96,10 @@ def test_id_to_token(self):
def test_token_to_id(self):
self.assertEqual(self.tokenizer.token_to_id("▁the"), 4)
self.assertEqual(self.tokenizer.token_to_id("▁round"), 10)
# Test any random OOV token.
self.assertEqual(self.tokenizer.token_to_id("<oov-token>"), 3)
# Test a special token.
self.assertEqual(self.tokenizer.token_to_id("<pad>"), 1)

def test_serialization(self):
config = keras.utils.serialize_keras_object(self.tokenizer)
Expand Down