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

[Sentencepiece] make sure legacy do not require protobuf #25684

Merged
merged 2 commits into from
Aug 25, 2023
Merged

[Sentencepiece] make sure legacy do not require protobuf #25684

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5a49a10
make sure legacy does not require `protobuf`
ArthurZucker Aug 23, 2023
3288408
Merge branch 'main' of https://github.com/huggingface/transformers in…
ArthurZucker Aug 23, 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
7 changes: 5 additions & 2 deletions src/transformers/convert_slow_tokenizer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,17 +27,20 @@
from tokenizers.models import BPE, Unigram, WordPiece

from .utils import is_protobuf_available, requires_backends
from .utils.import_utils import PROTOBUF_IMPORT_ERROR


def import_protobuf():
def import_protobuf(error_message=""):
if is_protobuf_available():
import google.protobuf

if version.parse(google.protobuf.__version__) < version.parse("4.0.0"):
from transformers.utils import sentencepiece_model_pb2
else:
from transformers.utils import sentencepiece_model_pb2_new as sentencepiece_model_pb2
return sentencepiece_model_pb2
return sentencepiece_model_pb2
else:
raise ImportError(PROTOBUF_IMPORT_ERROR.format(error_message))


class SentencePieceExtractor:
Expand Down
13 changes: 8 additions & 5 deletions src/transformers/models/llama/tokenization_llama.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -159,14 +159,17 @@ def __init__(
# Copied from transformers.models.t5.tokenization_t5.T5Tokenizer.get_spm_processor
def get_spm_processor(self):
tokenizer = spm.SentencePieceProcessor(**self.sp_model_kwargs)
if self.legacy: # no dependency on protobuf
tokenizer.Load(self.vocab_file)
return tokenizer

with open(self.vocab_file, "rb") as f:
sp_model = f.read()
model_pb2 = import_protobuf()
model_pb2 = import_protobuf(f"The new behaviour of {self.__class__.__name__} (with `self.legacy = False`)")
model = model_pb2.ModelProto.FromString(sp_model)
if not self.legacy:
normalizer_spec = model_pb2.NormalizerSpec()
normalizer_spec.add_dummy_prefix = False
model.normalizer_spec.MergeFrom(normalizer_spec)
normalizer_spec = model_pb2.NormalizerSpec()
normalizer_spec.add_dummy_prefix = False
model.normalizer_spec.MergeFrom(normalizer_spec)
sp_model = model.SerializeToString()
tokenizer.LoadFromSerializedProto(sp_model)
return tokenizer
Expand Down
13 changes: 8 additions & 5 deletions src/transformers/models/t5/tokenization_t5.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -195,14 +195,17 @@ def __init__(

def get_spm_processor(self):
tokenizer = spm.SentencePieceProcessor(**self.sp_model_kwargs)
if self.legacy: # no dependency on protobuf
tokenizer.Load(self.vocab_file)
return tokenizer

with open(self.vocab_file, "rb") as f:
sp_model = f.read()
model_pb2 = import_protobuf()
model_pb2 = import_protobuf(f"The new behaviour of {self.__class__.__name__} (with `self.legacy = False`)")
model = model_pb2.ModelProto.FromString(sp_model)
if not self.legacy:
normalizer_spec = model_pb2.NormalizerSpec()
normalizer_spec.add_dummy_prefix = False
model.normalizer_spec.MergeFrom(normalizer_spec)
normalizer_spec = model_pb2.NormalizerSpec()
normalizer_spec.add_dummy_prefix = False
model.normalizer_spec.MergeFrom(normalizer_spec)
sp_model = model.SerializeToString()
tokenizer.LoadFromSerializedProto(sp_model)
return tokenizer
Expand Down