image-match/python/py/Lib/site-packages/transformers/models/siglip2/tokenization_siglip2.py

#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
#           This file was automatically generated from src/transformers/models/siglip2/modular_siglip2.py.
#               Do NOT edit this file manually as any edits will be overwritten by the generation of
#             the file from the modular. If any change should be done, please apply the change to the
#                          modular_siglip2.py file directly. One of our CI enforces this.
#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
# Copyright 2025 The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.


from tokenizers import Tokenizer, decoders, normalizers, pre_tokenizers
from tokenizers.models import BPE

from ...tokenization_utils_tokenizers import TokenizersBackend


VOCAB_FILES_NAMES = {"tokenizer_file": "tokenizer.json"}


class Siglip2Tokenizer(TokenizersBackend):
    """
    Gemma tokenizer + SigLIP2 training default: lowercase normalization.
    """

    vocab_files_names = VOCAB_FILES_NAMES
    padding_side = "left"
    model_input_names = ["input_ids", "attention_mask"]
    model = BPE

    def __init__(
        self,
        vocab: str | dict[str, int] | None = None,
        merges: str | list[str] | None = None,
        unk_token: str = "<unk>",
        bos_token: str = "<bos>",
        eos_token: str = "<eos>",
        pad_token: str = "<pad>",
        mask_token: str = "<mask>",
        **kwargs,
    ):
        if vocab is None:
            vocab = {
                str(pad_token): 0,
                str(eos_token): 1,
                str(bos_token): 2,
                str(unk_token): 3,
                str(mask_token): 4,
            }
        self._vocab = vocab
        self._merges = merges or []

        self._tokenizer = Tokenizer(
            BPE(
                vocab=self._vocab,
                merges=self._merges,
                fuse_unk=True,
                unk_token=str(unk_token),
                dropout=None,
                byte_fallback=True,
            )
        )
        self._tokenizer.pre_tokenizer = pre_tokenizers.Split(
            pattern=" ", behavior="merged_with_previous", invert=False
        )

        self._tokenizer.decoder = decoders.Sequence(
            [decoders.Replace("▁", " "), decoders.ByteFallback(), decoders.Fuse()]
        )
        self._tokenizer.normalizer = normalizers.Replace(" ", "▁")
        super().__init__(
            unk_token=unk_token,
            bos_token=bos_token,
            eos_token=eos_token,
            pad_token=pad_token,
            mask_token=mask_token,
            **kwargs,
        )

        # Persist for save/load + push_to_hub dynamic tokenizer test
        if hasattr(self, "init_kwargs") and isinstance(self.init_kwargs, dict):
            self.init_kwargs.setdefault("tokenizer_class", self.__class__.__name__)

        backend = getattr(self, "_tokenizer", None)
        if backend is not None and backend.normalizer is not None:
            backend.normalizer = normalizers.Sequence([normalizers.Lowercase(), backend.normalizer])


__all__ = ["Siglip2Tokenizer"]