Buckets:
| import torch | |
| from transformers import T5Tokenizer, UMT5EncoderModel | |
| from .._mixin import OstrisTransformersMixin | |
| class PatchedT5Tokenizer(T5Tokenizer): | |
| def __init__( | |
| self, | |
| vocab=None, | |
| eos_token="</s>", | |
| unk_token="<unk>", | |
| pad_token="<pad>", | |
| _spm_precompiled_charsmap=None, | |
| extra_ids=100, | |
| additional_special_tokens=None, | |
| **kwargs, | |
| ): | |
| super().__init__( | |
| vocab=vocab, | |
| eos_token=eos_token, | |
| unk_token=unk_token, | |
| pad_token=pad_token, | |
| _spm_precompiled_charsmap=None, # this is passing a empty byte string for some reason now. | |
| extra_ids=extra_ids, | |
| additional_special_tokens=additional_special_tokens, | |
| **kwargs, | |
| ) | |
| class UMT5TextEncoder(UMT5EncoderModel, OstrisTransformersMixin): | |
| """UMT5-XXL text encoder (wan family).""" | |
| aitk_subfolder = "text_encoder" | |
| aitk_tokenizer_subfolder = "tokenizer" | |
| aitk_config_repo = "ai-toolkit/umt5_xxl_encoder" | |
| # comfy fp8 umt5 files store non-quantized tensors in a dtype mix; T5 | |
| # assumes a uniform dtype, so follow the load dtype | |
| aitk_cast_quantized_load = True | |
| aitk_comfy_repo = "Comfy-Org/Wan_2.1_ComfyUI_repackaged" | |
| # comfy umt5 files already use the transformers key layout; the | |
| # fp8_e4m3fn_scaled variant is the legacy scaled-fp8 format (handled by | |
| # the importer's float8 backend) | |
| aitk_comfy_weight_names = { | |
| "ai-toolkit/umt5_xxl_encoder": [ | |
| "split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors", | |
| "split_files/text_encoders/umt5_xxl_fp16.safetensors", | |
| ], | |
| } | |
| def get_transformer_block_names(cls): | |
| return ["encoder.block"] | |
| def convert_state_dict_on_load(cls, state_dict): | |
| # drop the embedded sentencepiece blob and materialize the tied | |
| # embed_tokens reference | |
| state_dict = dict(state_dict) | |
| state_dict.pop("spiece_model", None) | |
| if ( | |
| "encoder.embed_tokens.weight" not in state_dict | |
| and "shared.weight" in state_dict | |
| ): | |
| state_dict["encoder.embed_tokens.weight"] = state_dict["shared.weight"] | |
| return state_dict | |
| def load_tokenizer(cls, name_or_path=None, subfolder=None, **kwargs): | |
| # T5's tokenizer needs the _spm_precompiled_charsmap patch | |
| source = name_or_path if name_or_path is not None else cls.aitk_tokenizer_repo | |
| if subfolder is None: | |
| subfolder = cls.aitk_tokenizer_subfolder | |
| return PatchedT5Tokenizer.from_pretrained( | |
| source, subfolder=subfolder or "", **kwargs | |
| ) | |
Xet Storage Details
- Size:
- 2.73 kB
- Xet hash:
- 1e2397c5266f034fcf26933897dbe1cdcfcad0a539ac97dd92a85ccd169faea9
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.