medcat.components.addons.meta_cat.mctokenizers.bert_tokenizer
Classes:
-
TokenizerWrapperBERT–Wrapper around a huggingface BERT tokenizer so that it works with the
Attributes:
FAKE_TOKENIZER_PATH
module-attribute
FAKE_TOKENIZER_PATH = '#\n/fake-path-not-exist#/'
TokenizerWrapperBERT
TokenizerWrapperBERT(hf_tokenizers: Optional[BertTokenizerFast] = None)
Bases: TokenizerWrapperBase
Wrapper around a huggingface BERT tokenizer so that it works with the MetaCAT models.
Parameters:
-
–transformers.models.bert.tokenization_bert_fast.BertTokenizerFastA huggingface Fast BERT.
Methods:
-
create_new– -
get_pad_id– -
get_size– -
load– -
save– -
token_to_id–
Attributes:
-
name–
Source code in medcat-v2/medcat/components/addons/meta_cat/mctokenizers/bert_tokenizer.py
23 24 25 | |
name
class-attribute
instance-attribute
name = 'bert-tokenizer'
create_new
classmethod
create_new(model_variant: Optional[str]) -> TokenizerWrapperBERT
Source code in medcat-v2/medcat/components/addons/meta_cat/mctokenizers/bert_tokenizer.py
85 86 87 88 | |
get_pad_id
Source code in medcat-v2/medcat/components/addons/meta_cat/mctokenizers/bert_tokenizer.py
98 99 100 | |
get_size
get_size() -> int
Source code in medcat-v2/medcat/components/addons/meta_cat/mctokenizers/bert_tokenizer.py
90 91 92 | |
load
classmethod
load(dir_path: str, model_variant: Optional[str] = '', **kwargs) -> TokenizerWrapperBERT
Source code in medcat-v2/medcat/components/addons/meta_cat/mctokenizers/bert_tokenizer.py
65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 | |
save
save(dir_path: str) -> None
Source code in medcat-v2/medcat/components/addons/meta_cat/mctokenizers/bert_tokenizer.py
60 61 62 63 | |
token_to_id
Source code in medcat-v2/medcat/components/addons/meta_cat/mctokenizers/bert_tokenizer.py
94 95 96 | |