run_tc module
Fine-tuning the library models for named entity recognition on CoNLL-2003 (Bert or Roberta).
- class run_tc.DataTrainingArguments(dataset_name: Optional[str] = None, dataset_config_name: Optional[str] = None, train_file: Optional[str] = None, validation_file: Optional[str] = None, test_file: Optional[str] = None, text_column_name: Optional[str] = None, label_column_name: Optional[str] = None, max_seq_length: int = 512, pad_to_max_length: bool = False, overwrite_cache: bool = False, preprocessing_num_workers: Optional[int] = None, task_name: str = 'ner', early_stop: bool = False)[source]
Bases:
object
Arguments pertaining to what data we are going to input our model for training and eval.
- dataset_config_name: Optional[str] = None
- dataset_name: Optional[str] = None
- early_stop: bool = False
- label_column_name: Optional[str] = None
- max_seq_length: int = 512
- overwrite_cache: bool = False
- pad_to_max_length: bool = False
- preprocessing_num_workers: Optional[int] = None
- task_name: str = 'ner'
- test_file: Optional[str] = None
- text_column_name: Optional[str] = None
- train_file: Optional[str] = None
- validation_file: Optional[str] = None
- class run_tc.ModelArguments(model_name_or_path: str, config_name: Optional[str] = None, tokenizer_name: Optional[str] = None, use_fast: bool = False, cache_dir: Optional[str] = None, log_dir: Optional[str] = None, use_auth_token: bool = False)[source]
Bases:
object
Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
- cache_dir: Optional[str] = None
- config_name: Optional[str] = None
- log_dir: Optional[str] = None
- model_name_or_path: str
- tokenizer_name: Optional[str] = None
- use_auth_token: bool = False
- use_fast: bool = False