run_seq module

Finetuning the library models for sequence classification tasks.

class run_seq.DataTrainingArguments(task_name: Optional[str] = None, dataset_name: Optional[str] = None, dataset_config_name: Optional[str] = None, max_seq_length: int = 128, overwrite_cache: bool = False, pad_to_max_length: bool = True, max_train_samples: Optional[int] = None, max_eval_samples: Optional[int] = None, max_predict_samples: Optional[int] = None, train_file: Optional[str] = None, validation_file: Optional[str] = None, test_file: Optional[str] = None)[source]

Bases: object

Arguments pertaining to what data we are going to input our model for training and eval.

Using HfArgumentParser we can turn this class into argparse arguments to be able to specify them on the command line.

dataset_config_name: Optional[str] = None
dataset_name: Optional[str] = None
max_eval_samples: Optional[int] = None
max_predict_samples: Optional[int] = None
max_seq_length: int = 128
max_train_samples: Optional[int] = None
overwrite_cache: bool = False
pad_to_max_length: bool = True
task_name: Optional[str] = None
test_file: Optional[str] = None
train_file: Optional[str] = None
validation_file: Optional[str] = None
class run_seq.ModelArguments(model_name_or_path: str, config_name: Optional[str] = None, tokenizer_name: Optional[str] = None, cache_dir: Optional[str] = None, use_fast_tokenizer: bool = True, model_revision: str = 'main', use_auth_token: bool = False, log_dir: Optional[str] = None)[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
model_revision: str = 'main'
tokenizer_name: Optional[str] = None
use_auth_token: bool = False
use_fast_tokenizer: bool = True
class run_seq.TaskArguments(task: str, early_stop: bool = False)[source]

Bases: object

Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.

early_stop: bool = False
task: str
run_seq.main(args)[source]