deeppavlov.core.commands

Basic training and inference functions.

deeppavlov.core.commands.infer.build_model(config: Union[str, pathlib.Path, dict], mode: str = 'infer', load_trained: bool = False, download: bool = False, serialized: Optional[bytes] = None) → deeppavlov.core.common.chainer.Chainer

Build and return the model described in corresponding configuration file.

deeppavlov.core.commands.infer.interact_model(config: Union[str, pathlib.Path, dict]) → None

Start interaction with the model described in corresponding configuration file.

deeppavlov.core.commands.infer.predict_on_stream(config: Union[str, pathlib.Path, dict], batch_size: int = 1, file_path: Optional[str] = None) → None

Make a prediction with the component described in corresponding configuration file.

class deeppavlov.core.commands.train.Metric(name, fn, inputs)

fn

Alias for field number 1

inputs

Alias for field number 2

name

Alias for field number 0

deeppavlov.core.commands.train.fit_chainer(config: dict, iterator: Union[deeppavlov.core.data.data_learning_iterator.DataLearningIterator, deeppavlov.core.data.data_fitting_iterator.DataFittingIterator]) → deeppavlov.core.common.chainer.Chainer

Fit and return the chainer described in corresponding configuration dictionary.

deeppavlov.core.commands.train.get_iterator_from_config(config: dict, data: dict)

Create iterator (from config) for specified data.

deeppavlov.core.commands.train.prettify_metrics(metrics: List[Tuple[str, float]], precision: int = 4) → collections.OrderedDict

Prettifies the dictionary of metrics.

deeppavlov.core.commands.train.read_data_by_config(config: dict)

Read data by dataset_reader from specified config.

deeppavlov.core.commands.train.train_evaluate_model_from_config(config: [<class 'str'>, <class 'pathlib.Path'>, <class 'dict'>], iterator=None, *, to_train=True, to_validate=True, download=False, start_epoch_num=0, recursive=False) → Dict[str, Dict[str, float]]

Make training and evaluation of the model described in corresponding configuration file.