autointent.Pipeline#

class autointent.Pipeline(nodes)#

Pipeline optimizer class.

Parameters:: nodes (list[autointent.nodes.NodeOptimizer] | list[autointent.nodes.InferenceNode])

nodes#

set_config(config)#

Set configuration for the optimizer.

Parameters:: config (autointent.configs.LoggingConfig | autointent.configs.VectorIndexConfig | autointent.configs.EmbedderConfig) – Configuration
Return type:: None

classmethod from_search_space(search_space)#

Create pipeline optimizer from dictionary search space.

Parameters:: search_space (list[dict[str, Any]] | pathlib.Path | str) – Dictionary config
Return type:: Pipeline

classmethod default_optimizer(multilabel)#

Create pipeline optimizer with default search space for given classification task.

Parameters:: multilabel (bool) – Whether the task multi-label, or single-label.
Return type:: Pipeline

fit(dataset, force_multilabel=False)#

Optimize the pipeline from dataset.

Parameters:

dataset (autointent.Dataset) – Dataset for optimization
force_multilabel (bool) – Whether to force multilabel or not

Returns:

Context

Return type:

autointent.Context

classmethod from_dict_config(nodes_configs)#

Create inference pipeline from dictionary config.

Parameters:: nodes_configs (list[dict[str, Any]]) – list of dictionary config for nodes
Returns:: pipeline ready for inference
Return type:: Pipeline

classmethod from_config(nodes_configs)#

Create inference pipeline from config.

Parameters:: nodes_configs (list[autointent.configs.InferenceNodeConfig]) – list of config for nodes
Return type:: Pipeline

classmethod load(path)#

Load pipeline in inference mode.

This method loads fitted modules and tuned hyperparameters. :path: path to optimization run directory :return: initialized pipeline, ready for inference

Parameters:: path (str | pathlib.Path)
Return type:: Pipeline

predict(utterances)#

Predict the labels for the utterances.

Parameters:: utterances (list[str]) – list of utterances
Returns:: list of predicted labels
Return type:: numpy.typing.NDArray[Any]

predict_with_metadata(utterances)#

Predict the labels for the utterances with metadata.

Parameters:: utterances (list[str]) – list of utterances
Returns:: prediction output
Return type:: autointent._pipeline._schemas.InferencePipelineOutput