abacusai.api_class.batch_prediction =================================== .. py:module:: abacusai.api_class.batch_prediction Classes ------- .. autoapisummary:: abacusai.api_class.batch_prediction.BatchPredictionArgs abacusai.api_class.batch_prediction.ForecastingBatchPredictionArgs abacusai.api_class.batch_prediction.NamedEntityExtractionBatchPredictionArgs abacusai.api_class.batch_prediction.PersonalizationBatchPredictionArgs abacusai.api_class.batch_prediction.PredictiveModelingBatchPredictionArgs abacusai.api_class.batch_prediction.PretrainedModelsBatchPredictionArgs abacusai.api_class.batch_prediction.SentenceBoundaryDetectionBatchPredictionArgs abacusai.api_class.batch_prediction.ThemeAnalysisBatchPredictionArgs abacusai.api_class.batch_prediction.ChatLLMBatchPredictionArgs abacusai.api_class.batch_prediction.TrainablePlugAndPlayBatchPredictionArgs abacusai.api_class.batch_prediction.AIAgentBatchPredictionArgs abacusai.api_class.batch_prediction._BatchPredictionArgsFactory Module Contents --------------- .. py:class:: BatchPredictionArgs Bases: :py:obj:`abacusai.api_class.abstract.ApiClass` An abstract class for Batch Prediction args specific to problem type. .. py:attribute:: _support_kwargs :type: bool :value: True .. py:attribute:: kwargs :type: dict .. py:attribute:: problem_type :type: abacusai.api_class.enums.ProblemType :value: None .. py:method:: _get_builder() :classmethod: .. py:class:: ForecastingBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the FORECASTING problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation :type for_eval: bool :param predictions_start_date: The start date for predictions. Accepts timestamp integers and strings in many standard formats such as YYYY-MM-DD, YYYY-MM-DD HH:MM:SS, or YYYY-MM-DDTHH:MM:SS. If not specified, the prediction start date will be automatically defined. :type predictions_start_date: str :param use_prediction_offset: If True, use prediction offset. :type use_prediction_offset: bool :param start_date_offset: Sets prediction start date as this offset relative to the prediction start date. :type start_date_offset: int :param forecasting_horizon: The number of timestamps to predict in the future. Range: [1, 1000]. :type forecasting_horizon: int :param item_attributes_to_include_in_the_result: List of columns to include in the prediction output. :type item_attributes_to_include_in_the_result: list :param explain_predictions: If True, calculates explanations for the forecasted values along with predictions. :type explain_predictions: bool :param create_monitor: Controls whether to automatically create a monitor to calculate the drift each time the batch prediction is run. Defaults to true if not specified. :type create_monitor: bool .. py:attribute:: for_eval :type: bool :value: None .. py:attribute:: predictions_start_date :type: str :value: None .. py:attribute:: use_prediction_offset :type: bool :value: None .. py:attribute:: start_date_offset :type: int :value: None .. py:attribute:: forecasting_horizon :type: int :value: None .. py:attribute:: item_attributes_to_include_in_the_result :type: list :value: None .. py:attribute:: explain_predictions :type: bool :value: None .. py:attribute:: create_monitor :type: bool :value: None .. py:method:: __post_init__() .. py:class:: NamedEntityExtractionBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the NAMED_ENTITY_EXTRACTION problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation. :type for_eval: bool .. py:attribute:: for_eval :type: bool :value: None .. py:method:: __post_init__() .. py:class:: PersonalizationBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the PERSONALIZATION problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation. :type for_eval: bool :param number_of_items: Number of items to recommend. :type number_of_items: int :param item_attributes_to_include_in_the_result: List of columns to include in the prediction output. :type item_attributes_to_include_in_the_result: list :param score_field: If specified, relative item scores will be returned using a field with this name :type score_field: str .. py:attribute:: for_eval :type: bool :value: None .. py:attribute:: number_of_items :type: int :value: None .. py:attribute:: item_attributes_to_include_in_the_result :type: list :value: None .. py:attribute:: score_field :type: str :value: None .. py:method:: __post_init__() .. py:class:: PredictiveModelingBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the PREDICTIVE_MODELING problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation. :type for_eval: bool :param explainer_type: The type of explainer to use to generate explanations on the batch prediction. :type explainer_type: enums.ExplainerType :param number_of_samples_to_use_for_explainer: Number Of Samples To Use For Kernel Explainer. :type number_of_samples_to_use_for_explainer: int :param include_multi_class_explanations: If True, Includes explanations for all classes in multi-class classification. :type include_multi_class_explanations: bool :param features_considered_constant_for_explanations: Comma separate list of fields to treat as constant in SHAP explanations. :type features_considered_constant_for_explanations: str :param importance_of_records_in_nested_columns: Returns importance of each index in the specified nested column instead of SHAP column explanations. :type importance_of_records_in_nested_columns: str :param explanation_filter_lower_bound: If set explanations will be limited to predictions above this value, Range: [0, 1]. :type explanation_filter_lower_bound: float :param explanation_filter_upper_bound: If set explanations will be limited to predictions below this value, Range: [0, 1]. :type explanation_filter_upper_bound: float :param explanation_filter_label: For classification problems specifies the label to which the explanation bounds are applied. :type explanation_filter_label: str :param output_columns: A list of column names to include in the prediction result. :type output_columns: list :param explain_predictions: If True, calculates explanations for the predicted values along with predictions. :type explain_predictions: bool :param create_monitor: Controls whether to automatically create a monitor to calculate the drift each time the batch prediction is run. Defaults to true if not specified. :type create_monitor: bool .. py:attribute:: for_eval :type: bool :value: None .. py:attribute:: explainer_type :type: abacusai.api_class.enums.ExplainerType :value: None .. py:attribute:: number_of_samples_to_use_for_explainer :type: int :value: None .. py:attribute:: include_multi_class_explanations :type: bool :value: None .. py:attribute:: features_considered_constant_for_explanations :type: str :value: None .. py:attribute:: importance_of_records_in_nested_columns :type: str :value: None .. py:attribute:: explanation_filter_lower_bound :type: float :value: None .. py:attribute:: explanation_filter_upper_bound :type: float :value: None .. py:attribute:: explanation_filter_label :type: str :value: None .. py:attribute:: output_columns :type: list :value: None .. py:attribute:: explain_predictions :type: bool :value: None .. py:attribute:: create_monitor :type: bool :value: None .. py:method:: __post_init__() .. py:class:: PretrainedModelsBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the PRETRAINED_MODELS problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation. :type for_eval: bool :param files_output_location_prefix: The output location prefix for the files. :type files_output_location_prefix: str :param channel_id_to_label_map: JSON string for the map from channel ids to their labels. :type channel_id_to_label_map: str .. py:attribute:: for_eval :type: bool :value: None .. py:attribute:: files_output_location_prefix :type: str :value: None .. py:attribute:: channel_id_to_label_map :type: str :value: None .. py:method:: __post_init__() .. py:class:: SentenceBoundaryDetectionBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the SENTENCE_BOUNDARY_DETECTION problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation :type for_eval: bool :param explode_output: Explode data so there is one sentence per row. :type explode_output: bool .. py:attribute:: for_eval :type: bool :value: None .. py:attribute:: explode_output :type: bool :value: None .. py:method:: __post_init__() .. py:class:: ThemeAnalysisBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the THEME_ANALYSIS problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation. :type for_eval: bool :param analysis_frequency: The length of each analysis interval. :type analysis_frequency: str :param start_date: The end point for predictions. :type start_date: str :param analysis_days: How many days to analyze. :type analysis_days: int .. py:attribute:: for_eval :type: bool :value: None .. py:attribute:: analysis_frequency :type: str :value: None .. py:attribute:: start_date :type: str :value: None .. py:attribute:: analysis_days :type: int :value: None .. py:method:: __post_init__() .. py:class:: ChatLLMBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the ChatLLM problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation. :type for_eval: bool .. py:attribute:: for_eval :type: bool :value: None .. py:method:: __post_init__() .. py:class:: TrainablePlugAndPlayBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the TrainablePlugAndPlay problem type :param for_eval: If True, the test fold which was created during training and used for metrics calculation will be used as input data. These predictions are hence, used for model evaluation. :type for_eval: bool :param create_monitor: Controls whether to automatically create a monitor to calculate the drift each time the batch prediction is run. Defaults to true if not specified. :type create_monitor: bool .. py:attribute:: for_eval :type: bool :value: None .. py:attribute:: create_monitor :type: bool :value: None .. py:method:: __post_init__() .. py:class:: AIAgentBatchPredictionArgs Bases: :py:obj:`BatchPredictionArgs` Batch Prediction Config for the AIAgents problem type .. py:method:: __post_init__() .. py:class:: _BatchPredictionArgsFactory Bases: :py:obj:`abacusai.api_class.abstract._ApiClassFactory` Helper class that provides a standard way to create an ABC using inheritance. .. py:attribute:: config_abstract_class .. py:attribute:: config_class_key :value: 'problem_type' .. py:attribute:: config_class_map