actableai.utils.resources.predict.features package

Submodules

actableai.utils.resources.predict.features.association_rules module

class actableai.utils.resources.predict.features.association_rules.AssociationRulesFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Correlation Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

actableai.utils.resources.predict.features.base module

class actableai.utils.resources.predict.features.base.FeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType)

Bases: abc.ABC

Base class for features extractors

actableai.utils.resources.predict.features.bayesian_regression module

class actableai.utils.resources.predict.features.bayesian_regression.BayesianRegressionFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Regression Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

actableai.utils.resources.predict.features.classification module

class actableai.utils.resources.predict.features.classification.ClassificationFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Classification Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

class actableai.utils.resources.predict.features.classification.ClassificationTrainFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Classification Train Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['train_dataset_type', 'train_dataset_row_count', 'train_dataset_categorical_column_short_count', 'train_dataset_categorical_column_long_count', 'train_dataset_text_column_unique_short_count', 'train_dataset_text_column_unique_long_count', 'train_dataset_numbers_int_column_count', 'train_dataset_numbers_float_column_count', 'train_dataset_categorical_column_count', 'train_dataset_text_column_count', 'train_dataset_numbers_column_count', 'train_dataset_column_count', 'val_dataset_type', 'val_dataset_row_count', 'val_dataset_categorical_column_short_count', 'val_dataset_categorical_column_long_count', 'val_dataset_text_column_unique_short_count', 'val_dataset_text_column_unique_long_count', 'val_dataset_numbers_int_column_count', 'val_dataset_numbers_float_column_count', 'val_dataset_categorical_column_count', 'val_dataset_text_column_count', 'val_dataset_numbers_column_count', 'val_dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['train_dataset_type', 'train_dataset_row_count', 'train_dataset_categorical_column_short_count', 'train_dataset_categorical_column_long_count', 'train_dataset_text_column_unique_short_count', 'train_dataset_text_column_unique_long_count', 'train_dataset_numbers_int_column_count', 'train_dataset_numbers_float_column_count', 'train_dataset_categorical_column_count', 'train_dataset_text_column_count', 'train_dataset_numbers_column_count', 'train_dataset_column_count', 'val_dataset_type', 'val_dataset_row_count', 'val_dataset_categorical_column_short_count', 'val_dataset_categorical_column_long_count', 'val_dataset_text_column_unique_short_count', 'val_dataset_text_column_unique_long_count', 'val_dataset_numbers_int_column_count', 'val_dataset_numbers_float_column_count', 'val_dataset_categorical_column_count', 'val_dataset_text_column_count', 'val_dataset_numbers_column_count', 'val_dataset_column_count']}

actableai.utils.resources.predict.features.clustering module

class actableai.utils.resources.predict.features.clustering.ClusteringFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Clustering Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

class actableai.utils.resources.predict.features.clustering.DECAnchorClusteringFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

DEC Anchor Clustering Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

actableai.utils.resources.predict.features.common module

actableai.utils.resources.predict.features.common.extract_dataset_features(df_dataset: pandas.core.frame.DataFrame, prefix: str = '') → dict

Extract features from a dataset

Parameters

• df_dataset – The Pandas DataFrame dataset to extract the features from
• prefix – Prefix to add to all the features

Return type

The features extracted

actableai.utils.resources.predict.features.correlation module

class actableai.utils.resources.predict.features.correlation.CorrelationFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Correlation Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

actableai.utils.resources.predict.features.data_imputation module

class actableai.utils.resources.predict.features.data_imputation.DataImputationFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Data Imputation Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

actableai.utils.resources.predict.features.forecast module

class actableai.utils.resources.predict.features.forecast.ForecastFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Forecast Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

actableai.utils.resources.predict.features.intervention module

class actableai.utils.resources.predict.features.intervention.InterventionFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Correlation Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

actableai.utils.resources.predict.features.method module

class actableai.utils.resources.predict.features.method.MethodFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.base.FeaturesExtractor

Base class for method features extractors

actableai.utils.resources.predict.features.regression module

class actableai.utils.resources.predict.features.regression.RegressionFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Regression Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

class actableai.utils.resources.predict.features.regression.RegressionTrainFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Regression Train Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['train_dataset_type', 'train_dataset_row_count', 'train_dataset_categorical_column_short_count', 'train_dataset_categorical_column_long_count', 'train_dataset_text_column_unique_short_count', 'train_dataset_text_column_unique_long_count', 'train_dataset_numbers_int_column_count', 'train_dataset_numbers_float_column_count', 'train_dataset_categorical_column_count', 'train_dataset_text_column_count', 'train_dataset_numbers_column_count', 'train_dataset_column_count', 'val_dataset_type', 'val_dataset_row_count', 'val_dataset_categorical_column_short_count', 'val_dataset_categorical_column_long_count', 'val_dataset_text_column_unique_short_count', 'val_dataset_text_column_unique_long_count', 'val_dataset_numbers_int_column_count', 'val_dataset_numbers_float_column_count', 'val_dataset_categorical_column_count', 'val_dataset_text_column_count', 'val_dataset_numbers_column_count', 'val_dataset_column_count', 'test_dataset_type', 'test_dataset_row_count', 'test_dataset_categorical_column_short_count', 'test_dataset_categorical_column_long_count', 'test_dataset_text_column_unique_short_count', 'test_dataset_text_column_unique_long_count', 'test_dataset_numbers_int_column_count', 'test_dataset_numbers_float_column_count', 'test_dataset_categorical_column_count', 'test_dataset_text_column_count', 'test_dataset_numbers_column_count', 'test_dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['train_dataset_type', 'train_dataset_row_count', 'train_dataset_categorical_column_short_count', 'train_dataset_categorical_column_long_count', 'train_dataset_text_column_unique_short_count', 'train_dataset_text_column_unique_long_count', 'train_dataset_numbers_int_column_count', 'train_dataset_numbers_float_column_count', 'train_dataset_categorical_column_count', 'train_dataset_text_column_count', 'train_dataset_numbers_column_count', 'train_dataset_column_count', 'val_dataset_type', 'val_dataset_row_count', 'val_dataset_categorical_column_short_count', 'val_dataset_categorical_column_long_count', 'val_dataset_text_column_unique_short_count', 'val_dataset_text_column_unique_long_count', 'val_dataset_numbers_int_column_count', 'val_dataset_numbers_float_column_count', 'val_dataset_categorical_column_count', 'val_dataset_text_column_count', 'val_dataset_numbers_column_count', 'val_dataset_column_count', 'test_dataset_type', 'test_dataset_row_count', 'test_dataset_categorical_column_short_count', 'test_dataset_categorical_column_long_count', 'test_dataset_text_column_unique_short_count', 'test_dataset_text_column_unique_long_count', 'test_dataset_numbers_int_column_count', 'test_dataset_numbers_float_column_count', 'test_dataset_categorical_column_count', 'test_dataset_text_column_count', 'test_dataset_numbers_column_count', 'test_dataset_column_count']}

actableai.utils.resources.predict.features.sentiment_analysis module

class actableai.utils.resources.predict.features.sentiment_analysis.SentimentAnalysisFeaturesExtractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, function: Callable)

Bases: actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Sentiment Analysis Features Extractor

resource_predicted_features_filter = {ResourcePredictorType.MAX_GPU_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count'], ResourcePredictorType.MAX_MEMORY: ['dataset_type', 'dataset_row_count', 'dataset_categorical_column_short_count', 'dataset_categorical_column_long_count', 'dataset_text_column_unique_short_count', 'dataset_text_column_unique_long_count', 'dataset_numbers_int_column_count', 'dataset_numbers_float_column_count', 'dataset_categorical_column_count', 'dataset_text_column_count', 'dataset_numbers_column_count', 'dataset_column_count']}

Module contents

actableai.utils.resources.predict.features.extract_features(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, task: actableai.tasks.TaskType, function: Callable, *args, **kwargs) → dict

Extract features from function

Parameters

• resource_predicted – The resource to predict
• task – The task to predict for
• function – The function to extract the features from (used to get the signature)
• args – The arguments passed to the function
• kwargs – The named arguments passed to the function

Return type

The extracted features

actableai.utils.resources.predict.features.get_features_extractor(resource_predicted: actableai.utils.resources.predict.ResourcePredictorType, task: actableai.tasks.TaskType, function: Callable) → actableai.utils.resources.predict.features.method.MethodFeaturesExtractor

Get a features extractor object corresponding to the predicted resources and the task

Parameters

• resource_predicted – The resource to predict
• task – The task to predict for
• function – The function to extract the features from (used to get the signature)

Return type

Features extractor class