DataFusionTools package

Subpackages

• DataFusionTools.interpolation package
  • Submodules
  • DataFusionTools.interpolation.interpolation module
  • Module contents
• DataFusionTools.machine_learning package
  • Submodules
  • DataFusionTools.machine_learning.baseclass module
  • DataFusionTools.machine_learning.convolutional module
  • DataFusionTools.machine_learning.enumeration_classes module
  • DataFusionTools.machine_learning.mpl module
  • DataFusionTools.machine_learning.neural_networks module
  • DataFusionTools.machine_learning.random_forest module
  • DataFusionTools.machine_learning.support_vector_machine module
  • DataFusionTools.machine_learning.bayesian_neural_network module
  • Module contents
• DataFusionTools.d_series_parser
  • DataFusionTools.clustering module
  • DataFusionTools.d_stability_parser module
• DataFusionTools.sensitivity
• DataFusionTools.visualization

Submodules

DataFusionTools.data_input module

class datafusiontools._core.data_input.Data(location: datafusiontools._core.data_input.Geometry, independent_variable: typing.Optional[datafusiontools._core.data_input.Variable] = None, variables: typing.List[datafusiontools._core.data_input.Variable] = <property object>)

Bases: object

Class that contains values of one location

Parameters

• location – The location the data

• variables – Variables that are contain in the data

• independent_variable – Independent variable that is connected to all the data

get_variable(name: str)

Function that returns variable based on its name

independent_variable: Optional[datafusiontools._core.data_input.Variable] = None

location: datafusiontools._core.data_input.Geometry

update_variable(name: str, value: numpy.ndarray)

Function that updates variable

property variables

class datafusiontools._core.data_input.Geometry(x: float, y: float, z: Optional[float] = None, label: Optional[float] = None)

Bases: object

label: Optional[float] = None

x: float

y: float

z: Optional[float] = None

class datafusiontools._core.data_input.Variable(label: str, value: <built-in function array>)

Bases: object

label: str

value: numpy.array

datafusiontools._core.utils module

class datafusiontools._core.utils.AggregateMethod(value)

Bases: enum.Enum

An enumeration.

MAX = 'max'

MEAN = 'mean'

MIN = 'min'

SUM = 'sum'

class datafusiontools._core.utils.CreateInputsML

Bases: object

Utils class that creates features and targets for machine learning class

add_features(input: datafusiontools._core.data_input.Data, variable_names: List[str], use_independent_variable: bool = True, use_location_as_input: Tuple[bool, bool, bool] = (False, False, False))

Method that creates features based on the inputs given.

add_location_to_features(name_dict, value, length, dictionary=None)

add_targets(input: datafusiontools._core.data_input.Data, variable_names: List[str])

Static method that creates features based on the inputs given.

aggregate_extracted_features(agrregate_method: datafusiontools._core.utils.AggregateMethod, aggregate_variable: str, closer_extracted_features: List[datafusiontools._core.data_input.Data])

append_features(input: datafusiontools._core.data_input.Data, variable_names: List[str], use_independent_variable: bool = True, use_location_as_input: Tuple[bool, bool, bool] = (False, False, False))

Function that appends features in private class properties.

Parameters

• input – data class to be added as feature

• variable_names – list of strings that represents the features that should be extracted from the dataclass

• use_independent_variable – If true then the independent variable of the data class is used as a feature

• use_location_as_input – If true the location attribute of the data class is used as a feature

find_closer_points(input_data: List[datafusiontools._core.data_input.Data], combined_data: List[datafusiontools._core.data_input.Data], aggregate_method: datafusiontools._core.utils.AggregateMethod, aggregate_variable: str, number_of_points: int = 1, interpolate_on_independent_variable: bool = False, bounds_error: bool = False, fill_value: Union[str, numpy.array, List] = 'extrapolate')

Function that finds the closest point and aggregates results and returns those aggregated results

get_all_features(flatten: bool)

Function that returns all features in a form of a numpy.array

Parameters

flatten – the returned array is flattened per feature

get_all_targets(flatten: bool)

get_feature_names()

All the names of features are returned.

get_features(features: dict, flatten: bool)

Function that returns features from dict

Parameters

• features – a dictionary of features to be combined

• flatten – the returned array is flattened per feature

get_features_test(flatten: bool)

Function that returns features that are used for testing in a form of a numpy.array

Parameters

flatten – the returned array is flattened per feature

get_features_train(flatten: bool)

Function that returns features that are used for training in a form of a numpy.array

Parameters

flatten – the returned array is flattened per feature

get_features_validation(flatten: bool)

Function that returns features that are used for validation in a form of a numpy.array

Parameters

flatten – the returned array is flattened per feature

get_k_closest_features(point_compare: datafusiontools._core.data_input.Geometry, combined_data: List[datafusiontools._core.data_input.Data], number_of_points: int)

get_targets(targets: dict, flatten: bool)

get_targets_test(flatten: bool)

get_targets_train(flatten: bool)

get_targets_validation(flatten: bool)

interpolate_on_independent_variable(closer_extracted_features: List[datafusiontools._core.data_input.Data], main_features: datafusiontools._core.data_input.Data, aggregate_method: datafusiontools._core.utils.AggregateMethod, aggregate_variable: str, bounds_error: bool = False, fill_value: Union[str, numpy.array, List] = 'extrapolate')

split_train_test_data(train_percentage=0.7, validation_percentage_on_test=0.5)

Method that splits training and testing data.

Parameters

• train_percentage – Percentage of train samples

• validation_percentage_on_test – Percentage of validation samples taken from test samples

Module contents