dice_ml.utils package

Subpackages

• dice_ml.utils.sample_architecture package
  • Submodules
  • dice_ml.utils.sample_architecture.vae_model module
    • AutoEncoder
      • AutoEncoder.decoder()
      • AutoEncoder.encoder()
      • AutoEncoder.forward()
      • AutoEncoder.normal_likelihood()
      • AutoEncoder.sample_latent_code()
      • AutoEncoder.training
    • CF_VAE
      • CF_VAE.compute_elbo()
      • CF_VAE.decoder()
      • CF_VAE.encoder()
      • CF_VAE.forward()
      • CF_VAE.normal_likelihood()
      • CF_VAE.sample_latent_code()
      • CF_VAE.training
  • Module contents

Submodules

dice_ml.utils.exception module

Exceptions for the package.

exception dice_ml.utils.exception.SystemException

Bases: Exception

An exception indicating that some system exception happened during execution.

Parameters

exception_message (str) – A message describing the error.

dice_ml.utils.helpers module

This module containts helper functions to load data and get meta deta.

class dice_ml.utils.helpers.DataTransfomer(func=None, kw_args=None)

Bases: object

A class to transform data based on user-defined function to get predicted outcomes. This class calls FunctionTransformer of scikit-learn internally (https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.FunctionTransformer.html).

feed_data_params(data_interface)

initialize_transform_func()

inverse_transform(data)

transform(data)

dice_ml.utils.helpers.get_adult_data_info()

dice_ml.utils.helpers.get_adult_income_modelpath(backend='TF1')

dice_ml.utils.helpers.get_base_gen_cf_initialization(data_interface, encoded_size, cont_minx, cont_maxx, margin, validity_reg, epochs, wm1, wm2, wm3, learning_rate)

dice_ml.utils.helpers.get_custom_dataset_modelpath_pipeline()

dice_ml.utils.helpers.get_custom_dataset_modelpath_pipeline_binary()

dice_ml.utils.helpers.get_custom_dataset_modelpath_pipeline_multiclass()

dice_ml.utils.helpers.get_custom_dataset_modelpath_pipeline_regression()

dice_ml.utils.helpers.get_custom_vars_dataset_modelpath_pipeline()

dice_ml.utils.helpers.inverse_ohe_min_max_transformation(data, data_interface)

dice_ml.utils.helpers.load_adult_income_dataset(only_train=True)

Loads adult income dataset from https://archive.ics.uci.edu/ml/datasets/Adult and prepares

the data for data analysis based on https://rpubs.com/H_Zhu/235617

Return adult_data

returns preprocessed adult income dataset.

dice_ml.utils.helpers.load_custom_testing_dataset()

dice_ml.utils.helpers.load_custom_testing_dataset_binary()

dice_ml.utils.helpers.load_custom_testing_dataset_binary_str()

dice_ml.utils.helpers.load_custom_testing_dataset_multiclass()

dice_ml.utils.helpers.load_custom_testing_dataset_multiclass_str()

dice_ml.utils.helpers.load_custom_testing_dataset_regression()

dice_ml.utils.helpers.load_min_max_equal_dataset()

dice_ml.utils.helpers.load_outcome_not_last_column_dataset()

dice_ml.utils.helpers.ohe_min_max_transformation(data, data_interface)

the data is one-hot-encoded and min-max normalized and fed to the ML model

dice_ml.utils.helpers.save_adult_income_model(modelpath, test_fraction=0.2, random_state=0)

dice_ml.utils.neuralnetworks module

class dice_ml.utils.neuralnetworks.FFNetwork(*args: Any, **kwargs: Any)

Bases: Module

forward(x)

class dice_ml.utils.neuralnetworks.MulticlassNetwork(*args: Any, **kwargs: Any)

Bases: Module

forward(x)

dice_ml.utils.serialize module

class dice_ml.utils.serialize.DummyDataInterface(outcome_name, data_df=None)

Bases: object

to_json()

Module contents