interpret_community.common.constants module¶
Defines constants for interpret community.
- class interpret_community.common.constants.Attributes¶
Bases:
object
Provide constants for attributes.
- EXPECTED_VALUE = 'expected_value'¶
- class interpret_community.common.constants.DNNFramework¶
Bases:
object
Provide DNN framework constants.
- PYTORCH = 'pytorch'¶
- TENSORFLOW = 'tensorflow'¶
- class interpret_community.common.constants.Defaults¶
Bases:
object
Provide constants for default values to explain methods.
- AUTO = 'auto'¶
- DEFAULT_BATCH_SIZE = 100¶
- HDBSCAN = 'hdbscan'¶
- MAX_DIM = 50¶
- class interpret_community.common.constants.Dynamic¶
Bases:
object
Provide constants for dynamically generated classes.
- GLOBAL_EXPLANATION = 'DynamicGlobalExplanation'¶
- LOCAL_EXPLANATION = 'DynamicLocalExplanation'¶
- class interpret_community.common.constants.ExplainParams¶
Bases:
object
Provide constants for interpret community (init, explain_local and explain_global) parameters.
- BATCH_SIZE = 'batch_size'¶
- CLASSES = 'classes'¶
- CLASSIFICATION = 'classification'¶
- EVAL_DATA = 'eval_data'¶
- EVAL_Y_PRED = 'eval_y_predicted'¶
- EVAL_Y_PRED_PROBA = 'eval_y_predicted_proba'¶
- EXPECTED_VALUES = 'expected_values'¶
- EXPLAIN_SUBSET = 'explain_subset'¶
- EXPLANATION_ID = 'explanation_id'¶
- FEATURES = 'features'¶
- GLOBAL_IMPORTANCE_NAMES = 'global_importance_names'¶
- GLOBAL_IMPORTANCE_RANK = 'global_importance_rank'¶
- GLOBAL_IMPORTANCE_VALUES = 'global_importance_values'¶
- GLOBAL_NAMES = 'global_names'¶
- GLOBAL_RANK = 'global_rank'¶
- GLOBAL_VALUES = 'global_values'¶
- ID = 'id'¶
- INCLUDE_LOCAL = 'include_local'¶
- INIT_DATA = 'init_data'¶
- IS_ENG = 'is_engineered'¶
- IS_LOCAL_SPARSE = 'is_local_sparse'¶
- IS_RAW = 'is_raw'¶
- LOCAL_EXPLANATION = 'local_explanation'¶
- LOCAL_IMPORTANCE_VALUES = 'local_importance_values'¶
- METHOD = 'method'¶
- MODEL_ID = 'model_id'¶
- MODEL_TASK = 'model_task'¶
- MODEL_TYPE = 'model_type'¶
- NUM_CLASSES = 'num_classes'¶
- NUM_EXAMPLES = 'num_examples'¶
- NUM_FEATURES = 'num_features'¶
- PER_CLASS_NAMES = 'per_class_names'¶
- PER_CLASS_RANK = 'per_class_rank'¶
- PER_CLASS_VALUES = 'per_class_values'¶
- PROBABILITIES = 'probabilities'¶
- SAMPLING_POLICY = 'sampling_policy'¶
- SHAP_VALUES_OUTPUT = 'shap_values_output'¶
- classmethod get_private(explain_param)¶
Return the private version of the ExplainParams property.
- Parameters
cls (ExplainParams) – ExplainParams input class.
explain_param (str) – The ExplainParams property to get private version of.
- Returns
The private version of the property.
- Return type
- classmethod get_serializable()¶
Return only the ExplainParams properties that have meaningful data values for serialization.
- Parameters
cls (ExplainParams) – ExplainParams input class.
- Returns
A set of property names, e.g., ‘GLOBAL_IMPORTANCE_VALUES’, ‘MODEL_TYPE’, etc.
- Return type
- class interpret_community.common.constants.ExplainType¶
Bases:
object
Provide constants for model and explainer type information, useful for visualization.
- CLASSIFICATION = 'classification'¶
- DATA = 'data_type'¶
- EXPLAIN = 'explain_type'¶
- EXPLAINER = 'explainer'¶
- FUNCTION = 'function'¶
- GLOBAL = 'global'¶
- HAN = 'han'¶
- IS_ENG = 'is_engineered'¶
- IS_RAW = 'is_raw'¶
- LIME = 'lime'¶
- LOCAL = 'local'¶
- METHOD = 'method'¶
- MIMIC = 'mimic'¶
- MODEL = 'model_type'¶
- MODEL_CLASS = 'model_class'¶
- MODEL_TASK = 'model_task'¶
- PFI = 'pfi'¶
- REGRESSION = 'regression'¶
- SHAP = 'shap'¶
- SHAP_DEEP = 'shap_deep'¶
- SHAP_GPU_KERNEL = 'shap_gpu_kernel'¶
- SHAP_KERNEL = 'shap_kernel'¶
- SHAP_LINEAR = 'shap_linear'¶
- SHAP_TREE = 'shap_tree'¶
- TABULAR = 'tabular'¶
- class interpret_community.common.constants.ExplainableModelType(value)¶
-
Provide constants for the explainable model type.
- LINEAR_EXPLAINABLE_MODEL_TYPE = 'linear_explainable_model_type'¶
- TREE_EXPLAINABLE_MODEL_TYPE = 'tree_explainable_model_type'¶
- class interpret_community.common.constants.ExplanationParams¶
Bases:
object
Provide constants for explanation parameters.
- CLASSES = 'classes'¶
- EXPECTED_VALUES = 'expected_values'¶
- class interpret_community.common.constants.Extension¶
Bases:
object
Provide constants for extensions to interpret package.
- BLACKBOX = 'blackbox'¶
- GLASSBOX = 'model'¶
- GLOBAL = 'global'¶
- GREYBOX = 'specific'¶
- LOCAL = 'local'¶
- class interpret_community.common.constants.InterpretData¶
Bases:
object
Provide Data and Visualize constants for interpret core.
- BASE_VALUE = 'Base Value'¶
- EXPLANATION_CLASS_DIMENSION = 'explanation_class_dimension'¶
- EXPLANATION_TYPE = 'explanation_type'¶
- EXTRA = 'extra'¶
- FEATURE_LIST = 'feature_list'¶
- GLOBAL_FEATURE_IMPORTANCE = 'global_feature_importance'¶
- INTERCEPT = 'intercept'¶
- LOCAL_FEATURE_IMPORTANCE = 'local_feature_importance'¶
- MLI = 'mli'¶
- MULTICLASS = 'multiclass'¶
- NAMES = 'names'¶
- OVERALL = 'overall'¶
- PERF = 'perf'¶
- SCORES = 'scores'¶
- SINGLE = 'single'¶
- SPECIFIC = 'specific'¶
- TYPE = 'type'¶
- UNIVARIATE = 'univariate'¶
- VALUE = 'value'¶
- VALUES = 'values'¶
- class interpret_community.common.constants.LightGBMParams¶
Bases:
object
Provide constants for LightGBM.
- CATEGORICAL_FEATURE = 'categorical_feature'¶
- class interpret_community.common.constants.LightGBMSerializationConstants¶
Bases:
object
Provide internal class that defines fields used for MimicExplainer serialization.
- IDENTITY = '_identity'¶
- LOGGER = '_logger'¶
- MODEL_STR = 'model_str'¶
- MULTICLASS = 'multiclass'¶
- OBJECTIVE = 'objective'¶
- REGRESSION = 'regression'¶
- TREE_EXPLAINER = '_tree_explainer'¶
- enum_properties = ['_shap_values_output']¶
- nonify_properties = ['_logger', '_tree_explainer']¶
- save_properties = ['_lgbm']¶
- class interpret_community.common.constants.MimicSerializationConstants¶
Bases:
object
Provide internal class that defines fields used for MimicExplainer serialization.
- ALLOW_ALL_TRANSFORMATIONS = '_allow_all_transformations'¶
- FUNCTION = 'function'¶
- IDENTITY = '_identity'¶
- INITIALIZATION_EXAMPLES = 'initialization_examples'¶
- LOGGER = '_logger'¶
- MODEL = 'model'¶
- ORIGINAL_EVAL_EXAMPLES = '_original_eval_examples'¶
- PREDICT_PROBA_FLAG = 'predict_proba_flag'¶
- RESET_INDEX = 'reset_index'¶
- TIMESTAMP_FEATURIZER = '_timestamp_featurizer'¶
- enum_properties = ['_shap_values_output']¶
- nonify_properties = ['_logger', 'model', 'function', 'initialization_examples', '_original_eval_examples', '_timestamp_featurizer']¶
- save_properties = ['surrogate_model']¶
- class interpret_community.common.constants.ModelTask(value)¶
-
Provide model task constants. Can be ‘classification’, ‘regression’, or ‘unknown’.
By default the model domain is inferred if ‘unknown’, but this can be overridden if you specify ‘classification’ or ‘regression’.
- Classification = 'classification'¶
- Regression = 'regression'¶
- Unknown = 'unknown'¶
- class interpret_community.common.constants.ResetIndex(value)¶
-
Provide index column handling constants. Can be ‘ignore’, ‘reset’ or ‘reset_teacher’.
By default the index column is ignored, but you can override to reset it and make it a feature column that is then featurized to numeric, or reset it and ignore it during featurization but set it as the index when calling predict on the original model.
- Ignore = 'ignore'¶
- Reset = 'reset'¶
- ResetTeacher = 'reset_teacher'¶
- class interpret_community.common.constants.SHAPDefaults¶
Bases:
object
Provide constants for default values to SHAP.
- INDEPENDENT = 'independent'¶
- class interpret_community.common.constants.SKLearn¶
Bases:
object
Provide scikit-learn related constants.
- EXAMPLES = 'examples'¶
- LABELS = 'labels'¶
- PREDICTIONS = 'predictions'¶
- PREDICT_PROBA = 'predict_proba'¶
- class interpret_community.common.constants.Scipy¶
Bases:
object
Provide scipy related constants.
- CSR_FORMAT = 'csr'¶
- class interpret_community.common.constants.ShapValuesOutput(value)¶
-
Provide constants for the SHAP values output from the explainer.
Can be ‘default’, ‘probability’ or ‘teacher_probability’. If ‘teacher_probability’ is specified, we use the probabilities from the teacher model.
- DEFAULT = 'default'¶
- PROBABILITY = 'probability'¶
- TEACHER_PROBABILITY = 'teacher_probability'¶