Shapley Global Importance

API reference for sovai.extensions.shapley_global_importance

Module: sovai.extensions.shapley_global_importance

Classes

ClusteringExplainer

class ClusteringExplainer

Trains a classifier to predict cluster labels and provides SHAP explanations.

Attributes

• random_state

• model

• explainer

• scaler

Methods

__init__()

def __init__(self, random_state = 42)

Initializes the explainer and scaler.

Parameters

Parameter	Type	Description
random_state	—	Default: 42

---

fit()

def fit(self, X, y)

Fits the LGBM classifier and creates the SHAP explainer.

Parameters

Parameter	Type	Description
X	—	—
y	—	—

---

get_shap_values()

def get_shap_values(self, X)

Gets SHAP values using the trained explainer.

Parameters

Parameter	Type	Description
X	—	—

---

---

Functions

hash_of_df()

def hash_of_df(df, sample_size = 100)

Calculates a SHA256 hash of a sampled portion of a DataFrame.

Parameters

Parameter	Type	Description
df	—	—
sample_size	—	Default: 100

---

get_shap_values_for_dataset()

def get_shap_values_for_dataset(
    df,
    clustering_method = 'KMEANS',
    n_clusters = 10,
    random_state = 42,
    sample_size = 5000,
)

Performs clustering, trains a model, and calculates mean absolute SHAP values.

Parameters

Parameter	Type	Description
df	—	—
clustering_method	—	Default: 'KMEANS'
n_clusters	—	Default: 10
random_state	—	Default: 42
sample_size	—	Default: 5000

---

run_simulations_frame_global()

def run_simulations_frame_global(df, num_simulations = 4, clustering_method = 'KMEANS')

Runs multiple simulations of SHAP value calculation in parallel and averages.

Parameters

Parameter	Type	Description
df	—	—
num_simulations	—	Default: 4
clustering_method	—	Default: 'KMEANS'

---

run_simulations_global_importance()

def run_simulations_global_importance(
    df,
    num_simulations = 4,
    clustering_method = 'KMEANS',
)

Calculates overall feature importance based on averaged SHAP values.

Parameters

Parameter	Type	Description
df	—	—
num_simulations	—	Default: 4
clustering_method	—	Default: 'KMEANS'

---