Quickstart

Analyst Facade

from ScoreCardModel import ScoreCardWrapper

sc = ScoreCardWrapper(binning_strategy='quantile', base_points=600, pdo=20)
sc.fit(X_train, y_train)

scores = sc.predict(X_test)
card = sc.export_scorecard()
print(card.head(10))

Scikit-learn Pipeline

from sklearn.pipeline import Pipeline
from sklearn.linear_model import LogisticRegression
from ScoreCardModel import BinningTransformer, WOETransformer

pipeline = Pipeline([
    ('binning', BinningTransformer(strategy='tree', n_bins=5)),
    ('woe', WOETransformer(method='empirical_logit')),
    ('model', LogisticRegression())
])
pipeline.fit(X_train, y_train)

Scorecard Templates

from ScoreCardModel.templates import BaseScorecard

sc = BaseScorecard()
sc.fit(X_train, y_train)
scores = sc.predict(X_test)

Feature Selection

from ScoreCardModel.analytics.selection import rank_features, select_by_correlation

ranking = rank_features(X_train, y_train)
accept = ranking[ranking['Recommendation'] == 'Accept']['Feature'].tolist()
final = select_by_correlation(X_train[accept], max_corr=0.7)
pipeline.fit(X_train[final], y_train)

Automated Report

from ScoreCardModel.analytics.reporting import generate_report

generate_report(pipeline, X_train, y_train, X_test, y_test,
                output_path="scorecard_report.md")

Generates a Markdown report with embedded PNG plots, scorecard table, and executive summary — ready for regulatory review.

Interactive What-If (Jupyter)

# Requires: pip install scorecard-toolkit[interactive]
from ScoreCardModel.interactive import ScorecardWidget

widget = ScorecardWidget(pipeline, X_train)
widget.display()

Opens an interactive dashboard with sliders/dropdowns for each feature and a live waterfall chart that updates as you adjust values. See the Interactive Notebook for a full walkthrough.

Scorecard Table in Jupyter

from ScoreCardModel.score_card.transformers import ScoreCardTransformer

lr, bt, wt = pipeline.named_steps['model'], pipeline.named_steps['binning'], pipeline.named_steps['woe']
sct = ScoreCardTransformer(lr, bt, wt)
sct  # renders as a styled HTML table in the notebook