GLM Logistic Regression (Taiwan Credit)

Experiment initialization and data preparation

from piml import Experiment
from piml.models import GLMClassifier

exp = Experiment()
exp.data_loader(data="TaiwanCredit", silent=True)
exp.data_summary(feature_exclude=["LIMIT_BAL", "SEX", "EDUCATION", "MARRIAGE", "AGE"], silent=True)
exp.data_prepare(target="FlagDefault", task_type="classification", silent=True)

Train Model

exp.model_train(model=GLMClassifier(), name="GLM")

Evaluate predictive performance

exp.model_diagnose(model="GLM", show='accuracy_table')

          ACC      AUC      F1  LogLoss    Brier

Train  0.8083   0.7375  0.3745   0.4585   0.1431
Test   0.8150   0.7356  0.3764   0.4458   0.1385
Gap    0.0067  -0.0019  0.0019  -0.0127  -0.0046

Regression coefficient plot for numerical features

exp.model_interpret(model="GLM", show="glm_coef_plot", figsize=(5, 4))

Regression coefficient table for all features

exp.model_interpret(model="GLM", show="glm_coef_table")

Feature importance plot

exp.model_interpret(model="GLM", show="global_fi", figsize=(5, 4))

Local interpretation without centering

exp.model_interpret(model="GLM", show="local_fi", sample_id=0, centered=False, original_scale=False, figsize=(5, 4))

Local interpretation with original scale of x

exp.model_interpret(model="GLM", show="local_fi", sample_id=0, centered=False, original_scale=True, figsize=(5, 4))

Local interpretation with centering and original scale of x

exp.model_interpret(model="GLM", show="local_fi", sample_id=0, centered=True, original_scale=True, figsize=(5, 4))

Estimated memory usage: 34 MB