feature_scoring_flu_confidence.py

"""Feature scoring with confidence intervals."""

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sns

from ema_workbench import ema_logging, load_results
from ema_workbench.analysis.feature_scoring import (
    RuleInductionType,
    get_ex_feature_scores,
)

ema_logging.log_to_stderr(level=ema_logging.INFO)

# load data
fn = r"./data/1000 flu cases no policy.tar.gz"
x, outcomes = load_results(fn)

x = x.drop(["model", "policy"], axis=1)
y = np.max(outcomes["infected_fraction_R1"], axis=1)

all_scores = []
for _ in range(100):
    indices = np.random.choice(np.arange(0, x.shape[0]), size=x.shape[0])
    selected_x = x.iloc[indices, :]
    selected_y = y[indices]

    scores = get_ex_feature_scores(
        selected_x, selected_y, mode=RuleInductionType.REGRESSION
    )[0]
    all_scores.append(scores)
all_scores = pd.concat(all_scores, axis=1, sort=False)

sns.boxplot(data=all_scores.T)
plt.show()