validate_dataset.py

"""Validate synthetic disability rights & social protection dataset."""

from __future__ import annotations

from pathlib import Path

import matplotlib.pyplot as plt
import pandas as pd

SCENARIO_FILES = {
    "urban_formal_sector": "drsp_urban.csv",
    "periurban_informal": "drsp_periurban.csv",
    "rural_subsistence": "drsp_rural.csv",
}


def load_data() -> pd.DataFrame:
    frames = []
    for scenario, filename in SCENARIO_FILES.items():
        df = pd.read_csv(Path("data") / filename)
        frames.append(df)
    return pd.concat(frames, ignore_index=True)


def plot_validation(df: pd.DataFrame, output_path: Path) -> None:
    fig, axes = plt.subplots(4, 2, figsize=(14, 16))
    axes = axes.flatten()

    sp_cols = ["cash_transfer", "health_insurance", "disability_registered", "disability_card"]
    sp = df.groupby("scenario")[sp_cols].mean() * 100
    sp.plot(kind="bar", ax=axes[0])
    axes[0].set_title("Social Protection (%)")
    axes[0].legend(fontsize=6)

    edu_cols = ["ever_attended_school", "school_completed", "out_of_school"]
    edu = df.groupby("scenario")[edu_cols].mean() * 100
    edu.plot(kind="bar", ax=axes[1])
    axes[1].set_title("Education (%)")
    axes[1].legend(fontsize=7)

    emp_cols = ["employed", "formal_employment", "self_employed", "income_below_poverty"]
    emp = df.groupby("scenario")[emp_cols].mean() * 100
    emp.plot(kind="bar", ax=axes[2])
    axes[2].set_title("Employment & Poverty (%)")
    axes[2].legend(fontsize=6)

    rts_cols = ["crpd_aware", "dpo_member", "voted_last_election"]
    rts = df.groupby("scenario")[rts_cols].mean() * 100
    rts.plot(kind="bar", ax=axes[3])
    axes[3].set_title("Rights & Participation (%)")
    axes[3].legend(fontsize=7)

    disc_cols = ["discrimination_experienced", "abuse_experienced", "reported_discrimination"]
    disc = df.groupby("scenario")[disc_cols].mean() * 100
    disc.plot(kind="bar", ax=axes[4])
    axes[4].set_title("Discrimination & Abuse (%)")
    axes[4].legend(fontsize=7)

    bar_cols = ["stigma", "transport_barrier"]
    bar = df.groupby("scenario")[bar_cols].mean() * 100
    bar.plot(kind="bar", ax=axes[5])
    axes[5].set_title("Barriers (%)")
    axes[5].legend(fontsize=7)

    wb = df.groupby(["scenario", "wellbeing"]).size().groupby(level=0).apply(lambda s: s / s.sum())
    wb.unstack().plot(kind="bar", stacked=True, ax=axes[6])
    axes[6].set_title("Wellbeing Distribution")
    axes[6].legend(fontsize=7)

    dis = df.groupby(["scenario", "disability_type"]).size().groupby(level=0).apply(lambda s: s / s.sum())
    dis.unstack().plot(kind="bar", stacked=True, ax=axes[7])
    axes[7].set_title("Disability Type Distribution")
    axes[7].legend(fontsize=5)

    plt.tight_layout()
    fig.savefig(output_path, dpi=200)
    plt.close(fig)


def main() -> None:
    df = load_data()
    plot_validation(df, Path("validation_report.png"))
    print("Saved validation_report.png")


if __name__ == "__main__":
    main()