validate_dataset.py

#!/usr/bin/env python3
"""Validation & Diagnostic Visualization for Medical Oxygen Supply Dataset."""

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import os

SCENARIOS = ['referral_hospital', 'district_hospital', 'rural_health_centre']


def load_scenarios(data_dir='data'):
    dfs = {}
    for sc in SCENARIOS:
        path = os.path.join(data_dir, f'oxygen_{sc}.csv')
        if os.path.exists(path):
            dfs[sc] = pd.read_csv(path)
    return dfs


def make_report(dfs, output='validation_report.png'):
    fig, axes = plt.subplots(4, 2, figsize=(16, 24))
    fig.suptitle(
        'Medical Oxygen Supply — Validation Report\n'
        '(Referral Hospital → District Hospital → Rural Health Centre)',
        fontsize=15, fontweight='bold', y=0.99)
    colors = ['#2ecc71', '#f39c12', '#e74c3c']
    x = np.arange(len(SCENARIOS))
    labels = ['Referral Hosp', 'District Hosp', 'Rural HC']

    # Panel 1: Oxygen available today
    ax = axes[0, 0]
    avail = [dfs[sc]['oxygen_available_today'].mean()*100 for sc in SCENARIOS if sc in dfs]
    ax.bar(x, avail, color=colors, alpha=0.8)
    ax.set_xticks(x)
    ax.set_xticklabels(labels, fontsize=9)
    for i, v in enumerate(avail):
        ax.text(i, v + 1, f'{v:.0f}%', ha='center', fontsize=10, fontweight='bold')
    ax.set_ylabel('Availability (%)')
    ax.set_title('Oxygen Available on Day of Assessment')
    ax.set_ylim(0, 100)

    # Panel 2: Oxygen source distribution (district)
    ax = axes[0, 1]
    df = dfs.get('district_hospital', list(dfs.values())[0])
    src = df['primary_oxygen_source'].value_counts()
    ax.barh(range(len(src)), src.values, color='#3498db', alpha=0.7)
    ax.set_yticks(range(len(src)))
    ax.set_yticklabels([s.replace('_', ' ').title() for s in src.index], fontsize=8)
    ax.set_xlabel('Count')
    ax.set_title('Primary Oxygen Source (District Hospital)')

    # Panel 3: Pulse oximetry functional
    ax = axes[1, 0]
    pox = [dfs[sc]['pulse_oximeter_functional'].mean()*100 for sc in SCENARIOS if sc in dfs]
    ax.bar(x, pox, color=colors, alpha=0.8)
    ax.set_xticks(x)
    ax.set_xticklabels(labels, fontsize=9)
    for i, v in enumerate(pox):
        ax.text(i, v + 1, f'{v:.0f}%', ha='center', fontsize=10, fontweight='bold')
    ax.set_ylabel('Rate (%)')
    ax.set_title('Functional Pulse Oximetry')

    # Panel 4: Patients needing vs receiving oxygen
    ax = axes[1, 1]
    w = 0.3
    need = [dfs[sc]['patients_needing_oxygen'].mean() for sc in SCENARIOS if sc in dfs]
    recv = [dfs[sc]['patients_received_oxygen'].mean() for sc in SCENARIOS if sc in dfs]
    ax.bar(x - w/2, need, w, label='Needing O₂', color='#e74c3c', alpha=0.8)
    ax.bar(x + w/2, recv, w, label='Received O₂', color='#2ecc71', alpha=0.8)
    ax.set_xticks(x)
    ax.set_xticklabels(labels, fontsize=9)
    ax.set_ylabel('Mean Patients/Month')
    ax.set_title('Oxygen Need vs Delivery Gap')
    ax.legend(fontsize=8)

    # Panel 5: Hypoxemia deaths
    ax = axes[2, 0]
    deaths = [dfs[sc]['deaths_hypoxemia_related'].mean() for sc in SCENARIOS if sc in dfs]
    ax.bar(x, deaths, color=colors, alpha=0.8)
    ax.set_xticks(x)
    ax.set_xticklabels(labels, fontsize=9)
    for i, v in enumerate(deaths):
        ax.text(i, v + 0.1, f'{v:.1f}', ha='center', fontsize=10, fontweight='bold')
    ax.set_ylabel('Mean Deaths/Month')
    ax.set_title('Hypoxemia-Related Deaths (per observation)')

    # Panel 6: Shortage causes (district)
    ax = axes[2, 1]
    short_df = df[df['shortage_cause'] != 'not_applicable']
    if len(short_df) > 0:
        causes = short_df['shortage_cause'].value_counts().head(8)
        ax.barh(range(len(causes)), causes.values, color='#e74c3c', alpha=0.7)
        ax.set_yticks(range(len(causes)))
        ax.set_yticklabels([s.replace('_', ' ').title() for s in causes.index], fontsize=7)
        ax.set_xlabel('Count')
    ax.set_title('Top Oxygen Shortage Causes (District)')

    # Panel 7: Concentrator functional rate
    ax = axes[3, 0]
    conc_func = []
    for sc in SCENARIOS:
        if sc in dfs:
            has_conc = dfs[sc][dfs[sc]['concentrator_count'] > 0]
            if len(has_conc) > 0:
                rate = (has_conc['concentrator_functional'] / has_conc['concentrator_count']).mean() * 100
            else:
                rate = 0
            conc_func.append(rate)
    ax.bar(x, conc_func, color=colors, alpha=0.8)
    ax.set_xticks(x)
    ax.set_xticklabels(labels, fontsize=9)
    for i, v in enumerate(conc_func):
        ax.text(i, v + 1, f'{v:.0f}%', ha='center', fontsize=10, fontweight='bold')
    ax.set_ylabel('Functional Rate (%)')
    ax.set_title('Concentrator Functionality (among facilities with concentrators)')

    # Panel 8: Distance to refill & transport
    ax = axes[3, 1]
    dist = [dfs[sc]['distance_to_refill_km'].mean() for sc in SCENARIOS if sc in dfs]
    ax.bar(x, dist, color=colors, alpha=0.8)
    ax.set_xticks(x)
    ax.set_xticklabels(labels, fontsize=9)
    for i, v in enumerate(dist):
        ax.text(i, v + 2, f'{v:.0f}km', ha='center', fontsize=10, fontweight='bold')
    ax.set_ylabel('Distance (km)')
    ax.set_title('Mean Distance to Cylinder Refill Point')

    plt.tight_layout(rect=[0, 0, 1, 0.97])
    plt.savefig(output, dpi=150, bbox_inches='tight')
    print(f'Saved validation report to {output}')
    plt.close()


if __name__ == '__main__':
    dfs = load_scenarios()
    if dfs:
        make_report(dfs)