Upload folder using huggingface_hub

README.md

@@ -0,0 +1,73 @@
+---
+license: cc-by-4.0
+task_categories:
+  - tabular-classification
+language:
+  - en
+tags:
+  - disability
+  - intellectual-disability
+  - developmental-delays
+  - early-intervention
+  - special-education
+  - synthetic
+  - sub-saharan-africa
+pretty_name: Intellectual Disability & Developmental Delays (SSA)
+size_categories:
+  - 10K<n<100K
+configs:
+  - config_name: urban_specialist_centre
+    data_files: data/intdis_urban.csv
+    default: true
+  - config_name: district_integrated
+    data_files: data/intdis_district.csv
+  - config_name: rural_community
+    data_files: data/intdis_rural.csv
+---
+
+# Intellectual Disability & Developmental Delays in Sub-Saharan Africa
+
+## Abstract
+
+Synthetic dataset modelling intellectual disability causes, diagnosis, early intervention, education, and family support across three service tiers in SSA. Prevalence 1-3% globally, higher in LMICs; perinatal asphyxia, malnutrition, infections are major causes.
+
+## Parameterization Evidence
+
+| Parameter | Value | Source | Year |
+| --- | --- | --- | --- |
+| ID prevalence 1-3% globally; higher in LMICs | Burden | WHO | 2023 |
+| Causes: asphyxia, malaria, meningitis, malnutrition | Aetiology | Lancet | 2022 |
+| <0.1 psychiatrists per 100K in most SSA | Workforce | WHO Atlas | 2020 |
+
+## Validation
+
+![Validation Report](validation_report.png)
+
+## Usage
+
+```python
+from datasets import load_dataset
+ds = load_dataset("electricsheepafrica/intellectual-disability-developmental", "urban_specialist_centre")
+```
+
+## References
+
+1. WHO. Intellectual disability fact sheet. 2023.
+2. Lancet. Developmental disabilities in SSA. 2022.
+3. WHO Mental Health Atlas. 2020.
+
+## Citation
+
+```bibtex
+@dataset{electricsheepafrica_intellectual_disability_2025,
+  title={Intellectual Disability and Developmental Delays in Sub-Saharan Africa},
+  author={Electric Sheep Africa},
+  year={2025},
+  publisher={HuggingFace},
+  url={https://huggingface.co/datasets/electricsheepafrica/intellectual-disability-developmental}
+}
+```
+
+## License
+
+CC-BY-4.0

generate_dataset.py

@@ -0,0 +1,180 @@
+"""Generate synthetic intellectual disability & developmental delays dataset for SSA.
+
+Research-based parameterization:
+- WHO: Intellectual disability prevalence 1-3% globally; higher in
+  LMICs due to malnutrition, infections, birth complications.
+- Lancet: In SSA, causes include perinatal asphyxia, malaria,
+  meningitis, malnutrition, iodine deficiency, consanguinity.
+- Few specialized services; <0.1 psychiatrists per 100K in most SSA;
+  stigma major barrier; institutionalization declining but community
+  services lacking.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+
+SEED = 42
+N_PER_SCENARIO = 10_000
+
+YEAR_RANGE = np.arange(2010, 2025)
+YEAR_WEIGHTS = np.linspace(0.85, 1.3, len(YEAR_RANGE))
+YEAR_WEIGHTS = YEAR_WEIGHTS / YEAR_WEIGHTS.sum()
+
+SCENARIOS = {
+    "urban_specialist_centre": {
+        "setting_probs": {"specialist_centre": 0.30, "paediatric_hospital": 0.25,
+                          "special_school": 0.25, "private_clinic": 0.20},
+        "cause_probs": {"perinatal_asphyxia": 0.20, "genetic_chromosomal": 0.15,
+                        "meningitis_encephalitis": 0.12, "malnutrition": 0.10,
+                        "iodine_deficiency": 0.05, "congenital_infection": 0.08,
+                        "trauma": 0.05, "unknown": 0.25},
+        "diagnosis_rate": 0.40,
+        "early_intervention_pct": 0.15,
+        "special_education_pct": 0.30,
+        "family_support_pct": 0.20,
+    },
+    "district_integrated": {
+        "setting_probs": {"district_hospital": 0.30, "health_centre": 0.25,
+                          "community_programme": 0.25, "inclusive_school": 0.20},
+        "cause_probs": {"perinatal_asphyxia": 0.25, "malnutrition": 0.15,
+                        "meningitis_encephalitis": 0.12, "malaria_cerebral": 0.10,
+                        "iodine_deficiency": 0.08, "congenital_infection": 0.08,
+                        "genetic_chromosomal": 0.07, "unknown": 0.15},
+        "diagnosis_rate": 0.15,
+        "early_intervention_pct": 0.05,
+        "special_education_pct": 0.10,
+        "family_support_pct": 0.08,
+    },
+    "rural_community": {
+        "setting_probs": {"health_post": 0.30, "community_home": 0.30,
+                          "traditional_healer": 0.15, "cbr_programme": 0.25},
+        "cause_probs": {"perinatal_asphyxia": 0.25, "malnutrition": 0.20,
+                        "meningitis_encephalitis": 0.10, "malaria_cerebral": 0.10,
+                        "iodine_deficiency": 0.10, "congenital_infection": 0.05,
+                        "genetic_chromosomal": 0.05, "unknown": 0.15},
+        "diagnosis_rate": 0.05,
+        "early_intervention_pct": 0.02,
+        "special_education_pct": 0.03,
+        "family_support_pct": 0.03,
+    },
+}
+
+SCENARIO_FILES = {
+    "urban_specialist_centre": "intdis_urban.csv",
+    "district_integrated": "intdis_district.csv",
+    "rural_community": "intdis_rural.csv",
+}
+
+
+def _choice(rng, prob_map):
+    keys = list(prob_map.keys())
+    weights = np.array(list(prob_map.values()), dtype=float)
+    weights = weights / weights.sum()
+    return rng.choice(keys, p=weights)
+
+
+def _simulate_scenario(name, params, seed):
+    rng = np.random.default_rng(seed)
+    records = []
+
+    for idx in range(N_PER_SCENARIO):
+        year = int(rng.choice(YEAR_RANGE, p=YEAR_WEIGHTS))
+        setting = _choice(rng, params["setting_probs"])
+        age = int(np.clip(rng.lognormal(np.log(8), 0.8), 0, 40))
+        sex = rng.choice(["male", "female"], p=[0.55, 0.45])
+
+        cause = _choice(rng, params["cause_probs"])
+        severity = rng.choice(["mild", "moderate", "severe", "profound"],
+                              p=[0.40, 0.30, 0.20, 0.10])
+        comorbid_epilepsy = int(rng.random() < 0.20)
+        comorbid_cerebral_palsy = int(rng.random() < 0.12)
+        comorbid_autism = int(rng.random() < 0.05)
+        age_at_concern = int(np.clip(rng.exponential(2), 0, 10))
+
+        # Diagnosis & assessment
+        formally_diagnosed = int(rng.random() < params["diagnosis_rate"])
+        developmental_screening = int(rng.random() < params["diagnosis_rate"] * 1.5)
+        iq_assessed = int(formally_diagnosed and rng.random() < 0.30)
+        age_at_diagnosis = int(np.clip(age_at_concern + rng.exponential(3), 1, 15)) if formally_diagnosed else 0
+
+        # Interventions
+        early_intervention = int(age < 6 and rng.random() < params["early_intervention_pct"])
+        speech_therapy = int(rng.random() < 0.05)
+        occupational_therapy = int(rng.random() < 0.04)
+        behavioural_support = int(rng.random() < 0.05)
+        medication = int(comorbid_epilepsy and rng.random() < 0.40)
+
+        # Education
+        school_enrolled = int(age >= 5 and rng.random() < (0.60 if formally_diagnosed else 0.30))
+        special_education = int(school_enrolled and rng.random() < params["special_education_pct"])
+        inclusive_education = int(school_enrolled and not special_education and rng.random() < 0.15)
+        dropped_out = int(school_enrolled and rng.random() < 0.30)
+
+        # Family & social
+        family_support = int(rng.random() < params["family_support_pct"])
+        caregiver_burden = int(rng.random() < 0.60)
+        caregiver_mental_health = int(caregiver_burden and rng.random() < 0.30)
+        stigma_experienced = int(rng.random() < 0.45)
+        social_isolation = int(stigma_experienced and rng.random() < 0.50)
+        abuse_neglect_risk = int(rng.random() < 0.15)
+
+        # Barriers
+        financial_barrier = int(rng.random() < 0.55)
+        awareness_barrier = int(rng.random() < 0.50)
+        no_services_available = int(rng.random() < 0.45)
+
+        # Outcomes
+        daily_living_skills = rng.choice(["independent", "supervised", "dependent"],
+                                          p=[0.20, 0.40, 0.40] if severity in ("mild", "moderate")
+                                          else [0.05, 0.25, 0.70])
+        community_participation = int(rng.random() < (0.35 if family_support else 0.15))
+
+        record = {
+            "record_id": f"{name[:3].upper()}-{idx:05d}",
+            "scenario": name,
+            "year": year,
+            "setting": setting,
+            "age": age,
+            "sex": sex,
+            "cause": cause,
+            "severity": severity,
+            "comorbid_epilepsy": comorbid_epilepsy,
+            "comorbid_cerebral_palsy": comorbid_cerebral_palsy,
+            "formally_diagnosed": formally_diagnosed,
+            "developmental_screening": developmental_screening,
+            "early_intervention": early_intervention,
+            "speech_therapy": speech_therapy,
+            "medication": medication,
+            "school_enrolled": school_enrolled,
+            "special_education": special_education,
+            "inclusive_education": inclusive_education,
+            "dropped_out": dropped_out,
+            "family_support": family_support,
+            "caregiver_burden": caregiver_burden,
+            "stigma_experienced": stigma_experienced,
+            "abuse_neglect_risk": abuse_neglect_risk,
+            "financial_barrier": financial_barrier,
+            "no_services_available": no_services_available,
+            "daily_living_skills": daily_living_skills,
+            "community_participation": community_participation,
+        }
+        records.append(record)
+
+    return pd.DataFrame(records)
+
+
+def main():
+    output_dir = Path("data")
+    output_dir.mkdir(parents=True, exist_ok=True)
+    for idx, (name, params) in enumerate(SCENARIOS.items()):
+        df = _simulate_scenario(name, params, SEED + idx * 211)
+        df.to_csv(output_dir / SCENARIO_FILES[name], index=False)
+        print(f"Saved {name} -> {SCENARIO_FILES[name]}")
+
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,3 @@
+numpy>=1.24
+pandas>=2.0
+matplotlib>=3.7

validate_dataset.py

@@ -0,0 +1,88 @@
+"""Validate synthetic intellectual disability & developmental delays dataset."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import matplotlib.pyplot as plt
+import pandas as pd
+
+SCENARIO_FILES = {
+    "urban_specialist_centre": "intdis_urban.csv",
+    "district_integrated": "intdis_district.csv",
+    "rural_community": "intdis_rural.csv",
+}
+
+COLORS = {"urban_specialist_centre": "#e6550d", "district_integrated": "#756bb1", "rural_community": "#31a354"}
+
+
+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()
+
+    dx_cols = ["formally_diagnosed", "developmental_screening", "early_intervention"]
+    dx = df.groupby("scenario")[dx_cols].mean() * 100
+    dx.plot(kind="bar", ax=axes[0])
+    axes[0].set_title("Diagnosis & Early Intervention (%)")
+    axes[0].legend(fontsize=7)
+
+    cau = df.groupby(["scenario", "cause"]).size().groupby(level=0).apply(lambda s: s / s.sum())
+    cau.unstack().plot(kind="bar", stacked=True, ax=axes[1])
+    axes[1].set_title("Cause Distribution")
+    axes[1].legend(fontsize=4)
+
+    edu_cols = ["school_enrolled", "special_education", "inclusive_education", "dropped_out"]
+    edu = df.groupby("scenario")[edu_cols].mean() * 100
+    edu.plot(kind="bar", ax=axes[2])
+    axes[2].set_title("Education (%)")
+    axes[2].legend(fontsize=6)
+
+    soc_cols = ["family_support", "caregiver_burden", "stigma_experienced", "abuse_neglect_risk"]
+    soc = df.groupby("scenario")[soc_cols].mean() * 100
+    soc.plot(kind="bar", ax=axes[3])
+    axes[3].set_title("Family & Social (%)")
+    axes[3].legend(fontsize=6)
+
+    sev = df.groupby(["scenario", "severity"]).size().groupby(level=0).apply(lambda s: s / s.sum())
+    sev.unstack().plot(kind="bar", stacked=True, ax=axes[4])
+    axes[4].set_title("Severity Distribution")
+    axes[4].legend(fontsize=7)
+
+    bar_cols = ["financial_barrier", "no_services_available"]
+    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)
+
+    com_cols = ["comorbid_epilepsy", "comorbid_cerebral_palsy"]
+    com = df.groupby("scenario")[com_cols].mean() * 100
+    com.plot(kind="bar", ax=axes[6])
+    axes[6].set_title("Comorbidities (%)")
+    axes[6].legend(fontsize=7)
+
+    dls = df.groupby(["scenario", "daily_living_skills"]).size().groupby(level=0).apply(lambda s: s / s.sum())
+    dls.unstack().plot(kind="bar", stacked=True, ax=axes[7])
+    axes[7].set_title("Daily Living Skills")
+    axes[7].legend(fontsize=7)
+
+    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()