electricsheepafrica/africa-nutrition-in-central-african-republic

Central African Republic: Global Acute Malnutrition and Severe Acute Malnutrion data

Publisher: OCHA Central African Republic · Source: HDX · License: cc-by · Updated: 2025-04-09

---

Abstract

GAM and SAM data for Central African Republic.

Each row in this dataset represents geolocated point observations. Data was last updated on HDX on 2025-04-09. Geographic scope: CAF.

Curated into ML-ready Parquet format by Electric Sheep Africa.

---

Dataset Characteristics

Domain	Demographics and population
Unit of observation	Geolocated point observations
Rows (total)	80
Columns	21 (17 numeric, 4 categorical, 0 datetime)
Train split	64 rows
Test split	16 rows
Geographic scope	CAF
Publisher	OCHA Central African Republic
HDX last updated	2025-04-09

---

Variables

Geographic — popoulation_hno (range 310.0–6148641.39), population_0_59m (range 48.205–956113.7361), population_fefa (range 24.8–491891.3112).

Identifier / Metadata — pcode (CF224, CF612, CF122), sous_prefecture (Abba, Alindao, Mongoumba), idp_site (range 0.0–491891.3112), idp_fa (range 0.0–75530.4687), esa_source (HDX) and 1 others.

Other — pdi_site (range 0.0–120521.0), pdi_hors_site (range 0.0–368345.0), retournés (range 0.0–226912.0), hote (range 0.0–5432863.39), pdi_site_1 (range 0.0–0.488) and 7 others.

---

Quick Start

from datasets import load_dataset

ds    = load_dataset("electricsheepafrica/africa-nutrition-in-central-african-republic")
train = ds["train"].to_pandas()
test  = ds["test"].to_pandas()

print(train.shape)
train.head()

---

Schema

Column	Type	Null %	Range / Sample Values
pcode	object	10.0%	CF224, CF612, CF122
sous_prefecture	object	10.0%	Abba, Alindao, Mongoumba
popoulation_hno	float64	8.8%	310.0 – 6148641.39 (mean 168455.9285)
population_0_59m	float64	8.8%	48.205 – 956113.7361 (mean 26194.8969)
population_fefa	float64	8.8%	24.8 – 491891.3112 (mean 13476.4743)
pdi_site	float64	8.8%	0.0 – 120521.0 (mean 3301.9452)
pdi_hors_site	float64	8.8%	0.0 – 368345.0 (mean 10091.6438)
retournés	float64	8.8%	0.0 – 226912.0 (mean 6216.7671)
hote	float64	8.8%	0.0 – 5432863.39 (mean 148845.5723)
pdi_site_1	float64	10.0%	0.0 – 0.488 (mean 0.0263)
pdi_hors_site_1	float64	10.0%	0.0 – 1.0 (mean 0.092)
retournés_1	float64	8.8%	0.0 – 1076012.2433 (mean 14739.9371)
hote_1	float64	3.8%	0.0 – 956113.6755 (mean 13974.9842)
pin_nut	float64	3.8%	0.075 – 1222767.6945 (mean 37749.1405)
idp_site	float64	6.2%	0.0 – 491891.3112 (mean 7298.8955)
idp_fa	float64	8.8%	0.0 – 75530.4687 (mean 2069.3279)
retourne	float64	8.8%	0.0 – 49614.0877 (mean 1359.2901)
hôte	float64	8.8%	0.0 – 1069860.6808 (mean 29311.2515)
severite_projetee	float64	12.5%	2.0 – 4.0 (mean 2.8714)
esa_source	object	0.0%	HDX
esa_processed	object	0.0%	2026-04-18

---

Numeric Summary

Column	Min	Max	Mean	Median
popoulation_hno	310.0	6148641.39	168455.9285	55642.0
population_0_59m	48.205	956113.7361	26194.8969	8652.331
population_fefa	24.8	491891.3112	13476.4743	4451.36
pdi_site	0.0	120521.0	3301.9452	0.0
pdi_hors_site	0.0	368345.0	10091.6438	2366.0
retournés	0.0	226912.0	6216.7671	726.0
hote	0.0	5432863.39	148845.5723	40872.0
pdi_site_1	0.0	0.488	0.0263	0.0
pdi_hors_site_1	0.0	1.0	0.092	0.0547
retournés_1	0.0	1076012.2433	14739.9371	0.0152
hote_1	0.0	956113.6755	13974.9842	0.9056
pin_nut	0.075	1222767.6945	37749.1405	11818.8875
idp_site	0.0	491891.3112	7298.8955	0.0
idp_fa	0.0	75530.4687	2069.3279	545.363
retourne	0.0	49614.0877	1359.2901	167.343

---

Curation

Raw data was downloaded from HDX via the CKAN API and converted to Parquet. Column names were lowercased and standardised to snake_case. Common missing-value markers (N/A, null, none, -, unknown, no data, #N/A) were unified to NaN. 1 exact duplicate rows were removed. The dataset was split 80/20 into train and test partitions using a fixed random seed (42) and saved as Snappy-compressed Parquet.

---

Limitations

• Data originates from OCHA Central African Republic and has not been independently validated by ESA.
• Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
• Refer to the original HDX dataset page for the publisher's own methodology notes and caveats.

---

Citation

@dataset{hdx_africa_nutrition_in_central_african_republic,
  title     = {Central African Republic: Global Acute Malnutrition and Severe Acute Malnutrion data},
  author    = {OCHA Central African Republic},
  year      = {2025},
  url       = {https://data.humdata.org/dataset/nutrition-in-central-african-republic},
  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

Electric Sheep Africa — Africa's ML dataset infrastructure. Lagos, Nigeria.