electricsheepafrica/africa-health-facilities-kenya

Kenya Healthsites

Publisher: Global Healthsites Mapping Project · Source: HDX · License: ODbL · Updated: 2025-10-15

---

Abstract

This dataset shows the list of operating health facilities. Attributes included: Name,Nature of Facility, Activities, Lat, Long

Each row in this dataset represents tabular records. Data was last updated on HDX on 2025-10-15. Geographic scope: KEN.

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

---

Dataset Characteristics

Domain	Public health
Unit of observation	Tabular records
Rows (total)	2,558
Columns	18 (6 numeric, 11 categorical, 0 datetime)
Train split	2,046 rows
Test split	511 rows
Geographic scope	KEN
Publisher	Global Healthsites Mapping Project
HDX last updated	2025-10-15

---

Variables

Geographic — x (range 33.9758–41.5247), y (range -4.4–3.953), osm_type (node, way), amenity (hospital, pharmacy, clinic), operator_type (private, public, government).

Temporal — changeset_timestamp.

Identifier / Metadata — osm_id (range 25473029.0–13202466836.0), name (Chemist, Faith healer, chemist), changeset_id (range 3089987.0–173109831.0), uuid (fa79ca7f10794c288995b0ff692e9a6d, ea6fc01726de45b6ad6c3cb054fb02c8, 05f1aac2329f4b57b9b07bd473900252), esa_source (HDX) and 1 others.

Other — completeness (range 6.25–59.375), healthcare (hospital, counselling, clinic), operator (Ministry of Health, private, public), operational_status (operational, non_operational, open), opening_hours (24/7, Mo-Fr 08:00-17:00, sunrise-sunset) and 1 others.

---

Quick Start

from datasets import load_dataset

ds    = load_dataset("electricsheepafrica/africa-health-facilities-kenya")
train = ds["train"].to_pandas()
test  = ds["test"].to_pandas()

print(train.shape)
train.head()

---

Schema

Column	Type	Null %	Range / Sample Values
x	float64	27.1%	33.9758 – 41.5247 (mean 36.9558)
y	float64	27.1%	-4.4 – 3.953 (mean -0.8696)
osm_id	int64	0.0%	25473029.0 – 13202466836.0 (mean 5067761548.5801)
osm_type	object	0.0%	node, way
completeness	float64	0.0%	6.25 – 59.375 (mean 17.025)
amenity	object	1.1%	hospital, pharmacy, clinic
healthcare	object	47.4%	hospital, counselling, clinic
name	object	4.5%	Chemist, Faith healer, chemist
operator	object	78.5%	Ministry of Health, private, public
operator_type	object	56.8%	private, public, government
operational_status	object	75.1%	operational, non_operational, open
opening_hours	object	77.1%	24/7, Mo-Fr 08:00-17:00, sunrise-sunset
changeset_id	int64	0.0%	3089987.0 – 173109831.0 (mean 115881603.6591)
changeset_version	int64	0.0%	1.0 – 16.0 (mean 2.2455)
changeset_timestamp	datetime64[ns, UTC]	0.0%
uuid	object	0.0%	fa79ca7f10794c288995b0ff692e9a6d, ea6fc01726de45b6ad6c3cb054fb02c8, 05f1aac2329f4b57b9b07bd473900252
esa_source	object	0.0%	HDX
esa_processed	object	0.0%

---

Numeric Summary

Column	Min	Max	Mean	Median
x	33.9758	41.5247	36.9558	36.8451
y	-4.4	3.953	-0.8696	-1.265
osm_id	25473029.0	13202466836.0	5067761548.5801	4546095039.5
completeness	6.25	59.375	17.025	15.625
changeset_id	3089987.0	173109831.0	115881603.6591	134436534.0
changeset_version	1.0	16.0	2.2455	2.0

---

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. 19 column(s) with >80% missing values were removed: source, speciality, beds, staff_doctors, staff_nurses, health_amenity_type.... 1 column(s) were cast from string to numeric or datetime based on parse-success rate (>85% threshold). 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 Global Healthsites Mapping Project and has not been independently validated by ESA.
• Automated cleaning cannot correct for misreported values, definitional inconsistencies, or sampling bias in the original collection.
• The following columns have >20% missing values and should be treated with caution in modelling: x, y, healthcare, operator, operator_type, operational_status, opening_hours.
• Refer to the original HDX dataset page for the publisher's own methodology notes and caveats.

---

Citation

@dataset{hdx_africa_health_facilities_kenya,
  title     = {Kenya Healthsites},
  author    = {Global Healthsites Mapping Project},
  year      = {2025},
  url       = {https://data.humdata.org/dataset/kenya-healthsites},
  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

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