Add README.md

README.md

@@ -1,106 +1,197 @@
 ---
+annotations_creators:
+- no-annotation
+language_creators:
+- found
+language:
+- en
+license: cc-by-4.0
+multilinguality:
+- monolingual
+size_categories:
+- 1K<n<10K
+source_datasets:
+- original
+task_categories:
+- tabular-classification
+- other
+task_ids: []
+tags:
+- africa
+- humanitarian
+- hdx
+- electric-sheep-africa
+- facilities-infrastructure
+- health
+- health-facilities
+- hxl
+- zwe
+pretty_name: "Zimbabwe: Health facilities"
 dataset_info:
-  features:
-  - name: id1
-    dtype: float64
-  - name: id
-    dtype: float64
-  - name: province
-    dtype: string
-  - name: district
-    dtype: string
-  - name: longitude
-    dtype: float64
-  - name: latitude
-    dtype: float64
-  - name: elevation
-    dtype: float64
-  - name: updated
-    dtype: float64
-  - name: nameoffaci
-    dtype: string
-  - name: ownership
-    dtype: float64
-  - name: yearbuilt
-    dtype: float64
-  - name: typeoffaci
-    dtype: string
-  - name: numofdocto
-    dtype: float64
-  - name: numofnurse
-    dtype: float64
-  - name: numofnur_1
-    dtype: float64
-  - name: numofpcn
-    dtype: float64
-  - name: numofehts
-    dtype: float64
-  - name: numofpharm
-    dtype: float64
-  - name: numoflabte
-    dtype: float64
-  - name: numofbeds
-    dtype: float64
-  - name: numofmater
-    dtype: float64
-  - name: numofgener
-    dtype: float64
-  - name: cathmentpo
-    dtype: float64
-  - name: distneares
-    dtype: float64
-  - name: hascommuni
-    dtype: float64
-  - name: hascommu_1
-    dtype: float64
-  - name: haswaterpi
-    dtype: float64
-  - name: haswaterun
-    dtype: float64
-  - name: haselectri
-    dtype: float64
-  - name: haselect_1
-    dtype: float64
-  - name: distnear_1
-    dtype: float64
-  - name: hassanitat
-    dtype: float64
-  - name: hassanit_1
-    dtype: float64
-  - name: hassanit_2
-    dtype: float64
-  - name: hassecurit
-    dtype: float64
-  - name: hassecur_1
-    dtype: float64
-  - name: hasroadtar
-    dtype: float64
-  - name: hasroadgra
-    dtype: float64
-  - name: hasinciner
-    dtype: float64
-  - name: hasautoway
-    dtype: float64
-  - name: hasdental
-    dtype: float64
-  - name: esa_source
-    dtype: string
-  - name: esa_processed
-    dtype: string
   splits:
-  - name: train
-    num_bytes: 509345
-    num_examples: 1352
-  - name: test
-    num_bytes: 126959
-    num_examples: 338
-  download_size: 136701
-  dataset_size: 636304
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-  - split: test
-    path: data/test-*
+    - name: train
+      num_examples: 1352
+    - name: test
+      num_examples: 338
 ---
+
+# Zimbabwe: Health facilities
+
+**Publisher:** OCHA Regional Office for Southern and Eastern Africa (ROSEA) · **Source:** [HDX](https://data.humdata.org/dataset/zimbabwe-health-facilities) · **License:** `cc-by` · **Updated:** 2025-11-13
+
+---
+
+## Abstract
+
+List of Health facilities in Zimbabwe
+
+Each row in this dataset represents subnational administrative unit observations. Data was last updated on HDX on 2025-11-13. Geographic scope: **ZWE**.
+
+*Curated into ML-ready Parquet format by [Electric Sheep Africa](https://huggingface.co/electricsheepafrica).*
+
+---
+
+## Dataset Characteristics
+
+| | |
+|---|---|
+| **Domain** | Public health |
+| **Unit of observation** | Subnational administrative unit observations |
+| **Rows (total)** | 1,690 |
+| **Columns** | 43 (37 numeric, 6 categorical, 0 datetime) |
+| **Train split** | 1,352 rows |
+| **Test split** | 338 rows |
+| **Geographic scope** | ZWE |
+| **Publisher** | OCHA Regional Office for Southern and Eastern Africa (ROSEA) |
+| **HDX last updated** | 2025-11-13 |
+
+---
+
+## Variables
+
+**Geographic** — `province` (Manicaland, Mashonaland East, Midlands), `district` (Mutasa, Chipinge, Mutare), `longitude` (range 25.8258–33.0345), `latitude` (range -22.3173–-15.7006), `yearbuilt` (range 0.0–2011.0) and 2 others.
+
+**Temporal** — `updated` (range 1998.0–2013.0).
+
+**Identifier / Metadata** — `id1` (range 1.0–1686.0), `id` (range 0.0–1685.0), `nameoffaci` (ZRP, Chivi, Shamva), `esa_source` (HDX), `esa_processed` (2026-04-18).
+
+**Other** — `elevation` (range 0.0–1569.0), `ownership` (range 0.0–9.0), `numofdocto` (range 0.0–3.0), `numofnurse` (range 0.0–41.0), `numofnur_1` (range 0.0–21.0) and 25 others.
+
+---
+
+## Quick Start
+
+```python
+from datasets import load_dataset
+
+ds    = load_dataset("electricsheepafrica/africa-zimbabwe-health-facilities")
+train = ds["train"].to_pandas()
+test  = ds["test"].to_pandas()
+
+print(train.shape)
+train.head()
+```
+
+---
+
+## Schema
+
+| Column | Type | Null % | Range / Sample Values |
+|---|---|---|---|
+| `id1` | float64 | 0.1% | 1.0 – 1686.0 (mean 845.1445) |
+| `id` | float64 | 2.0% | 0.0 – 1685.0 (mean 662.1087) |
+| `province` | object | 0.0% | Manicaland, Mashonaland East, Midlands |
+| `district` | object | 0.0% | Mutasa, Chipinge, Mutare |
+| `longitude` | float64 | 0.1% | 25.8258 – 33.0345 (mean 30.5756) |
+| `latitude` | float64 | 0.1% | -22.3173 – -15.7006 (mean -18.8716) |
+| `elevation` | float64 | 0.9% | 0.0 – 1569.0 (mean 87.397) |
+| `updated` | float64 | 0.9% | 1998.0 – 2013.0 (mean 2001.2042) |
+| `nameoffaci` | object | 0.1% | ZRP, Chivi, Shamva |
+| `ownership` | float64 | 10.7% | 0.0 – 9.0 (mean 5.0199) |
+| `yearbuilt` | float64 | 4.7% | 0.0 – 2011.0 (mean 297.0981) |
+| `typeoffaci` | object | 1.0% | Clinic, Rural Health Centre, Council Clinic |
+| `numofdocto` | float64 | 4.0% | 0.0 – 3.0 (mean 0.0265) |
+| `numofnurse` | float64 | 4.0% | 0.0 – 41.0 (mean 0.4418) |
+| `numofnur_1` | float64 | 4.0% | 0.0 – 21.0 (mean 0.244) |
+| `numofpcn` | float64 | 4.0% | 0.0 – 39.0 (mean 0.1041) |
+| `numofehts` | float64 | 4.0% | 0.0 – 4.0 (mean 0.0444) |
+| `numofpharm` | float64 | 4.0% | 0.0 – 2.0 (mean 0.0105) |
+| `numoflabte` | float64 | 4.0% | 0.0 – 5.0 (mean 0.008) |
+| `numofbeds` | float64 | 4.0% | 0.0 – 2500.0 (mean 2.6402) |
+| `numofmater` | float64 | 4.0% | 0.0 – 20.0 (mean 0.1763) |
+| `numofgener` | float64 | 4.0% | 0.0 – 140.0 (mean 0.6245) |
+| `cathmentpo` | float64 | 4.0% | 0.0 – 1616300.0 (mean 2154.8712) |
+| `distneares` | float64 | 4.0% | 0.0 – 565.0 (mean 5.4855) |
+| `hascommuni` | float64 | 4.0% |  |
+| `hascommu_1` | float64 | 4.0% |  |
+| `haswaterpi` | float64 | 4.0% |  |
+| `haswaterun` | float64 | 4.0% |  |
+| `haselectri` | float64 | 4.0% |  |
+| `haselect_1` | float64 | 4.0% |  |
+| `distnear_1` | float64 | 4.0% |  |
+| `hassanitat` | float64 | 4.0% |  |
+| `hassanit_1` | float64 | 4.0% |  |
+| `hassanit_2` | float64 | 4.0% |  |
+| `hassecurit` | float64 | 4.0% |  |
+| `hassecur_1` | float64 | 4.0% |  |
+| `hasroadtar` | float64 | 4.0% |  |
+| `hasroadgra` | float64 | 4.0% |  |
+| `hasinciner` | float64 | 4.0% |  |
+| `hasautoway` | float64 | 4.1% |  |
+| `hasdental` | float64 | 4.0% |  |
+| `esa_source` | object | 0.0% | HDX |
+| `esa_processed` | object | 0.0% | 2026-04-18 |
+
+---
+
+## Numeric Summary
+
+| Column | Min | Max | Mean | Median |
+|---|---|---|---|---|
+| `id1` | 1.0 | 1686.0 | 845.1445 | 845.0 |
+| `id` | 0.0 | 1685.0 | 662.1087 | 652.5 |
+| `longitude` | 25.8258 | 33.0345 | 30.5756 | 30.8722 |
+| `latitude` | -22.3173 | -15.7006 | -18.8716 | -18.7445 |
+| `elevation` | 0.0 | 1569.0 | 87.397 | 0.0 |
+| `updated` | 1998.0 | 2013.0 | 2001.2042 | 1998.0 |
+| `ownership` | 0.0 | 9.0 | 5.0199 | 5.0 |
+| `yearbuilt` | 0.0 | 2011.0 | 297.0981 | 0.0 |
+| `numofdocto` | 0.0 | 3.0 | 0.0265 | 0.0 |
+| `numofnurse` | 0.0 | 41.0 | 0.4418 | 0.0 |
+| `numofnur_1` | 0.0 | 21.0 | 0.244 | 0.0 |
+| `numofpcn` | 0.0 | 39.0 | 0.1041 | 0.0 |
+| `numofehts` | 0.0 | 4.0 | 0.0444 | 0.0 |
+| `numofpharm` | 0.0 | 2.0 | 0.0105 | 0.0 |
+| `numoflabte` | 0.0 | 5.0 | 0.008 | 0.0 |
+
+---
+
+## 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`. 2 column(s) with >80% missing values were removed: `comments`, `type_edite`. 4 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 OCHA Regional Office for Southern and Eastern Africa (ROSEA) 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](https://data.humdata.org/dataset/zimbabwe-health-facilities) for the publisher's own methodology notes and caveats.
+
+---
+
+## Citation
+
+```bibtex
+@dataset{hdx_africa_zimbabwe_health_facilities,
+  title     = {Zimbabwe: Health facilities},
+  author    = {OCHA Regional Office for Southern and Eastern Africa (ROSEA)},
+  year      = {2025},
+  url       = {https://data.humdata.org/dataset/zimbabwe-health-facilities},
+  note      = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
+}
+```
+
+---
+
+*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — Africa's ML dataset infrastructure. Lagos, Nigeria.*