electricsheepafrica/africa-sudan-settlement-26-july-2020

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- 10K<n<100K
source_datasets:
- original
task_categories:
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- geodata
- populated-places-settlements
- sdn
pretty_name: "Sudan: Settlements"
dataset_info:
splits:
- name: train
num_examples: 9584
- name: test
num_examples: 2396
---

# Sudan: Settlements

**Publisher:** OCHA Sudan · **Source:** [HDX](https://data.humdata.org/dataset/sudan-settlement-26-july-2020) · **License:** `cc-by` · **Updated:** 2025-04-10

---

## Abstract

The settlements dataset contains the location of cities, towns and villages in Sudan.

Each row in this dataset represents subnational administrative unit observations. Temporal coverage is indicated by the `date`, `validon_1` column(s). Geographic scope: **SDN**.

*Curated into ML-ready Parquet format by [Electric Sheep Africa](https://huggingface.co/electricsheepafrica).*

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Humanitarian and development data |
| **Unit of observation** | Subnational administrative unit observations |
| **Rows (total)** | 11,980 |
| **Columns** | 30 (8 numeric, 20 categorical, 2 datetime) |
| **Train split** | 9,584 rows |
| **Test split** | 2,396 rows |
| **Geographic scope** | SDN |
| **Publisher** | OCHA Sudan |
| **HDX last updated** | 2025-04-10 |

---

## Variables

**Geographic** — `type` (range 0.0–5.0), `type_text` (Village, Secondary Town, Principal Town), `x` (range 21.8816–38.5167), `y` (range 8.9832–22.9667), `old_admin2` (Ed Damer, Abyei PCA Area, Mellit) and 2 others.

**Temporal** — `date`.

**Identifier / Metadata** — `fid_sdn_ad` (range 0.0–188.0), `old_fid_su` (range 0.0–11999.0), `objectid_1` (range 0.0–11997.0), `featureref` (Hashaba, Unknown Name, Amara), `adm2_pcode` (SD16011, SD19001, SD02129) and 5 others.

**Other** — `featurenam` (Hashaba, Unknown Name, Amara), `feature_pc` (SD19001001, SD14037018, SD14037009), `adm2_en` (Ad Damar, Abyei PCA area, Melit), `adm2_ar` (Γò¬┬║Γöÿ├ñΓò¬┬╗Γò¬┬║Γöÿ├áΓò¬ΓûÆ, Γò¬├æΓò¬┬╗Γò¬┬║Γò¬ΓûÆΓöÿ├¿Γò¬ΓîÉ Γò¬├║Γò¬┬┐Γöÿ├¿Γöÿ├¿, Γöÿ├áΓöÿ├ñΓöÿ├¿Γò¬Γòû), `old_admi_1` (SD16011, SD19101, SD02129) and 7 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-sudan-settlement-26-july-2020")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()
```

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `fid_sdn_ad` | int64 | 0.0% | 0.0 – 188.0 (mean 93.7586) |
| `old_fid_su` | int64 | 0.0% | 0.0 – 11999.0 (mean 6003.4255) |
| `objectid_1` | int64 | 0.0% | 0.0 – 11997.0 (mean 6001.4207) |
| `featurenam` | object | 0.0% | Hashaba, Unknown Name, Amara |
| `featureref` | object | 0.0% | Hashaba, Unknown Name, Amara |
| `feature_pc` | object | 0.0% | SD19001001, SD14037018, SD14037009 |
| `type` | int64 | 0.0% | 0.0 – 5.0 (mean 4.9144) |
| `type_text` | object | 0.0% | Village, Secondary Town, Principal Town |
| `x` | float64 | 0.0% | 21.8816 – 38.5167 (mean 29.3264) |
| `y` | float64 | 0.0% | 8.9832 – 22.9667 (mean 13.9491) |
| `adm2_en` | object | 0.0% | Ad Damar, Abyei PCA area, Melit |
| `adm2_ar` | object | 0.0% | الدامر, إدارية أبيي, مليط |
| `adm2_pcode` | object | 0.0% | SD16011, SD19001, SD02129 |
| `old_admin2` | object | 0.0% | Ed Damer, Abyei PCA Area, Mellit |
| `old_admi_1` | object | 0.0% | SD16011, SD19101, SD02129 |
| `adm1_en` | object | 0.0% | North Darfur, South Darfur, South Kordofan |
| `adm1_ar` | object | 0.0% | |
| `adm1_pcode` | object | 0.0% | |
| `old_admin1` | object | 0.0% | |
| `old_admi_2` | object | 0.0% | |
| `admin0name` | object | 0.0% | |
| `adm0_en` | object | 0.0% | |
| `adm0_ar` | object | 0.0% | |
| `adm0_pcode` | object | 0.0% | |
| `shape_leng` | float64 | 0.0% | 0.4312 – 18.5256 (mean 4.7566) |
| `shape_area` | float64 | 0.0% | 0.0099 – 16.5707 (mean 1.1699) |
| `date` | datetime64[ns] | 0.0% | |
| `validon_1` | datetime64[ns] | 0.0% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `fid_sdn_ad` | 0.0 | 188.0 | 93.7586 | 96.0 |
| `old_fid_su` | 0.0 | 11999.0 | 6003.4255 | 6007.5 |
| `objectid_1` | 0.0 | 11997.0 | 6001.4207 | 6005.5 |
| `type` | 0.0 | 5.0 | 4.9144 | 5.0 |
| `x` | 21.8816 | 38.5167 | 29.3264 | 30.0833 |
| `y` | 8.9832 | 22.9667 | 13.9491 | 13.4 |
| `shape_leng` | 0.4312 | 18.5256 | 4.7566 | 3.6929 |
| `shape_area` | 0.0099 | 16.5707 | 1.1699 | 0.4911 |

---

## 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`. 3 column(s) with >80% missing values were removed: `feature_ar`, `adm2_ref`, `validto_1`. 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 Sudan 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/sudan-settlement-26-july-2020) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_sudan_settlement_26_july_2020,
title = {Sudan: Settlements},
author = {OCHA Sudan},
year = {2025},
url = {https://data.humdata.org/dataset/sudan-settlement-26-july-2020},
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.*