electricsheepafrica/africa-cod-em-zmb

---
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:
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- administrative-boundaries-divisions
- zmb
pretty_name: "Zambia - Subnational Edge-matched Administrative Boundaries"
dataset_info:
splits:
- name: train
num_examples: 1482
- name: test
num_examples: 370
---

# Zambia - Subnational Edge-matched Administrative Boundaries

**Publisher:** OCHA Field Information Services Section (FISS) · **Source:** [HDX](https://data.humdata.org/dataset/cod-em-zmb) · **License:** `cc-by-igo` · **Updated:** 2025-06-24

---

## Abstract

Zambia administrative level 0-4 shapefiles, geodatabase, gazetteer and geoservices

COD-EM datasets do not replace the authoritative COD-AB available [https://data.humdata.org/dataset/cod-ab-zmb](here); however COD-EM datasets may be preferred for cartographic purposes. See caveats.

Vetting by Information Technology Outreach Services (ITOS) with funding from USAID.

Each row in this dataset represents time-series observations. Temporal coverage is indicated by the `date`, `validon` column(s). Geographic scope: **ZMB**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Humanitarian and development data |
| **Unit of observation** | Time-series observations |
| **Rows (total)** | 1,853 |
| **Columns** | 17 (3 numeric, 12 categorical, 2 datetime) |
| **Train split** | 1,482 rows |
| **Test split** | 370 rows |
| **Geographic scope** | ZMB |
| **Publisher** | OCHA Field Information Services Section (FISS) |
| **HDX last updated** | 2025-06-24 |

---

## Variables

**Temporal** — `date`.

**Identifier / Metadata** — `adm4_pcode` (ZM101001001001, ZM108002109009, ZM108002109007), `adm3_pcode` (ZM108004111, ZM102003019, ZM107001095), `adm2_pcode` (ZM105006, ZM102004, ZM102009), `adm1_pcode` (ZM102, ZM109, ZM110), `adm0_pcode` (ZM) and 3 others.

**Other** — `adm4_en` (Luangwa, Kawama, Kafue), `adm3_en` (Kasempa, Kalulushi, Chilubi), `adm2_en` (Lusaka, Kitwe, Mufulira), `adm1_en` (Copperbelt, Southern, Western), `adm0_en` (Zambia) and 3 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-cod-em-zmb")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `adm4_en` | object | 0.0% | Luangwa, Kawama, Kafue |
| `adm4_pcode` | object | 0.0% | ZM101001001001, ZM108002109009, ZM108002109007 |
| `adm3_en` | object | 0.0% | Kasempa, Kalulushi, Chilubi |
| `adm3_pcode` | object | 0.0% | ZM108004111, ZM102003019, ZM107001095 |
| `adm2_en` | object | 0.0% | Lusaka, Kitwe, Mufulira |
| `adm2_pcode` | object | 0.0% | ZM105006, ZM102004, ZM102009 |
| `adm1_en` | object | 0.0% | Copperbelt, Southern, Western |
| `adm1_pcode` | object | 0.0% | ZM102, ZM109, ZM110 |
| `adm0_en` | object | 0.0% | Zambia |
| `adm0_pcode` | object | 0.0% | ZM |
| `date` | datetime64[ns] | 0.0% | |
| `validon` | datetime64[ns] | 0.0% | |
| `shape_length` | float64 | 0.0% | 0.0156 – 5.2961 (mean 0.874) |
| `shape_area` | float64 | 0.0% | 0.0 – 0.9305 (mean 0.0339) |
| `area_sqkm` | float64 | 0.0% | 0.131 – 11075.2633 (mean 405.2959) |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `shape_length` | 0.0156 | 5.2961 | 0.874 | 0.7729 |
| `shape_area` | 0.0 | 0.9305 | 0.0339 | 0.0198 |
| `area_sqkm` | 0.131 | 11075.2633 | 405.2959 | 236.884 |

---

## 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: `adm4_ref`, `validto`. 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 Field Information Services Section (FISS) 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/cod-em-zmb) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_cod_em_zmb,
title = {Zambia - Subnational Edge-matched Administrative Boundaries},
author = {OCHA Field Information Services Section (FISS)},
year = {2025},
url = {https://data.humdata.org/dataset/cod-em-zmb},
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.*