electricsheepafrica/africa-cadre-harmonise

---
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:
- tabular-classification
- tabular-regression
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- food-security
- integrated-food-security-phase-classification-ipc
- sahel
- ben
- bfa
- cpv
- cmr
- caf
pretty_name: "West & Central Africa Food Security Data - Cadre Harmonise (CH) and Integrated Food Security Phase Classification (IPC) data"
dataset_info:
splits:
- name: train
num_examples: 32270
- name: test
num_examples: 8067
---

# West & Central Africa Food Security Data - Cadre Harmonise (CH) and Integrated Food Security Phase Classification (IPC) data

**Publisher:** Food Security and Nutrition Working Group, West and Central Africa · **Source:** [HDX](https://data.humdata.org/dataset/cadre-harmonise) · **License:** `cc-by` · **Updated:** 2026-02-06

---

## Abstract

The Cadre Harmonisé (CH) and Integrated Food Security Phase Classification (IPC) are analytical frameworks which synthesize indicators of food and nutrition security outcomes and the inference of contributing factors into scales and figures representing the nature and severity of crisis and implications for strategic response in food security and nutrition.

There is also a [global Acute Food Insecurity Country dataset](https://data.humdata.org/dataset/global-acute-food-insecurity-country-data).

Each row in this dataset represents geolocated point observations. Data was last updated on HDX on 2026-02-06. Geographic scope: **BEN, BFA, CPV, CMR, CAF, TCD, CIV, GMB, and 11 others**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Food security and nutrition |
| **Unit of observation** | Geolocated point observations |
| **Rows (total)** | 40,338 |
| **Columns** | 29 (14 numeric, 15 categorical, 0 datetime) |
| **Train split** | 32,270 rows |
| **Test split** | 8,067 rows |
| **Geographic scope** | BEN, BFA, CPV, CMR, CAF, TCD, CIV, GMB, and 11 others |
| **Publisher** | Food Security and Nutrition Working Group, West and Central Africa |
| **HDX last updated** | 2026-02-06 |

---

## Variables

**Geographic** — `exercise_code` (range 1.0–3.0), `exercise_label` (Sep-Dec, Jan-May, Jun-Aug), `exercise_year` (range 2014.0–2025.0), `chtype`, `reference_year` (range 2014.0–2026.0) and 1 others.

**Temporal** — `usethisperiod`.

**Outcome / Measurement** — `phase_class` (range 0.0–4.0), `phase1` (range 0.0–5972708.0), `phase2` (range 0.0–2919820.0), `phase3` (range 0.0–1898477.0), `phase4` (range 0.0–710000.0) and 4 others.

**Identifier / Metadata** — `adm0_name` (Nigeria, Chad, Niger), `adm1_name` (Kano, Borno, Katsina), `adm1_5_name` (Borno Central, Borno South, Kano Central), `adm2_name` (Bamako, Dagana, Tillaberi), `reference_code` (range 1.0–3.0) and 3 others.

**Other** — `adm0_pcod2` (NG, TD, NE), `adm0_pcod3` (NGA, TCD, NER), `adm1_pcod2` (NG20, NG08, NG21), `adm1_5_pcod2` (NG00801, NG02002, NG02001), `adm2_pcod2` (ML0901, NE0612, NE0801).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-cadre-harmonise")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `adm0_name` | object | 0.0% | Nigeria, Chad, Niger |
| `adm0_pcod2` | object | 0.0% | NG, TD, NE |
| `adm0_pcod3` | object | 0.0% | NGA, TCD, NER |
| `adm1_name` | object | 0.0% | Kano, Borno, Katsina |
| `adm1_pcod2` | object | 0.0% | NG20, NG08, NG21 |
| `adm1_5_name` | object | 77.5% | Borno Central, Borno South, Kano Central |
| `adm1_5_pcod2` | object | 78.5% | NG00801, NG02002, NG02001 |
| `adm2_name` | object | 8.5% | Bamako, Dagana, Tillaberi |
| `adm2_pcod2` | object | 8.8% | ML0901, NE0612, NE0801 |
| `exercise_code` | int64 | 0.0% | 1.0 – 3.0 (mean 1.4864) |
| `exercise_label` | object | 0.0% | Sep-Dec, Jan-May, Jun-Aug |
| `exercise_year` | int64 | 0.0% | 2014.0 – 2025.0 (mean 2020.9766) |
| `chtype` | object | 0.0% | |
| `reference_code` | int64 | 0.0% | 1.0 – 3.0 (mean 2.2424) |
| `reference_label` | object | 0.0% | |
| `reference_year` | int64 | 0.0% | 2014.0 – 2026.0 (mean 2021.2441) |
| `population` | float64 | 0.3% | 0.0 – 6492074.0 (mean 346270.3344) |
| `phase_class` | float64 | 1.0% | 0.0 – 4.0 (mean 1.8226) |
| `phase1` | float64 | 0.4% | 0.0 – 5972708.0 (mean 238333.4619) |
| `phase2` | float64 | 0.4% | 0.0 – 2919820.0 (mean 78166.1273) |
| `phase3` | float64 | 0.9% | 0.0 – 1898477.0 (mean 26295.6842) |
| `phase4` | float64 | 1.3% | 0.0 – 710000.0 (mean 2036.8538) |
| `phase5` | float64 | 1.9% | 0.0 – 66383.0 (mean 13.5931) |
| `phase35` | float64 | 0.9% | 0.0 – 2495141.0 (mean 28498.323) |
| `foodconsumption_phase` | float64 | 78.8% | 1.0 – 5.0 (mean 1.7312) |
| `livelihoods_phase` | float64 | 79.7% | 1.0 – 4.0 (mean 2.1738) |
| `usethisperiod` | object | 0.0% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `exercise_code` | 1.0 | 3.0 | 1.4864 | 1.0 |
| `exercise_year` | 2014.0 | 2025.0 | 2020.9766 | 2021.0 |
| `reference_code` | 1.0 | 3.0 | 2.2424 | 3.0 |
| `reference_year` | 2014.0 | 2026.0 | 2021.2441 | 2022.0 |
| `population` | 0.0 | 6492074.0 | 346270.3344 | 243380.5 |
| `phase_class` | 0.0 | 4.0 | 1.8226 | 2.0 |
| `phase1` | 0.0 | 5972708.0 | 238333.4619 | 148302.0 |
| `phase2` | 0.0 | 2919820.0 | 78166.1273 | 48280.0 |
| `phase3` | 0.0 | 1898477.0 | 26295.6842 | 13811.0 |
| `phase4` | 0.0 | 710000.0 | 2036.8538 | 0.0 |
| `phase5` | 0.0 | 66383.0 | 13.5931 | 0.0 |
| `phase35` | 0.0 | 2495141.0 | 28498.323 | 14368.0 |
| `foodconsumption_phase` | 1.0 | 5.0 | 1.7312 | 2.0 |
| `livelihoods_phase` | 1.0 | 4.0 | 2.1738 | 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`. 34 column(s) with >80% missing values were removed: `adm0_5_name`, `adm0_5_pcod2`, `adm2_5_name`, `adm2_5_pcod2`, `adm3_name`, `adm3_pcod2`.... 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 Food Security and Nutrition Working Group, West and Central Africa 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: `adm1_5_name`, `adm1_5_pcod2`, `foodconsumption_phase`, `livelihoods_phase`.
- This dataset spans 19 countries; geographic and methodological inconsistencies across national boundaries may affect cross-country comparability.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/cadre-harmonise) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_cadre_harmonise,
title = {West & Central Africa Food Security Data - Cadre Harmonise (CH) and Integrated Food Security Phase Classification (IPC) data},
author = {Food Security and Nutrition Working Group, West and Central Africa},
year = {2026},
url = {https://data.humdata.org/dataset/cadre-harmonise},
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.*