electricsheepafrica/africa-mozambique-cyclone-idai-4w

---
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
- cyclones-hurricanes-typhoons
- hxl
- who-is-doing-what-and-where-3w-4w-5w
- moz
pretty_name: "Mozambique Cyclone Idai 3W/4W"
dataset_info:
splits:
- name: train
num_examples: 14275
- name: test
num_examples: 3568
---

# Mozambique Cyclone Idai 3W/4W

**Publisher:** OCHA Mozambique · **Source:** [HDX](https://data.humdata.org/dataset/mozambique-cyclone-idai-4w) · **License:** `cc-by-igo` · **Updated:** 2025-04-08

---

## Abstract

Mozambique 3W/4W (Response Tracking) for the the Cyclone Idai response.

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

*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)** | 17,844 |
| **Columns** | 25 (3 numeric, 21 categorical, 1 datetime) |
| **Train split** | 14,275 rows |
| **Test split** | 3,568 rows |
| **Geographic scope** | MOZ |
| **Publisher** | OCHA Mozambique |
| **HDX last updated** | 2025-04-08 |

---

## Variables

**Geographic** — `activity_activity_code` (HEA 0179, HEA 0047, HEA 0083), `org_type_name_short` (INGO, Gov., UN), `org_type_name_role` (Lead Organisation, Implementing partner, ONGs internacionais), `activity_activity_type_name` (Food distribution In Kind, Distribuição de alimentos em espécie, Water), `activity_activity_name` (2.4 - Rehabilitation and disinfection of existing water points, Mother and Child Health (Nutrition and EPI), 4.2 - Household level hygiene promotion activities) and 2 others.

**Temporal** — `date_report`, `date_start` (range 49.0–412453.0), `date_end` (range -509603.0–44499.0).

**Identifier / Metadata** — `org_name` (WFP, COSACA, UNICEF), `org_name_short` (WFP, MISAU, COSACA), `sector_cluster_name` (Saúde, Segurança Alimentar, Água, Saneamento e Higiene), `output_name` (food aid, Mixed food, Learning Materials), `adm1_name` and 9 others.

**Other** — `reached_ind` (range 0.0–1500000.0).

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-mozambique-cyclone-idai-4w")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `date_report` | datetime64[ns] | 2.2% | |
| `activity_activity_code` | object | 0.0% | HEA 0179, HEA 0047, HEA 0083 |
| `org_name` | object | 0.5% | WFP, COSACA, UNICEF |
| `org_name_short` | object | 3.9% | WFP, MISAU, COSACA |
| `org_type_name_short` | object | 6.1% | INGO, Gov., UN |
| `org_type_name_role` | object | 2.4% | Lead Organisation, Implementing partner, ONGs internacionais |
| `sector_cluster_name` | object | 0.0% | Saúde, Segurança Alimentar, Água, Saneamento e Higiene |
| `activity_activity_type_name` | object | 34.2% | Food distribution In Kind, Distribuição de alimentos em espécie, Water |
| `activity_activity_name` | object | 53.4% | 2.4 - Rehabilitation and disinfection of existing water points, Mother and Child Health (Nutrition and EPI), 4.2 - Household level hygiene promotion activities |
| `output_name` | object | 53.5% | food aid, Mixed food, Learning Materials |
| `modality_name` | object | 64.7% | em espécie (in-kind), dinheiro em espécie, cupom |
| `adm1_name` | object | 1.9% | |
| `adm1_code` | object | 4.5% | |
| `adm2_name` | object | 11.9% | |
| `adm2_code` | object | 14.5% | |
| `adm3_name` | object | 58.2% | |
| `adm3_code` | object | 62.0% | |
| `loc_name` | object | 63.5% | |
| `reached_ind` | float64 | 50.8% | 0.0 – 1500000.0 (mean 5572.9276) |
| `beneficiary_type_name` | object | 78.4% | |
| `date_start` | float64 | 53.0% | 49.0 – 412453.0 (mean 43168.0132) |
| `date_end` | float64 | 64.8% | -509603.0 – 44499.0 (mean 43449.7293) |
| `status_name` | object | 8.1% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `reached_ind` | 0.0 | 1500000.0 | 5572.9276 | 500.0 |
| `date_start` | 49.0 | 412453.0 | 43168.0132 | 43556.0 |
| `date_end` | -509603.0 | 44499.0 | 43449.7293 | 43564.0 |

---

## 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`. 4,993 exact duplicate rows were removed. 3 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 Mozambique 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: `activity_activity_type_name`, `activity_activity_name`, `output_name`, `modality_name`, `adm3_name`, `adm3_code`, `loc_name`, `reached_ind`....
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/mozambique-cyclone-idai-4w) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_mozambique_cyclone_idai_4w,
title = {Mozambique Cyclone Idai 3W/4W},
author = {OCHA Mozambique},
year = {2025},
url = {https://data.humdata.org/dataset/mozambique-cyclone-idai-4w},
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.*