electricsheepafrica/africa-uga-rainfall-subnational

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- 100K<n<1M
source_datasets:
- original
task_categories:
- tabular-regression
- other
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- climate-weather
- environment
- uga
pretty_name: "Uganda: Rainfall Indicators at Subnational Level"
dataset_info:
splits:
- name: train
num_examples: 363816
- name: test
num_examples: 90954
---

# Uganda: Rainfall Indicators at Subnational Level

**Publisher:** WFP - World Food Programme · **Source:** [HDX](https://data.humdata.org/dataset/uga-rainfall-subnational) · **License:** `cc-by` · **Updated:** 2026-04-03

---

## Abstract

This dataset contains dekadal rainfall indicators, computed from Climate Hazards Group InfraRed Precipitation satellite imagery with insitu Station data (CHIRPS) version 2 and the CHIRPS-GEFS short term rainfall forecasts, aggregated by subnational administrative units.

Included indicators are (for each dekad):

- 10 day rainfall [mm] (`rfh`)
- rainfall 1-month rolling aggregation [mm] (`r1h`)
- rainfall 3-month rolling aggregation [mm] (`r3h`)
- rainfall long term average [mm] (`rfh_avg`)
- rainfall 1-month rolling aggregation long term average [mm] (`r1h_avg`)
- rainfall 3-month rolling aggregation long term average [mm] (`r3h_avg`)
- rainfall anomaly [%] (`rfq`)
- rainfall 1-month anomaly [%] (`r1q`)
- rainfall 3-month anomaly [%] (`r3q`)

The administrative units used for aggregation are based on WFP data and contain a Pcode reference attributed to each unit. The number of input pixels used to create the aggregates, is provided in the `n_pixels` column. Finally, the `type` column indicates if the value is based on a forecast, a preliminary or a final product.

Forecasts are issued on the 6th, 16th, and 26th of each month for the upcoming 10-day period (dekad), then updated with improved versions on the 1st, 11th, and 21st.
Preliminary observations replace the previous dekad’s forecast on the 3rd, 13th, and 23rd, and are later replaced by final observations—published mid-month (13th or 23rd)—covering all three dekads of the prior month. Please find a summary below:

Publication Day: Forecast type, Covers (Dekad)

- 1st: Updated forecast, 1–10 of the same month
- 6th: Initial forecast, 11–20 of the same month
- 11th: Updated forecast, 1–10 of the same month
- 16th: Initial forecast, 21–end of the same month
- 21st: Updated forecast, 11–20 of the same month
- 26th: Initial forecast, 1–10 of the following month

For more on CHIRPS-GEFS forecasts, see: https://www.chc.ucsb.edu/data/chirps-gefs

For further details, please see the methodology section.

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

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Climate and environment |
| **Unit of observation** | Time-series observations |
| **Rows (total)** | 454,770 |
| **Columns** | 17 (12 numeric, 4 categorical, 1 datetime) |
| **Train split** | 363,816 rows |
| **Test split** | 90,954 rows |
| **Geographic scope** | UGA |
| **Publisher** | WFP - World Food Programme |
| **HDX last updated** | 2026-04-03 |

---

## Variables

**Geographic** — `n_pixels` (range 1.0–232.0).

**Temporal** — `date`.

**Identifier / Metadata** — `adm_id` (range 743.0–999581.0), `pcode` (UG2, UG3, UG4), `esa_source` (HDX), `esa_processed` (2026-04-06).

**Other** — `adm_level` (range 1.0–2.0), `rfh` (range 0.0–245.0), `rfh_avg` (range 1.9951–118.9667), `r1h` (range 2.1277–605.625), `r1h_avg` (range 8.1333–327.1833) and 6 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-uga-rainfall-subnational")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `date` | datetime64[ns] | 0.0% | |
| `adm_level` | int64 | 0.0% | 1.0 – 2.0 (mean 1.5986) |
| `adm_id` | int64 | 0.0% | 743.0 – 999581.0 (mean 172943.5448) |
| `pcode` | object | 0.0% | UG2, UG3, UG4 |
| `n_pixels` | float64 | 0.0% | 1.0 – 232.0 (mean 48.0932) |
| `rfh` | float64 | 0.0% | 0.0 – 245.0 (mean 35.2426) |
| `rfh_avg` | float64 | 0.0% | 1.9951 – 118.9667 (mean 35.0317) |
| `r1h` | float64 | 0.1% | 2.1277 – 605.625 (mean 105.7384) |
| `r1h_avg` | float64 | 0.1% | 8.1333 – 327.1833 (mean 105.1156) |
| `r3h` | float64 | 0.5% | 12.9573 – 1086.0 (mean 317.3341) |
| `r3h_avg` | float64 | 0.5% | 43.8943 – 828.7167 (mean 315.4759) |
| `rfq` | float64 | 0.0% | 9.9979 – 678.1157 (mean 100.6825) |
| `r1q` | float64 | 0.1% | 10.3232 – 496.0638 (mean 100.6239) |
| `r3q` | float64 | 0.5% | 18.077 – 339.4637 (mean 100.6378) |
| `version` | object | 0.0% | final, prelim, forecast |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-06 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `adm_level` | 1.0 | 2.0 | 1.5986 | 2.0 |
| `adm_id` | 743.0 | 999581.0 | 172943.5448 | 28492.0 |
| `n_pixels` | 1.0 | 232.0 | 48.0932 | 36.0 |
| `rfh` | 0.0 | 245.0 | 35.2426 | 31.0 |
| `rfh_avg` | 1.9951 | 118.9667 | 35.0317 | 35.6833 |
| `r1h` | 2.1277 | 605.625 | 105.7384 | 101.8734 |
| `r1h_avg` | 8.1333 | 327.1833 | 105.1156 | 108.2019 |
| `r3h` | 12.9573 | 1086.0 | 317.3341 | 315.4078 |
| `r3h_avg` | 43.8943 | 828.7167 | 315.4759 | 321.2959 |
| `rfq` | 9.9979 | 678.1157 | 100.6825 | 91.8974 |
| `r1q` | 10.3232 | 496.0638 | 100.6239 | 96.5192 |
| `r3q` | 18.077 | 339.4637 | 100.6378 | 98.5857 |

---

## 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`. 1 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 WFP - World Food Programme 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/uga-rainfall-subnational) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_uga_rainfall_subnational,
title = {Uganda: Rainfall Indicators at Subnational Level},
author = {WFP - World Food Programme},
year = {2026},
url = {https://data.humdata.org/dataset/uga-rainfall-subnational},
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.*

annotations_creators:
- 无注释
language_creators:
- 现有采集
language:
- 英语（en）
license: cc-by-4.0
multilinguality:
- 单语言
size_categories:
- 10万 < 样本数 < 100万
source_datasets:
- 原创数据集
task_categories:
- 表格回归
- 其他
task_ids: []
tags:
- 非洲
- 人道主义
- HDX（Humanitarian Data Exchange）
- Electric Sheep Africa
- 气候与天气
- 环境
- UGA（乌干达国家代码）
pretty_name: "乌干达：次国家级降水指标数据集"
dataset_info:
splits:
- name: train
num_examples: 363816
- name: test
num_examples: 90954

# 乌干达：次国家级降水指标数据集

**发布方：世界粮食计划署（World Food Programme, WFP） · 来源：** [HDX](https://data.humdata.org/dataset/uga-rainfall-subnational) · **许可协议：** `cc-by` · **更新时间：** 2026-04-03

---

## 摘要

本数据集包含旬度降水指标，其计算基于结合了地面站数据的气候灾害组红外降水（Climate Hazards Group InfraRed Precipitation, CHIRPS）版本2数据集，以及CHIRPS-GEFS短期降水预报数据，所有数据均按次国家级行政单元进行聚合。

包含的指标如下（针对每个旬度）：

- 10日降水量（单位：毫米）（`rfh`）
- 1个月滚动聚合降水量（单位：毫米）（`r1h`）
- 3个月滚动聚合降水量（单位：毫米）（`r3h`）
- 降水长期平均值（单位：毫米）（`rfh_avg`）
- 1个月滚动聚合降水量长期平均值（单位：毫米）（`r1h_avg`）
- 3个月滚动聚合降水量长期平均值（单位：毫米）（`r3h_avg`）
- 降水距平百分率（`rfq`）
- 1个月降水距平百分率（`r1q`）
- 3个月降水距平百分率（`r3q`）

用于聚合的行政单元基于WFP数据构建，每个单元均配有Pcode编码参考。用于生成聚合数据的输入像素数量将在`n_pixels`列中提供。此外，`version`列用于标注该数值是基于预报、预发布还是最终产品。

预报于每月6日、16日、26日发布，覆盖未来10天的旬期，随后在每月1日、11日、21日更新为优化版本。
预发布观测数据会在每月3日、13日、23日替换上一旬的预报数据，后续再由最终观测数据替换——最终观测数据于月中（13日或23日）发布，覆盖上月的全部三个旬期。具体发布安排如下：

发布日期 | 预报类型 | 覆盖旬期
- 1日：更新版预报，当月1-10日
- 6日：初始预报，当月11-20日
- 11日：更新版预报，当月1-10日
- 16日：初始预报，当月21日至当月月末
- 21日：更新版预报，当月11-20日
- 26日：初始预报，下月1-10日

如需了解更多CHIRPS-GEFS预报的相关信息，请访问：https://www.chc.ucsb.edu/data/chirps-gefs

如需进一步细节，请参阅方法学章节。

本数据集的每一行均代表时序观测数据。时间覆盖范围由`date`列标注。地理覆盖范围：**UGA（乌干达）**。

*本数据集由[Electric Sheep Africa](https://huggingface.co/electricsheepafrica)整理为适合机器学习的Parquet格式。*

---

## 数据集特征

| | |
|---|---|
| **领域** | 气候与环境 |
| **观测单元** | 时序观测数据 |
| **总行数** | 454,770 |
| **列数** | 17列（12列数值型、4列分类型、1列日期型） |
| **训练集划分** | 363,816行 |
| **测试集划分** | 90,954行 |
| **地理覆盖范围** | UGA（乌干达） |
| **发布方** | WFP - 世界粮食计划署 |
| **HDX最后更新时间** | 2026-04-03 |

---

## 变量

**地理相关变量** — `n_pixels`（取值范围1.0–232.0）。

**时间相关变量** — `date`。

**标识符/元数据** — `adm_id`（取值范围743.0–999581.0）、`pcode`（格式为UG2、UG3、UG4）、`esa_source`（固定为HDX）、`esa_processed`（处理时间：2026-04-06）。

**其他变量** — `adm_level`（取值范围1.0–2.0）、`rfh`（取值范围0.0–245.0）、`rfh_avg`（取值范围1.9951–118.9667）、`r1h`（取值范围2.1277–605.625）、`r1h_avg`（取值范围8.1333–327.1833）以及另外6个同类变量。

---

## 快速开始

python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-uga-rainfall-subnational")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

print(train.shape)
train.head()

---

## Schema

| 列名 | 数据类型 | 空值占比 | 取值范围/示例值 |
|---|---|---|---|
| `date` | datetime64[ns] | 0.0% | 无 |
| `adm_level` | int64 | 0.0% | 1.0 – 2.0（均值1.5986） |
| `adm_id` | int64 | 0.0% | 743.0 – 999581.0（均值172943.5448） |
| `pcode` | object | 0.0% | UG2、UG3、UG4 |
| `n_pixels` | float64 | 0.0% | 1.0 – 232.0（均值48.0932） |
| `rfh` | float64 | 0.0% | 0.0 – 245.0（均值35.2426） |
| `rfh_avg` | float64 | 0.0% | 1.9951 – 118.9667（均值35.0317） |
| `r1h` | float64 | 0.1% | 2.1277 – 605.625（均值105.7384） |
| `r1h_avg` | float64 | 0.1% | 8.1333 – 327.1833（均值105.1156） |
| `r3h` | float64 | 0.5% | 12.9573 – 1086.0（均值317.3341） |
| `r3h_avg` | float64 | 0.5% | 43.8943 – 828.7167（均值315.4759） |
| `rfq` | float64 | 0.0% | 9.9979 – 678.1157（均值100.6825） |
| `r1q` | float64 | 0.1% | 10.3232 – 496.0638（均值100.6239） |
| `r3q` | float64 | 0.5% | 18.077 – 339.4637（均值100.6378） |
| `version` | object | 0.0% | final（最终）、prelim（预发布）、forecast（预报） |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-06 |

---

## 数值摘要

| 列名 | 最小值 | 最大值 | 均值 | 中位数 |
|---|---|---|---|---|
| `adm_level` | 1.0 | 2.0 | 1.5986 | 2.0 |
| `adm_id` | 743.0 | 999581.0 | 172943.5448 | 28492.0 |
| `n_pixels` | 1.0 | 232.0 | 48.0932 | 36.0 |
| `rfh` | 0.0 | 245.0 | 35.2426 | 31.0 |
| `rfh_avg` | 1.9951 | 118.9667 | 35.0317 | 35.6833 |
| `r1h` | 2.1277 | 605.625 | 105.7384 | 101.8734 |
| `r1h_avg` | 8.1333 | 327.1833 | 105.1156 | 108.2019 |
| `r3h` | 12.9573 | 1086.0 | 317.3341 | 315.4078 |
| `r3h_avg` | 43.8943 | 828.7167 | 315.4759 | 321.2959 |
| `rfq` | 9.9979 | 678.1157 | 100.6825 | 91.8974 |
| `r1q` | 10.3232 | 496.0638 | 100.6239 | 96.5192 |
| `r3q` | 18.077 | 339.4637 | 100.6378 | 98.5857 |

---

## 数据整理

原始数据通过CKAN API从HDX下载，并转换为Parquet格式。列名统一转换为小写蛇形命名法。常见的缺失值标记（`N/A`、`null`、`none`、`-`、`unknown`、`no data`、`#N/A`）被统一替换为`NaN`。根据解析成功率（阈值>85%），将1列从字符串类型转换为数值型或日期型。本数据集以固定随机种子（42）按80/20的比例划分为训练集和测试集，并以Snappy压缩的Parquet格式存储。

---

## 局限性

- 本数据源自世界粮食计划署（WFP），Electric Sheep Africa未对其进行独立验证。
- 自动化清洗流程无法修正原始数据收集中的错报值、定义不一致或采样偏差问题。
- 如需了解发布方的方法学说明与免责声明，请参阅[原始HDX数据集页面](https://data.humdata.org/dataset/uga-rainfall-subnational)。

---

## 引用

bibtex
@dataset{hdx_africa_uga_rainfall_subnational,
title = {Uganda: Rainfall Indicators at Subnational Level},
author = {WFP - World Food Programme},
year = {2026},
url = {https://data.humdata.org/dataset/uga-rainfall-subnational},
note = {Repackaged for machine learning by Electric Sheep Africa (https://huggingface.co/electricsheepafrica)}
}

---

*[Electric Sheep Africa](https://huggingface.co/electricsheepafrica) — 非洲机器学习数据集基础设施。尼日利亚拉各斯。*