electricsheepafrica/africa-daily-cross-border-trade-for-uganda-6824

---
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
- eastern-africa
- trade
- uga
pretty_name: "Uganda Daily FEWS NET Cross Border Trade Data"
dataset_info:
splits:
- name: train
num_examples: 17112
- name: test
num_examples: 4278
---

# Uganda Daily FEWS NET Cross Border Trade Data

**Publisher:** FEWS NET · **Source:** [HDX](https://data.humdata.org/dataset/daily_cross_border_trade_for_uganda_6824) · **License:** `cc-by` · **Updated:** 2026-03-30

---

## Abstract

Uganda Daily cross border trade data collected by FEWS NET since 2010.

Each row in this dataset represents first-level administrative unit observations. Temporal coverage is indicated by the `start_date`, `period_date` column(s). Geographic scope: **UGA**.

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Humanitarian and development data |
| **Unit of observation** | First-level administrative unit observations |
| **Rows (total)** | 21,390 |
| **Columns** | 40 (7 numeric, 31 categorical, 2 datetime) |
| **Train split** | 17,112 rows |
| **Test split** | 4,278 rows |
| **Geographic scope** | UGA |
| **Publisher** | FEWS NET |
| **HDX last updated** | 2026-03-30 |

---

## Variables

**Geographic** — `reporting_country` (Uganda, South Sudan, Rwanda), `reporting_country_code` (UG, SS, RW), `source_country_code` (UG, CD, KE), `destination_country_code` (SS, UG, CD), `flow_type` and 10 others.

**Temporal** — `start_date`, `period_date`, `value_one_month_ago` (range 0.0073–244536280.5), `pct_change_from_one_month_ago` (range -99.9679–751585.3933).

**Outcome / Measurement** — `value` (range 0.0–978145122.0).

**Identifier / Metadata** — `source` (Uganda, Democratic Republic of the Congo, Kenya), `indicator_name` (TradeFlowQuantity), `source_organization`, `source_document`, `dataseries_name` and 4 others.

**Other** — `border_point` (Mpondwe, Nimule, Gatuna), `destination` (South Sudan, Uganda, Democratic Republic of the Congo), `cpcv2` (R01701AA, R01122AC, P23161AA), `product` (Beans (mixed), Maize Grain (White), Rice (Milled)), `collection_status` and 6 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-daily-cross-border-trade-for-uganda-6824")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `reporting_country` | object | 0.0% | Uganda, South Sudan, Rwanda |
| `reporting_country_code` | object | 0.0% | UG, SS, RW |
| `border_point` | object | 0.0% | Mpondwe, Nimule, Gatuna |
| `source` | object | 0.0% | Uganda, Democratic Republic of the Congo, Kenya |
| `source_country_code` | object | 0.0% | UG, CD, KE |
| `destination` | object | 0.0% | South Sudan, Uganda, Democratic Republic of the Congo |
| `destination_country_code` | object | 0.0% | SS, UG, CD |
| `cpcv2` | object | 0.0% | R01701AA, R01122AC, P23161AA |
| `product` | object | 0.0% | Beans (mixed), Maize Grain (White), Rice (Milled) |
| `indicator_name` | object | 0.0% | TradeFlowQuantity |
| `start_date` | datetime64[ns] | 0.0% | |
| `period_date` | datetime64[ns] | 0.0% | |
| `value` | float64 | 0.0% | 0.0 – 978145122.0 (mean 1060618.3002) |
| `flow_type` | object | 0.0% | |
| `trade_type` | object | 0.0% | |
| `collection_status` | object | 0.0% | |
| `source_organization` | object | 0.0% | |
| `source_document` | object | 0.0% | |
| `dataseries_name` | object | 0.0% | |
| `dataseries` | int64 | 0.0% | 6544634.0 – 7402478.0 (mean 6696235.9707) |
| `unit` | object | 0.0% | |
| `unit_type` | object | 0.0% | |
| `unit_name` | object | 0.0% | |
| `status` | object | 0.0% | |
| `common_unit` | object | 0.0% | |
| `common_unit_quantity` | float64 | 0.0% | 0.0 – 10399759600.0 (mean 3841847.8136) |
| `reporting_country_geographic_group` | object | 0.0% | |
| `reporting_country_fewsnet_region` | object | 0.0% | |
| `source_geographic_group` | object | 0.0% | |
| `source_fewsnet_region` | object | 0.0% | |
| `destination_geographic_group` | object | 0.0% | |
| `destination_fewsnet_region` | object | 0.0% | |
| `value_one_month_ago` | float64 | 61.5% | 0.0073 – 244536280.5 (mean 683881.2778) |
| `value_one_year_ago` | float64 | 76.7% | 0.0 – 244536280.5 (mean 825279.2189) |
| `pct_change_from_one_month_ago` | float64 | 61.5% | -99.9679 – 751585.3933 (mean 1110.2398) |
| `pct_change_from_one_year_ago` | float64 | 76.7% | -99.9943 – 3199900.0 (mean 6942.3517) |
| `collection_schedule` | object | 0.0% | |
| `data_usage_policy` | object | 0.0% | |
| `esa_source` | object | 0.0% | |
| `esa_processed` | object | 0.0% | |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `value` | 0.0 | 978145122.0 | 1060618.3002 | 0.405 |
| `dataseries` | 6544634.0 | 7402478.0 | 6696235.9707 | 6628067.0 |
| `common_unit_quantity` | 0.0 | 10399759600.0 | 3841847.8136 | 8.0 |
| `value_one_month_ago` | 0.0073 | 244536280.5 | 683881.2778 | 200.0 |
| `value_one_year_ago` | 0.0 | 244536280.5 | 825279.2189 | 248.6875 |
| `pct_change_from_one_month_ago` | -99.9679 | 751585.3933 | 1110.2398 | 245.3733 |
| `pct_change_from_one_year_ago` | -99.9943 | 3199900.0 | 6942.3517 | 253.3684 |

---

## 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`. 8 column(s) with >80% missing values were removed: `id`, `value_two_years_ago`, `value_three_years_ago`, `value_four_years_ago`, `value_five_years_ago`, `two_year_average`.... 2 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 FEWS NET 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: `value_one_month_ago`, `value_one_year_ago`, `pct_change_from_one_month_ago`, `pct_change_from_one_year_ago`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/daily_cross_border_trade_for_uganda_6824) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_daily_cross_border_trade_for_uganda_6824,
title = {Uganda Daily FEWS NET Cross Border Trade Data},
author = {FEWS NET},
year = {2026},
url = {https://data.humdata.org/dataset/daily_cross_border_trade_for_uganda_6824},
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.*