electricsheepafrica/africa-kenya-kisumu-county-crop-production-data-2014-2016

---
annotations_creators:
- no-annotation
language_creators:
- found
language:
- en
license: cc-by-4.0
multilinguality:
- monolingual
size_categories:
- n<1K
source_datasets:
- original
task_categories:
- tabular-classification
- tabular-regression
task_ids: []
tags:
- africa
- humanitarian
- hdx
- electric-sheep-africa
- food-security
- livelihoods
- ken
pretty_name: "Kenya - Kisumu county Crop production Data 2014-2016"
dataset_info:
splits:
- name: train
num_examples: 326
- name: test
num_examples: 81
---

# Kenya - Kisumu county Crop production Data 2014-2016

**Publisher:** Kenya Open Data Initiative (inactive) · **Source:** [HDX](https://data.humdata.org/dataset/kenya-kisumu-county-crop-production-data-2014-2016) · **License:** `cc-by` · **Updated:** 2023-03-03

---

## Abstract

A monthly report of Crop Farming in Kisumu County and their status from the year 2014 to 2016.

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

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

---

## Dataset Characteristics

| | |
|---|---|
| **Domain** | Food security and nutrition |
| **Unit of observation** | Time-series observations |
| **Rows (total)** | 408 |
| **Columns** | 13 (5 numeric, 7 categorical, 1 datetime) |
| **Train split** | 326 rows |
| **Test split** | 81 rows |
| **Geographic scope** | KEN |
| **Publisher** | Kenya Open Data Initiative (inactive) |
| **HDX last updated** | 2023-03-03 |

---

## Variables

**Geographic** — `county` (Kisumu., Kisumu), `commodity_type` ( Maize , Wheat , Rice ).

**Temporal** — `date`.

**Demographic** — `crop_stage` (No Crop, Harvested, Flowering).

**Identifier / Metadata** — `esa_source` (HDX), `esa_processed` (2026-04-09).

**Other** — `house_hold_stock_tonnes` (range 0.0–30151.0), `production_estimates_tonnes` (range 0.0–84960.0), `millers_tonnes` (range 0.0–13950.0), `traders_tonnes` (range 0.0–16737.0), `national_cereals_produce_board_tonnes` (range 0.0–4571.0) and 2 others.

---

## Quick Start

```python
from datasets import load_dataset

ds = load_dataset("electricsheepafrica/africa-kenya-kisumu-county-crop-production-data-2014-2016")
train = ds["train"].to_pandas()
test = ds["test"].to_pandas()

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

---

## Schema

| Column | Type | Null % | Range / Sample Values |
|---|---|---|---|
| `county` | object | 0.0% | Kisumu., Kisumu |
| `date` | datetime64[ns] | 0.0% | |
| `commodity_type` | object | 0.0% | Maize , Wheat , Rice |
| `house_hold_stock_tonnes` | int64 | 0.0% | 0.0 – 30151.0 (mean 923.049) |
| `production_estimates_tonnes` | int64 | 0.0% | 0.0 – 84960.0 (mean 2272.8382) |
| `millers_tonnes` | float64 | 0.2% | 0.0 – 13950.0 (mean 451.4201) |
| `traders_tonnes` | int64 | 0.0% | 0.0 – 16737.0 (mean 790.0931) |
| `national_cereals_produce_board_tonnes` | float64 | 0.2% | 0.0 – 4571.0 (mean 49.2334) |
| `crop_stage` | object | 6.1% | No Crop, Harvested, Flowering |
| `crop_conditions` | object | 8.1% | No Crop, Good, Average |
| `remarks` | object | 71.6% | Drought set in at critical stages of growth . Computation of stocks held by NCPB is done at national level, Expected to do well with el-nino rains. Computation of stocks held by NCPB is done at national level, Expected to well with el-nino rains |
| `esa_source` | object | 0.0% | HDX |
| `esa_processed` | object | 0.0% | 2026-04-09 |

---

## Numeric Summary

| Column | Min | Max | Mean | Median |
|---|---|---|---|---|
| `house_hold_stock_tonnes` | 0.0 | 30151.0 | 923.049 | 0.0 |
| `production_estimates_tonnes` | 0.0 | 84960.0 | 2272.8382 | 0.0 |
| `millers_tonnes` | 0.0 | 13950.0 | 451.4201 | 0.0 |
| `traders_tonnes` | 0.0 | 16737.0 | 790.0931 | 0.0 |
| `national_cereals_produce_board_tonnes` | 0.0 | 4571.0 | 49.2334 | 0.0 |

---

## 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`. 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 Kenya Open Data Initiative (inactive) 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: `remarks`.
- Refer to the [original HDX dataset page](https://data.humdata.org/dataset/kenya-kisumu-county-crop-production-data-2014-2016) for the publisher's own methodology notes and caveats.

---

## Citation

```bibtex
@dataset{hdx_africa_kenya_kisumu_county_crop_production_data_2014_2016,
title = {Kenya - Kisumu county Crop production Data 2014-2016},
author = {Kenya Open Data Initiative (inactive)},
year = {2023},
url = {https://data.humdata.org/dataset/kenya-kisumu-county-crop-production-data-2014-2016},
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.*