# Crop Type Classification Dataset for Western Cape, South Africa (Spot the Crop)

## Facts

- Provider: Western Cape Department of Agriculture and Radiant Earth Foundation
- License: CC-BY-4.0 (https://radiantearth.blob.core.windows.net/mlhub/ref-south-africa-crops-competition-v1/README.md)
- Format: JSON (in sampled files)
- Stored at: Azure Blob Storage, https://radiantearth.blob.core.windows.net/mlhub/ref-south-africa-crops-competition-v1
- Access: public, readable without an account
- DOI: https://doi.org/10.34911/rdnt.j0co8q
- Category: Earth Observation & Geospatial
- Source: https://doi.org/10.34911/rdnt.j0co8q
- Page: https://scigantic.com/archive/c851b04a-880f-53a0-84e8-1f223e90da1f
- Keywords: Earth Observation & Geospatial, satellite imagery, time series, JSON, Azure Blob Storage

## About

Field-level crop type labels for the Western Cape province of South Africa with Sentinel-1 and Sentinel-2 time series over the 2017 growing season, released by the Western Cape Department of Agriculture and Radiant Earth Foundation for the Radiant Earth Spot the Crop challenge on Zindi. The task is to classify each field by crop type (cereals, oilseed, pasture, fallow and perennial crops such as vines) from images collected during the season. Fields are rasterised to 256 x 256 pixel tiles; each tile has a Sentinel-2 L2A time series (13 bands plus cloud mask) and a Sentinel-1 backscatter series, a field id raster and, for training tiles, a crop label raster. Measured 2026-11-15: more than 40,000 files (at least 1.6 GB listed) under train/ and test/ (each with imagery/ and labels/), plus field_info_train.csv, field_info_test.csv, labels.json and sample_submission.csv. Read with rasterio and pandas. Released under CC BY 4.0; cite Western Cape Department of Agriculture and Radiant Earth Foundation (2021), doi:10.34911/rdnt.j0co8q.

## Load the data

```python
import adlfs
fs = adlfs.AzureBlobFileSystem(account_name='radiantearth', anon=True)
fs.ls('mlhub/ref-south-africa-crops-competition-v1/')[:10]
```

Lists the files with anonymous access. Runs on any machine, with no cloud account.

## Data structure

File types in sampled files: JSON

Example file: `.source/metadata.json`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/c851b04a-880f-53a0-84e8-1f223e90da1f
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/c851b04a-880f-53a0-84e8-1f223e90da1f
- Over MCP: `get_schema_card` with id `c851b04a-880f-53a0-84e8-1f223e90da1f` at https://api.scigantic.com/mcp

## Related topics

- [Earth Observation & Geospatial datasets](https://scigantic.com/topics/earth-observation-and-geospatial)
- [JSON datasets](https://scigantic.com/topics/json)
- [Satellite imagery datasets](https://scigantic.com/topics/satellite-imagery)
- [Time series datasets](https://scigantic.com/topics/time-series)
