vancouverevs/afriadapt-agri-train-v2

AfriAdapt Agriculture Training Dataset (v2)

High-quality instruction-response dataset focused on practical agricultural advice for smallholder farmers in East and Southern Africa.

This is version 2 of the dataset, significantly expanded for the Adaption AutoScientist Challenge 2026.

Dataset Summary

• —Domain: Agriculture & Climate-smart farming
• —Target users: Smallholder farmers and agricultural extension workers
• —Geographic focus: Kenya, Tanzania, Uganda, Ethiopia and similar regions
• —Language: English
• —Size: ~1,538 high-quality examples
• —Format: Instruction → Output pairs
• —Version: 2.0

What's New in v2

• —Significantly larger dataset (previously 176 → now ~1,538 examples)
• —Broader coverage of practical farming scenarios
• —Improved diversity across climate stress, pests & diseases, soil fertility, water management, and mixed cropping
• —Higher consistency in response quality

Intended Use

This dataset is designed for:

• —Fine-tuning / adapting language models for agricultural advisory tasks
• —Improving model performance on practical, resource-constrained farming scenarios
• —Research on climate-resilient and locally relevant agricultural AI
• —Use with AutoScientist and similar adaptation frameworks

Data Fields

Example

Instruction:

In the semi-arid region of Machakos County, Kenya, smallholder farmer Jane is experiencing severe drought that has led to moisture stress in her maize crop...

Output:

To manage moisture stress in her maize crop, Jane can take the following steps: first, she should mulch her farm using available organic materials...

Data Creation

The dataset was generated using the AfriAdapt pipeline with strong emphasis on:

• —Realistic smallholder constraints
• —Local context (East & Southern Africa)
• —Practical and safe recommendations
• —Climate resilience and limited-resource decision making

License

Apache 2.0

Citation

@dataset{afriadapt_agri_train_v2_2026,
  title={AfriAdapt Agriculture Training Dataset v2},
  author={AfriAdapt},
  year={2026},
  url={https://huggingface.co/datasets/vancouverevs/afriadapt-agri-train-v2}
}