Resources

Open data, tools and compute

What the Lab shares with researchers, developers and educators.

Open datasets

Labelled image datasets collected in Tanzanian farms with smartphone cameras, published in Data in Brief.

102,668 images

  • Maize · 2023

    Maize leaf imagery dataset

    18,148 images

    Maize Lethal Necrosis and Maize Streak Virus

    Use for: Disease detection, classification, segmentation

    Neema Mduma, Hudson Laizer

    Data paper
  • Maize · 2024

    Maize leaf imagery dataset, expanded season

    9,356 images

    New farms and the November–January season

    Use for: Extends the 2023 maize dataset

    Neema Mduma, Flavia Mayo

    Data paper
  • Banana · 2023

    Banana leaf and stem imagery dataset

    16,092 images

    Black Sigatoka and Fusarium Wilt Race 1

    Use for: Object detection, classification, segmentation

    Neema Mduma, Judith Leo

    Data paper
  • Common bean · 2024

    Common bean leaf imagery dataset

    59,072 images

    Bean rust and bean anthracnose

    Use for: Early disease detection and diagnosis tools

    Hudson Laizer, Neema Mduma, Dina Machuve, Reinfrid Maganga

    Data paper

Shared GPU compute

The Lab runs NVIDIA A30 and A6000 GPU servers, shared through JupyterHub, for model development and evaluation. This year the Lab received 27 requests for shared services and 16 researchers used its infrastructure.

Request access

Learning materials

Written by Lab members for students and practitioners.

Practical Machine Learning: A Beginner's Guide with Ethical Insights
Ally S. Nyamawe, Mohamedi M. Mjahidi, Noe E. Nnko, Salim A. Diwani, Godbless G. Minja, Kulwa Malyango · Taylor & Francis, 2025

DOI

Partners and funders

Implementation partners include TMDA, AGCOT and the EAC AI Alliance.