Contribute artifacts

Contribute datasets, pipelines, and tools

Tools gallery

Commons Connect and Know Your Landscape are two key tools built on the CoRE stack datasets and analytics. And many contributors are building new tools!

CoRE insights

A neurosymbolic approach to diagnosing stresses in a landscape.
Try it out. Blog. Github. Dev call presentation.
Contributors: Aaditeshwar Seth, Shivani A. Mehta, Riti Verma, Immanuel Shadrach

CoRE stack Village Analytics Tool

Efficient generation of reports and data stories at a village level. Works for all villages in India.
Try it out. Methodology. Github. Dev call presentation.
Contributors: Aaman Sheikh, Saksham Kushwaha, M. Rayhan Khan

District Agriculture Insights

Detailed insights for a district on changes in cropping intensity, primary crops grown, and distances to markets.
Try it out. Methodology. Github. Dev call presentation.
Contributor: Sanket Gharat

Tracker for Protected Areas

Track the area deforested, degraded, afforested, and improved, within various protected areas in India.
Try it out. Blog. Github.
Contributor: Aaditeshwar Seth.

To contribute, check the APIs and artifacts to learn what kind of outputs are readily available. Think of new ideas and let us know of your innovations! Write to us at contact@core-stack.org to discuss any ideas if you want to brainstorm. And join the weekly developer calls at 3pm IST on Fridays – details here.

All CoRE stack artefacts are in the open-source [Github repo].

Datasets

The CoRE stack can be conceptualized as being comprised of three layers: Datasets built through rigorously evaluated methodologies, Analytics derived from these datasets, and Tools built using these datasets and analytics. While some datasets are static, most of them are dynamic and generated through automated pipelines that process geospatial data, apply machine learning models, etc. Likewise for analytics computed on the datasets. Take a look at various datasets and methodologies that are available in the CoRE stack.

We welcome you to contribute your own datasets and methodologies too and enrich the CoRE stack. This detailed guide explains how to write pipelines to extract data from underlying assets or run models to create new data. A wishlist of new datasets we would like to pull is also available. Write to us at contact@core-stack.org to discuss how you can contribute and add to this. Join the weekly developer calls at 3pm IST on Fridays to work together – details here.

News events Datasets: Human-wildlife conflicts, Avian Influenza, Crop damage

Event extraction using GDELT, followed by structured data extraction using LLMs, and geo-tagging of the locations mentioned in the articles.
Try it out. Blog. Github.
Contributor: Aaditeshwar Seth, IIT Delhi

All CoRE stack artefacts are in the open-source [Github repo].

Pipelines

Analytics and dynamic datasets that need to be updated regularly require software pipelines that can periodically download data like satellite imagery, process it locally on on cloud infrastructures like GEE, and generate and export new datasets. The CoRE stack implements several dozen pipelines that refresh rainfall data, generate land-use layers, compute change statistics, etc. See the datasets and methodologies to understand what kind of pipelines have been implemented.

We welcome you to contribute your own pipelines and enrich the CoRE stack. This detailed guide explains how to write pipelines to generate and refresh data. A wishlist of new pipelines we would like to pull is also available. Write to us at contact@core-stack.org to discuss how you can contribute and add to this. Join the weekly developer calls at 3pm IST on Fridays – details here.

All CoRE stack artefacts are in the open-source [Github repo, deep-wiki documentation].