Model Zoo

[brianterry-ama]

Problem Statement

In order to evaluate distributed cloud native datasets we need a set representative examples for testing various typical use cases across the spectrum of dataset file size.

Proposed Solution

Create a file zoo with various and common formats and sizes to test against

Potential sources

• Larger Datasets: https://huggingface.co/Major-TOM
• Small Cogs: https://planetarycomputer.microsoft.com/dataset/io-lulc-annual-v02
• Zarr: https://github.com/google-research/arco-era5