Clay Foundation Model — Clay Foundation Model

The article discusses the Clay Foundation Model, an open-source AI model for Earth that uses a Vision Transformer architecture to understand geospatial and temporal relations on Earth Observation data. The model is trained using a Self-supervised learning (SSL) method with a Masked Autoencoder (MAE). The Clay model can generate semantic embeddings for any location and time, fine-tune for tasks such as classification, regression, and generative tasks, and serve as a backbone for other models.

The article also provides resources related to the Clay model. The website for the project is madewithclay.org, and the model's code and documentation are available on Github and the project's site, respectively. The model's weights can be found on Hugging Face. Future plans include maintaining a set of embeddings on Source Cooperative. The Clay project is sponsored by the non-profit Radiant Earth Foundation.

Key takeaways:

Clay is an open source AI model for Earth that uses a Vision Transformer architecture to understand geospatial and temporal relations on Earth Observation data.
The Clay model can be used to generate semantic embeddings for any location and time, fine-tune the model for downstream tasks, and as a backbone for other models.
The Clay model's code, weights, and documentation are available on their website, Github, Hugging Face, and on-site respectively.
Clay is a project sponsored by the non-profit Radiant Earth Foundation.

Clay Foundation Model — Clay Foundation Model

Key takeaways:

Comments (0)

Newsletter