combine an LSTM with a Gaussian Process

[aminims]

Dear Hakeem,

Thanks for the neat code you provided in the GitHub repository. I have a question: as far as I understand, your code uses Gaussian Process Regression with GPyTorch for prediction. I would like to combine an LSTM with a Gaussian Process — in other words, I want to use an LSTM as a feature extractor and feed its output into the Gaussian Process Regression model. Can you provide me an example this?

Thank you in advance,
Shoaib Amini