I am currently training a Gaussian process regression model on the first 30% of a dataset to predict the last 70%. As i have numerous datasets which are somewhat similar, so i would like to use an already trained model, and fit it to a new dataset, where the parameters and weights are not just overwritten, but instead tuned to the new data, from the old model. When looking through the doumentation from SKlearns Gaussian process regression, i can't seem to figure out how to do this. I found that other models have the partial.fit() function, but this doesn't exist for the GPR.
So i would like to know if there is a workaround for this, or if there actually is a function which can achieve what i am trying to do.
