Ignacio Montes Álvarez

Thanks both of you for answering! I would like to give a little bit more of context on my dataset, the 13 vectors of 100 points (then as suggested, `torch.Size([1500,...

I managed to train the model with the following architecture and it seems to be optimizing! ``` import gpytorch N_INTEGRALS = 13 class MultitaskGPModel(gpytorch.models.ExactGP): def __init__(self, X_train, train_y, likelihood): super(MultitaskGPModel,...

Thanks for your answer! I did some experiments in the inference part but I end up with the same memory issues so that seems to not be helping me. Also...

Sorry with 'testing loop' I just meant the inference process. However, I have managed to infer one datapoint at a time, which is really slow but it is better than...

It would be nice to know whether 'encoder-decoder' architectures are supported as I also want to use LMCocktail for my research 😃 @RKoopal your implementation looks nice, curious to hear...