Help modeling Multimodal Time Series data

Hey, I was wondering if anyone has any experience with time series modeling? I am currently working on a regression task using some multimodal data and I’m not sure what the best practices are for that kinda task and also what are good time-series models to use for the kind of data I have.

I’m just looking for general pointers to get started :slight_smile: I have tried basic MLP but that does not do very well.

The picture is one sample of the dataset (out of 100000).

A summary of the dataset:


#3 waves of length 201: Pvp Pvs

#4 summaries: map sbp dbp sv

#1 parameters: HR


#12 outputs: Ees_LV Ees_RV Vs Eas Cvs Eap Cvp Tes τ Rap Ras Brv