I'm working on a new project on climate data. I suppose my output $Y(H*W*1)$ is a consequence of a given data scenario $X(C=4*T*H*W)$ and an initial state $Y0(H*W*1)$. I've chosen a temporal convolutional network
(more or less this one)
For X spatio-temporal feature extraction. The fact is I have data with different timesteps (for example 1 dataset is 500 years long and as a 10 years timestep, and another is 100 years long ans has a 1 year timestep).
I wonder if someone has an idea of how I can include length and timestep as inputs of my model in order my model will be able to deal with different length and timesteps?
Thank you very much for your help, Marin