Advice on models and processes for sequential data

Deep Learning
1

Hello all,

I’m new to deep learning and am looking for some advice on models and processes to be using for my task.

I have sequential data consisting of eight variables at every time step as input; and the output is one variable, also at every time step.

I am starting with a simple RNN using LSTMs. Is there any other models or processes that I should be aware of that may improve performance?

Thanks very much!

2

Hi @stoops_ML ,
You may check the time series/ sequential data thread.
Another resource that may be of interest to you is tsai. It is a time series and sequential data open-source library built on top of Pytorch/ fastai. It contains a number of tutorial notebooks that may help get started.

1 Like