Closed v0lta closed 3 years ago
For RNNs to work we will probably also require more data, I am not sure if we want to do this here, we could also choose to be happy with what we have
I think a next step, could be to ditch the vector approach and go try sequential models. These expect an input from each channel at every step. Any Ideas what a good way to sync all channels might be?
The sync between both servers is already done. I took care of that in the preprocessing.
For RNNs to work we will probably also require more data, I am not sure if we want to do this here, we could also choose to be happy with what we have
Perhaps not for the paper but it sounds interesting. I am collecting more samples right now. I expect to reach at least 300 samples
I would close this issue since the time series sync was already implemented.
I think a next step, could be to ditch the vector approach and go try sequential models. These expect an input from each channel at every step. Any Ideas what a good way to sync all channels might be?