You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Right now, the nn interface only allows TaggedTable as an input.
For TimeSeries I will add a temporary solution, with to_tagged_table, which converts the TimeSeries into a tagged table. This can also be useful for the future implementation of the InputLayer of the nn Interface.
Desired solution
Right now just a temporary solution and the Integration into the InputLayer Interface.
The lag parameter could also take place here
The text was updated successfully, but these errors were encountered:
This is not a good way to solve the problem, as the windows are not held as data internally and are more used a schema for the data to load into the tensors.
Seems related to our discussion about input/output layers that handle the conversion from/to tensors. Also seems related to @Marsmaennchen221's work on a DataSet class for images.
Is your feature request related to a problem?
Right now, the nn interface only allows
TaggedTable
as an input.For
TimeSeries
I will add a temporary solution, withto_tagged_table
, which converts theTimeSeries
into a tagged table. This can also be useful for the future implementation of the InputLayer of the nn Interface.Desired solution
Right now just a temporary solution and the Integration into the InputLayer Interface.
The lag parameter could also take place here
The text was updated successfully, but these errors were encountered: