https://github.com/nixtla logo
#neural-forecast
Title
# neural-forecast
m

marah othman

05/04/2023, 11:27 AM
i have questions : do we treat the problem of non stationary time series? in the library also why we put the input size -1 as default
k

Kin Gtz. Olivares

05/04/2023, 2:34 PM
Hi @marah othman, We use by default in the RNN-Based methods (RNN, GRU, LSTM, TCN,...) all the available history for the series at the time of the prediction y[:t]. If you want to reduce the inference time you can use the
input_size
, to make the inputs of fixed size y[t-input_size:t].
m

marah othman

05/05/2023, 5:48 AM
do we treat the problem of non stationary time series? in the library . and what about this