[Question] WGAN-GP for timeseries data

[pbezz1]

Can I use WGAN-GP for timeseries data such as stock data? If yes do I need to preprocess the data to create sequences of say 24 data points in the same way described for the TimeGAN approach? Any help or direction is much appreciated. Thanks.

[jfsantos-ds]

Hey Patrick! For this you would need either an RNN synthesizer ran in a stateful manner or a conditional generator. WGAN_GP does not meet any of these! See this synthesizer as a vanilla GAN with Wasserstein loss and Gradient Penalty instead. (Not indicated for time series)