The next model in the FluxArchitectures repository is the “Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction”, based on the paper by Qin et. al., 2017. It claims to have a better performance than the previously implemented LSTNet, with the additional advantage that an attention mechanism automatically tries to determine important parts of the time series, instead of introducing parameters that need to be optimized by the user.

JuliaCon 2020 will now be a fully online conference. With all the uncertainties due to Coronavirus, this seems like the best option to keep everyone safe, and provide some certainly in planning.  Many details are still to be finalised, and we’ll say more soon. In particular, details about the schedule, tickets and registration will be communicated in the next two weeks. Watch this space!  Talk reviews have been completed, and acceptance decisions will be made very soon.  Watch this space!  Regar...