The result of DA-RNN look bad so how to improve it to get result like the paper?

sunfanyunn / DARNN

A Dual-Stage Attention-Based Recurrent Neural Network for Time Series Prediction

112 stars 23 forks source link

Open lzh66 opened 4 years ago

sunfanyunn commented 4 years ago

Results reported on the paper are obtained through extensive parameter search.