Reconstruct shifted values after training a dynamic neural network ?
显示 更早的评论
I'm implementing a dynmaic neural network that has an input delay and also a feedback delay. After training and simulating my neural net I observed that the output has been shifted respectively with the number of steps of the my time delay. Is there a way to reconstruct the lost values?
采纳的回答
Greg Heath
2017-9-20
0 个投票
They are not lost.
It takes multiple input values to yield 1 output value.
Therefore you cannot have as many output time values as input time values.
Hope this helps.
Thank you for fomally accepting my answer
Greg
更多回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
