how to achieve parameter sharing in the deep learning design?

Question

zini 2024-8-17，11:15

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2145939-how-to-achieve-parameter-sharing-in-the-deep-learning-design

回答： Amith 2024-8-17，11:37

in the deep learing design, is there a way to achieve an intermidiate layer, and this layer can accept varied length of inputs, each input is processed using the same learnable parameters.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Amith 2024-8-17，11:37

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/2145939-how-to-achieve-parameter-sharing-in-the-deep-learning-design#answer_1500069

Hi Zini,

If the Deep Learning Toolbo doesn't offer the specific layer you need for your task, you have the option to create your own custom layer. You can learn more about defining a custom deep layer with formatted inputs by visiting this link:

https://www.mathworks.com/help/deeplearning/ug/define-custom-deep-learning-layer-with-formatted-inputs.html

Hope this helps!