Deleting some layers from trained neural network model

4 次查看（过去 30 天）
Abdelwahab Afifi
Abdelwahab Afifi 2020-6-22
回答： Bhargavi Maganuru 2020-7-6
I have trained a big neural network NN which has the structure [20 2 20] as shown in the first figure. After training, I wanna separate/copy/keep the first part of the network [20 2] and save its wights, biases, and connections as shown in the second figure. How can I do so?

请先登录，再回答此问题。

回答（1 个）

Bhargavi Maganuru
Hi,
You can get weights, biases of feedforward network using network object parameters IW,LW,b
You can refer following links for more information
https://www.mathworks.com/help/deeplearning/ug/neural-network-object-properties.html#bss4hk6-60
https://www.mathworks.com/help/deeplearning/ug/neural-network-subobject-properties.html#bss4hk6-136

类别

在 Help Center 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by