How can pass the output of the fully connected layer of a deep learning model to the sequence input layer of another deep learning model?

Question

mahmoud Bassiouni 2022-12-20

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1881967-how-can-pass-the-output-of-the-fully-connected-layer-of-a-deep-learning-model-to-the-sequence-input

编辑： James Gross 2022-12-21

I need to known how to pass the features extracted from the fully connected layers of a pre-trained model for example (Alexnet) to

a sequence input layer of an LSTM model any Help.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

mahmoud Bassiouni 2022-12-20

So that the Alexnet and the LSTM model become one model

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

James Gross 2022-12-21

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1881967-how-can-pass-the-output-of-the-fully-connected-layer-of-a-deep-learning-model-to-the-sequence-input#answer_1133152

编辑：James Gross 2022-12-21

Hi Mahmoud,

Considering the sequence input layer and LSTMs are meant to be used for sequential data while AlexNet is for images, it sounds like you would like to use a pre-trained image classification model for video classification. If this is the case, the following documention example demonstrates precisely how to do this using GoogleNet.

https://www.mathworks.com/help/deeplearning/ug/classify-videos-using-deep-learning.html

You should be able to easily modify this example for a pre-trained image classification model of your choosing.

I hope this information helps!

James