Incorporate trained LSTM network in simulink

1 回表示 (過去 30 日間)
Anjali Rijil
Anjali Rijil 2019 年 5 月 1 日
回答済み: Arkadiy Turevskiy 2021 年 10 月 18 日
I have to validate my trained network on the system. For this I need to make a simulink model. I have to get the trained network in simulink. gensim does not work since it can only be used for NARX trained network, but I have my network trained using LSTM. Please help.
  1 件のコメント
Prajith Chilummula
Prajith Chilummula 2019 年 5 月 10 日
As an implementation method, you load the weight information of each unit, generate a calculation formula as an m script using it, and introduce the generated m script as a Matlab Function block.
The below link might be helpful:
.

サインインしてコメントする。

回答 (1 件)

Arkadiy Turevskiy
Arkadiy Turevskiy 2021 年 10 月 18 日
Hi,
Starting with 21a you can use Stateful Predict and Stateful Clasiffy blocks.
Take a look at this example too.
Hope this helps.
Arkadiy

カテゴリ

Help Center および File ExchangeSequence and Numeric Feature Data Workflows についてさらに検索

タグ

Community Treasure Hunt

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

Start Hunting!

Translated by