can an example is provided to use Attention mechanism in time series sequence data? and also how to use it with LSTM?.

I have large time series sequence data, and I want to use attention mechanism for this data, and also concatenate the output of the attention mechanism with LSTM. Can any one help me in this regards by solving an example?

Answer 1

[Hornett]

Hi,

You can create a custom deep learning layer, then add it to your LSTM network and use trainnet to train the LSTM with attention mechanism.

Take a look at the following MATLAB answer: https://in.mathworks.com/matlabcentral/answers/1743390-how-to-create-an-attention-layer-for-deep-learning-networks

Hope it helps!