how do i define a convolution layer for time series data matlab

7 views (last 30 days)
how do i define a convolution layer for time series data matlab using deep learning designer app?When i use sequence input layer before convolution layer , it shows input size mismatch error

Accepted Answer

Davide Fantin
Davide Fantin on 24 May 2021
As of MATLAB R2021a, defining a convolution layer for time series data is not directly supported in Deep Network Designer.
Currently, the 1-D convolution operation is available through the custom training loop workflow with a "model as function" approach and you can find an example available in the doc here: https://www.mathworks.com/help/deeplearning/ug/sequence-to-sequence-classification-using-1-d-convolutions.html
The Development team is aware of this limitation and will consider to address it in future releases.
Solutions that you can try:
  • Recurrent Neural Network (lstmLayer, gruLayer, ...), available in Deep Network Designer App
  • Adaptation of Feed-Forward Networks, i.e. folding/unfolding of sequences and a fullyConnectedLayer in between, available in Deep Network Designer App.
  • Convolutional Neural Network approach as described in the aforementioned example, using a custom training loop (hence not using Deep Network Designer)
Hope this helps!

More Answers (1)

Girijashankar Sahoo
Girijashankar Sahoo on 24 May 2021
Choose length of input is same as length of comvolution layer
  1 Comment
NN
NN on 24 May 2021
Thanks for the suggestion , but please let me know how i can do it .?i am attaching the layer details,which parameter should i change in accordance with number of features as in sequence input data?

Sign in to comment.

Community Treasure Hunt

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

Start Hunting!