Lets say we have an autoencoder with three layers. And the data we feed to autoencoder is text data. So in the middle layer, we will have a reduced dimension of the input data. let's say 20 neurons in the middle layer with consider 20 features in each (largest weight features).

Consider that I have already saved the weight changes of these features over 50 epochs.

For example: computer 0.01, 0.04, 0.5, -0.3, 0.8 ...

Is there any cool tools or any cool approach to visualize the change of the weight associated to these specific features?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.