What are some common strategies to visualise a weight matrix in a convolutional neural network?

I.e. visualise a matrix with the form (kH, inD, outD) in 1D convolution, and (kH, kW, inD, outD) in 2D convolution, where k is the kernel sizes and in/out is the input/output dimensions.

Preferably in Numpy/Matplotlib

Thanks Fred

  • 1
    $\begingroup$ As currently written, this question appears to be entirely about programming in Python and not about statistics, making it off-topic here. However you may be able to edit the question to make it on-topic by making it less about syntax in a particular language/package and more generally about data visualization strategies for your particular problem (which ideally you'd describe in a bit more detail). $\endgroup$ – Jake Westfall Aug 3 '17 at 21:20

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.