Hi,
I was wondering if there is a way to get more data from the built-in Machine Learning functions (especially on training); it would be useful to see how the given parameters are tuned or what the training functions output per iteration (e.g. gradient while calculating the gradient descent, or cluster centroids while performing k-means clustering).
Summarising the model does not go that much in detail, does anyone know how (and if) that is possible to do? What approaches could be taken to track those values?
Thank you very much in advance!
Flaminia
