Hello I am a M.Sc student in ML and and I’m trying to get some into my DNN. I have seen some articles on using t-sne for vizualizeing the fully layers of a CNN.

What do you think of the approach for a DNN trained on structured (tabular) ? Any practical on how to apply it? Alternatively would you recommend some other method to vizualize and understand the internals of a DNN?



Source link
thanks you RSS link
( https://www.reddit.com/r//comments/9nhw2t/d__for_insight_into__/)

LEAVE A REPLY

Please enter your comment!
Please enter your name here