You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Many years ago I did document classification using the FreeViz part of the Linear Projection widget. Back then the optimization ran until manual stopping and there were numerous controls (to me most important were the separation force controls) to guide the optimization algorithm when it got stuck in local minima. I could easily achieve 99% accuracy in text classification where the most complex models alone could reach up to 60-70%.
Replicating this has been impossible since FreeViz became auto-stopping and omitted the controls. Is there a chance for the old controls to be implemented in recent versions?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hi!
Many years ago I did document classification using the FreeViz part of the Linear Projection widget. Back then the optimization ran until manual stopping and there were numerous controls (to me most important were the separation force controls) to guide the optimization algorithm when it got stuck in local minima. I could easily achieve 99% accuracy in text classification where the most complex models alone could reach up to 60-70%.
Replicating this has been impossible since FreeViz became auto-stopping and omitted the controls. Is there a chance for the old controls to be implemented in recent versions?
Beta Was this translation helpful? Give feedback.
All reactions