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
Some possible issues to brainstorm (and I am interested in inputs from anyone that likes it as well).
Since a user can generate 100s of results in a single experiment, and that might make scrolling horizontally uncomfortable, and cause issues with the diffing, what if we let the user generate the results then select two of them to compare (like when you compare two different products on Amazon)?
If we were to display more than two results side by side, should we diff from the first result or from the previous result?
I think diffing the inference parameters is a great idea, BTW!
It might be nice to be able to:
Here's a crappy mock I whipped up:
Just an idea - feel free to close if not desirable.
The text was updated successfully, but these errors were encountered: