I think this was the one. Thank you!https://twitter.com/_joaogui1/status/1332122280521052161?s=19 …
Show this thread
Wasn't there a recent study that provided data about the effectiveness of model architecture choices made by experts vs beginners? Anyone has the link?
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Please share the article once you find it, I am interested to read it and read more about model architecture choices made by experts.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Also quite interested. Please keep us updated
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I mean, this is more about other hyperparameters, I believe they kept the architecture fixed
End of conversation
New conversation -
-
Would also love to see this. Intuitively I bet there's less difference than people might expect
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
My experience has almost always been that the first architecture you try on a project is almost always the best one
-
Depending on the architecture, who really has the resources to run a formal CV-GridSearch?
- Show replies
New conversation -
-
-
Why don't we have a good model to predict architecture basing on dataset samples and training trials?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.