That wasn't crystal clear to me. Now it is! Thanks for reminding!
Some people complain that, as Keras users, they still have to write TF code for research use cases. That's precisely how it's meant to be!pic.twitter.com/WAw3Ea0AbU
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I find keras a too opaque layer on tensorflow, which makes exactly that too cumbersome. Maybe that's just me, but mixing tf+K was a pain
-
Which is why I don't use keras. Maybe just a matter of improving docs on, last time I checked them on that matter, they weren't of much help
End of conversation
New conversation -
-
-
do you plan on Keras supporting pytorch and dynamic computational graphs?
-
Dynamic graphs are coming to TF, and thus to Keras as well. I don't expect many users to need it though (some specialized NLP models...)
- Show replies
New conversation -
-
-
Being able to interface with the lower level language as needed is a godsend
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
It’s as fun as dropping to assembly
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I don't understand people who complain.
-
No lazy losers in Japan, like we got here?
- Show replies
New conversation -
-
-
LOL. I wanna meet some of them
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.