Isn't this like autograph in TF? https://www.tensorflow.org/guide/autograph
Today I had the opportunity to dig deeper into MXNet's approach to hybrid eager/graph execution. Really interesting. https://mxnet.incubator.apache.org/tutorials/gluon/hybrid.html …
-
-
-
This is more like the difference between graph and eager execution in TF.
End of conversation
New conversation -
-
-
This Tweet is unavailable.
-
At MXNet we certainly believe so, though
@jekbradbury might disagree? (because, PL for the win?). Nonetheless, we have a fair amount of experience with this approach now, and it enabled us to do some interesting stuff. happy to meet and share a deep-dive if helpful@fchollet - Show replies
-
-
-
@ApacheMXNet Hybridization is the way to go!Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
It looks pretty neat.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Reading their updated tutorial. I’m not sure I prefer this style over either TF 2.0 or PyTorch 1.0. The hybridize() method reminds me of http://sess.run () in the manner it “hides” things. https://gluon.mxnet.io/chapter07_distributed-learning/hybridize.html …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
*whispers* pytoooooorch
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.