Why not make it standard to publish code? I feel paper format and goals doesn't use today's tools & reproducibility of results in science is still an issue. DL model is just 1 piece, training regime, loss, etc. matter a lot.
DLPaper2Code: learning to automatically generate Keras or Caffe code for deep learning papers. Not a joke! https://arxiv.org/abs/1711.03543 pic.twitter.com/ANFMv1OHvd
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Deep learning is getting seriously deep. How recursive can it get? For example, can another work "learn to write deep learning papers by training on deep learning papers and using DLPaper2Code" ?
#DeepLearning#AI -
Use AutoML to find this model and write the paper
End of conversation
New conversation -
-
-
@runmoemen
@mrgemy95 well this one wasn’t on our “fakes” list :”D - End of conversation
New conversation -
-
-
I would love to see the output, if the input would be it's own paper.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Any idea what the URL is for their model repo? From the paper: "the system URL is not provided in this version of the paper" (!)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Found the URL for the platform described in the paper: https://darviz.mybluemix.net Very cool project.
@ibmdarviz -
Just super awesome. Thanks for pointing this one out. Psyched to give this thing a try.https://www.youtube.com/watch?v=cfQ7GaeZgQw …
End of conversation
New conversation -
-
-
I'm going to need more coffee
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.