"The war between ML frameworks has raged on since the rebirth of deep learning." That's a bit too dramatic though, no?https://twitter.com/gradientpub/status/1182327049307844613 …
-
Can you explain why you think its cherrypicked? I "cherry picked" the conference list, but I believe it should roughly match consensus. If you have other conferences you think I should add to the figures I'd be glad to do so.
FYI the actual split in the research world is ~30-40% PT vs ~60-70% TF/Keras. This is the range of numbers you get by looking at 1) ArXiv mentions data, 2) Colab usage, 3) Actual research surveys. All the sources I can think of end up in this range. Extremely dishonest article.
-
-
Academic research is also literally the only usage category where you will see high PyTorch numbers (above 30%). And it's a very narrow category (~5% of TF usage).
-
Do you have any proof for these claims? At least from
@cHHillee we know how the numbers were derived. Also, what are your responses to a) his ablation study to include Keras to TF (no significant increase) and b) that he tackles more than NLP?
End of conversation
New conversation -
-
-
Seriously, Collab and surveys? The article is not dishonest at all. The top conferences is what matters for bleeding edge. Community has written off TF/Keras. It’s a giant mess even in TF2 with perf issues and fragmented API. Listen to trends instead of burying head in sand.
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.