Looking at recent growth stats for deep learning frameworks. Fast-growing: TF, Keras, Caffe, MXNet. Medium growth: CNTK, DL4J, Theano.
-
Still seeing great recent work in torch
it has lots of inertia (FB notably) but hardly any new adopters, and lots of people moving away. Net total is negative
-
-
Must be the language. LuaJIT with cuDNN5 very fast, many available libraries, and integrates well with Python.
-
"supports cuDNN5, very fast, integrates with Python" is true of ten or so libs. Maybe ~3 will survive in the long term.
End of conversation
New conversation -
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.