News for hackers. Stuff that matters.

A comment on Tensorflow sucks. Parent comment

Re:Tensorflow sucks by vaughngh on Sunday October 08, 2017 @08:22PM (#15430766)
I have tensors in later layers whose shapes depend on the training of the previous layers.

Rad! Do you have any examples (or literature) that explains when this is beneficial?

[ Parent | on HN ]
Re:Tensorflow sucks by gormanc on Sunday October 08, 2017 @08:34PM (#15430815)
Not yet! I'm not using convnets or backprop or anything so I don't think it would be beneficial that way, but you could get something similar to what I'm doing by looking at Fritzke's Growing Neural Gas[1]

[1] http://papers.nips.cc/paper/893-a-growing-neural-gas-network...

[ Parent | on HN ]
Re:Tensorflow sucks by vaughngh on Tuesday October 10, 2017 @12:12AM (#15439282)
Neat, thanks for the link.
[ Parent | on HN ]