News for hackers. Stuff that matters.
Re:Pyro: PyTorch-Based Deep Universal Probabilistic Programming by l5870uoo9y on Friday November 03, 2017 @11:14AM (#15619770)
Anyone have a qualified opinion on how Pytorch compares to Tensorflow?
Re:Pyro: PyTorch-Based Deep Universal Probabilistic Programming by lowpro on Friday November 03, 2017 @12:06PM (#15620208)
Siraj Raval covers PyTorch very well in a 5 min video (https://youtu.be/nbJ-2G2GXL0).

It comes down to design decisions, of which I'm not qualified to go into. This article made the front page last week, about the downsides to Tensorflow which people rarely talk about: http://nicodjimenez.github.io/2017/10/08/tensorflow.html

And this interview with a Tensorflow engineer (10 mins) explains a little bit about those design decisions (https://youtu.be/axRHotkkTVI).

Re:Pyro: PyTorch-Based Deep Universal Probabilistic Programming by amelius on Friday November 03, 2017 @12:09PM (#15620229)
I guess whatever qualities both platforms have does not really matter compared to how much traction they get. If the world converges to one platform, then you should probably go with that platform, no matter how good the other platform is. The reason is that things are just moving too fast, and you don't want to spend your time porting new research from one platform to the other.

That said, I'm curious why PyTorch (or specifically Autograd) couldn't have been built on top of TF.

Re:Pyro: PyTorch-Based Deep Universal Probabilistic Programming by newlyretired on Friday November 03, 2017 @11:58AM (#15620136)
Re:Pyro: PyTorch-Based Deep Universal Probabilistic Programming by steev on Friday November 03, 2017 @12:57PM (#15620583)
I would roughly classify them based on use cases:

* If you are working on research such as optimization or other improvements on the algorithms of training neural networks, PyTorch is a better option as it is (in my experience) much more understandable and easily modified.

* If you are experimenting with network architectures and aren't going to be mucking around with the internals (e.g., developing a new optimization algorithm), Tensorflow is a better option.