News for hackers. Stuff that matters.

A comment on Jeff Dean explains TensorFlow [video]. Parent comment

Re:Jeff Dean explains TensorFlow [video] by mailshanx on Tuesday November 10, 2015 @01:56AM (#10537959)
I think the fundamental differentiator might be how "production ready" TensorFlow is - the promise of simply declaring an ML pipeline and have that run in very heterogeneous compute environments from mobile phones to GPU blades to plain-old clusters, if fulfilled, can indeed be a huge game changer. The promise is that you literally do not have to write any new code when you are done with a research project / a series of experiments and are ready to deploy your pipeline at a large scale. None of Theano / Torch etc make that promise.
[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by argonaut on Tuesday November 10, 2015 @02:37AM (#10538046)
That's not really a fundamental differentiator. Torch/Theano are definitely production ready. I think the portability is definitely an advantage, though.
[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by albertzeyer on Tuesday November 10, 2015 @04:15AM (#10538289)
I think you would not use Theano in an end-user product. It's made for developers to run on developer machines. It's very fragile. It has a very long start-up time, might be in the order of several minutes at the first start.

Maybe it would work good enough in a service backend. But even there it would not scale that well. For example, it doesn't support multi-threading (running a theano.function from multiple threads at the same time).

[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by argonaut on Tuesday November 10, 2015 @05:50AM (#10538617)
Good point. Torch, then.
[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by mtw on Tuesday November 10, 2015 @07:35AM (#10538927)
Torch is used in production by Facebook , deepmind and possibly baidu, amongst others. Facebook especially have released code to make torch run much faster on AWS with GPU cards. Also no startup time. The design is done with a high level language (lua) while computation done mostly in C. I'd be very surprised if tensorflow is actually faster than torch on a single machine
[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by AMEDICALRe on Tuesday November 10, 2015 @08:49AM (#10539214)
Torch has extremely difficult learning curve due to Lua. With Tensor Flow underlying engine in C++ it is likely as efficient as Torch. Special extension such as Nvidia Cudnn could also be used with tensorflow.
[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by pritambaral on Tuesday November 10, 2015 @09:27AM (#10539435)
What do you mean by "diffcult learning curve due to Lua"? Lua isn't difficult. Lua is easier than JavaScript!

Or perhaps you mean to say that Torch itself is difficult to learn because of design choices that were made in order to use Lua?

[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by olooney on Tuesday November 10, 2015 @10:14AM (#10539760)
Learning an entirely new language for a project, no matter how simple that language is, is certainly a barrier to entry.
[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by catwell on Wednesday November 11, 2015 @01:21PM (#10548194)
If you have already used another dynamic, imperative language, you can probably learn enough Lua to use Torch effectively in 30 minutes.

Seriously, there are three features in the Lua language which are not trivial: metatables, coroutines and _ENV. None of those are needed to use Torch.

It will take more time to learn Torch-specific APIs, but the same problem exists with the other ML frameworks.

[ Parent | on HN ]
Re:Jeff Dean explains TensorFlow [video] by argonaut on Tuesday November 10, 2015 @11:51PM (#10544880)
If somebody finds learning Lua to be difficult, then learning C++, learning basic statistics, or learning machine learning will be impossible for them.
[ Parent | on HN ]