News for hackers. Stuff that matters.
Re:Announcing TensorFlow 0.8 – now with distributed computin... by modeless on Wednesday April 13, 2016 @04:07PM (#11491739)
Very cool! Any progress on Windows support?
Re:Announcing TensorFlow 0.8 – now with distributed computin... by vrv on Wednesday April 13, 2016 @07:56PM (#11493333)
Re:Announcing TensorFlow 0.8 – now with distributed computin... by hebdo on Wednesday April 13, 2016 @06:03PM (#11492675)
I doubt it is a priority. But I can certainly recommend Amazon GPU-enabled instances (~0.6$/hour/GPU, not that much actually).
Re:Announcing TensorFlow 0.8 – now with distributed computin... by modeless on Wednesday April 13, 2016 @07:04PM (#11493080)
Linux is great for training, however I would like to deploy my models to run locally on user machines which are running Windows. Theano supports Windows but TensorFlow doesn't.
Re:Announcing TensorFlow 0.8 – now with distributed computin... by AgentME on Wednesday April 13, 2016 @10:17PM (#11493896)
Can models be ported between systems easily?
Re:Announcing TensorFlow 0.8 – now with distributed computin... by babo on Wednesday April 13, 2016 @06:11PM (#11492732)
g2 instances has a GPU which is not compatible with the stock tensorflow, you must rebuild it from source. Do you have a workaround for that?
Re:Announcing TensorFlow 0.8 – now with distributed computin... by vrv on Thursday April 14, 2016 @02:01AM (#11494593)
I believe our published wheels now include the code for cuda compute 3.0, so it should work out of the box now.

(as long as the images have cudnn v4 and cuda 7.5 installed, I think :)

Re:Announcing TensorFlow 0.8 – now with distributed computin... by babo on Friday April 15, 2016 @07:07AM (#11503405)
Great news, I'll try today!
Re:Announcing TensorFlow 0.8 – now with distributed computin... by manav on Wednesday April 13, 2016 @06:24PM (#11492829)
I find the Amazon GPU prices pretty high in the long run. The g2.2xlarge is around 3x slower than a GTX 980.
Re:Announcing TensorFlow 0.8 – now with distributed computin... by p1esk on Wednesday April 13, 2016 @07:11PM (#11493113)
It should be a priority. The main reason I picked Theano (and still use it) is Windows support.