Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by arbre on Wednesday November 11, 2015 @12:09PM
(#10547672)
This is very exciting to see these distributed deep learning frameworks open sourced by top companies. What I do not understand is why amazon AWS and other top cloud service are not integrating these frameworks in their services. Training a distributed neural net should be as simple as defining the model and specifying the resources.
[ Parent
| on HN ]
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by vegabook on Wednesday November 11, 2015 @12:12PM
(#10547695)
agreed - though I'm somewhat saddened to see that this library doesn't seem to be getting the same positive reception (in terms of scoring / comments) as Tensorflow did yesterday. Am I right and if so, is there a technical reason for that?
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by cbgb on Wednesday November 11, 2015 @01:38PM
(#10548323)
One technical reason may be that C++ (I only looked at LightLDA) has a higher learning curve than Python, which the Tensorflow docs stress.
Though Tensorflow stayed at #1 for a while, it only garnered around 200 comments, which is very high for a ML topic on HN (in my anecdotal experience, ML topics are highly rated but under-commented.) I imagine the audience for this library to be very small compared to Tensorflow, which likely included more ML/Google FOSS enthusiasts than day-to-day practitioners. Looks like this library is firmly targeted toward the latter.
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by graffitici on Wednesday November 11, 2015 @12:17PM
(#10547720)
Not sure if this was ready for public release.. The chess demo released this morning seems more polished than this library.. Also, probably not the best decision to post it a day after the TensorFlow release..
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by math_and_stuff on Wednesday November 11, 2015 @12:17PM
(#10547722)
Less fancy landing page? No fancy names in the title? Marketing works.
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by blazespin on Wednesday November 11, 2015 @02:13PM
(#10548553)
NO GPU, not for production use.
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by nightski on Wednesday November 11, 2015 @04:54PM
(#10549608)
You do realize there are a fair amount of machine learning algorithms that do not run efficiently on the GPU right? Deep learning isn't the only method out there...
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by blazespin on Wednesday November 11, 2015 @05:36PM
(#10549888)
Fair enough! None that I'm interested in, however.
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by gcr on Wednesday November 11, 2015 @07:45PM
(#10550530)
Come on, keep an open mind! Random forests still work great!
There's still some love for genetic algorithms somewhere, right?
...Right ??
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by irascible on Thursday November 12, 2015 @08:14AM
(#10552889)
Someone make me a painting of a shriveled genetic algorithm dying in a random forest..
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by vegabook on Wednesday November 11, 2015 @04:29PM
(#10549417)
fair enough. By contrast it seems to support clusters.
Re:DMTK: Distributed Machine Learning Toolkit from Microsoft
by x0x0 on Wednesday November 11, 2015 @02:47PM
(#10548791)
distributed ml tends not to work well on aws. Many (most?) distributed ml algorithms are essentially limited by network speed. In a p2p topology, many are effectively limited by the slowest peer-to-peer link of the (nodes choose 2) links. aws networking is mediocre at best; even the rack-local instances are relatively slow.
cred: helped build a distributed in-memory ml toolkit