I've seen that there are four neural net packages in R:
neuralneuralnetnnetRSNNSH2O
What are the advantages/disadvantages of those compared with each other?
As I found out NeuralNetTools only provides additional tools.
$\begingroup$
$\endgroup$
2
-
2$\begingroup$ I would also add h2o's deeplearning to the list. It's one of the most powerful implementations of NNets accessible from R. And I think it scales up really well. $\endgroup$– wacaxDec 23, 2015 at 1:02
-
$\begingroup$ I primarily use h2o now, built in parallel processing and ram allocation is very nice. $\endgroup$– TBSRounderDec 23, 2015 at 18:45
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
take a look at these blogs- R for Deep Learning - 1 and R for Deep Learning - 2.
Hope it helps!
-
1$\begingroup$ Not the most in-depth analysis of the packages themselves, but +1 for some information at least :-). I got the a little the impression the take away is "everything sucks", but maybe I am over-critical. $\endgroup$– Make42Oct 15, 2016 at 14:40
-
$\begingroup$ Glad, it could help a lil, atleast. But no worries will try to find something more specific and will update you on that. Cheers! :). $\endgroup$– AbhishekOct 15, 2016 at 14:44