Re: Speed up LDA in Mahit 0.9

2015-05-08 Thread Yutaka Mandai
If it's small enough to fit in memory, setting MAHOUT_LOCAL=TRUE should drive you crazy! I've suffered a lot from running LDA(CVB0) on even on EMR. If you believe your data is small enough, then the local is the best. Regards,,, Y.Mandai iPhoneから送信 2015/05/07 20:12、mw m...@plista.com のメッセージ:

Re: Speed up LDA in Mahit 0.9

2015-05-08 Thread Andrew Musselman
I'd also recommend getting the newest version of Mahout, 0.10. On Fri, May 8, 2015 at 7:15 AM, Yutaka Mandai 20525entrad...@gmail.com wrote: If it's small enough to fit in memory, setting MAHOUT_LOCAL=TRUE should drive you crazy! I've suffered a lot from running LDA(CVB0) on even on EMR. If

Replacement for DefaultAnalyzer

2015-05-08 Thread Lewis John Mcgibbney
Hi Folks, I'm making an upgrade from Mahout 0.7 -- 0.9. I am experiencing the same problem as experienced in the following post [0]. Can someone please suggest what I should replace DefaultAnalyzer with? I am aware that it was removed from the Mahout API in 0.8? In the meantime I am going to tst