mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ted Dunning <>
Subject Re: spark-itemsimilarity out of memory problem
Date Tue, 23 Dec 2014 17:29:08 GMT
On Tue, Dec 23, 2014 at 9:16 AM, Pat Ferrel <> wrote:

> To use the hadoop mapreduce version (Ted’s suggestion) you’ll loose the
> cross-cooccurrence indicators and you’ll have to translate your IDs into
> Mahout IDs. This means mapping user and item IDs from your values into
> non-negative integers representing the row (user) and column (item) numbers.

I don't think that I was sufficiently discouraging about the map-reduce
version.  To be avoided if feasible.

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message