clear query| facets| time Search criteria: author:"Dmitriy Lyubimov".   Results from 1 to 10 from 71 (0.0s).
Loading phrases to help you
refine your search...
[MAHOUT-1774] Enable maven compilation with JDK 1.8 - Mahout - [issue]
http://issues.apache.org/jira/browse/MAHOUT-1774    Author: Dmitriy Lyubimov , 2015-11-07, 02:32
[MAHOUT-1781] Dense matrix view multiplication is 4x slower than non-view one - Mahout - [issue]
...if mxA and mxB are two in-core DenseMatrix matrices, then mxA(::,: %% mxB(::,: takes 4x the time of mxA %% mxB.possibly an issue of dot products on VectorViews vs. DenseVectors.dot product o...
http://issues.apache.org/jira/browse/MAHOUT-1781    Author: Dmitriy Lyubimov , 2016-04-11, 08:42
[MAHOUT-1791] Automatic threading for java based mmul in the front end and the backend. - Mahout - [issue]
...As we know, we are still struggling with decisions which path to take for bare metal accelerations in in-core math. Meanwhile, a simple no-brainer improvement though is to add decision paths...
http://issues.apache.org/jira/browse/MAHOUT-1791    Author: Dmitriy Lyubimov , 2018-03-03, 22:19
[MAHOUT-1792] Documentation, website-related stuff. - Mahout - [issue]
http://issues.apache.org/jira/browse/MAHOUT-1792    Author: Dmitriy Lyubimov , 2018-03-03, 21:10
[MAHOUT-1566] Regular ALS factorizer with convergence test. - Mahout - [issue]
...ALS-related: let's start with unweighed, unregularized implementation....
http://issues.apache.org/jira/browse/MAHOUT-1566    Author: Dmitriy Lyubimov , 2015-04-13, 10:20
[MAHOUT-1571] Functional Views are not serialized as dense/sparse correctly - Mahout - [issue]
...per Sebastian Schelter all entries of a TransposeView (and possibly other views) of a sparse matrix are serialized, resulting in OOM...
http://issues.apache.org/jira/browse/MAHOUT-1571    Author: Dmitriy Lyubimov , 2015-04-13, 10:20
[MAHOUT-1572] blockify() to detect (naively) the data sparsity in the loaded data - Mahout - [issue]
...per Sebastian Schelter:.bq a dense matrix is converted into a SparseRowMatrix with dense row vectors by blockify(), after serialization this becomes a dense matrix in sparse format (triggeri...
http://issues.apache.org/jira/browse/MAHOUT-1572    Author: Dmitriy Lyubimov , 2016-04-19, 15:37
[MAHOUT-1573] More explicit parallelism adjustments in math-scala DRM apis; elements of automatic parallelism management - Mahout - [issue]
...(1) add minSplit parameter pass-thru to drmFromHDFS to be able to explicitly increase parallelism. (2) add parrallelism readjustment parameter to a checkpoint() call. This implies shuffle-le...
http://issues.apache.org/jira/browse/MAHOUT-1573    Author: Dmitriy Lyubimov , 2015-04-13, 10:21
[MAHOUT-1583] cbind() operator for Scala DRMs - Mahout - [issue]
...Another R-like operator, cbind (stitching two matrices together). Seems to come up now and then. Just like with elementwise operations, and, perhaps some other, it will have two physical imp...
http://issues.apache.org/jira/browse/MAHOUT-1583    Author: Dmitriy Lyubimov , 2015-04-13, 10:20
[MAHOUT-1597] A + 1.0 (element-wise scala operation) gives wrong result if rdd is missing rows, Spark side - Mahout - [issue]
...    // Concoct an rdd with missing rows    val aRdd: DrmRdd[Int] = sc.parallelize(      0 -> dvec(1, 2, 3) ::          3 -> ...
http://issues.apache.org/jira/browse/MAHOUT-1597    Author: Dmitriy Lyubimov , 2015-04-13, 10:21