zhengruifeng commented on pull request #32857: URL: https://github.com/apache/spark/pull/32857#issuecomment-858366098
[test](https://github.com/apache/spark/pull/32857/commits/7e8f195fc056ce39b5032fb85706615ada1f66c4) : ``` test("sparse gemm") { val rng = new scala.util.Random(0) val m = 1e3.toInt val n = 1e3.toInt val k = 1e3.toInt val alpha = rng.nextDouble val beta = rng.nextDouble val a = Array.tabulate(m * k) { _ => if (rng.nextDouble < 0.9) 0.0 else rng.nextDouble } val A = Matrices.dense(m, k, a).toSparse val B = Matrices.dense(k, n, Array.fill(k * n)(rng.nextDouble)).toDense val C = Matrices.dense(m, n, Array.fill(k * n)(rng.nextDouble)).toDense val tic1 = System.nanoTime Iterator.range(0, 1000).foreach { i => BLAS.gemm(alpha, A, B, beta, C) } val toc1 = System.nanoTime // scalastyle:off println println(f"new sparse gemm duration: ${(toc1 - tic1) * 1e-9}%.3f") // scalastyle:on println } ``` results: master: 92.080; this pr: 79.319 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
