Last active
August 29, 2015 14:02
-
-
Save myui/84795a5afc0d6b16e2ad to your computer and use it in GitHub Desktop.
spark training on kddcup2012 track2 dataset
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import org.apache.spark.mllib.util.MLUtils | |
import org.apache.spark.mllib.classification.LogisticRegressionWithSGD | |
val training = MLUtils.loadLibSVMFile(sc, "hdfs://dm01:8020/user/hive/warehouse/kdd12track2.db/training_libsvmfmt_10k", multiclass = false, numFeatures = 16777216, minPartitions = 64) | |
//val training = MLUtils.loadLibSVMFile(sc, "hdfs://dm01:8020/user/hive/warehouse/kdd12track2.db/training_libsvmfmt_10k", multiclass = false) | |
val model = LogisticRegressionWithSGD.train(training, numIterations = 1) | |
//val model = LogisticRegressionWithSGD.train(training, numIterations = 20) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The dataset used for the training.
https://dl.dropboxusercontent.com/u/13123103/spark/training_libsvmfmt_10k.t
We evaluated spark 1.0 on 33 nodes in which each executor uses 7GB of memory.
The Hadoop version used in the evaluation is CDH3u6.
Spark seems too slow (does not finish in at least in 30m!) though Liblinear requires just 2m39s for convergence with 11 iterations.