Thanks. Now I know how to broadcast the dataset but I still wonder after broadcasting the dataset how can I apply my algorithm to training the model in the wokers. To describe my question in detail, The following code is used to train LDA(Latent Dirichlet Allocation) model with JGibbLDA in single machine, it iterate to sample the topic and train the model. After broadcasting the dataset, how can I keep the code running in Spark? Thank you. LDACmdOption ldaOption = new LDACmdOption(); //to set the parameters of LDA ldaOption.est = true; ldaOption.estc = false; ldaOption.modelName = "model-final";//the name of the output file ldaOption.dir = "/usr/Java"; ldaOption.dfile = "newDoc.dat"//this is the input data file ldaOption.alpha = 0.5; ldaOption.beta = 0.1; ldaOption.K = 10;// the numbers of the topic ldaOption.niters = 1000;//the times of iteration topicNum = ldaOption.K; Estimator estimator = new Estimator(); estimator.init(ldaOption); estimator.estimate();
-- View this message in context: http://apache-spark-user-list.1001560.n3.nabble.com/How-to-process-multiple-classification-with-SVM-in-MLlib-tp7174p7368.html Sent from the Apache Spark User List mailing list archive at Nabble.com.