Hi, I have a library of clustering algorithms that I'm trying to run in the SparkR interactive shell. (I am working on a proof of concept for a document classification tool.) Each algorithm takes a term document matrix in the form of a dataframe. When I pass the method a local dataframe, the clustering algorithm works correctly, but when I pass it a spark rdd, it gives an error trying to coerce the data into a vector. Here is the code, that I'm calling within SparkR:
# get matrix from a file file <- "/Applications/spark-1.5.0-bin-hadoop2.6/examples/src/main/resources/matrix.csv" #read it into variable raw_data <- read.csv(file,sep=',',header=FALSE) #convert to a local dataframe localDF = data.frame(raw_data) # create the rdd rdd <- createDataFrame(sqlContext,localDF) #call the algorithm with the localDF - this works result <- galileo(localDF, model='hclust',dist='euclidean',link='ward',K=5) #call with the rdd - this produces error result <- galileo(rdd, model='hclust',dist='euclidean',link='ward',K=5) Error in as.vector(data) : no method for coercing this S4 class to a vector I get the same error if I try to directly call as.vector(rdd) as well. Is there a reason why this works for localDF and not rdd? Should I be doing something else to coerce the object into a vector? Thanks, Ellen -- View this message in context: http://apache-spark-user-list.1001560.n3.nabble.com/SparkR-calling-as-vector-with-rdd-dataframe-causes-error-tp24717.html Sent from the Apache Spark User List mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
