The step 3 is just to evaluate my recommender.
I had follow the same steps using CLI and everything works well : (mahout
splitDataset , mahout parallelALS, mahout recommendfactorized). I think
that is something wrong with the RecommederJob class. Here is my java
code:
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
import org.apache.hadoop.conf.Configuration;
import org.apache.mahout.cf.taste.hadoop.als.DatasetSplitter;
import org.apache.mahout.cf.taste.hadoop.als.ParallelALSFactorizationJob;
import org.apache.mahout.cf.taste.hadoop.als.RecommenderJob;
public class Test {
public static String baseDirectory =
MainAlsRecommender.class.getResource("/reco").getPath();
public static String inputSplitDirectory = baseDirectory
+"/ratings.csv";
public static String outputSplitDirectory = baseDirectory ;
public static String inputAlsDirectory = baseDirectory
+"/trainingSet/";
public static String outputAlsDirectory = baseDirectory +"/als/out";
public static String recommenderJobUserFeatures = baseDirectory
+"/als/out/U/";
public static String recommenderJobItemFeatures = baseDirectory
+"/als/out/M/";
public static String recommenderJobInput = baseDirectory
+"/als/out/userRatings/";
public static String recommenderJobOutput = baseDirectory
+"/recommendations/";
public static String similarityMatrixDirectory = baseDirectory
+"/similarityMatrix";
public static String tmpDirectory = baseDirectory +"/tmp";
public static String similarityClassname = "SIMILARITY_COSINE";
public static void main(String[] args) throws Exception {
Configuration config = new Configuration();
DatasetSplitter datasetSplitter = new DatasetSplitter();
String[] datasetSplitterArgs = {"--input",
inputSplitDirectory, "--output", outputSplitDirectory,
"--trainingPercentage", "0.9",
"--probePercentage","0.1"};
datasetSplitter.setConf(config);
datasetSplitter.run(datasetSplitterArgs);
//
ParallelALSFactorizationJob parallelALSFactorizationJob = new
ParallelALSFactorizationJob();
String[] parallelAlsFactorisationArgs = {"--input",
inputAlsDirectory, "--output", outputAlsDirectory, "--numFeatures", "20",
"--numIterations","10","--lambda","0.065"};
parallelALSFactorizationJob.setConf(config);
parallelALSFactorizationJob.run(parallelAlsFactorisationArgs);
// Mahout’s recommendfactorized function will compute
recommendations based on ratings matrix
RecommenderJob recommenderJob = new RecommenderJob();
String[] recommenderJobArgs = {"--input", recommenderJobInput,
"--output",recommenderJobOutput,"--userFeatures",
recommenderJobUserFeatures ,
"--itemFeatures",recommenderJobItemFeatures,
"--numRecommendations","6","--maxRating","5"};
recommenderJob.setConf(config);
recommenderJob.run(recommenderJobArgs);
}
}
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Thanks you very much
2015-03-19 15:53 GMT+01:00 Pat Ferrel <[email protected]>:
> I think that means you are using the wrong input to #4. It looks like you
> are trying to evaluate the ALS recs then create a recommender? To use ALS
> for recs try the CLI first see “mahout recommendfactorized” with input =
> (userid, itemid, rating). If #4 is to create all recs for all known users
> then it is independent of all other jobs since it takes interactions or
> ratings as input. not the output f a previous job.
>
>
> On Mar 19, 2015, at 7:11 AM, James Parker <[email protected]> wrote:
>
> Hi,
> I try ALS recommender using Java code. This is the steps i follow:
> 1. I create DatasetSplitter Job
>
> 2. I create ParallelALSFactorizationJob Job
>
> 3. I create FactorizationEvaluator Job
>
> 4. I create RecommenderJob Job
>
> Everything is ok for 1 to 3. But when i run the RecommenderJob , i have
> the following error
>
> -------------------------------------------------------------------------------------------------------------------------------
> java.lang.Exception: java.lang.RuntimeException:
> java.lang.ClassCastException:
> org.apache.mahout.math.map.OpenIntObjectHashMap cannot be cast to
> org.apache.mahout.common.Pair
> at
> org.apache.hadoop.mapred.LocalJobRunner$Job.run(LocalJobRunner.java:403)
> .
> .
>
> ----------------------------------------------------------------------------------------------------------------------------
>
>
> I'm suprised to see that if i run steps 1 to 3 using my java code, and the
> last step using the command line "mahout recommendfactorized ........"
> with the files generates with my java code steps 1 to 3), recommendations
> are correctly generated.
>
> Thanks you very much
>
> Evans
>
>
