Follow the instructions of nebula-java to build and install nebula-java client.
${SPARK_HOME}/bin/spark-submit --class com.epik.kbgateway.DomainKnowledge2LogFile --master yarn --deploy-mode cluster --driver-memory 256M --driver-java-options "-Dspark.testing.memory=536870912" --executor-memory 6g --num-executors 4 --executor-cores 2 /root/epik-logfile-encoder-job.jar -f /cn_dbpedia_input/baike_triples.txt -d cn_dbpedia -t /epik_log_output