Spark、ADAM 和 Zeppelin

Question

正在尝试使用 ADAM 和 Zeppelin 进行基因组分析。我不确定我这样做是否正确，但运行进入以下问题。

%dep
z.reset()
z.addRepo("Spark Packages Repo").url("http://dl.bintray.com/spark-packages/maven")
z.load("com.databricks:spark-csv_2.10:1.2.0")   
z.load("mysql:mysql-connector-java:5.1.35")  
z.load("org.bdgenomics.adam:adam-core_2.10:0.20.0")
z.load("org.bdgenomics.adam:adam-cli_2.10:0.20.0")
z.load("org.bdgenomics.adam:adam-apis_2.10:0.20.0")

%spark

import org.bdgenomics.adam.rdd.ADAMContext._
import org.bdgenomics.adam.rdd.ADAMContext
import org.bdgenomics.adam.projections.{ AlignmentRecordField, Projection }
import org.apache.spark.SparkContext
import org.apache.spark.SparkConf
import org.bdgenomics.adam.rdd.ADAMContext
import org.bdgenomics.adam.rdd.ADAMContext._
import org.bdgenomics.adam.projections.Projection
import org.bdgenomics.adam.projections.AlignmentRecordField
import scala.io.Source
import org.apache.spark.rdd.RDD
import org.bdgenomics.formats.avro.Genotype
import scala.collection.JavaConverters._
import org.bdgenomics.formats.avro._
import org.apache.spark.SparkContext._
import org.apache.spark.mllib.linalg.{ Vector => MLVector, Vectors }
import org.apache.spark.mllib.clustering.{ KMeans, KMeansModel }

val ac = new ADAMContext(sc)

我得到以下错误输出

import org.bdgenomics.adam.rdd.ADAMContext._
import org.bdgenomics.adam.rdd.ADAMContext
import org.bdgenomics.adam.projections.{AlignmentRecordField, Projection}
import org.apache.spark.SparkContext
import org.apache.spark.SparkConf
import org.bdgenomics.adam.rdd.ADAMContext
import org.bdgenomics.adam.rdd.ADAMContext._
import org.bdgenomics.adam.projections.Projection
import org.bdgenomics.adam.projections.AlignmentRecordField
import scala.io.Source
import org.apache.spark.rdd.RDD
import org.bdgenomics.formats.avro.Genotype
import scala.collection.JavaConverters._
import org.bdgenomics.formats.avro._
import org.apache.spark.SparkContext._
import org.apache.spark.mllib.linalg.{Vector=>MLVector, Vectors}
import org.apache.spark.mllib.clustering.{KMeans, KMeansModel}
res7: org.apache.spark.SparkContext = org.apache.spark.SparkContext@62ec8142
<console>:188: error: constructor ADAMContext in class ADAMContext cannot be accessed in class $iwC
              new ADAMContext(sc)

知道去哪里找吗？我是否缺少任何依赖项？ ^

Answer 1

根据您使用的版本中的文件 ADAMContext.scala。构造函数是私有的。

class ADAMContext private (@transient val sc: SparkContext) 
    extends Serializable with Logging {
    ...
}

您可以这样使用。

import org.bdgenomics.adam.rdd.ADAMContext._

val adamContext: ADAMContext = z.sc

它将使用对象 ADAMContext 中的隐式转换

object ADAMContext {
    implicit def sparkContextToADAMContext(sc: SparkContext): ADAMContext = 
        new ADAMContext(sc)
}

Answer 2

它在不使用 Z 参考的情况下工作！！

val ac:ADAMContext  = sc
val genotypes: RDD[Genotype] = ac.loadGenotypes("/tmp/ADAM2").rdd

输出

ac: org.bdgenomics.adam.rdd.ADAMContext = org.bdgenomics.adam.rdd.ADAMContext@2c60ef7e

genotypes: 
org.apache.spark.rdd.RDD[org.bdgenomics.formats.avro.Genotype] = MapPartitionsRDD[3] at map at ADAMContext.scala:207

我曾尝试在 adam-shell 提示符下执行此操作，我不记得必须使用隐式转换。不过它使用的是 0.19 版本的 ADAM。

Spark、ADAM 和 Zeppelin

Spark, ADAM and Zeppelin

apache-spark

apache-zeppelin