BIGDL ON CDH
BigDL是一种基于Apache Spark的分布式深度学习框架
安装
下载源码
https://github.com/intel-analytics/BigDL
编译:
CDH集群安装了spark2.2,采用spark2.2编译。执行如下命令
- 安装maven(apache maven官网下载解压配置环境变量即可,此处省略)
- 进入/home/bigdl/BigDL-master,执行如下命令编译:
bash make-dist.sh -P spark_2.2
3,编译完成,执行spark命令测试,如下:
spark2-shell --properties-file /home/bigdl/BigDL-master/dist/conf/spark-bigdl.conf \
--jars /home/bigdl/BigDL-master/dist/lib/bigdl-0.5.0-SNAPSHOT-jar-with-dependencies.jar
scala>import com.intel.analytics.bigdl.utils.Engine
scala>Engine.init
scala> import com.intel.analytics.bigdl.tensor.Tensor
import com.intel.analytics.bigdl.tensor.Tensor
scala> Tensor[Double](2,2).fill(1.0)
res9: com.intel.analytics.bigdl.tensor.Tensor[Double] =
1.0 1.0
1.0 1.0
[com.intel.analytics.bigdl.tensor.DenseTensor of size 2x2]
官方文档:https://bigdl-project.github.io/master/#ScalaUserGuide/install-build-src/
下载官方编译的版本
https://bigdl-project.github.io/0.6.0/#release-download/
服务器解压安装
unzip dist-spark-2.2.0-scala-2.11.8-all-0.6.0-dist.zip -d bigdl
配置环境变量
export SPARK_HOME=/opt/cloudera/parcels/SPARK2/lib/spark2
export BIGDL_HOME=/root/bigdl/