找到来填坑,安装的真费劲//发现只需要改一点点就可以了
环境
- 环境:windows10
- python :Anaconda
- spark:spark-2.4.4-bin-hadoop2.7.tgz
- hadoop:hadoop-2.7.0.tar
- scala:scala-2.11.12
添加kernel
- 路径:D:\code\anaconda\share\jupyter\kernels
- 参考python3,新建pyspark文件夹和kernel.json文件
- json文件配置如下
{
"argv": [
"D:\\code\\anaconda\\python.exe",
"-m",
"ipykernel_launcher",
"-f",
"{connection_file}"
],
"display_name": "PySpark3",
"language": "python",
"env": {
"SPARK_HOME": "D:\\code\\spark",
"PYTHONPATH": "D:\\code\\spark\\python;D:\\code\\spark\\python\\lib\\py4j-0.10.7-src.zip",
"PYTHONSTARTUP": "D:\\code\\spark\\python\\pyspark\\shell.py ",
"PYSPARK_SUBMIT_ARGS": "pyspark-shell"
}
}
坑
之前一直参考mac的配置方法,结果PYTHONPATH设置的有问题。mac里两个路径之间是冒号:,window是分号:
2019-12-06 于南京市江宁区九龙湖