spark dataframe 类型转换

读一张表，对其进行二值化特征转换。可以二值化要求输入类型必须double类型，类型怎么转换呢？

直接利用spark column 就可以进行转换：

DataFrame dataset = hive.sql("select age,sex,race from hive_race_sex_bucktizer ");

/**

* 类型转换

dataset = dataset.select(dataset.col("age").cast(DoubleType).as("age"),dataset.col("sex"),dataset.col("race"));

是不是很简单。想起之前的类型转换做法，遍历并创建另外一个满足类型要求的RDD，然后根据RDD创建Datafame，好复杂！！！！

JavaRDD<Row> parseDataset =   dataset.toJavaRDD().map(new Function<Row,Row>() {
 
    @Override
    public Row call(Row row) throws Exception {
        System.out.println(row);
        long age = row.getLong(row.fieldIndex("age"));
        String sex = row.getAs("sex");
        String race =row.getAs("race");
        double raceV  = -1;
        if("white".equalsIgnoreCase(race)){
            raceV = 1;
        } else if("black".equalsIgnoreCase(race)) {
            raceV = 2;
        } else if("yellow".equalsIgnoreCase(race)) {
            raceV = 3;
        } else if("Asian-Pac-Islander".equalsIgnoreCase(race)) {
            raceV = 4;
        }else if("Amer-Indian-Eskimo".equalsIgnoreCase(race)) {
            raceV = 3;
        }else {
            raceV = 0;
        }
         
        return RowFactory.create(age,("male".equalsIgnoreCase(sex)?1:0),raceV);
    }
});
 
StructType schema = new StructType(new StructField[]{
         createStructField("_age", LongType, false),
          createStructField("_sex", IntegerType, false),
          createStructField("_race", DoubleType, false)
        });
 
DataFrame  df  =  hive.createDataFrame(parseDataset, schema);