PySpark 连接两个 RDD 结果为空 RDD

Question

我是 Spark 新手，正在尝试在我的数据集上编辑和应用此电影推荐教程 (https://databricks-training.s3.amazonaws.com/movie-recommendation-with-mllib.html)。但它一直抛出此错误：

ValueError: Can not reduce() empty RDD

这是计算模型的均方根误差的函数：

def computeRmse(model, data, n):
    """
    Compute RMSE (Root Mean Squared Error).
    """
    predictions = model.predictAll(data.map(lambda x: (x[0], x[1])))

    print predictions.count()
    print predictions.first()
    print "predictions above"

    print data.count()
    print data.first()
    print "validation data above"


    predictionsAndRatings = predictions.map(lambda x: ((x[0], x[1]), x[2])) \
#LINE56
   .join(data.map(lambda line: line.split(‘,’) ).map(lambda x: ((x[0], x[1]), x[2]))) \
  .values()    

   print predictionsAndRatings.count()
   print "predictions And Ratings above" 
#LINE63
return sqrt(predictionsAndRatings.map(lambda x: (x[0] - x[1]) **    2).reduce(add) / float(n))

模型 = ALS.train（训练、等级、numIter、lambda）。数据是验证数据集。训练和验证集最初来自 ratings.txt 文件，格式为：userID,productID,rating,ratingopID

这些是部分输出：

879
...
Rating(user=0, product=656, rating=4.122132631144641)
predictions above
...
1164
...
(u'640085', u'1590', u'5')
validation data above    
...
16/08/26 12:47:18 INFO DAGScheduler: Registering RDD 259 (join at     /path/myapp/MyappALS.py:56)
16/08/26 12:47:18 INFO DAGScheduler: Got job 20 (count at         /path/myapp/MyappALS.py:59) with 12 output partitions 
16/08/26 12:47:18 INFO DAGScheduler: Final stage: ResultStage 238 (count at /path/myapp/MyappALS.py:59)
16/08/26 12:47:18 INFO DAGScheduler: Parents of final stage:     List(ShuffleMapStage 237)
16/08/26 12:47:18 INFO DAGScheduler: Missing parents: List(ShuffleMapStage 237)
16/08/26 12:47:18 INFO DAGScheduler: Submitting ShuffleMapStage 237     (PairwiseRDD[259] at join at /path/myapp/MyappALS.py:56), which has no     missing parents    
....

0
predictions And Ratings above

...
Traceback (most recent call last):
File "/path/myapp/MyappALS.py", line 130, in <module>
validationRmse = computeRmse(model, validation, numValidation)
File "/path/myapp/MyappALS.py", line 63, in computeRmse
return sqrt(predictionsAndRatings.map(lambda x: (x[0] - x[1]) ** 2).reduce(add) / float(n))
File "/spark/python/lib/pyspark.zip/pyspark/rdd.py", line 805, in reduce
ValueError: Can not reduce() empty RDD

所以从 count() 我确定初始 RDD 不为空。比 INFO log Registering RDD 259 (join at /path/myapp/MyappALS.py:56) 是否意味着加入作业已启动？

我是不是遗漏了什么？谢谢。

Answer 1

当我将 int() 添加到 :

时，该错误消失了

predictionsAndRatings = predictions.map(lambda x: ((x[0], x[1]), x[2])) \ .join(data.map(lambda x: ((int(x[0]), int(x[1])), int(x[2])))) \ .values()

我们认为这是因为 pediction 是从给出 tupple 的方法 predictAll 输出的，但是其他数据是由算法手动解析的

PySpark 连接两个 RDD 结果为空 RDD

PySpark join two RDD results in an empty RDD

apache-spark

rdd

pyspark

apache-spark-mllib