IllegalArgumentException: u'requirement failed' on kmeans.fit

Question

使用 zeppelin 笔记本中的 spark，我从昨天开始收到此错误。 这是我的代码：

from pyspark.ml.clustering import KMeans
from pyspark.ml.feature import VectorAssembler

df = sqlContext.table("rfmdata_clust")

k = 4

# Set Kmeans input/output columns
vecAssembler = VectorAssembler(inputCols=["v1_clust", "v2_clust", "v3_clust"], outputCol="features")
featuresDf = vecAssembler.transform(df)

# Run KMeans
kmeans = KMeans().setInitMode("k-means||").setK(k)
model = kmeans.fit(featuresDf)
resultDf = model.transform(featuresDf)

# KMeans WSSSE
wssse = model.computeCost(featuresDf)
print("Within Set Sum of Squared Errors = " + str(wssse))

这是错误：

Traceback (most recent call last):
  File "/tmp/zeppelin_pyspark-8890997346928959256.py", line 346, in <module>
    raise Exception(traceback.format_exc())
Exception: Traceback (most recent call last):
  File "/tmp/zeppelin_pyspark-8890997346928959256.py", line 334, in <module>
    exec(code)
  File "<stdin>", line 8, in <module>
  File "/usr/lib/spark/python/pyspark/ml/base.py", line 64, in fit
    return self._fit(dataset)
  File "/usr/lib/spark/python/pyspark/ml/wrapper.py", line 236, in _fit
    java_model = self._fit_java(dataset)
  File "/usr/lib/spark/python/pyspark/ml/wrapper.py", line 233, in _fit_java
    return self._java_obj.fit(dataset._jdf)
  File "/usr/lib/spark/python/lib/py4j-0.10.4-src.zip/py4j/java_gateway.py", line 1133, in __call__
    answer, self.gateway_client, self.target_id, self.name)
  File "/usr/lib/spark/python/pyspark/sql/utils.py", line 79, in deco
    raise IllegalArgumentException(s.split(': ', 1)[1], stackTrace)
IllegalArgumentException: u'requirement failed'

引发错误的行是 kmeans.fit() 行。 我检查了 rfmdata_clust 数据框，它似乎一点也不奇怪。

df.printSchema() 给出：

root
 |-- id: string (nullable = true)
 |-- v1_clust: double (nullable = true)
 |-- v2_clust: double (nullable = true)
 |-- v3_clust: double (nullable = true)

featuresDf.printSchema() 给出：

root
 |-- id: string (nullable = true)
 |-- v1_clust: double (nullable = true)
 |-- v2_clust: double (nullable = true)
 |-- v3_clust: double (nullable = true)
 |-- features: vector (nullable = true)

另一个有趣的点是在 featuresDf 的定义下面添加featuresDf = featuresDf.limit(10000) 可以使代码无错误地运行。可能和数据的大小有关？

Answer 1

希望这个问题已经解决了，如果没有，请试试这个

    df=df.na.fill(1)

这会将所有使用 NaN 的值填充为 1，当然您可以选择任何其他值。 该错误是由于特征向量中有 NaN 的事实。 您可能需要导入必要的包。 This 也应该有帮助。

如果失败请告诉我。