MLlib MatrixFactorizationModel recommendProducts(u

2019-06-23 18:54发布

I trained a MatrixFactorizationModel model using ALS.train() and now using model.recommendProducts(user, num) to get the top recommended products, but the code fails on some users with the following error:

  user_products = model.call("recommendProducts", user, prodNum)
  File "/usr/lib/spark/python/pyspark/mllib/common.py", line 136, in call
    return callJavaFunc(self._sc, getattr(self._java_model, name), *a)
  File "/usr/lib/spark/python/pyspark/mllib/common.py", line 113, in callJavaFunc
    return _java2py(sc, func(*args))
  File "/usr/lib/spark/python/lib/py4j-0.8.2.1-src.zip/py4j/java_gateway.py", line 538, in __call__
  File "/usr/lib/spark/python/lib/py4j-0.8.2.1-src.zip/py4j/protocol.py", line 300, in get_return_value
py4j.protocol.Py4JJavaError: An error occurred while calling o68.recommendProducts.
: java.util.NoSuchElementException: next on empty iterator
    at scala.collection.Iterator$$anon$2.next(Iterator.scala:39)
    at scala.collection.Iterator$$anon$2.next(Iterator.scala:37)
    at scala.collection.IndexedSeqLike$Elements.next(IndexedSeqLike.scala:64)
    at scala.collection.IterableLike$class.head(IterableLike.scala:91)
    at scala.collection.mutable.WrappedArray.scala$collection$IndexedSeqOptimized$$super$head(WrappedArray.scala:34)
    at scala.collection.IndexedSeqOptimized$class.head(IndexedSeqOptimized.scala:120)
    at scala.collection.mutable.WrappedArray.head(WrappedArray.scala:34)
    at org.apache.spark.mllib.recommendation.MatrixFactorizationModel.recommendProducts(MatrixFactorizationModel.scala:117)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:606)
    at py4j.reflection.MethodInvoker.invoke(MethodInvoker.java:231)
    at py4j.reflection.ReflectionEngine.invoke(ReflectionEngine.java:379)
    at py4j.Gateway.invoke(Gateway.java:259)
    at py4j.commands.AbstractCommand.invokeMethod(AbstractCommand.java:133)
    at py4j.commands.CallCommand.execute(CallCommand.java:79)
    at py4j.GatewayConnection.run(GatewayConnection.java:207)
    at java.lang.Thread.run(Thread.java:745)

As you can see in first line above, I am running

user_products = model.call("recommendProducts", user, prodNum)

instead of

user_products = model.recommendProducts(user, prodNum)

because the latter is not implemented in 1.3.0 pyspark which I am using. Anyhow, it correctly returns prediction for some users, but then it fails on others.

I understand that it probably does not have the exact number of predictions I am requesting, I would expect it would return fewer.

1条回答
等我变得足够好
2楼-- · 2019-06-23 19:33

Short answer:

  • You have trained model on ratings where userID was in range [0;N]
  • You have asked for a recommendation for a userID=N+x, where x is a positive integer number. This caused the exception.

The rest of the answer:

  • You are allowed to ask for recommendations within the userID space used during the training phase
  • You are allowed to ask for as many recommendations as the product space used during the training phase

When you want to make recommendation for a new user you have to add information about his or her taste to a training dataset.

You can find examples covering Collaborative Filtering in Spark Summit 2014 Hands-on Exercises.

(If I missed anything please correct me in comment below)

查看更多
登录 后发表回答