2

I need to use sparknlp to do lemmatization in python, i want to use the pretrained pipeline, however need to do it offline. what is the correct way to do this? i am not able to find any python example.

I am passing token as the inputcol for lemmatization and lemma as the outputcol. Following is my code:

documentAssembler = DocumentAssembler().setInputCol("Transcript").setOutputCol("document")
tokenizer = Tokenizer().setInputCols(['document']).setOutputCol('token')
lemmatizer = LemmatizerModel().load("xx").setInputCols(["token"]).setOutputCol("lemma")

error message:

Py4JJavaError: An error occurred while calling None.com.johnsnowlabs.nlp.annotators.LemmatizerModel.
: java.lang.NoClassDefFoundError: Could not initialize class com.johnsnowlabs.util.ConfigHelper$
    at com.johnsnowlabs.nlp.serialization.Feature.<init>(Feature.scala:22)
    at com.johnsnowlabs.nlp.serialization.MapFeature.<init>(Feature.scala:145)
    at com.johnsnowlabs.nlp.annotators.LemmatizerModel.<init>(LemmatizerModel.scala:49)
    at sun.reflect.NativeConstructorAccessorImpl.newInstance0(Native Method)
    at sun.reflect.NativeConstructorAccessorImpl.newInstance(NativeConstructorAccessorImpl.java:62)
    at sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
    at java.lang.reflect.Constructor.newInstance(Constructor.java:423)
    at py4j.reflection.MethodInvoker.invoke(MethodInvoker.java:247)
    at py4j.reflection.ReflectionEngine.invoke(ReflectionEngine.java:357)
    at py4j.Gateway.invoke(Gateway.java:238)
    at py4j.commands.ConstructorCommand.invokeConstructor(ConstructorCommand.java:80)
    at py4j.commands.ConstructorCommand.execute(ConstructorCommand.java:69)
    at py4j.GatewayConnection.run(GatewayConnection.java:238)
    at java.lang.Thread.run(Thread.java:748)
Fiona
  • 85
  • 1
  • 7

1 Answers1

0

Not sure why you're getting this specific error (how did you initialize Spark?), but this is another issue.

To use the lemmatizer offline you first need to download a pretrained one, for example from here, and unzip it. Then change your code to

lemmatizer = LemmatizerModel().load("/path/to/unzipped/model").setInputCols...
shay__
  • 3,815
  • 17
  • 34