Meaning of output/training status of 256 in Stanford NLP NER?

Question

I have a Python program where I am using os.sys to train the Stanford NER from the command line. This returns an output/training status which I save in the variable "status", and it is usually 0. However, I just ran it and got an output of 256, as well as not creating a file for the trained model. This error is only occurring for larger sets of training data. I searched through the documentation on the Stanford NLP website and there doesn't seem to be info on the meanings of the outputs or why increasing training data might affect the training. Thanks in advance for any help and problem code is below.

cmdToSys = "java -mx20g -cp stanford-corenlp-4.2.2.jar edu.stanford.nlp.ie.crf.CRFClassifier -prop " + self.trainPropFileName + " -ner.useSUTime false test -ner.applyNumericClassifiers false test "

status = os.system(cmdToSys)

note: self.trainPropFileName is just the property file

Answer 1

Status is an exit code, and non-zero exit codes mean your program failed. This is not a Stanford NLP convention, it's how all programs work on Unix/Linux.

There should be an error somewhere, maybe you ran out of memory? You'll have to track that down to find out what's wrong.