Question

I’m currently parsing Arabic text using the following command:

java -mx1500m edu.stanford.nlp.parser.lexparser.LexicalizedParser \
  -cp "$scriptdir/*:" -sentences newline -outputFormat "penn,wordsAndTags" \
  edu/stanford/nlp/models/lexparser/arabicFactored.ser.gz $FILE

This outputs tokens, tags, and parse trees, plus a global overview on the processing speed similar to this output:

Parsed 280 words in 10 sentences (1.95 wds/sec; 0.07 sents/sec).

Is there an option or command line switch to trigger output of processing time on a per-sentence level? Or would I have to add that to the code myself?

Haven’t found such an option neither in the FAQ nor on the Stanford NLP website, so I’d appreciate if someone could clarify on this issue.

Answer 1

No, at present there isn’t. You’d need to add the code yourself.