I am processing a NYT portion of English Gigaword Corpus with Stanford CoreNLP. While it is still ongoing, the following message is logged several times:
Unknown variable: WEEKDAY
Each time this message is logged, memory consumption increases. Now about 23.8 GB. Does anyone know what this is about?
I am using Stanford CoreNLP 3.6.0 from Github with a commit identifier 4fd28dc4848616e568a2dd6eeb09b9769d1e3f4eand the following models stanford-english-corenlp-2016-01-10-models. My pipeline is as follows: "annotators", "tokenize, ssplit, pos, lemma, ner, parse, depparse, mention, coref".
I already know this question. But no one answered.
source
share