jar - How to use Python API of Stanford NER? -
this question has answer here:
- how install , invoke stanford nertagger? 5 answers
version
- windows 7
- python 2.7.10
- nltk 3.1
- stanford ner 3.6 (2015-12-09)
issue
i trained custom ner model stanford ner , got serialized model. tried using model carry out ner on unseen corpus via python api provided nltk.
according documentation, should specify path model , path stanford-ner.jar
. however, need specify both path stanford-ner.jar
, path slf4j-api.jar
because stanford ner requires logging module.
i not figure out how specify 2 paths in nltk api. constructor takes 2 arguments first 1 path/to/model
, second 1 path/to/jar
. tried concatenating 2 jar paths, putting them in list , tuple none of methods worked.
how can tell nltk find both jars in order invoke prediction?
could provide code tried? looking @ source code nltk.tag.stanford module looks have initiate stanfordnertagger model , path stanford-ner jar, , logger added classpath automatically.
this __init__()
method stanfordtagger superclass right after setting model , engine. looks logger jar inside parent folder of stanford-ner jar path provide , adds path implicitly calling find_jars_within_path()
nltk.internals
, under hood appending folder os.path
Comments
Post a Comment