emorynlp / nlp4j-old

NLP tools developed by Emory University.
Other
60 stars 19 forks source link

only POS tagger #17

Closed DeepInEvil closed 8 years ago

DeepInEvil commented 8 years ago

Hi,

the command line tagging takes a lot of time,not suitable for real-time systems. Please provide API for single taggers like only for POS.

jdchoi77 commented 8 years ago

What is your input format? I believe NLPDecoder can be used for this but if you let me know the format, I can provide more fine-grained API.

DeepInEvil commented 8 years ago

we are calling the decoder with a text file, it is taking around 2 mins 30 seconds to process

On Wed, Jun 8, 2016 at 11:27 PM, Jinho D. Choi notifications@github.com wrote:

What is your input format? I believe NLPDecoder https://github.com/emorynlp/nlp4j/blob/master/src/main/java/edu/emory/mathcs/nlp/decode/NLPDecoder.java can be used for this but if you let me know the format, I can provide more fine-grained API.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/emorynlp/nlp4j/issues/17#issuecomment-224675396, or mute the thread https://github.com/notifications/unsubscribe/ANLoeue3kt7jwNdPnnS7T8Rc1aZyRF1mks5qJwKfgaJpZM4Iw9tY .

jdchoi77 commented 8 years ago

I think most of the time is perhaps spent for model loading. Here is the API you can use for this purpose:

https://github.com/emorynlp/nlp4j-demo/blob/master/src/main/java/edu/emory/mathcs/nlp/demo/NLPDecodeRaw.java

Here is the configuration file you can use just for pos tagging: https://github.com/emorynlp/nlp4j-demo/blob/master/src/main/resources/configuration/config-decode-en-pos.xml

The following line loads all the models, which should be done only once:

NLPDecoder decoder = new NLPDecoder(IOUtils.getInputStream(configurationFile));

Thanks.

DeepInEvil commented 8 years ago

Thanks a lot for the reply, I will check and let you know if we have further issues

Regards, Debanjan

On Wed, Jun 8, 2016 at 11:55 PM, Jinho D. Choi notifications@github.com wrote:

I think most of the time is perhaps spent for model loading. Here is the API you can use for this purpose:

https://github.com/emorynlp/nlp4j-demo/blob/master/src/main/java/edu/emory/mathcs/nlp/demo/NLPDecodeRaw.java

Here is the configuration file you can use just for pos tagging:

https://github.com/emorynlp/nlp4j-demo/blob/master/src/main/resources/configuration/config-decode-en-pos.xml

The following line loads all the models, which should be done only once:

NLPDecoder decoder = new NLPDecoder(IOUtils.getInputStream(configurationFile));

Thanks.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/emorynlp/nlp4j/issues/17#issuecomment-224683466, or mute the thread https://github.com/notifications/unsubscribe/ANLoevpQfdbT1GpK4ZmvqqbetnH3bMz1ks5qJwkVgaJpZM4Iw9tY .