nreimers / truecaser Goto Github PK

Hi, thanks for making this repo. I want to ask whether a CRF model is better than this n-gram model. Becuase I saw that stanford nlp implemented a CRF truecaser.

Hi,

First of all, thank you for sharing your work with the community!

Can you provide details (maybe a paper) that can be used to cite your work in another scientific work? I am working on my master thesis and would like to attribute the credit to your work. :)

Cheers,
Rosko

Thanks for this! We have found it useful, and would (potentially) like to cite it. When discussing it in our paper, it would be helpful to know what data the included model was trained on. Could you say what data you used?

There is probably a bug in TrainFunctions.py.
In loop that creates trigrams missed assignment to word variable (word = sentence[tokenIdx]). So this loop uses the last value from previous loop.

its very slow , how to improve the performance of this application