I get a segfault on this line:

sum_w = numpy.dot(param[events.keys(),].T, events.values())

I'm not sure why. The values look good.

Hi. How did you build model.latin.20120315 model? What is the input corpus to build the model?
Thanks

Hi,

I was curious if there are any plans to port this to Java at some point or to bring the short text matching functionality to language-detector?

Thanks.

Hi
It is a great work. But I do hope you can share how to build the model for other languages which r not convered in your trained model. Thanks

Hello,

I would like to ask you how it is possible use ldig to detect language from short text. Is it possible to use it as a command line tool where I would load a model directory and some text dto detect?

Something like this:
$ /c/Python27/python ldig.py -m models/ldig.model.small/ldig.model.small/model.small/ -t "Dneska mi je krasne"
and the program will return cs

I am confused by using the data file - I need only detect languages for a list of text, I don't need to learn my own model or measure the accuracy by estimating label.

Thank you.