jollyday | Jollyday - A holiday API

This one is the source of your trouble:

I downloaded the distribution folder from here: http://stanfordnlp.github.io/CoreNLP/download.html

I unzipped it and cd'd into stanford-corenlp-full-2016-10-31

I entered this command:

You need to make sure you use the proper dependencies with the proper version.

If you use Stanford CoreNLP 3.7.0, make sure you also have the latest lib and liblocal folders.

I believe this error is because you have an incompatible dependency jar somewhere.

Update: This is an error due to Java 9. Add this flag

I'm not sure, what's going on. What version of CoreNLP are you using? Working at the command-line, I couldn't reproduce this problem. See below. So, my first two questions are: Are you using the latest CoreNLP? (If not, try upgrading.) Does this work for you at the command-line? (In which case, it's only the use in IntelliJ that is problematic - not sure what would be going on there but something to do with how reflection and classloaders work?)

You are creating a new thread at each loop, the number of threads exceed the number of your CPU cores. Since the CPU tries to be fair to each instance having say 150000 threads means that each thread only executes for sometime before being switched to the background for new thread.

Try instead creating 2 threads or 8 threads which would yield much better results.

With the current code base if you have the slf4j-api.jar in your CLASSPATH and run the 3.7.0 segmenter you will get this error. I'm going to push a code change to fix this but for the time being if you remove the slf4j-api.jar from the CLASSPATH this error should go away.

The Stanford NER classifier is a java program. The NLTK's module is only an interface to the java executable. So you train a model exactly as you did before (or as you saw done in the link you provide).

In your code, you are confusing the training of a model with its use to chunk new text. The .prop file contains instructions for training a new model; it is not itself a model. This is what I recommend:

1. Forget about python/nltk for the moment, and train a new model from the Windows command line (CMD prompt or whatever): Follow the how-to you mention in your question, to generate a serialized model (.ser file) named ner-model.ser.gz or whatever you decide to call it from your .prop file.

2. In your python code, set the path_to_model variable to point to the .ser file you generated in step 1.

If you really want to control the training process from python, you could use the subprocess module to issue the appropriate command line commands. But it sounds like you don't really need this; just try to understand what these steps do so that you can carry them out properly.