Error in nltk.tag.accuracy

[Richard Parker]

Hi,

I was trying to implement the accuracy measure of NLTK tag,
print nltk.tag.accuracy(unigram_chunker, chunk_data)
it is giving me following error message,
Traceback (most recent call last):
  File "<pyshell#3>", line 1, in <module>
    print nltk.tag.accuracy(unigram_chunker, chunk_data)
AttributeError: 'module' object has no attribute 'accuracy'

I am using NLTK3.1
[>>> print('The nltk version is {}.'.format(nltk.__version__))
The nltk version is 3.1]
My Python version is Python 2.7.11 (v2.7.11:6d1b6a68f775, Dec  5 2015, 20:32:19) [MSC v.1500 32 bit (Intel)] on win32,
I am using Windows 10.

If any one may kindly suggest how may I fix this error.

Regards,
RP

[Steven Bird]

Please try:

from nltk.metrics import accuracy

And then just call:

print accuracy(unigram_chunker, chunk_data)

--
You received this message because you are subscribed to the Google Groups "nltk-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nltk-users+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Richard Parker]

Dear Sir,

The unigram chunker module is generating some error, so I could not test it. 

But if you wrote it must be correct. I will check and get back to you in few days.

Regards,

RP.

[Alexis]

I think Steven must have confused the APIs for nltk.chunk.accuracy( ) and nltk.metrics.accuracy( ). The former is called with a chunker as its first argument, the second is called with two sets of tagged data. 

>>> help(nltk.metrics.accuracy)

accuracy(reference, test)

    Given a list of reference values and a corresponding list of test values, ...

Use `help()` to figure out which one is suited to your needs, and how to call it.

Alexis

Dr. Alexis Dimitriadis | Assistant Professor and Senior Research Fellow | Utrecht Institute of Linguistics OTS | Utrecht University | Trans 10, 3512 JK Utrecht, room 2.33 | +31 30 253 65 68 | a.dimi...@uu.nl | www.hum.uu.nl/medewerkers/a.dimitriadis

[Jeff Silverman]

Hello,

I am still having problems.  I am a bit of a newbie.

(Pdb) from nltk.metrics import accuracy

(Pdb) print (nltk.tag.accuracy.__doc__)

*** AttributeError: 'module' object has no attribute 'accuracy'

(Pdb) dir(accuracy)

['__call__', '__class__', '__closure__', '__code__', '__defaults__', '__delattr__', '__dict__', '__doc__', '__format__', '__get__', '__getattribute__', '__globals__', '__hash__', '__init__', '__module__', '__name__', '__new__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__sizeof__', '__str__', '__subclasshook__', 'func_closure', 'func_code', 'func_defaults', 'func_dict', 'func_doc', 'func_globals', 'func_name']

(Pdb) 

(Pdb)  print('The nltk version is {}.'.format(nltk.__version__))

The nltk version is 3.2.1.

(Pdb) import sys

(Pdb) print sys.version

2.7.12 (default, Sep 29 2016, 13:30:34) 

[GCC 6.2.1 20160916 (Red Hat 6.2.1-2)]

(Pdb) 

I am getting my instructions from how to train a nltk-chunker

Thank you

Jeff