from panns import *

# create an index of Euclidean distance
p = PannsIndex('euclidean')

# generate a 1000 x 100 dataset
for i in xrange(1000):
   v = gaussian_vector(100)
   p.add_vector(v)

# build an index of 50 trees and save to a file
p.build(int(50))
#p.save('test.idx')
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-9-b53e159ae9fd> in <module>()
    10 
    11 # build an index of 50 trees and save to a file
---> 12 p.build(int(50))
    13 #p.save('test.idx')

/Users/*****/Library/Enthought/Canopy_64bit/User/lib/python2.7/site-packages/panns/index.pyc in build(self, c)
   132         """
   133         num_prj = int(2 ** (numpy.log2(len(self.mtx) / self.K) + 1))
--> 134         self.prj = self.random_directions(num_prj)
   135         if self.parallel:
   136             self.mmap_core_data()

/Users/******/Library/Enthought/Canopy_64bit/User/lib/python2.7/site-packages/panns/index.pyc in random_directions(self, c)
   258         c: the number of principle components needed.
   259         """
--> 260         return [ gaussian_vector(self.dim, True, self.typ) for _ in xrange(c) ]
   261 
   262

/Users/******/Library/Enthought/Canopy_64bit/User/lib/python2.7/site-packages/panns/utils.pyc in gaussian_vector(size, normalize, dtype)
   112     normalize: the vector length is normalized to 1 if True.
   113     """
--> 114     v = numpy.random.normal(0,1,size)
   115     if normalize:
   116         v = v / linalg.norm(v)

/Users/****/Library/Enthought/Canopy_64bit/User/lib/python2.7/site-packages/numpy/random/mtrand.so in mtrand.RandomState.normal (numpy/random/mtrand/mtrand.c:12052)()

/Users/****/Library/Enthought/Canopy_64bit/User/lib/python2.7/site-packages/numpy/random/mtrand.so in mtrand.cont2_array_sc (numpy/random/mtrand/mtrand.c:2757)()

TypeError: an integer is required