In [8]:

    

from __future__ import print_function
import sdm as sdmlib
from sdm import utils
import numpy as np
import matplotlib.mlab as mlab
from IPython.display import clear_output
import matplotlib.pyplot as plt


    

In [9]:

    

#bits, radius = 1000, 451
bits, radius = 256, 103
p = utils.calculate_probability(bits, radius)
#p = 0.00107185004892

sample = 1000000
scanner_type = sdmlib.SDM_SCANNER_OPENCL

address_space = sdmlib.AddressSpace.init_random(bits, sample)
address_space.opencl_init();


    

In [10]:

    

p


    

    
Out[10]:





0.0010668455863780898

In [11]:

    

counter = []


    

In [12]:

    

for i in range(10000):
    if i%100 == 0:
        clear_output(wait=True)
        print(i)
    bs = sdmlib.Bitstring.init_random(bits)
    result = address_space.scan_opencl2(bs, radius)
    counter.append(len(result))


    

    




9900

In [13]:

    

mu = sample*p
sigma = (sample*p*(1-p))**(0.5)
x = np.linspace(0, 2000, 2000)
y = mlab.normpdf(x, mu, sigma)
print(mu)
print(sigma)


    

    




1066.8455863780898
32.64517463382476

In [14]:

    

plt.figure(figsize=(8, 6), dpi=100)
plt.hist(counter, bins=50, density=True)
label = u'$\mu={:.2f}$, $\sigma={:.2f}$'.format(mu, sigma)
plt.plot(x, y, 'r', linewidth=2.0, label=label)
plt.xlim(mu-4.5*sigma, mu+4.5*sigma)
plt.legend()
plt.show()


    

In [ ]: