def test_all_pairs_equality(self):
'''Test if parallel and sequential version of all_pairs deliver equal results'''
p = all_pairs(self.catalogue).tolist()
pp = all_pairs(self.catalogue, parallel=True).tolist()
np.testing.assert_array_equal(p, pp)
def test_all_pairs_unicode(self):
'''Test all_pairs with strange unicode characters'''
all_pairs(['ae', u'\xc3\xa4', 'ss', u'\xc3\x9f'])
def test_all_pairs_sequential(self):
'''Test if sequential version of all_pairs works'''
p = all_pairs(self.catalogue)
np.testing.assert_array_equal(p, self.all_pairs)
def test_all_pairs_parallel(self):
'''Test if parallel version of all_pairs works'''
p = all_pairs(self.catalogue, parallel=True)
np.testing.assert_array_equal(p, self.all_pairs)
("http://www.gutenberg.org/cache/epub/5323/pg5323.txt", "Effi Bri."),
("http://www.gutenberg.org/files/26686/26686-0.txt", "Birnbaum"),
("http://www.gutenberg.org/files/21593/21593-0.txt", "Urteil"),
("http://www.gutenberg.org/cache/epub/22367/pg22367.txt", "Verwand.")]
# get it from the interwebs
catalogue = []
for url, name in urls:
headers = {'User-Agent': 'Mozilla/5.0'}
req = urllib2.Request(url, None, headers)
catalogue.append(urllib2.urlopen(req).read())
# calc similarity matrix
M = all_pairs(catalogue,
distance=distances.jaccard,
dist_kwargs=dict(mode=1),
parallel=True)
# plot similarity matrix
pylab.figure(1)
pylab.title("similarity matrix")
pylab.imshow(M, aspect='auto', interpolation="nearest", cmap="Reds")
pylab.colorbar()
# plot complete linkage
pylab.figure(2)
pylab.title("complete linkage clustering")
hcluster.dendrogram(cluster(M, method='complete'), leaf_label_func=lambda i: urls[i][1])
# finally show
pylab.show()
def calc():
log.debug('Calculating similarity matrix for key %s...', self.key)
return all_pairs([s.source for s in self.submissions])
def test_all_pairs_empty(self):
'''Test all_pairs with empty input'''
all_pairs(['Something', 'nothing', ''])
}
} // end of class HelloWorld
""", """
public class HelloWorld {
public static void main(String[] args) {
System.out.println("Hello World!");
}
} // end of class HelloWorld
""", """
public class HelloUniverse {
public static void main(String[] args) {
String message = "Hello World!";
System.out.println(message);
}
}
"""
]
print all_pairs(catalogue)
print all_pairs(catalogue, parallel=True)
print all_pairs(catalogue, distance=distances.kolmogorov)
# Example code for loading a pickle file of submissions
# d = pickle.load(open('submissions.pkl'))
#
# for a in (a for a in d if a['submissions']):
# catalogue = [s['source'] for s in a['submissions'] if s['source']]
# print all_pairs(catalogue, parallel=True)
def test_all_pairs_equality(self):
'''Test if parallel and sequential version of all_pairs deliver equal results'''
p = all_pairs(self.catalogue).tolist()
pp = all_pairs(self.catalogue, parallel=True).tolist()
np.testing.assert_array_equal(p, pp)
def test_all_pairs_unicode(self):
'''Test all_pairs with strange unicode characters'''
all_pairs(['ae', u'\xc3\xa4', 'ss', u'\xc3\x9f'])
def test_all_pairs_parallel(self):
'''Test if parallel version of all_pairs works'''
p = all_pairs(self.catalogue, parallel=True)
np.testing.assert_array_equal(p, self.all_pairs)
def test_all_pairs_sequential(self):
'''Test if sequential version of all_pairs works'''
p = all_pairs(self.catalogue)
np.testing.assert_array_equal(p, self.all_pairs)
System.out.println("Hello World!");
}
} // end of class HelloWorld
""", """
public class HelloWorld {
public static void main(String[] args) {
System.out.println("Hello World!");
}
} // end of class HelloWorld
""", """
public class HelloUniverse {
public static void main(String[] args) {
String message = "Hello World!";
System.out.println(message);
}
}
"""]
print(all_pairs(catalogue))
print(all_pairs(catalogue, parallel=True))
print(all_pairs(catalogue, distance=distances.kolmogorov))
# Example code for loading a pickle file of submissions
# d = pickle.load(open('submissions.pkl'))
#
# for a in (a for a in d if a['submissions']):
# catalogue = [s['source'] for s in a['submissions'] if s['source']]
# print all_pairs(catalogue, parallel=True)
def test_all_pairs_empty(self):
'''Test all_pairs with empty input'''
all_pairs(['Something', 'nothing', ''])
def test_all_pairs_kwargs(self):
'''Test kwargs acceptance of all_pairs'''
all_pairs(self.catalogue, distance=distances.jaccard, dist_kwargs=dict(mode=2))
all_pairs(self.catalogue, dist_kwargs=dict(jaccard_mode=2))
def test_all_pairs_kwargs(self):
'''Test kwargs acceptance of all_pairs'''
all_pairs(self.catalogue,
distance=distances.jaccard,
dist_kwargs=dict(mode=2))
all_pairs(self.catalogue, dist_kwargs=dict(jaccard_mode=2))
("http://www.gutenberg.org/cache/epub/5323/pg5323.txt", "Effi Bri."),
("http://www.gutenberg.org/files/26686/26686-0.txt", "Birnbaum"),
("http://www.gutenberg.org/files/21593/21593-0.txt", "Urteil"),
("http://www.gutenberg.org/cache/epub/22367/pg22367.txt", "Verwand.")]
# get it from the interwebs
catalogue = []
for url, name in urls:
headers = {'User-Agent': 'Mozilla/5.0'}
req = urllib2.Request(url, None, headers)
catalogue.append(urllib2.urlopen(req).read())
# calc similarity matrix
M = all_pairs(catalogue,
distance=distances.jaccard,
dist_kwargs=dict(mode=1),
parallel=True)
# plot similarity matrix
pylab.figure(1)
pylab.title("similarity matrix")
pylab.imshow(M, aspect='auto', interpolation="nearest", cmap="Reds")
pylab.colorbar()
# plot complete linkage
pylab.figure(2)
pylab.title("complete linkage clustering")
hcluster.dendrogram(cluster(M, method='complete'),
leaf_label_func=lambda i: urls[i][1])
# finally show
