def testRef():
e15 = ser.loadDict('dicts/15_wiki_full.pkl')
p15 = ser.loadDict('dicts/people_ed_15.pkl')
s = set()
for p in p15:
s.add(p['id'])
ed15 = filter(lambda x: x['id'] in s, e15)
ser.ed15 = ed15
getActualReferences(ed15, 'mention_references_15.html')
break
return res
def loadOccupations():
files = ['../data/occp/occupations', '../data/occp/nobility']
res = set()
for name in files:
f = open(name)
for line in f:
res.add(line[:-1].lower())
f.close()
return res
if __name__=='__main__':
dfd = ser.loadDict('dicts/15_dfd.pkl')
wts = ser.getWikiTitles()
#arts = ser.loadDict('dicts/newcontext_people_9_v2.pkl')
arts = ser.loadDict('dicts/newcontext_people_15_v1.pkl')
c = 0
for a in arts:
if 'candids' not in a:
break
c+=1
c = 37000
print 'starting at', c
matchPickledArticles(arts, wts, dfd, N15, 'dicts/newcontext3_people_15.txt', start=c)
print 'shelve hit', wiki.SHM.hit
print 'shelve miss', wiki.SHM.miss
def load_existing_names(self, name):
return ser.loadDict(name)
def get_wt():
dfd = ser.loadDict('_data/15_dfd.pkl')
occp = loadOccupations()
fn = lambda x, y: a_m.getTFScore(x,y,dfd,81697, a_m.getWTFIDFd, PARAMS, occp=occp)
return fn
def __init__(self, exist_brit_path, edition, wikidealer, pickle_path):
Dealer.__init__(self,exist_brit_path, edition)
self.wikidealer = wikidealer
self.arts = ser.loadDict(pickle_path)