def __init__(self, d):
self.d = d
self.Data = [random.gauss(0, 1) for i in range(self.d)]
if True == gol.get_value('DEBUG'):
print('CosineLsh:\td: ' + str(self.d))
print('CosineLsh Data: ')
print(self.Data)
def resize(self, L):
if True == gol.get_value('DEBUG'):
print 'LshWrapper resize:\tnew L: ' + str(L) + '\tL: ' + str(self.L) + '\tk: ' + str(self.k)
# shrink the number of hash tables to be used
if L < self.L:
self.hash_tables = self.hash_tables[:L]
else:
# initialise a new hash table for each hash function
hash_funcs = [[self.__creat_ht__() for h in xrange(self.k)] for l in xrange(self.L, L)]
self.hash_tables.extend([(g, defaultdict(lambda:[])) for g in hash_funcs]) #g是一个hash表
if True == gol.get_value('DEBUG'):
print 'resize wrapper hashtable: '
for ht, ct in self.hash_tables:
print 'ht: ' + str(type(ht)) + '\tct: ' + str(type(ct))
print ht
print ct
self.L = L
def __init__(self, r, d): # hash family 各参数
self.r = r
self.d = d
self.b = random.uniform(0, self.r) # 0 < b < r //随机生成b,0 < b < r
self.Data = [random.gauss(0, 1) for i in xrange(self.d)] #高斯分布
if True == gol.get_value('DEBUG'):
print 'L2Lsh:\tr: ' + str(self.r) + '\td: ' + str(self.d) + '\tb: ' + str(self.b)
print 'L2Lsh Data: '
print self.Data
def __init__(self, lsh_type, d, r = 1.0, k = 2, L = 2):
self.type = lsh_type
self.d = d
self.r = r
self.k = k
self.L = 0
self.hash_tables = []
self.resize(L)
if True == gol.get_value('DEBUG'):
print 'LshWrapper init:\ttype: ' + str(self.type) + '\td: ' + str(self.d) + '\tr: ' + str(self.r) + '\tk: ' + str(self.k) + '\tL: ' + str(self.L)
def index(self, datas):
# index the supplied datas
self.datas = datas
for ht, ct in self.hash_tables:
for ix, p in enumerate(self.datas):
ct[self.hash(ht, p)].append(ix)
# reset stats
self.tot_touched = 0
self.num_queries = 0
if True == gol.get_value('DEBUG'):
print('index wrapper hashtable: ')
for ht, ct in self.hash_tables:
print('ht: ' + str(type(ht)) + '\tct: ' + str(type(ct)))
print(ht)
print(ct)