def compute(self,
c=0.15,
epsilon=1e-6,
max_iters=100,
handles_deadend=True):
'''
Compute the PageRank score vector (global ranking)
inputs
c : float
restart probability
epsilon : float
error tolerance for power iteration
max_iters : int
maximum number of iterations for power iteration
handles_deadend : bool
if true, it will handle the deadend issue in power iteration
otherwise, it won't, i.e., no guarantee for sum of RWR scores
to be 1 in directed graphs
outputs
r : ndarray
PageRank score vector
'''
self.normalize()
q = np.ones((self.n, 1))
q = q / self.n
r, residuals = iterator.iterate(self.nAT, q, c, epsilon, max_iters,
handles_deadend)
return r
def compute(self,
seeds,
c=0.15,
epsilon=1e-9,
max_iters=100,
handles_deadend=True):
'''
Compute the PPR score vector w.r.t. the seed node
inputs
seeds : list
list of seeds
c : float
restart probability
epsilon : float
error tolerance for power iteration
max_iters : int
maximum number of iterations for power iteration
handles_deadend : bool
if true, it will handle the deadend issue in power iteration
otherwise, it won't, i.e., no guarantee for sum of RWR scores
to be 1 in directed graphs
outputs
r : ndarray
PPR score vector
'''
self.normalize()
seeds = [seed - self.base for seed in seeds]
if len(seeds) is 0:
raise ValueError('Seeds are empty')
if min(seeds) < 0 or max(seeds) >= self.n:
raise ValueError('Out of range of seed node id')
# q = np.zeros((self.n, 1))
q = np.zeros(self.n)
q[seeds] = 1.0 / len(seeds)
r, residuals = iterator.iterate(self.nAT, q, c, epsilon, max_iters,
handles_deadend)
return r
def compute(self,
seed,
c=0.15,
epsilon=1e-6,
max_iters=100,
handles_deadend=True):
'''
Compute the RWR score vector w.r.t. the seed node
inputs
seed : int
seed (query) node id
c : float
restart probability
epsilon : float
error tolerance for power iteration
max_iters : int
maximum number of iterations for power iteration
handles_deadend : bool
if true, it will handle the deadend issue in power iteration
otherwise, it won't, i.e., no guarantee for sum of RWR scores
to be 1 in directed graphs
outputs
r : ndarray
RWR score vector
'''
self.normalize()
# adjust range of seed node id
seed = seed - self.base
# q = np.zeros((self.n, 1))
q = np.zeros(self.n)
if seed < 0 or seed >= self.n:
raise ValueError('Out of range of seed node id')
q[seed] = 1.0
r, residuals = iterator.iterate(self.nAT, q, c, epsilon, max_iters,
handles_deadend)
return r