def estimateAverageDegree(top = 10000, filename = None):
steps = []
if filename:
steps = rw.readStepsFromFile(filename, top)
else:
steps = rw.readSteps()
deg = {k:v for k,v in degree.getDegree(withCount = True)}
fest = open(data.ESITMATE_FILE, 'w')
rd = 0.0
counter = 0.0
for tid in steps:
d = deg[tid]
if d == 0:
continue
counter += 1.0
rd += 1.0 / d
est = counter / rd
fest.write('%lf\n' % est)
fest.close()
@printRunningTime
def rse(steps, rankList):
eachSize = len(steps) / SAMPLE_TIME / SAMPLE_COUNT
outputs = []
for k in range(TOP):
f = open('sample-%d.txt' % (k + 1), 'w')
outputs.append(f)
for i in range(1, SAMPLE_TIME + 1):
size = eachSize * i
print 'counting:', size
for j in range(TOP):
freq = count(steps[j * size : (j + 1) * size])
ranks = getRank(freq, rankList)
for k, r in enumerate(ranks):
outputs[k].write(str(r))
if j != TOP - 1:
outputs[k].write(',')
for output in outputs:
output.write('\n')
for output in outputs:
output.close()
if __name__ == '__main__':
steps = rw.readSteps()
rank = rank.getRank()
rse(steps, rank[:TOP])
import rank
import rw
from collections import Counter
fout = open('rank-1.txt', 'w')
steps = rw.readSteps()
size = len(steps) / 10000
tid = str(rank.getRank(1)[0])
for i in range(10000):
print 'size', size * i
counter = Counter([str(s) for s in steps[i * size:(i + 1) * size]])
rank = 0
count = 0
for r, item in enumerate(counter.most_common()):
if item[0] == tid:
rank = r + 1
count = item[1]
print 'find'
break
fout.write('%d\t%d\n' % (rank, count))
fout.close()
