herrbuerger

from annoy import AnnoyIndex
index = AnnoyIndex(1)

for x in range(10000):
  id = "%d000%d" % (x,x)
  index.add_item(int(id),[x])

herrbuerger

48000693
49000693
50000693
51000693
52000693
53000693
54000693
55000693
56000693
57000693

herrbuerger

from math import sqrt

def confidence(ups, downs):
    n = ups + downs

    if n == 0:
        return 0

    z = 1.96 #1.96 = 95%
    phat = float(ups) / n

herrbuerger

➜  sample  imagedups . --search -f 92bc12786d430083be1e166a2174b4888ba0b4ba.jpg
92bc12786d430083be1e166a2174b4888ba0b4ba.jpg
Traceback (most recent call last):
  File "/usr/local/bin/imagedups", line 228, in <module>
    prog.main()
  File "/usr/local/bin/imagedups", line 60, in main
    self.cmd_search()
  File "/usr/local/bin/imagedups", line 90, in cmd_search
    self.compare_with_db(hashdb, self.args.samplefile)
  File "/usr/local/bin/imagedups", line 193, in compare_with_db