Zaki's test wrong

[valtheval]

Hello,
It looks like spade does not pass the "zaki's test" : if I use Zaki's text file example I end up with the following :

from pycspade.helpers import spade, print_result
from time import time
result = spade("zaki.txt", support=0.5, parse=True)
print_result(result)

gives :

   Occurs     Accum   Support    Confid      Lift              Sequence
        2         2 0.5000000 0.5000000 0.5000000            (2)->(1)
        2         2 0.5000000 0.5000000 0.5000000            (6)->(1)
        4         4 1.0000000       N/A       N/A                    (6)
        4         8 1.0000000       N/A       N/A                    (1)
        4         4 1.0000000       N/A       N/A                    (2)

Many sequences are missing :
(4), (1 2), (1 6), (2 6), (4)-(1), (4)-(2), (4)-(6), (1 2 6), (2 6)-(1), (4)-(2 6), (4)-(2)-(1), (4)-(6)-(1), (4)-(2 6)-(1)

Any suggestion ?

[roggenkamps]

I get the same results, but I have not looked into it yet as I just got it working this afternoon.

R has a similar package, arulesSequences that is also a wrapper around Zaki's original code.