Python Algorithm Implementation

It is not difficult for an experienced Python programmer to implement a Python algorithm. We have introduced this operation method to you today. You can learn this as a reference object and sum up experience and skills in practice to master this technology.

Python algorithm operation code example:

 
 

  
  
# -*- coding: UTF8 -*-  
  
  
import sys  
  
  
import copy  
  
  
def init_pass(T):  
  
  
C = {}  
  
  
for t in T:  
  
  
for i in t:  
  
  
if i in C.keys():  
  
  
C[i] += 1  
  
  
else:  
  
  
C[i] = 1  
  
  
return C  
  
  
def candidate_gen(F):  
  
  
C = []  
  
  
k = len(F[0]) + 1  
  
  
for f1 in F:  
  
  
for f2 in F:  
  
  
if f1[k-2] < f2[k-2]:  
  
  
c = copy.copy(f1)  
  
  
c.append(f2[k-2])  
  
  
flag = True 
  
  
for i in range(0,k-1):  
  
  
s = copy.copy(c)  
  
  
s.pop(i)  
  
  
if s not in F:  
  
  
flag = False 
  
  
break  
  
  
if flag and c not in C:  
  
  
C.append(c)  
  
  
return C  
  
  
def compare_list(A,B):  
  
  
if len(A) <= len(B):  
  
  
for a in A:  
  
  
if a not in B:  
  
  
return False  
  
  
else:  
  
  
for b in B:  
  
  
if b not in A:  
  
  
return False  
  
  
return True  
  
  
def apriori(T, minsup):  
  
  
C = []  
  
  
init = init_pass(T)  
  
  
keys = init.keys()  
  
  
keys.sort()  
  
  
C.append(keys)  
  
  
n = len(T)  
  
  
F = [[]]  
  
  
for f in C[0]:  
  
  
if init[f]*1.0/n >= minsup:  
  
  
F[0].append([f])  
  
  
k = 1 
  
  
while F[k-1] != []:  
  
  
C.append(candidate_gen(F[k-1]))  
  
  
F.append([])  
  
  
for c in C[k]:  
  
  
count = 0;  
  
  
for t in T:  
  
  
if compare_list(c,t):  
  
  
count += 1  
  
  
if count*1.0/n >= minsup:  
  
  
F[k].append(c)  
  
  
k += 1  
  
  
U = []  
  
  
for f in F:  
  
  
for x in f:  
  
  
U.append(x)  
  
  
return U  
  
  
T = [['A','B','C','D'],['B','C','E'],['A','B','C','E'],
['B','D','E'],['A','B','C','D']]  
  
  
F = apriori(T, 0.9)  
  
  
print F 
 
 

This section describes how to compile Python algorithm code.