I'm trying to write the Quine-McCluskey algorithm in python, but I wanted to see if there were any versions out there that I might use instead. A google search showed few useful results. I'm looking for 4x4 map reduction, not 2x2 or 3x3. Any ideas or references?
Add a comment
|
2 Answers
def combine(m, n):
a = len(m)
c = ''
count = 0
for i in range(a):
if(m[i] == n[i]):
c += m[i]
elif(m[i] != n[i]):
c += '-'
count += 1
if(count > 1):
return None
else:
return c
def find_prime_implicants(data):
newList = list(data)
size = len(newList)
IM = []
im = []
im2 = []
mark = [0]*size
m = 0
for i in range(size):
for j in range(i+1, size):
c = combine( str(newList[i]), str(newList[j]) )
if c != None:
im.append(str(c))
mark[i] = 1
mark[j] = 1
else:
continue
mark2 = [0]*len(im)
for p in range(len(im)):
for n in range(p+1, len(im)):
if( p != n and mark2[n] == 0):
if( im[p] == im[n]):
mark2[n] = 1
for r in range(len(im)):
if(mark2[r] == 0):
im2.append(im[r])
for q in range(size):
if( mark[q] == 0 ):
IM.append( str(newList[q]) )
m = m+1
if(m == size or size == 1):
return IM
else:
return IM + find_prime_implicants(im2)
minterms = set(['1101', '1100', '1110', '1111', '1010', '0011', '0111', '0110'])
minterms2 = set(['0000', '0100', '1000', '0101', '1100', '0111', '1011', '1111'])
minterms3 = set(['0001', '0011', '0100', '0110', '1011', '0000', '1000', '1010', '1100', '1101'])
print 'PI(s):', find_prime_implicants(minterms)
print 'PI2(s):', find_prime_implicants(minterms2)
print 'PI3(s):', find_prime_implicants(minterms3)
2 Comments
HighwayJohn
Does this algorithm only calculates the prime implicates or also the essential prime implicates?
fuglede
@HighwayJohn: Whatever comes out, it's certainly not minimal; I have a 10-bit example with 149 ones, where the algorithm here boils it down to 137 summands, but where I know the upper limit for a minimized sum is 119.
In the Wikipedia of which you gave the link, there are some "External links" at the bottom, among which are these, interesting relatively to your project:
" Python Implementation by Robert Dick "
Wouldn't this fulfil your need ?
" A series of two articles describing the algorithm(s) implemented in R: first article and second article. The R implementation is exhaustive and it offers complete and exact solutions. It processes up to 20 input variables. "
You could use the rpy Python interface to R language to run the R code of the Quine-McCluskey algorithm. Note that there is a rewrite of rpy : rpy2
Also, why not, write yourself a new Python script, using the enhancement of the algorithm done by Adrian Duşa in 2007 , lying in the second article ?
1 Comment
eqb
vi is up, and I'm coding my heart out, thanks to the reference to that second article. Thank you! :)