# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, # MA 02110-1301, USA. import dissociated_press as dp from time import sleep from sys import argv if len(argv) == 1: infile = "PLOMDATA" else: infile = argv[1] DEBUG = False N = 2 d = dp.dictionary(debug=DEBUG) f = open(infile,"r") input = [x[:-1] for x in f.readlines() if x.endswith("\n")] f.close() for i, l in enumerate(input): if DEBUG: print l d.dissociate(l, N=N) if i%100 == 0: print i try: while 1: sentence = d.associate()
#!/usr/bin/python # -*- coding: utf-8 -*- import dissociated_press as d s = d.sentence("Der behandschuhte Mann haut ein Kind.") t = d.sentence("Der Mann kotzt.") u = d.sentence("Ein Kind kotzt.") v = d.sentence("") dict = d.dictionary() s.dissociate(dict) t.dissociate(dict) u.dissociate(dict) v.dissociate(dict) for i in range(0,20): print dict.associate()
# Date: Sat Apr 25 09:54:24 +0000 2009 def parsedate(line): try: return datetime.strptime(re.sub(r"(st|nd|rd|th),", ",", line),"Date: %I:%M %p %b %d, %Y\n") except ValueError: return datetime.strptime(line,"Date: %a %b %d %H:%M:%S +0000 %Y\n") infile = config.local.tweetdata f = open(infile,"r") # initialize distr = {} # the distribution of the time of day of the tweets distrN = 0 # for probability distribution normalization d = diss.dictionary(debug=DEBUG) # THE dictionary input = [] # for comparison to avoid simple reposts for line in f: if line[:6] == "Date: ": t = parsedate(line).time() try: distr[(t.minute + t.hour*60)/(BINWIDTH)] += 1 except KeyError: distr[(t.minute + t.hour*60)/(BINWIDTH)] = 1 distrN += 1 elif line[:6] == "Text: ": d.dissociate(line[6:],N=N) input.append(line[6:]) f.close() # the real main loop while not TEST:
#!/usr/bin/python # -*- coding: utf-8 -*- import dissociated_press as dp from sys import argv import cProfile as profile if len(argv) == 1: infile = "PLOMDATA" else: infile = argv[1] N = 2 d = dp.dictionary(debug=False) f = open(infile,"r") input = f.readlines() f.close() profile_runs = [ 'for i, l in enumerate(input): d.dissociate(l, N=N)', 'for i in xrange(1000): d.associate()' ] for p in profile_runs: print p profile.run(p) print "========================"