1 ##################################################################
3 ##################################################################
5 # - more elaborate pseudo-code in the io()-function
8 # - CNF_Rule has a function LHS() which returns head, in DMV_Rule this
9 # returns the pair of (bars, head).
12 # - CNF_Rule has __eq__ and __ne__ defined, so that we can use == and
13 # != on two such rules
16 # import numpy # numpy provides Fast Arrays, for future optimization
17 # import pprint # for pretty-printing
21 # some of dmv-module bleeding in here... todo: prettier (in inner())
26 '''Easily turn on/off inline debug printouts with this global. There's
27 a lot of cluttering debug statements here, todo: clean up'''
33 '''The PCFG used in the I/O-algorithm.
38 Todo: as of now, this allows duplicate rules... should we check
39 for this? (eg. g = Grammar([x,x],[]) where x.prob == 1 may give
40 inner probabilities of 2.)'''
45 return [rule for rule in self.all_rules() if rule.LHS() == LHS]
53 def __init__(self, p_rules, p_terminals, numtag, tagnum):
54 '''rules and p_terminals should be arrays, where p_terminals are of
55 the form [preterminal, terminal], and rules are CNF_Rule's.'''
56 self.__p_rules = p_rules # todo: could check for summing to 1 (+/- epsilon)
57 self.__numtag = numtag
58 self.__tagnum = tagnum
59 self.p_terminals = p_terminals
65 '''A single CNF rule in the PCFG, of the form
67 where these are just integers
68 (where do we save the connection between number and symbol?
69 symbols being 'vbd' etc.)'''
70 def __eq__(self, other):
71 return self.LHS() == other.LHS() and self.R() == other.R() and self.L() == other.L()
72 def __ne__(self, other):
73 return self.LHS() != other.LHS() or self.R() != other.R() or self.L() != other.L()
75 return "%s -> %s %s [%.2f]" % (self.LHS(), self.L(), self.R(), self.prob)
76 def __init__(self, LHS, L, R, prob):
82 "Return a probability, doesn't care about attachment..."
91 def inner(s, t, LHS, g, sent, chart):
92 ''' Give the inner probability of having the node LHS cover whatever's
93 between s and t in sentence sent, using grammar g.
95 Returns a pair of the inner probability and the chart
97 For DMV, LHS is a pair (bar, h), but this function ought to be
100 e() is an internal function, so the variable chart (a dictionary)
101 is available to all calls of e().
103 Since terminal probabilities are just simple lookups, they are not
104 put in the chart (although we could put them in there later to
112 '''Chart has lists of probability and whether or not we've attached
113 yet to L and R, each entry is a list [p, Rattach, Lattach], where if
114 Rattach==True then the rule has a right-attachment or there is one
115 lower in the tree (meaning we're no longer adjacent).'''
116 if (s, t, LHS) in chart:
117 return chart[(s, t, LHS)]
119 debug( "trying from %d to %d with %s" % (s,t,LHS) )
121 if (LHS, O(s)) in g.p_terminals:
122 prob = g.p_terminals[LHS, O(s)] # b[LHS, O(s)]
124 prob = 0.0 # todo: is this the right way to deal with lacking rules?
125 print "\t LACKING TERMINAL:"
126 debug( "\t terminal: %s -> %s : %.1f" % (LHS, O(s), prob) )
127 # terminals have no attachment
130 if (s,t,LHS) not in chart:
131 # by default, not attachment yet
132 chart[(s,t,LHS)] = 0.0 #, False, False]
133 for rule in g.rules(LHS): # summing over j,k in a[LHS,j,k]
134 debug( "\tsumming rule %s" % rule )
137 for r in range(s, t): # summing over r = s to r = t-1
140 p = rule.p("todo","todo")
141 chart[(s, t, LHS)] += p * p_L * p_R
142 debug( "\tchart[(%d,%d,%s)] = %.2f" % (s,t,LHS, chart[(s,t,LHS)]) )
143 return chart[(s, t, LHS)]
146 inner_prob = e(s,t,LHS)
149 for k,v in chart.iteritems():
150 print "\t%s -> %s_%d ... %s_%d : %.1f" % (k[2], O(k[0]), k[0], O(k[1]), k[1], v)
151 print "---CHART:end---"
152 return [inner_prob, chart] # inner_prob == chart[(s,t,LHS)]
161 if __name__ == "__main__":
162 print "IO-module tests:"
164 s = CNF_Rule(0,1,2, 1.0) # s->np vp
165 np = CNF_Rule(1,3,4, 0.3) # np->n p
166 b[1, 'n'] = 0.7 # np->'n'
167 b[3, 'n'] = 1.0 # n->'n'
168 b[4, 'p'] = 1.0 # p->'p'
169 vp = CNF_Rule(2,5,1, 0.1) # vp->v np (two parses use this rule)
170 vp2 = CNF_Rule(2,2,4, 0.9) # vp->vp p
171 b[5, 'v'] = 1.0 # v->'v'
173 g = Grammar([s,np,vp,vp2], b, {0:'s',1:'np',2:'vp',3:'n',4:'p',5:'v'})
181 test1 = inner(0,0, 1, g, ['n'], {})
183 print "should be 0.70 : %.2f" % test1[0]
187 test2 = inner(0,2, 2, g, ['v','n','p'], test1[1])
188 print "should be 0.?? (.09??) : %.2f" % test2[0]
189 print "------ trying the same again:----------"
190 test2 = inner(0,2, 2, g, ['v','n','p'], test2[1])
191 print "should be 0.?? (.09??) : %.2f" % test2[0]
194 ##################################################################
195 # just junk from here on down: #
196 ##################################################################
199 # "(pseudo-code / wishful thinking) "
200 # g = initialize(corpus) # or corpus.tagset ?
202 # P = {('v','n','p'):0.09}
203 # # P is used in v_q, w_q (expectation), so each sentence in the
204 # # corpus needs some initial P.
206 # # --- Maximation: ---
208 # # actually, this step (from Lari & Young) probably never happens
209 # # with DMV, since instead of the a[i,j,k] and b[i,m] vectors, we
210 # # have P_STOP and P_CHOOSE... or, in a sense it happens only we
211 # # calculate P_STOP and P_CHOOSE..somehow.
212 # for rule in g.p_rules:
215 # for pre_term in range(len(g.p_terminals)):
216 # ptnum[pre_term] = 0
217 # ptden[pre_term] = 0
219 # # we could also flip this to make rules the outer loop, then we
220 # # wouldn't have to initialize den/num in loops of their own
221 # for sent in corpus:
222 # for rule in g.p_rules # Equation 20
223 # for s in range(len(sent)):
224 # for t in range(s, len(sent)):
225 # rule.num += w(s,t, rule.LHS(),rule.L,rule.R, g, sent, P[sent])
226 # rule.den += v(s,t, rule.LHS(), g, sent, P[sent])
227 # # todo: do we need a "new-prob" vs "old-prob" distinction here?
228 # probably, since we use inner/outer which checks rule.prob()
229 # # todo: also, this wouldn't work, since for each sentence, we'd
230 # # discard the old probability; should rules be the outer
232 # rule.prob = rule.num / rule.den
233 # for pre_term in range(len(g.p_terminals)): # Equation 21
236 # for s in range(len(sent)):
237 # for t in range(s, len(sent)):
238 # num += v(t,t,pre_term, g, sent, P[sent])
239 # den += v(s,t,pre_term, g, sent, P[sent])
241 # for rule in g.rules:
242 # rule.prob = rule.num / rule.den
243 # for pre_term in range(len(g.p_terminals)):
244 # g.p_terminals[pre_term] = ptnum[pre_term] / ptden[pre_term]
247 # # --- Expectation: ---
248 # for sent in corpus: # Equation 11
249 # inside = inner(0, len(sent), ROOT, g, sent)
250 # P[sent] = inside[0]
252 # # todo: set inner.chart to {} again, how?
254 # # todo: need a old-P new-P distinction to check if we're below
255 # # threshold difference
258 # def w(s,t, LHS,L,R, g, sent, P_sent):
260 # rule = g.rule(LHS, L, R)
261 # for r in range(s, t):
262 # w += rule.prob() * inner(s,r, L, g, sent) * inner(r+1, t, R, g, sent) * outer(s,t,LHS,g,sent)
265 # def v(s,t, LHS, g, sent, P_sent):
266 # return ( inner(s,t, LHS, g, sent) * outer(s,t, LHS, g, sent) ) / P_sent