1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
|
#!/usr/bin/env ruby
require 'trollop'
require 'tempfile'
require 'open3'
SMT_SEMPARSE = 'python /workspace/grounded/smt-semparse-cp/decode_sentence.py /workspace/grounded/smt-semparse-cp/working/full_dataset'
EVAL_PL = '/workspace/grounded/wasp-1.0/data/geo-funql/eval/eval.pl'
CDEC = "/toolbox/cdec-dtrain/bin/cdec"
# execute
def exec natural_language_string, reference_output, no_output=false
func = `#{SMT_SEMPARSE} "#{natural_language_string}"`.strip
output = `echo "execute_funql_query(#{func}, X)." | swipl -s #{EVAL_PL} 2>&1 | grep "X ="`.strip.split('X = ')[1]
puts " nrl: #{natural_language_string}" if !no_output
puts " mrl: #{func}" if !no_output
puts " output: #{output}" if !no_output
puts " correct?: #{output==reference_output}" if !no_output
return output==reference_output, func, output
end
# decoder interaction/translations
class Translation
attr_accessor :s, :f, :rank, :model, :score
def initialize kbest_line, rank=-1
a = kbest_line.split ' ||| '
@s = a[1].strip
h = {}
a[2].split.each { |i|
name, value = i.split '='
value = value.to_f
h[name] = value
}
@f = NamedSparseVector.new h
@rank = rank
@model = a[3].to_f
@score = -1.0
end
def to_s
"#{@rank} ||| #{@s} ||| #{@model} ||| #{@score} ||| #{@f.to_s}"
end
end
def predict_translation s, k, ini, w
o, s = Open3.capture2 "echo \"#{s}\" | #{CDEC} -c #{ini} -r -k #{k} -w #{w} 2>/dev/null"
j = -1
return o.split("\n").map{|i| j+=1; Translation.new(i, j)}
end
# scoring (per-sentence BLEU)
def ngrams_it(s, n, fix=false)
a = s.strip.split
a.each_with_index { |tok, i|
tok.strip!
0.upto([n-1, a.size-i-1].min) { |m|
yield a[i..i+m] if !(fix||(a[i..i+m].size>n))
}
}
end
def brevity_penalty hypothesis, reference
a = hypothesis.split; b = reference.split
return 1.0 if a.size>b.size
return Math.exp(1.0 - b.size.to_f/a.size);
end
def per_sentence_bleu hypothesis, reference, n=4
h_ng = {}; r_ng = {}
(1).upto(n) {|i| h_ng[i] = []; r_ng[i] = []}
ngrams_it(hypothesis, n) {|i| h_ng[i.size] << i}
ngrams_it(reference, n) {|i| r_ng[i.size] << i}
m = [n, reference.split.size].min
weight = 1.0/m
add = 0.0
sum = 0
(1).upto(m) { |i|
counts_clipped = 0
counts_sum = h_ng[i].size
h_ng[i].uniq.each {|j| counts_clipped += r_ng[i].count(j)}
add = 1.0 if i >= 2
sum += weight * Math.log((counts_clipped + add)/(counts_sum + add));
}
return brevity_penalty(hypothesis, reference) * Math.exp(sum)
end
def score_translations list_of_translations, reference
list_of_translations.each { |i| i.score = per_sentence_bleu i.s, reference}
end
# hope and fear
def hope_and_fear kbest, action
max = -1.0/0
max_idx = -1
kbest.each_with_index { |i,j|
if action=='hope' && i.model + i.score > max
max_idx = j; max = i.model + i.score
end
if action=='fear' && i.model - i.score > max
max_idx = j; max = i.model - i.score
end
}
return kbest[max_idx]
end
# update
def update w, hope, fear, eta
diff = hope.f - fear.f
diff *= eta
w += diff
return w
end
# weights
class NamedSparseVector
attr_accessor :h
def initialize init=nil
@h = {}
@h = init if init
@h.default = 0.0
end
def + other
new_h = Hash.new
new_h.update @h
ret = NamedSparseVector.new new_h
other.each_pair { |k,v| ret[k]+=v }
return ret
end
def from_file fn
f = File.new(fn, 'r')
while line = f.gets
name, value = line.strip.split
value = value.to_f
@h[name] = value
end
end
def to_file
s = []
@h.each_pair { |k,v| s << "#{k} #{v}" }
s.join("\n")+"\n"
end
def - other
new_h = Hash.new
new_h.update @h
ret = NamedSparseVector.new new_h
other.each_pair { |k,v| ret[k]-=v }
return ret
end
def * scalar
raise ArgumentError, "Arg is not numeric #{scalar}" unless scalar.is_a? Numeric
ret = NamedSparseVector.new
@h.keys.each { |k| ret[k] = @h[k]*scalar }
return ret
end
def dot other
sum = 0.0
@h.each_pair { |k,v|
sum += v * other[k]
}
return sum
end
def [] k
@h[k]
end
def []= k, v
@h[k] = v
end
def each_pair
@h.each_pair { |k,v| yield k,v }
end
def to_s
@h.to_s
end
def length
Math.sqrt(@h.values.map{|i|i*i}.inject(:+))
end
def normalize!
l = length
@h.each_pair { |k,v|
@h[k] = v/l
}
end
def size
@h.keys.size
end
end
# map models score to [0,1]
def adj_model kbest, factor
min = kbest.map{|i|i.model}.min
max = kbest.map{|i|i.model}.max
kbest.each {|i| i.model = factor*((i.model-min)/(max-min))}
end
class Stats
def initialize name
@name = name
@with_parse = 0.0
@with_output = 0.0
@correct_output = 0.0
end
def update feedback, func, output
@with_parse +=1 if func!="None"
@with_output +=1 if output!="null"
@correct_output += 1 if feedback==true
end
def print total
<<-eos
[#{@name}]
with parse #{((@with_parse/total)*100).round 2} abs:#{@with_parse}
with output #{((@with_output/total)*100).round 2} abs:#{@with_output}
with correct output #{((@correct_output/total)*100).round 2} abs:#{@correct_output}
eos
end
end
def _print rank, string, model, score
puts "rank=#{rank} string='#{string}' model=#{model} score=#{score}"
end
def bag_of_words s, stopwords=[]
s.split.uniq.sort.reject{|v| stopwords.include? v}
end
def main
opts = Trollop::options do
opt :k, "k", :type => :int, :required => true
opt :input, "'foreign' input", :type => :string, :required => true
opt :references, "(parseable) references", :type => :string, :required => true
opt :gold, "gold output", :type => :string, :require => true
opt :gold_mrl, "gold parse", :type => :string, :short => '-h', :require => true
opt :init_weights, "initial weights", :type => :string, :required => true, :short => '-w'
opt :cdec_ini, "cdec config file", :type => :string, :default => './cdec.ini'
opt :eta, "learning rate", :type => :float, :default => 0.01
opt :no_update, "don't update weights", :type => :bool, :default => false
opt :stop_after, "stop after x examples", :type => :int, :default => -1
opt :output_weights, "output file for final weights", :type => :string, :required => true
opt :scale_model, "scale model score by this factor", :type => :float, :default => 1.0, :short => '-m'
opt :normalize, "normalize weights after each update", :type => :bool, :default => false, :short => '-l'
opt :print_kbests, "print full kbest lists", :type => :bool, :default => false, :short => '-j'
opt :hope2, "select hope from the first X items in kbest that executes", :type => :int, :default => 0, :short => '-x'
opt :fear2, "skip example if fear executes", :type => :bool, :default => false
end
puts "cfg"
opts.each_pair {|k,v| puts "#{k}\t#{v}"}
puts
input = File.new(opts[:input], 'r').readlines.map{|i|i.strip}
references = File.new(opts[:references], 'r').readlines.map{|i|i.strip}
gold = File.new(opts[:gold], 'r').readlines.map{|i|i.strip}
gold_mrl = File.new(opts[:gold_mrl], 'r').readlines.map{|i|i.strip}
stopwords = File.new('stopwords.en', 'r').readlines.map{|i|i.strip}
# init weights
w = NamedSparseVector.new
w.from_file opts[:init_weights]
without_translations = 0
count = 0
top1_stats = Stats.new 'top1'
hope_stats = Stats.new 'hope'
fear_stats = Stats.new 'fear'
refs_stats = Stats.new 'refs'
type1_updates = 0
type2_updates = 0
top1_hit = 0
top1_variant = 0
top1_real_variant = 0
hope_hit = 0
hope_variant = 0
hope_real_variant = 0
kbest_sz = 0
last_wf = ''
input.each_with_index { |i,j|
count += 1
# write current weights to file
tmp_file = Tempfile.new('rampion')
tmp_file_path = tmp_file.path
last_wf = tmp_file.path
tmp_file.write w.to_file
tmp_file.close
# get kbest list for current input
kbest = predict_translation i, opts[:k], opts[:cdec_ini], tmp_file_path
kbest_sz += kbest.size
if kbest.size==0
without_translations += 1
next
end
score_translations kbest, references[j]
if opts[:print_kbests]
puts "KBEST"
kbest.each_with_index { |k,l|
_print l, k.s, k.model, k.score
}
end
adj_model kbest, opts[:scale_model]
# get feedback
puts "EXAMPLE #{j}"
puts "GOLD MRL: #{gold_mrl[j]}"
puts "GOLD OUTPUT #{gold[j]}"
# fear
fear = hope_and_fear kbest, 'fear'
if opts[:fear2]
f, g, o = exec fear.s, gold[j], true
if f
puts "FEAR EXECUTED, skipping example\n\n"
next
end
end
# top1
puts "---top1"
_print 0, kbest[0].s, kbest[0].model, kbest[0].score
feedback, func, output = exec kbest[0].s, gold[j]
# hope2
parses = []
if opts[:hope2]>0
already_seen = {}
puts "<<KBEST EXEC"
(1).upto([opts[:hope2]-1, kbest.size-1].min) { |l|
f, g, o = exec kbest[l].s, gold[j], true
words = bag_of_words kbest[l].s, stopwords
parses << f
puts "#{f} | #{l} | #{kbest[l].s} #{words.to_s}" if !already_seen.has_key? words
already_seen[words] = true
}
puts ">>>"
end
top1_stats.update feedback, func, output
# hope & update
ref_words = bag_of_words references[j], stopwords
hope = nil
if feedback==true
if kbest[0].s == references[j]
top1_hit +=1
else
top1_variant += 1
top1_real_variant += 1 if bag_of_words(kbest[0].s,stopwords)!=ref_words
end
#references[j] = kbest[0].s
hope = kbest[0]
type1_updates += 1
else
if opts[:hope2]>0
c=-1; found = parses.detect{|b| c+=1; b }
hope = kbest[c] if found
if !found
puts "NO GOOD HOPE, skipping example\n\n"
next
end
else
hope = hope_and_fear kbest, 'hope'
end
if hope.s == references[j]
hope_hit += 1
else
hope_variant += 1
hope_real_variant += 1 if bag_of_words(hope.s,stopwords)!=ref_words
end
type2_updates += 1
end
puts "---hope"
_print hope.rank, hope.s, hope.model, hope.score
feedback, func, output = exec hope.s, gold[j]
hope_stats.update feedback, func, output
puts "---fear"
_print fear.rank, fear.s, fear.model, fear.score
feedback, func, output = exec fear.s, gold[j]
fear_stats.update feedback, func, output
puts "---reference"
_print 'x', references[j], 'x', 1.0
feedback, func, output = exec references[j], gold[j]
refs_stats.update feedback, func, output
puts
w = update w, hope, fear, opts[:eta] if !opts[:no_update]
w.normalize! if opts[:normalize]
break if opts[:stop_after]>0 && (j+1)==opts[:stop_after]
}
FileUtils::cp(last_wf, opts[:output_weights])
puts "#{count} examples"
puts " type1 updates: #{type1_updates}"
puts " type2 updates: #{type2_updates}"
puts " top1 hits: #{top1_hit}"
puts " top1 variant: #{top1_variant}"
puts "top1 real variant: #{top1_real_variant}"
puts " hope hits: #{hope_hit}"
puts " hope variant: #{hope_variant}"
puts "hope real variant: #{hope_real_variant}"
puts " kbest size: #{(kbest_sz/count).round 2}"
puts "#{((without_translations.to_f/count)*100).round 2}% without translations (abs: #{without_translations})"
puts top1_stats.print count
puts hope_stats.print count
puts fear_stats.print count
puts refs_stats.print count
end
main
|