forked from YerevaNN/Dynamic-memory-networks-in-Theano
-
Notifications
You must be signed in to change notification settings - Fork 0
/
dmn_batch.py
452 lines (350 loc) · 21.1 KB
/
dmn_batch.py
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
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
import numpy as np
import theano
import theano.tensor as T
from theano.compile.nanguardmode import NanGuardMode
import lasagne
from lasagne import layers
from lasagne import nonlinearities
import cPickle as pickle
import utils
import nn_utils
floatX = theano.config.floatX
class DMN_batch:
def __init__(self, babi_train_raw, babi_test_raw, word2vec, word_vector_size, dim,
mode, answer_module, input_mask_mode, memory_hops, batch_size, l2,
normalize_attention):
self.vocab = {}
self.ivocab = {}
self.word2vec = word2vec
self.word_vector_size = word_vector_size
self.dim = dim
self.mode = mode
self.answer_module = answer_module
self.input_mask_mode = input_mask_mode
self.memory_hops = memory_hops
self.batch_size = batch_size
self.l2 = l2
self.normalize_attention = normalize_attention
self.max_fact_count = 0
self.train_input, self.train_q, self.train_answer, self.train_fact_count, self.train_input_mask = self._process_input(babi_train_raw)
self.test_input, self.test_q, self.test_answer, self.test_fact_count, self.test_input_mask = self._process_input(babi_test_raw)
self.vocab_size = len(self.vocab)
self.input_var = T.tensor3('input_var') # (batch_size, seq_len, glove_dim)
self.q_var = T.tensor3('question_var') # as self.input_var
self.answer_var = T.ivector('answer_var') # answer of example in minibatch
self.fact_count_var = T.ivector('fact_count_var') # number of facts in the example of minibatch
self.input_mask_var = T.imatrix('input_mask_var') # (batch_size, indices)
print "==> building input module"
self.W_inp_res_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.word_vector_size)), borrow=True)
self.W_inp_res_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_inp_res = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.W_inp_upd_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.word_vector_size)), borrow=True)
self.W_inp_upd_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_inp_upd = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.W_inp_hid_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.word_vector_size)), borrow=True)
self.W_inp_hid_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_inp_hid = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
input_var_shuffled = self.input_var.dimshuffle(1, 2, 0)
inp_dummy = theano.shared(np.zeros((self.dim, self.batch_size), dtype=floatX))
inp_c_history, _ = theano.scan(fn=self.input_gru_step,
sequences=input_var_shuffled,
outputs_info=T.zeros_like(inp_dummy))
inp_c_history_shuffled = inp_c_history.dimshuffle(2, 0, 1)
inp_c_list = []
inp_c_mask_list = []
for batch_index in range(self.batch_size):
taken = inp_c_history_shuffled[batch_index].take(self.input_mask_var[batch_index, :self.fact_count_var[batch_index]], axis=0)
inp_c_list.append(T.concatenate([taken, T.zeros((self.max_fact_count - taken.shape[0], self.dim), floatX)]))
inp_c_mask_list.append(T.concatenate([T.ones((taken.shape[0],), np.int32), T.zeros((self.max_fact_count - taken.shape[0],), np.int32)]))
self.inp_c = T.stack(inp_c_list).dimshuffle(1, 2, 0)
inp_c_mask = T.stack(inp_c_mask_list).dimshuffle(1, 0)
q_var_shuffled = self.q_var.dimshuffle(1, 2, 0)
q_dummy = theano.shared(np.zeros((self.dim, self.batch_size), dtype=floatX))
q_q_history, _ = theano.scan(fn=self.input_gru_step,
sequences=q_var_shuffled,
outputs_info=T.zeros_like(q_dummy))
self.q_q = q_q_history[-1]
print "==> creating parameters for memory module"
self.W_mem_res_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.W_mem_res_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_mem_res = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.W_mem_upd_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.W_mem_upd_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_mem_upd = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.W_mem_hid_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.W_mem_hid_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_mem_hid = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.W_b = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.W_1 = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, 7 * self.dim + 2)), borrow=True)
self.W_2 = theano.shared(lasagne.init.Normal(0.1).sample((1, self.dim)), borrow=True)
self.b_1 = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.b_2 = theano.shared(lasagne.init.Constant(0.0).sample((1,)), borrow=True)
print "==> building episodic memory module (fixed number of steps: %d)" % self.memory_hops
memory = [self.q_q.copy()]
for iter in range(1, self.memory_hops + 1):
current_episode = self.new_episode(memory[iter - 1])
memory.append(self.GRU_update(memory[iter - 1], current_episode,
self.W_mem_res_in, self.W_mem_res_hid, self.b_mem_res,
self.W_mem_upd_in, self.W_mem_upd_hid, self.b_mem_upd,
self.W_mem_hid_in, self.W_mem_hid_hid, self.b_mem_hid))
last_mem = memory[-1]
print "==> building answer module"
self.W_a = theano.shared(lasagne.init.Normal(0.1).sample((self.vocab_size, self.dim)), borrow=True)
if self.answer_module == 'feedforward':
self.prediction = nn_utils.softmax(T.dot(self.W_a, last_mem))
elif self.answer_module == 'recurrent':
self.W_ans_res_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim + self.vocab_size)), borrow=True)
self.W_ans_res_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_ans_res = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.W_ans_upd_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim + self.vocab_size)), borrow=True)
self.W_ans_upd_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_ans_upd = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
self.W_ans_hid_in = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim + self.vocab_size)), borrow=True)
self.W_ans_hid_hid = theano.shared(lasagne.init.Normal(0.1).sample((self.dim, self.dim)), borrow=True)
self.b_ans_hid = theano.shared(lasagne.init.Constant(0.0).sample((self.dim,)), borrow=True)
def answer_step(prev_a, prev_y):
a = self.GRU_update(prev_a, T.concatenate([prev_y, self.q_q]),
self.W_ans_res_in, self.W_ans_res_hid, self.b_ans_res,
self.W_ans_upd_in, self.W_ans_upd_hid, self.b_ans_upd,
self.W_ans_hid_in, self.W_ans_hid_hid, self.b_ans_hid)
y = nn_utils.softmax(T.dot(self.W_a, a))
return [a, y]
# TODO: add conditional ending
dummy = theano.shared(np.zeros((self.vocab_size, self.batch_size), dtype=floatX))
results, updates = theano.scan(fn=self.answer_step,
outputs_info=[last_mem, T.zeros_like(dummy)], #(last_mem, y)
n_steps=1)
self.prediction = results[1][-1]
else:
raise Exception("invalid answer_module")
self.prediction = self.prediction.dimshuffle(1, 0)
self.params = [self.W_inp_res_in, self.W_inp_res_hid, self.b_inp_res,
self.W_inp_upd_in, self.W_inp_upd_hid, self.b_inp_upd,
self.W_inp_hid_in, self.W_inp_hid_hid, self.b_inp_hid,
self.W_mem_res_in, self.W_mem_res_hid, self.b_mem_res,
self.W_mem_upd_in, self.W_mem_upd_hid, self.b_mem_upd,
self.W_mem_hid_in, self.W_mem_hid_hid, self.b_mem_hid,
self.W_b, self.W_1, self.W_2, self.b_1, self.b_2, self.W_a]
if self.answer_module == 'recurrent':
self.params = self.params + [self.W_ans_res_in, self.W_ans_res_hid, self.b_ans_res,
self.W_ans_upd_in, self.W_ans_upd_hid, self.b_ans_upd,
self.W_ans_hid_in, self.W_ans_hid_hid, self.b_ans_hid]
print "==> building loss layer and computing updates"
self.loss_ce = T.nnet.categorical_crossentropy(self.prediction, self.answer_var).mean()
if self.l2 > 0:
self.loss_l2 = self.l2 * nn_utils.l2_reg(self.params)
else:
self.loss_l2 = 0
self.loss = self.loss_ce + self.loss_l2
updates = lasagne.updates.adadelta(self.loss, self.params)
if self.mode == 'train':
print "==> compiling train_fn"
self.train_fn = theano.function(inputs=[self.input_var, self.q_var, self.answer_var, self.fact_count_var, self.input_mask_var],
outputs=[self.prediction, self.loss],
updates=updates,
#mode=NanGuardMode(nan_is_error=True, inf_is_error=True, big_is_error=False)
)
print "==> compiling test_fn"
self.test_fn = theano.function(inputs=[self.input_var, self.q_var, self.answer_var, self.fact_count_var, self.input_mask_var],
outputs=[self.prediction, self.loss, self.inp_c, self.q_q, last_mem],
#mode=NanGuardMode(nan_is_error=True, inf_is_error=True, big_is_error=False)
)
if self.mode == 'train':
print "==> computing gradients (for debugging)"
gradient = T.grad(self.loss, self.params)
self.get_gradient_fn = theano.function(inputs=[self.input_var, self.q_var, self.answer_var, self.fact_count_var, self.input_mask_var], outputs=gradient)
def GRU_update(self, h, x, W_res_in, W_res_hid, b_res,
W_upd_in, W_upd_hid, b_upd,
W_hid_in, W_hid_hid, b_hid):
""" mapping of our variables to symbols in DMN paper:
W_res_in = W^r
W_res_hid = U^r
b_res = b^r
W_upd_in = W^z
W_upd_hid = U^z
b_upd = b^z
W_hid_in = W
W_hid_hid = U
b_hid = b^h
"""
z = T.nnet.sigmoid(T.dot(W_upd_in, x) + T.dot(W_upd_hid, h) + b_upd.dimshuffle(0, 'x'))
r = T.nnet.sigmoid(T.dot(W_res_in, x) + T.dot(W_res_hid, h) + b_res.dimshuffle(0, 'x'))
_h = T.tanh(T.dot(W_hid_in, x) + r * T.dot(W_hid_hid, h) + b_hid.dimshuffle(0, 'x'))
return z * h + (1 - z) * _h
def save_params(self, file_name, epoch, **kwargs):
with open(file_name, 'w') as save_file:
pickle.dump(
obj = {
'params' : [x.get_value() for x in self.params],
'epoch' : epoch,
'gradient_value': (kwargs['gradient_value'] if 'gradient_value' in kwargs else 0)
},
file = save_file,
protocol = -1
)
def load_state(self, file_name):
print "==> loading state %s" % file_name
with open(file_name, 'r') as load_file:
dict = pickle.load(load_file)
loaded_params = dict['params']
for (x, y) in zip(self.params, loaded_params):
x.set_value(y)
def _empty_word_vector(self):
return np.zeros((self.word_vector_size,), dtype=floatX)
def input_gru_step(self, x, prev_h):
return self.GRU_update(prev_h, x, self.W_inp_res_in, self.W_inp_res_hid, self.b_inp_res,
self.W_inp_upd_in, self.W_inp_upd_hid, self.b_inp_upd,
self.W_inp_hid_in, self.W_inp_hid_hid, self.b_inp_hid)
def new_attention_step(self, ct, prev_g, mem, q_q):
cWq = T.dot(T.ones((1, self.batch_size), dtype=floatX), T.dot(T.dot(ct.T, self.W_b), q_q) * T.eye(n=self.batch_size, m=self.batch_size, dtype=floatX))
cWm = T.dot(T.ones((1, self.batch_size), dtype=floatX), T.dot(T.dot(ct.T, self.W_b), mem) * T.eye(n=self.batch_size, m=self.batch_size, dtype=floatX))
z = T.concatenate([ct, mem, q_q, ct * q_q, ct * mem, T.abs_(ct - q_q), T.abs_(ct - mem), cWq, cWm], axis=0)
l_1 = T.dot(self.W_1, z) + self.b_1.dimshuffle(0, 'x')
l_1 = T.tanh(l_1)
l_2 = T.dot(self.W_2, l_1) + self.b_2.dimshuffle(0, 'x')
G = T.nnet.sigmoid(l_2)[0]
return G
def new_episode_step(self, ct, g, prev_h):
gru = self.GRU_update(prev_h, ct,
self.W_mem_res_in, self.W_mem_res_hid, self.b_mem_res,
self.W_mem_upd_in, self.W_mem_upd_hid, self.b_mem_upd,
self.W_mem_hid_in, self.W_mem_hid_hid, self.b_mem_hid)
h = g * gru + (1 - g) * prev_h
return h
def new_episode(self, mem):
g, g_updates = theano.scan(fn=self.new_attention_step,
sequences=self.inp_c,
non_sequences=[mem, self.q_q],
outputs_info=T.zeros_like(self.inp_c[0][0]))
if (self.normalize_attention):
g = nn_utils.softmax(g)
e, e_updates = theano.scan(fn=self.new_episode_step,
sequences=[self.inp_c, g],
outputs_info=T.zeros_like(self.inp_c[0]))
e_list = []
for index in range(self.batch_size):
e_list.append(e[self.fact_count_var[index] - 1, :, index])
return T.stack(e_list).dimshuffle((1, 0))
def _process_input(self, data_raw):
max_inp_len = 0
max_q_len = 0
self.max_fact_count = 0
for x in data_raw:
inp = x["C"].lower().split(' ')
inp = [w for w in inp if len(w) > 0]
q = x["Q"].lower().split(' ')
q = [w for w in q if len(w) > 0]
if (self.input_mask_mode == 'word'):
fact_count = len(inp)
elif (self.input_mask_mode == 'sentence'):
fact_count = len([0 for w in inp if w == '.'])
else:
raise Exception("unknown input_mask_mode")
max_inp_len = max(max_inp_len, len(inp))
max_q_len = max(max_q_len, len(q))
self.max_fact_count = max(self.max_fact_count, fact_count)
questions = []
inputs = []
answers = []
fact_counts = []
input_masks = []
for x in data_raw:
inp = x["C"].lower().split(' ')
inp = [w for w in inp if len(w) > 0]
q = x["Q"].lower().split(' ')
q = [w for w in q if len(w) > 0]
if (self.input_mask_mode == 'word'):
fact_count = len(inp)
if (self.input_mask_mode == 'sentence'):
fact_count = len([0 for w in inp if w == '.'])
inp_vector = [utils.process_word(word = w,
word2vec = self.word2vec,
vocab = self.vocab,
ivocab = self.ivocab,
word_vector_size = self.word_vector_size,
to_return = "word2vec") for w in inp]
while(len(inp_vector) < max_inp_len):
inp_vector.append(self._empty_word_vector())
q_vector = [utils.process_word(word = w,
word2vec = self.word2vec,
vocab = self.vocab,
ivocab = self.ivocab,
word_vector_size = self.word_vector_size,
to_return = "word2vec") for w in q]
while(len(q_vector) < max_q_len):
q_vector.append(self._empty_word_vector())
if (self.input_mask_mode == 'word'):
input_mask = range(len(inp))
if (self.input_mask_mode == 'sentence'):
input_mask = [index for index, w in enumerate(inp) if w == '.']
while(len(input_mask) < self.max_fact_count):
input_mask.append(-1)
inputs.append(inp_vector)
questions.append(q_vector)
# NOTE: here we assume the answer is one word!
answers.append(utils.process_word(word = x["A"],
word2vec = self.word2vec,
vocab = self.vocab,
ivocab = self.ivocab,
word_vector_size = self.word_vector_size,
to_return = "index"))
fact_counts.append(fact_count)
input_masks.append(input_mask)
inputs = np.array(inputs).astype(floatX)
questions = np.array(questions).astype(floatX)
answers = np.array(answers).astype(np.int32)
fact_counts = np.array(fact_counts).astype(np.int32)
input_masks = np.array(input_masks).astype(np.int32)
return inputs, questions, answers, fact_counts, input_masks
def get_batches_per_epoch(self, mode):
if (mode == 'train'):
return len(self.train_input) / self.batch_size
elif (mode == 'test'):
return len(self.test_input) / self.batch_size
else:
raise Exception("unknown mode")
def step(self, batch_index, mode):
if mode == "train" and self.mode == "test":
raise Exception("Cannot train during test mode")
if mode == "train":
theano_fn = self.train_fn
inputs = self.train_input
qs = self.train_q
answers = self.train_answer
fact_counts = self.train_fact_count
input_masks = self.train_input_mask
if mode == "test":
theano_fn = self.test_fn
inputs = self.test_input
qs = self.test_q
answers = self.test_answer
fact_counts = self.test_fact_count
input_masks = self.test_input_mask
start_index = batch_index * self.batch_size
inp = inputs[start_index:start_index+self.batch_size]
q = qs[start_index:start_index+self.batch_size]
ans = answers[start_index:start_index+self.batch_size]
fact_count = fact_counts[start_index:start_index+self.batch_size]
input_mask = input_masks[start_index:start_index+self.batch_size]
skipped = 0
grad_norm = float('NaN')
if mode == 'train':
gradient_value = self.get_gradient_fn(inp, q, ans, fact_count, input_mask)
grad_norm = np.max([utils.get_norm(x) for x in gradient_value])
if (np.isnan(grad_norm)):
print "==> gradient is nan at index %d." % batch_index
print "==> skipping"
skipped = 1
if skipped == 0:
ret = theano_fn(inp, q, ans, fact_count, input_mask)
else:
ret = [float('NaN'), float('NaN')]
param_norm = np.max([utils.get_norm(x.get_value()) for x in self.params])
return {"prediction": ret[0],
"answers": ans,
"current_loss": ret[1],
"skipped": skipped,
"grad_norm": grad_norm,
"param_norm": param_norm,
"log": "",
}