forked from rayyuewang/gelearn
/
theano_learner.py
131 lines (112 loc) · 3.73 KB
/
theano_learner.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
from learner import Learner
import scipy.sparse as sp
import theano.tensor as T
import numpy as np
from theano import sparse
import theano
import sys
def get_label_map(labels):
m = {}
for i in range(len(labels)):
m[labels[i]] = i
return m
def get_feature_map(dat):
m = {}
i = 0
for doc_id, fea_dic in dat.items():
for f, v in fea_dic.items():
if f not in m:
m[f] = i
i += 1
return m
def extract_rows_containing_feature(dat, fea, fea_map):
num_row = 0
row = []
col = []
val = []
for doc_id, fea_dic in dat.items():
if fea in fea_dic: # add the entire row
for f, v in fea_dic.items():
row.append( num_row )
col.append( fea_map[f] )
val.append( v )
num_row += 1
spmat = sp.coo_matrix((val, (row, col)), shape=(num_row, len(fea_map))).tocsr()
return spmat, num_row
class TheanoLearner(Learner):
def __init__(self, data, init_model, param):
Learner.__init__(self, data, init_model, param)
# print 'Hey, TheanoLearner is initializing!'
### mapping from input data to continuous integer indices
# label_set['label'] = j
self.label_map = get_label_map(self.labels)
# feature_set['fea_id'] = k
self.feature_map = get_feature_map(data.dat)
# we separate the part building a training function from the actual learning.
# before learning happens, we have to have the training function.
# that's the trick to learn from different data sets.
def get_train_function(self):
# specify the computational graph
target = T.matrix('target')
weight = theano.shared(np.random.randn(len(self.feature_map), len(self.label_map)), name='weight')
feat_mat = sparse.csr_matrix(name='feat_mat')
mask_mat = sparse.csr_matrix(name='mask_mat')
sum_pred = sparse.dot( mask_mat, T.nnet.softmax( sparse.dot(feat_mat, weight) ) )
pred = sum_pred / sum_pred.sum(axis=1).reshape((sum_pred.shape[0], 1))
objective = T.nnet.categorical_crossentropy(pred, target).sum() + self.param.l2_regularization * (weight ** 2).sum()
grad_weight = T.grad(objective, weight)
# print 'Compiling function ...'
# compile the function
train = theano.function(inputs = [feat_mat, mask_mat, target], outputs = [objective, weight], updates=[(weight, weight - 0.1*grad_weight)] )
return train
def learn(self, train):
# print 'hey I am learning!'
# construct K matrices for K labeled features
mats = []
block_size = []
for fea in self.data.labeled_features:
mat, num_row = extract_rows_containing_feature(self.data.dat, fea, self.feature_map)
mats.append(mat)
block_size.append(num_row)
stack_mat = sp.vstack(mats, format='csr')
# construct the mask matrix
row = []
col = []
val = []
accu_c = 0
for r in range(len(block_size)):
for c in range(block_size[r]):
row.append(r)
col.append(c + accu_c)
val.append(1.)
accu_c += block_size[r]
mask_mat = sp.coo_matrix((val, (row, col)), shape=(len(block_size), accu_c)).tocsr()
# construct target: labeled features
target = []
for fea, label_dist in self.data.labeled_features.items():
dist = [1e-10]*len(self.labels)
for l in self.labels:
dist[ self.label_map[l] ] = label_dist[l]
target.append(dist)
target = np.asarray(target)
# print target
w = None
old_obj = -1
for i in range(20000):
obj, w = train(stack_mat, mask_mat, target)
if abs(obj - old_obj) < 5e-4: # stop training when the objective does not change much
break
old_obj = obj
# if i % 1000 == 0:
# sys.stderr.write('{}\t{}\r'.format(i, obj))
# print 'obj', obj
# print 'prd', prd
raw_model = w
# translate raw model
model = {}
for lbl in self.labels:
model[lbl] = {}
for fea, row_idx in self.feature_map.items():
for lbl, col_idx in self.label_map.items():
model[lbl][fea] = raw_model[row_idx, col_idx]
self.final_model = model