-
Notifications
You must be signed in to change notification settings - Fork 32
/
encode_qa_and_text.py
222 lines (176 loc) · 7.77 KB
/
encode_qa_and_text.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
#!/usr/bin/python
import os
import sys
import ipdb
import pickle
import word2vec as w2v
import scipy.sparse as sps
sys.path.append('/u/makarand/Codes/UToronto/skip-thoughts')
import skipthoughts
import numpy as np
import progressbar as pb
pb_widgets = ['Encoding: ', pb.Percentage(), ' ', pb.Bar(), ' ', pb.ETA()]
# Local imports
import utils
import tfidf as tfidfcalc
# import video-text embedding
# sys.path.append('/ais/guppy9/movie2text/video_text_embed')
# from sentence_encoder import Encoder as VisTextEncoder
from movieqa_importer import MovieQA
mqa = MovieQA.DataLoader()
def check_save_directory(filename=None, dirname=None):
"""Make the folder where descriptors are saved if it doesn't exist.
"""
if filename:
dirname = filename.rsplit('/', 1)[0]
if not os.path.isdir(dirname):
os.makedirs(dirname)
def encode_tfidf_model(document_type, word_thresh=1):
"""Load TF-IDF model.
"""
tfidf_fname = utils.TFIDF_TEMPLATE %(document_type, word_thresh)
check_save_directory(filename=tfidf_fname)
if os.path.exists(tfidf_fname):
with open(tfidf_fname, 'rb') as fid:
TFIDF = pickle.load(fid)
else:
# read the story and gather words
story, _ = mqa.get_story_qa_data('full', document_type)
sorted_movies = sorted(story.keys())
all_words_use = []
for imdb_key in sorted_movies:
all_words_use.append([])
for sentence in story[imdb_key]:
norm_sentence = utils.normalize_stemming(utils.normalize_alphanumeric(sentence.lower()))
all_words_use[-1].extend(norm_sentence.split(' '))
# compute TFIDF
TFIDF = tfidfcalc.TFIDF(sorted_movies)
TFIDF.get_filtered_vocabulary(all_words_use, word_thresh=word_thresh)
TFIDF.compute_tfidf(all_words_use)
# dump to pickle file for future
with open(tfidf_fname, 'wb') as fid:
pickle.dump(TFIDF, fid)
return TFIDF
def load_model(desc, tfidf_doc='split_plot', tfidf_wthr=1):
"""Load appropriate model based on descriptor type.
"""
model = None
if desc.startswith('tfidf'):
model = encode_tfidf_model(tfidf_doc, tfidf_wthr)
desc = desc + '-' + tfidf_doc + '-' + str(tfidf_wthr)
elif desc == 'word2vec':
model = w2v.load('models/movie_plots_1364.d-300.mc1.w2v', kind='bin')
elif desc == 'skipthought':
model = skipthoughts.load_model()
elif desc == 'vis-text-embed':
raise ValueError('Visual-Text embeddings are not yet supported.')
# model = VisTextEncoder()
return model, desc
def encode_sentences(desc, sentence_list, model, imdb_key=None, is_qa=False):
"""Encode a list of sentences given the model.
"""
if desc == 'skipthought':
# encode a sentence list directly
features = skipthoughts.encode(model, sentence_list, verbose=False)
elif desc == 'vis-text-embed':
# normalize sentence lists
norm_sentence_list = [utils.normalize_alphanumeric(sentence.lower()) for sentence in sentence_list]
# allows to encode a sentence list directly
features = model.encode(norm_sentence_list)
elif desc.startswith('tfidf'):
desc_dim = len(model.vocab)
midx = model.doc_names.index(imdb_key)
# use scipy sparse matrix when encoding stories, otherwise too huge!
if is_qa:
features = np.zeros((len(sentence_list), desc_dim), dtype='float32')
else:
features = sps.dok_matrix((len(sentence_list), desc_dim), dtype='float32')
for s, sentence in enumerate(sentence_list):
# NOTE: use both alphanumeric and stemming normalization
sentence = utils.normalize_stemming(utils.normalize_alphanumeric(sentence.lower())).split(' ')
# for each word in the normalized sentence
for word in sentence:
if word not in model.vocab: continue
widx = model.vocab.index(word)
features[s,widx] = model.tfidf[widx][midx]
if is_qa: # if not sparse, use numpy.linalg.norm
features[s] /= (np.linalg.norm(features[s]) + 1e-6)
else: # if sparse, use scipy.sparse.linalg.norm
features[s] /= (sps.linalg.norm(features[s]) + 1e-6)
elif desc == 'word2vec':
desc_dim = model.get_vector(model.vocab[-1]).shape[0]
features = np.zeros((len(sentence_list), desc_dim), dtype='float32')
for s, sentence in enumerate(sentence_list):
# NOTE: use only alphanumeric normalization, no stemming
sentence = utils.normalize_alphanumeric(sentence.lower()).split(' ')
# for each word in the normalized sentence
for word in sentence:
if word not in model.vocab: continue
features[s] += model.get_vector(word)
features[s] /= (np.linalg.norm(features[s]) + 1e-6)
return features
def encode_documents(document_type, desc, model):
"""Encode sentences from the documents using the descriptor.
"""
story, _ = mqa.get_story_qa_data('full', document_type)
check_save_directory(filename=utils.DOC_DESC_TEMPLATE %(document_type, desc, ''))
pbar = pb.ProgressBar(widgets=pb_widgets, maxval=len(story)).start()
for i, imdb_key in enumerate(story.keys()):
pbar.update(i)
npy_fname = utils.DOC_DESC_TEMPLATE % (document_type, desc, imdb_key)
# if word2vec file exists continue
if os.path.exists(npy_fname): continue
# create a list of sentences
sentence_list = story[imdb_key]
# encode sentences
story_features = encode_sentences(desc, sentence_list, model, imdb_key=imdb_key, is_qa=False)
# save features, use pickle saver in case of sparse matrix
if type(story_features) == sps.dok.dok_matrix:
with open(npy_fname, 'wb') as fid:
pickle.dump(story_features, fid)
else:
np.save(npy_fname, story_features)
pbar.finish()
def encode_qa(desc, model):
"""Encode question and answer using the descriptor.
"""
_, QA = mqa.get_story_qa_data('full', 'split_plot')
check_save_directory(filename=utils.QA_DESC_TEMPLATE %(desc, ''))
pbar = pb.ProgressBar(widgets=pb_widgets, maxval=len(QA)).start()
for i, qa in enumerate(QA):
pbar.update(i)
npy_fname = utils.QA_DESC_TEMPLATE % (desc, qa.qid)
# if word2vec file exists continue
if os.path.exists(npy_fname): continue
# create a list of sentences
sentence_list = [qa.question]
sentence_list.extend([ans for ans in qa.answers if ans])
# encode sentences, and save features
qa_features = encode_sentences(desc, sentence_list, model, imdb_key=qa.imdb_key, is_qa=True)
np.save(npy_fname, qa_features)
pbar.finish()
def one_pass_encoding(model, desc):
"""Encode all questions and story types using this model.
"""
### Encode all questions
print 'Encoding QA | desc: %s' %(desc)
encode_qa(desc, model)
### Encode all documents
for doc in reversed(documents):
print 'Encoding %s | desc: %s' %(doc.upper(), desc)
encode_documents(doc, desc, model)
if __name__ == '__main__':
### Variable types
documents = ['split_plot', 'script', 'subtitle', 'dvs']
descriptors = ['tfidf', 'word2vec', 'skipthought'] #, 'vis-text-embed']
# For each descriptor type
for desc in descriptors:
### Load encoding model
if desc == 'tfidf':
model, desc = load_model(desc, tfidf_doc='split_plot', tfidf_wthr=1)
one_pass_encoding(model, desc)
model, desc = load_model(desc, tfidf_doc='subtitle', tfidf_wthr=5)
one_pass_encoding(model, desc)
else:
model, desc = load_model(desc)
one_pass_encoding(model, desc)