-
Notifications
You must be signed in to change notification settings - Fork 2
/
preprocess.py
197 lines (157 loc) · 8.96 KB
/
preprocess.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
import os
import numpy as np
import torch
from torch import nn
from torch.nn import functional as F
import torch.utils.data as td
import torchvision as tv
import pandas as pd
from PIL import Image
from matplotlib import pyplot as plt
import socket
import getpass
import nntools as nt
import json
import re
from collections import defaultdict
from nltk.stem.porter import *
import string
from nltk.tokenize import word_tokenize
#preprocessing uses code found: https://github.com/zcyang/imageqa-san/blob/master/data_vqa/process_function.py
def process_sentence(sentence):
periodStrip = re.compile("(?!<=\d)(\.)(?!\d)")
commaStrip = re.compile("(\d)(\,)(\d)")
punct = [';', r"/", '[', ']', '"', '{', '}',
'(', ')', '=', '+', '\\', '_', '-',
'>', '<', '@', '`', ',', '?', '!']
contractions = {"aint": "ain't", "arent": "aren't", "cant": "can't", "couldve": "could've", "couldnt": "couldn't", \
"couldn'tve": "couldn't've", "couldnt've": "couldn't've", "didnt": "didn't", "doesnt": "doesn't", "dont": "don't", "hadnt": "hadn't", \
"hadnt've": "hadn't've", "hadn'tve": "hadn't've", "hasnt": "hasn't", "havent": "haven't", "hed": "he'd", "hed've": "he'd've", \
"he'dve": "he'd've", "hes": "he's", "howd": "how'd", "howll": "how'll", "hows": "how's", "id've": "i'd've", "i'dve": "i'd've", \
"im": "i'm", "ive": "i've", "isnt": "isn't", "itd": "it'd", "itd've": "it'd've", "it'dve": "it'd've", "itll": "it'll", "let's": "let's", \
"maam": "ma'am", "mightnt": "mightn't", "mightnt've": "mightn't've", "mightn'tve": "mightn't've", "mightve": "might've", \
"mustnt": "mustn't", "mustve": "must've", "neednt": "needn't", "notve": "not've", "oclock": "o'clock", "oughtnt": "oughtn't", \
"ow's'at": "'ow's'at", "'ows'at": "'ow's'at", "'ow'sat": "'ow's'at", "shant": "shan't", "shed've": "she'd've", "she'dve": "she'd've", \
"she's": "she's", "shouldve": "should've", "shouldnt": "shouldn't", "shouldnt've": "shouldn't've", "shouldn'tve": "shouldn't've", \
"somebody'd": "somebodyd", "somebodyd've": "somebody'd've", "somebody'dve": "somebody'd've", "somebodyll": "somebody'll", \
"somebodys": "somebody's", "someoned": "someone'd", "someoned've": "someone'd've", "someone'dve": "someone'd've", \
"someonell": "someone'll", "someones": "someone's", "somethingd": "something'd", "somethingd've": "something'd've", \
"something'dve": "something'd've", "somethingll": "something'll", "thats": "that's", "thered": "there'd", "thered've": "there'd've", \
"there'dve": "there'd've", "therere": "there're", "theres": "there's", "theyd": "they'd", "theyd've": "they'd've", \
"they'dve": "they'd've", "theyll": "they'll", "theyre": "they're", "theyve": "they've", "twas": "'twas", "wasnt": "wasn't", \
"wed've": "we'd've", "we'dve": "we'd've", "weve": "we've", "werent": "weren't", "whatll": "what'll", "whatre": "what're", \
"whats": "what's", "whatve": "what've", "whens": "when's", "whered": "where'd", "wheres": "where's", "whereve": "where've", \
"whod": "who'd", "whod've": "who'd've", "who'dve": "who'd've", "wholl": "who'll", "whos": "who's", "whove": "who've", "whyll": "why'll", \
"whyre": "why're", "whys": "why's", "wont": "won't", "wouldve": "would've", "wouldnt": "wouldn't", "wouldnt've": "wouldn't've", \
"wouldn'tve": "wouldn't've", "yall": "y'all", "yall'll": "y'all'll", "y'allll": "y'all'll", "yall'd've": "y'all'd've", \
"y'alld've": "y'all'd've", "y'all'dve": "y'all'd've", "youd": "you'd", "youd've": "you'd've", "you'dve": "you'd've", \
"youll": "you'll", "youre": "you're", "youve": "you've"}
inText = sentence.replace('\n', ' ')
inText = inText.replace('\t', ' ')
inText = inText.strip()
outText = inText
for p in punct:
if (p + ' ' in inText or ' ' + p in inText) or \
(re.search(commaStrip, inText) != None):
outText = outText.replace(p, '')
else:
outText = outText.replace(p, ' ')
outText = periodStrip.sub("", outText, re.UNICODE)
outText = outText.lower().split()
for wordId, word in enumerate(outText):
if word in contractions:
outText[wordId] = contractions[word]
outText = ' '.join(outText)
return outText
def process_answer(answer):
articles = ['a', 'an', 'the']
manualMap = { 'none': '0', 'zero': '0', 'one': '1', 'two': '2', 'three':
'3', 'four': '4', 'five': '5', 'six': '6', 'seven': '7',
'eight': '8', 'nine': '9', 'ten': '10' }
new_answer = process_sentence(answer)
outText = []
for word in new_answer.split():
if word not in articles:
word = manualMap.setdefault(word, word)
outText.append(word)
return ' '.join(outText)
def myimshow(image, ax=plt):
ax.figure()
image = image.to('cpu').numpy()
image = np.moveaxis(image, [0, 1, 2], [2, 0, 1])
image = (image + 1) / 2
image[image < 0] = 0
image[image > 1] = 1
h = ax.imshow(image)
ax.axis('off')
return h
# custom torch Dataset
class MSCOCODataset(td.Dataset):
def __init__(self, images_dir, q_dir, ans_dir, mode='train', image_size=(448, 448), top_num=1000):
super(MSCOCODataset, self).__init__()
self.mode = mode
self.image_size = image_size
self.root_image = os.path.join(images_dir, "%s2014" % mode)
self.top_num=top_num
root_q = os.path.join(q_dir + "%s2014_questions.json" % mode)
root_ans = os.path.join(ans_dir + "%s2014_annotations.json" % mode)
with open(root_q) as f:
self.q_json = json.load(f)['questions']
with open(root_ans) as f:
self.a_json = json.load(f)['annotations']
# answering parsing
self.answers = []
self.vocab_a = defaultdict(int)
for a in self.a_json:
processed = process_answer(a['multiple_choice_answer'])
self.answers.append(processed)
if len(processed.split(" ")) == 1:
self.vocab_a[processed] += 1
self.vocab_a = sorted(self.vocab_a.items(), key=lambda x : x[1], reverse=True)
print(len(self.vocab_a))
self.vocab_a = {self.vocab_a[i][0] : i for i in range(top_num)}
self.top_answers = []
self.top_questions = []
self.top_images = []
for i, each in enumerate(self.answers):
if all(word in self.vocab_a for word in each.split(" ")) and (len(each.split(" ")) == 1):
self.top_answers.append(each)
self.top_questions.append(process_sentence(self.q_json[i]['question']))
self.top_images.append(str(self.q_json[i]['image_id']))
# question parsing
self.vocab_q = set()
for q in self.top_questions:
for each in q.split(" "):
self.vocab_q.add(each)
self.vocab_q = {word : i+1 for i, word in enumerate(self.vocab_q)}
self.vocab_q['#'] = 0 # add padding
self.seq_question = max([len(x.split(" ")) for x in self.top_questions])
def __len__(self):
return len(self.top_questions)
def __repr__(self):
return "MSCOCODataset(mode={}, image_size={})" . \
format(self.mode, self.image_size)
def one_hot_answer(self, inp, mapping):
return torch.Tensor([mapping[inp]])
def one_hot_question(self, inp, mapping):
vec = torch.zeros(len(inp.split(" ")))
for i, word in enumerate(inp.split(" ")):
vec[i] = mapping[word]
return vec
def __getitem__(self, idx):
q = self.top_questions[idx]
a = self.top_answers[idx]
img_id = self.top_images[idx]
img_path = os.path.join(self.root_image, "COCO_%s2014_%s.jpg" % (self.mode, img_id.zfill(12)))
img = Image.open(img_path).convert("RGB")
# normalize each image
transform = tv.transforms.Compose([tv.transforms.CenterCrop(self.image_size),
tv.transforms.ToTensor(),
tv.transforms.Normalize((0.5, 0.5, 0.5),
(0.5, 0.5, 0.5))])
x = transform(img)
one_hot_q = self.one_hot_question(q, self.vocab_q)
one_hot_ans = self.one_hot_answer(a, self.vocab_a)
target_q = torch.zeros(self.seq_question)
target_q[:one_hot_q.shape[0]] = one_hot_q
return x, target_q, len(one_hot_q), one_hot_ans