VQA/data_preprocessing.py at master · haochen07/VQA · GitHub

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

import json
import os
import argparse
import pdb


def main(params):
	train = []
	test = []
	imdir='v7w_%s.jpg'
	print 'Loading annotations and questions...'
	data = json.load(open('dataset_v7w_%s.json' %(params['data_set']), 'r'))["images"]
	# pdb.set_trace()
	pdb.set_trace()

	for image in data:
		# print image.keys()
		for QA in image['qa_pairs']:
			correct_ans = QA['answer']
			question_id = QA['qa_id']
			image_path = imdir%(QA["image_id"])
			question = QA['question']
			# add correct answer
			if image['split'] == 'test':
				test.append({'ques_id': question_id, 'img_path': image_path, 'question': question, 'MC_ans': correct_ans, 'ans': 1})
			else:
				train.append({'ques_id': question_id, 'img_path': image_path, 'question': question, 'MC_ans': correct_ans, 'ans': 1})
			mc_ans = QA['multiple_choices']
			if len(mc_ans) != 3:
				print len(mc_ans)
			# add wrong answers
			for wrong_ans in mc_ans:
				if image['split'] == 'test':
					test.append({'ques_id': question_id, 'img_path': image_path, 'question': question, 'MC_ans': wrong_ans, 'ans': 0})
				else:
					train.append({'ques_id': question_id, 'img_path': image_path, 'question': question, 'MC_ans': wrong_ans, 'ans': 0})

	# subtype = 'val2014'
	# for i in range(len(val_anno['annotations'])):
	#     ans = val_anno['annotations'][i]['multiple_choice_answer']
	#     question_id = val_anno['annotations'][i]['question_id']
	#     image_path = imdir%(subtype, subtype, val_anno['annotations'][i]['image_id'])

	#     question = val_ques['questions'][i]['question']
	#     mc_ans = val_ques['questions'][i]['multiple_choices']

	#     test.append({'ques_id': question_id, 'img_path': image_path, 'question': question, 'MC_ans': mc_ans})


	json.dump(train, open('vqa_raw_train.json', 'w'))
	json.dump(test, open('vqa_raw_test.json', 'w'))

if __name__ == "__main__":
	parser = argparse.ArgumentParser()
	parser.add_argument('--data_set', default = 'telling',help = 'which data set, telling or pointing')
	args = parser.parse_args()
	params = vars(args) # convert to ordinary dict
	main(params)