def create_dictionary(dataroot):
dictionary = Dictionary()
questions = []
files = [
'v2_OpenEnded_mscoco_train2014_questions.json',
'v2_OpenEnded_mscoco_val2014_questions.json',
'v2_OpenEnded_mscoco_test2015_questions.json',
'v2_OpenEnded_mscoco_test-dev2015_questions.json'
]
for path in files:
question_path = os.path.join(dataroot, path)
qs = json.load(open(question_path))['questions']
for q in qs:
dictionary.tokenize(q['question'], True)
return dictionary
def make_json(logits, qIds, dataloader):
utils.assert_eq(logits.size(0), len(qIds))
results = []
for i in range(logits.size(0)):
result = {}
result['question_id'] = qIds[i].item()
result['answer'] = get_answer(logits[i], dataloader)
results.append(result)
return results
if __name__ == '__main__':
args = parse_args()
torch.backends.cudnn.benchmark = True
dictionary = Dictionary.load_from_file('data/dictionary.pkl')
eval_dset = VQAFeatureDataset_withmask(args.split, dictionary)
n_device = torch.cuda.device_count()
batch_size = args.batch_size * n_device
constructor = 'build_%s' % args.model
model = getattr(caq_model, constructor)(train_dset, args.num_hid).cuda()
eval_loader = DataLoader(eval_dset, batch_size, shuffle=False, num_workers=1, collate_fn=utils.trim_collate)
def process(args, model, eval_loader):
model_path = args.input+'/model%s.pth' % \
('' if 0 > args.epoch else '_epoch%d' % args.epoch)
print('loading %s' % model_path)
model_data = torch.load(model_path)