det_field = COCOControlSequenceField(detections_path=os.path.join(coco_root, 'coco_detections.hdf5'),
classes_path=os.path.join(coco_root, 'object_class_list.txt'),
pad_init=False, padding_idx=-1, all_boxes=False, fix_length=20)
text_field = TextField(init_token='<bos>', eos_token='<eos>', lower=True, remove_punctuation=True, fix_length=20)
dataset = COCOEntities(image_field, det_field, text_field,
img_root='',
ann_root=os.path.join(coco_root, 'annotations'),
entities_file=os.path.join(coco_root, 'coco_entities.json'),
id_root=os.path.join(coco_root, 'annotations'))
train_dataset, val_dataset, _ = dataset.splits
text_field.build_vocab(train_dataset, val_dataset, min_freq=5)
test_dataset = COCOEntities(image_field, det_field, RawField(),
img_root='',
ann_root=os.path.join(coco_root, 'annotations'),
entities_file=os.path.join(coco_root, 'coco_entities.json'),
id_root=os.path.join(coco_root, 'annotations'),
filtering=True)
_, val_dataset, _ = test_dataset.splits
if opt.sample_rl or opt.sample_rl_nw:
train_dataset.fields['text'] = RawField()
train_dataset_raw = PairedDataset(train_dataset.examples, {'image': image_field, 'detection': det_field, 'text': RawField()})
ref_caps_train = list(train_dataset_raw.text)
cider_train = evaluation.Cider(evaluation.PTBTokenizer.tokenize(ref_caps_train))
dataloader_train = DataLoader(train_dataset, batch_size=opt.batch_size, shuffle=False, num_workers=opt.nb_workers)
idx_vs_path=os.path.join(flickr_root, 'idx_2_vs_flickr.json'),
cap_verb_path=os.path.join(flickr_root, 'cap_2_verb_flickr.json'),
cap_classes_path=os.path.join(flickr_root, 'cap_2_classes_flickr.json'),
idx_v_og_path=os.path.join(flickr_root, 'idx_2_v_og_flickr.json'),
vocab_list_path=os.path.join(flickr_root, 'vocab_tv_flickr.json'),
fix_length=10,
visual=False)
text_field = TextField(init_token='<bos>',
eos_token='<eos>',
lower=True,
remove_punctuation=True,
fix_length=20)
test_dataset = FlickrEntities(image_field,
RawField(),
det_field,
img_root='',
ann_file=os.path.join(
flickr_root, 'flickr30k_annotations.json'),
entities_root=flickr_entities_root,
verb_filter=True)
train_dataset, val_dataset, test_dataset = test_dataset.splits
test_dataset = DictionaryDataset(test_dataset.examples, test_dataset.fields,
'image')
dataloader_test = DataLoader(test_dataset,
batch_size=opt.batch_size,
num_workers=opt.nb_workers)
train_dataset = DictionaryDataset(train_dataset.examples, train_dataset.fields,
'image')
precomp_glove_path=os.path.join(coco_root, 'object_class_glove.pkl'),
vocab_path=os.path.join(coco_root, 'vocab_tv.json'),
vlem_2_v_og_path=os.path.join(coco_root, 'vlem_2_vog_coco.json'),
cls_seq_path=os.path.join('saved_data/coco', 'img_cap_v_2_class_self.json'),
fix_length=10, max_detections=20, gt_verb=opt.gt)
text_field = TextField(init_token='<bos>', eos_token='<eos>', lower=True, remove_punctuation=True, fix_length=20)
# define the datasets
dataset = COCOEntities(image_field, det_field, text_field,
img_root='',
ann_root=os.path.join(coco_root, 'annotations'),
entities_file=os.path.join(coco_root, 'coco_entities.json'),
id_root=os.path.join(coco_root, 'annotations'))
test_dataset = COCOEntities(image_field, det_field, RawField(),
img_root='',
ann_root=os.path.join(coco_root, 'annotations'),
entities_file=os.path.join(coco_root, 'coco_entities.json'),
id_root=os.path.join(coco_root, 'annotations'),
filtering=True,
det_filtering=opt.det)
train_dataset, val_dataset, _ = dataset.splits
text_field.build_vocab(train_dataset, val_dataset, min_freq=5)
# define the dataloader
_, _, test_dataset = test_dataset.splits
test_dataset = DictionaryDataset(test_dataset.examples, test_dataset.fields, 'image')
dataloader_test = DataLoader(test_dataset, batch_size=opt.batch_size, num_workers=opt.nb_workers)