def setup(self, bottom, top):
if self.phase == 0: # train phase
import train_config
config = train_config.Config()
else: # val or test phase
import test_config
config = test_config.Config()
self.N = config.N
self.context_dim = config.context_dim
self.spatial_dim = config.spatial_dim
self.HW = config.spatial_pool_map * config.spatial_pool_map
self.T = config.T
self.key_word_thresh = config.key_word_thresh
self.hard_word_att_idx = []
# query_aware_context features for every image location
top[0].reshape(self.N, self.context_dim + self.spatial_dim, self.HW)
def setup(self, bottom, top):
if self.phase == 0: # train phase
import train_config
config = train_config.Config()
else: # val or test phase
import test_config
config = test_config.Config()
self.N = config.N
self.context_dim = config.context_dim
self.spatial_dim = config.spatial_dim
self.HW = config.spatial_pool_map * config.spatial_pool_map
self.T = config.T
self.key_word_thresh = config.key_word_thresh
self.hard_word_att_idx = []
# query-aware appear pool for every word
top[0].reshape(self.N, self.context_dim, self.T)
# query-aware spatial position pool for every word
top[1].reshape(self.N, self.spatial_dim, self.T)
scores_net.forward()
scores_val = scores_net.blobs['scores'].data.copy()
scores_val = scores_val[:num_proposal, ...].reshape(-1)
# Sort the scores for the proposals
if config.use_nms:
top_ids = eval_tools.nms(proposal.astype(np.float32),
scores_val, config.nms_thresh)
else:
top_ids = np.argsort(scores_val)[::-1]
# Evaluate on bounding boxes
for n_eval_num in range(len(eval_bbox_num_list)):
eval_bbox_num = eval_bbox_num_list[n_eval_num]
bbox_correct[n_eval_num] += \
np.any(proposal_IoUs[top_ids[:eval_bbox_num]] >= config.correct_iou_thresh)
bbox_total += 1
print('Final results on the whole test set')
result_str = ''
for n_eval_num in range(len(eval_bbox_num_list)):
result_str += 'recall@%s = %f\n' % \
(str(eval_bbox_num_list[n_eval_num]), bbox_correct[n_eval_num]/bbox_total)
print(result_str)
if __name__ == '__main__':
config = test_config.Config()
inference(config)
from fastNLP.core.losses import LossBase
from fastNLP.core.metrics import MetricBase
from fastNLP.core.optimizer import Optimizer
from fastNLP.core.batch import Batch
from fastNLP.core.sampler import RandomSampler
from fastNLP import Trainer
from fastNLP import Tester
from copy import deepcopy
from fastNLP import CrossEntropyLoss
from fastNLP import AccuracyMetric
from fastNLP.core import Adam
from fastNLP.core import SGD
from fastNLP.core.callback import EarlyStopCallback
opt = test_config.Config()
def test():
model_path = opt.model_path
test_data = pickle.load(open(opt.test_data_path, 'rb'))
vocab = pickle.load(open(opt.vocab, 'rb'))
word2idx = vocab.word2idx
idx2word = vocab.idx2word
input_size = len(word2idx)
vocab_size = opt.class_num
class_num = opt.class_num
embedding_dim = opt.embedding_dim