def test(test_exe, test_program, test_out, args):
featureLs = None
featureRs = None
out_feature, test_reader, flods, flags = test_out
for idx, data in enumerate(test_reader()):
res = []
res.append(
test_exe.run(test_program,
feed={u'image_test': data[0][u'image_test1']},
fetch_list=out_feature))
res.append(
test_exe.run(test_program,
feed={u'image_test': data[0][u'image_test2']},
fetch_list=out_feature))
res.append(
test_exe.run(test_program,
feed={u'image_test': data[0][u'image_test3']},
fetch_list=out_feature))
res.append(
test_exe.run(test_program,
feed={u'image_test': data[0][u'image_test4']},
fetch_list=out_feature))
featureL = np.concatenate((res[0][0], res[1][0]), 1)
featureR = np.concatenate((res[2][0], res[3][0]), 1)
if featureLs is None:
featureLs = featureL
else:
featureLs = np.concatenate((featureLs, featureL), 0)
if featureRs is None:
featureRs = featureR
else:
featureRs = np.concatenate((featureRs, featureR), 0)
result = {'fl': featureLs, 'fr': featureRs, 'fold': flods, 'flag': flags}
scipy.io.savemat(args.feature_save_dir, result)
ACCs = evaluation_10_fold(args.feature_save_dir)
with open(os.path.join(args.save_ckpt, 'log.txt'), 'a+') as f:
f.writelines('eval model {}\n'.format(args.model))
for i in range(len(ACCs)):
print('{} {}'.format(i + 1, ACCs[i] * 100))
with open(os.path.join(args.save_ckpt, 'log.txt'), 'a+') as f:
f.writelines('{} {}\n'.format(i + 1, ACCs[i] * 100))
print('--------')
print('AVE {}'.format(np.mean(ACCs) * 100))
with open(os.path.join(args.save_ckpt, 'log.txt'), 'a+') as f:
f.writelines('--------\n')
f.writelines('AVE {}\n'.format(np.mean(ACCs) * 100))
return np.mean(ACCs) * 100
else:
featureLs = np.concatenate((featureLs, featureL), 0)
if featureRs is None:
featureRs = featureR
else:
featureRs = np.concatenate((featureRs, featureR), 0)
result = {
'fl': featureLs,
'fr': featureRs,
'fold': folds,
'flag': flags
}
# save tmp_result
scipy.io.savemat('./result/tmp_result.mat', result)
accs = evaluation_10_fold('./result/tmp_result.mat')
_print(' ave: {:.4f}'.format(np.mean(accs) * 100))
# save model
if epoch % SAVE_FREQ == 0:
msg = 'Saving checkpoint: {}'.format(epoch)
_print(msg)
if multi_gpus:
net_state_dict = net.module.state_dict()
else:
net_state_dict = net.state_dict()
if not os.path.exists(save_dir):
os.mkdir(save_dir)
torch.jit.save(torch.jit.script(net),
os.path.join(save_dir, 'mobileface%03d.pt' % epoch))
torch.save({
res = [net(d).data.cpu().numpy() for d in data]
featureL = np.concatenate((res[0], res[1]), 1)
featureR = np.concatenate((res[2], res[3]), 1)
if featureLs is None:
featureLs = featureL
else:
featureLs = np.concatenate((featureLs, featureL), 0)
if featureRs is None:
featureRs = featureR
else:
featureRs = np.concatenate((featureRs, featureR), 0)
result = {'fl': featureLs, 'fr': featureRs, 'fold': folds, 'flag': flags}
# save tmp_result
scipy.io.savemat('./result/tmp_result.mat', result)
accs = evaluation_10_fold('./result/tmp_result.mat')
_print(' ave: {:.4f}'.format(np.mean(accs) * 100))
# save model
if epoch % SAVE_FREQ == 0:
msg = 'Saving checkpoint: {}'.format(epoch)
_print(msg)
if multi_gpus:
net_state_dict = net.module.state_dict()
else:
net_state_dict = net.state_dict()
if not os.path.exists(save_dir):
os.mkdir(save_dir)
torch.save({
'epoch': epoch,
'net_state_dict': net_state_dict},