“NameError:name'images\u train'未定义”，但已定义

Question

这是我的密码：

def get_parser():
    parser.add_argument('--batch_size', default=64, help='batch size to train network')
    return args
while True:
    try:
        images_train = np.load(path + str(batch_idx)+'images.npy')
        labels_train = np.load(path + str(batch_idx)+'labels.npy')
        break
    except:
        batch_idx = sorted(os.listdir(path))
        try:
            batch_idx = int(batch_idx[0].split('images.npy')[0])
        except:
            batch_idx = int(batch_idx[0].split('labels.npy')[0])
        batch_idx += 400 
batch_idx += 1
feed_dict = {images: images_train[0:args.batch_size], 
             labels: labels_train[0:args.batch_size], trainable: True}

文件为npz格式，形状为（64112112,3） 代码可以用args.batch\u大小=64，但将出现错误args.batch\u大小=32，如下所示

Traceback (most recent call last):

  File "<ipython-input-1-646ff1e212a5>", line 1, in <module>
    runfile('D:/InsightFace_TF-master/train_nets_npy_2.py', wdir='D:/InsightFace_TF-master')

  File "C:\ProgramData\Anaconda3\lib\site-packages\spyder\utils\site\sitecustomize.py", line 705, in runfile
    execfile(filename, namespace)

  File "C:\ProgramData\Anaconda3\lib\site-packages\spyder\utils\site\sitecustomize.py", line 102, in execfile
    exec(compile(f.read(), filename, 'exec'), namespace)

  File "D:/InsightFace_TF-master/train_nets_npy_2.py", line 298, in <module>
    feed_dict = {images: images_train[0:args.batch_size], labels: labels_train[0:args.batch_size], trainable: True}

NameError: name 'images_train' is not defined