大家好，我正在使用DeeplabV3Plus架构和Tensorflow（Keras）进行语义分割。我在另一个数据集上做得很好，但现在我想用我自己的数据集来做。但在加载数据的第一步，它显示了一个奇怪的错误。函数是
tf.数据.数据集.来自_Tensor_切片
其误差为：

ValueError                                Traceback (most recent call last)
~\AppData\Local\Temp\ipykernel_20192\306109049.py in <module>
     57 
     58 train_dataset = data_generator(train_images, train_masks)
---> 59 val_dataset = data_generator(val_images, val_masks)
     60 
     61 print("Train Dataset:", train_dataset)

~\AppData\Local\Temp\ipykernel_20192\306109049.py in data_generator(image_list, mask_list)
     50 
     51 def data_generator(image_list, mask_list):
---> 52     dataset = tf.data.Dataset.from_tensor_slices((image_list, mask_list))
     53     dataset = dataset.map(load_data, num_parallel_calls=tf.data.experimental.AUTOTUNE)
     54     dataset = dataset.batch(BATCH_SIZE, drop_remainder=True)

~\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\data\ops\dataset_ops.py in from_tensor_slices(tensors, name)
    812       Dataset: A `Dataset`.
    813     """
--> 814     return TensorSliceDataset(tensors, name=name)
    815 
    816   class _GeneratorState(object):

~\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\data\ops\dataset_ops.py in __init__(self, element, is_files, name)
   4720       batch_dim.assert_is_compatible_with(
   4721           tensor_shape.Dimension(
-> 4722               tensor_shape.dimension_value(t.get_shape()[0])))
   4723 
   4724     variant_tensor = gen_dataset_ops.tensor_slice_dataset(

~\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\framework\tensor_shape.py in assert_is_compatible_with(self, other)
    298     if not self.is_compatible_with(other):
    299       raise ValueError("Dimensions %s and %s are not compatible" %
--> 300                        (self, other))
    301 
    302   def merge_with(self, other):

ValueError: Dimensions 37 and 50 are not compatible

错误是"尺寸37和50不兼容"，我搜索了这个，但找不到解决方案。代码：

import os
import cv2
import numpy as np
from glob import glob
from scipy.io import loadmat
import matplotlib.pyplot as plt

import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers

IMAGE_SIZE = 512
BATCH_SIZE = 4
NUM_CLASSES = 20
DATA_DIR = r'C:/Users/Joshi/Desktop/CARLA_0.9.13/WindowsNoEditor/PythonAPI/examples/out'
NUM_TRAIN_IMAGES = 250
NUM_VAL_IMAGES = 50

train_images = sorted(glob(os.path.join(DATA_DIR, "out/*")))[:NUM_TRAIN_IMAGES]
train_masks = sorted(glob(os.path.join(DATA_DIR, "Seman/*")))[:NUM_TRAIN_IMAGES]
val_images = sorted(glob(os.path.join(DATA_DIR, "out/*")))[
    NUM_TRAIN_IMAGES : NUM_VAL_IMAGES + NUM_TRAIN_IMAGES
]
val_masks = sorted(glob(os.path.join(DATA_DIR, "Seman/*")))[
    NUM_TRAIN_IMAGES : NUM_VAL_IMAGES + NUM_TRAIN_IMAGES
]



def read_image(image_path, mask=False):
    image = tf.io.read_file(image_path)
    if mask:
        image = tf.image.decode_png(image, channels=1)
        image.set_shape([None, None, 1])
        image = tf.image.resize(images=image, size=[IMAGE_SIZE, IMAGE_SIZE])
    else:
        image = tf.image.decode_png(image, channels=3)
        image.set_shape([None, None, 3])
        image = tf.image.resize(images=image, size=[IMAGE_SIZE, IMAGE_SIZE])
        image = image / 127.5 - 1
    return image

def load_data(image_list, mask_list):
    image = read_image(image_list)
    mask = read_image(mask_list, mask=True)
    return image, mask

def data_generator(image_list, mask_list):
    dataset = tf.data.Dataset.from_tensor_slices((image_list, mask_list))
    dataset = dataset.map(load_data, num_parallel_calls=tf.data.experimental.AUTOTUNE)
    dataset = dataset.batch(BATCH_SIZE, drop_remainder=True)
    return dataset

train_dataset = data_generator(train_images, train_masks)
val_dataset = data_generator(val_images, val_masks)

print("Train Dataset:", train_dataset)
print("Val Dataset:", val_dataset)