运行tensorflow代码时出现的错误ImportError: cannot import name 'tokenizer_from_json'

import keras
import numpy as np
import tensorflow as tf
from keras_preprocessing import image
from tensorflow.keras import datasets, layers, optimizers, Sequential, metrics

#导入数据
(X_train,y_train), (X_test,y_test) = datasets.mnist.load_data()

#数据转换：归一化
X_train = tf.convert_to_tensor(X_train, dtype=tf.float32)/255.
dataset = tf.data.Dataset.from_tensor_slices((X_train, y_train))
dataset = dataset.batch(32).repeat(10)

#获取图片的大小
in_shape = X_train.shape[1:] #形状为(28,28)

#获取数字图片的种类
n_classes = len(np.unique(y_train)) #类别数为10
model = Sequential() #搭建空顺序模型
model.add( layers.Flatten(input_shape = in_shape))
model.add( layers.Dense(n_classes, activation = 'softmax'))

#设置优化器，学习率为0.01
optimizer = optimizer.SGD(lr = 0.01)

#设置算法性能的评估标准：分类精确度
acc_meter = metrics.Accuracy

for step, (x,y) in enumerate(dataset):
    with tf.GradientTape() as tape:
        #计算模型输出
        out = model(x)
        #将标签转化成独热编码
        y_onehot = tf.one_hot(y, depth=10)
        #计算损失
        loss = tf.square(out - y_onehot)
        #计算损失均值
        loss = tf.reduce_sum(loss) / 32
        
    acc_meter.update_state(tf.argmax(out, axis=1), y)
    grads = tape.gradient(loss, model.trainable_variables)
    optimizer.apply_gradients(zip(grades, model.trainable_variables))
    
    if step % 200 == 0:
        print('step {0}, loss:{1:.3f},ass:{2:.2f} %'.format(step, float(loss), acc_meter.result().numpy() * 100))
        acc_meter.reset_states()

产生的错误如图：