Machine learning 如何从代码中删除model.fit_generator（）函数导致的ValueError？

我的代码是：

# train our model
print('Dataset: ', len(train_imgs))
print('Descriptions: train=', len(train_descriptions))
print('Photos: train=', len(train_features))
print('Vocabulary Size:', vocab_size)
print('Description Length: ', max_length)
model = define_model(vocab_size, max_length)
epochs = 10
steps = len(train_descriptions)

# making a directory models to save our models
os.mkdir("E:\CAPSTONE PRJ\models")
for i in range(epochs):
    generator = data_generator(train_descriptions, train_features, tokenizer, max_length)
    model.fit_generator(generator, epochs=1, steps_per_epoch= steps, verbose=1)
    model.save("models/model_" + str(i) + ".h5")

即将出现的错误是：

ValueError: No gradients provided for any variable: ['embedding_9/embeddings:0', 'dense_27/kernel:0', 'dense_27/bias:0', 'lstm_9/lstm_cell_10/kernel:0', 'lstm_9/lstm_cell_10/recurrent_kernel:0', 'lstm_9/lstm_cell_10/bias:0', 'dense_28/kernel:0', 'dense_28/bias:0', 'dense_29/kernel:0', 'dense_29/bias:0'].

---

如何删除此错误？

包括

define\u model

的代码，并删除与问题无关的代码，如print语句和os.mkdir/model.save