How to cascade two pretrained networks? - python-3.x

Question

Very new to deep learning, I was researching into cascading multiple neural networks and was wondering how to actually put it into code. For a lesser example, say I imported a vgg and resnet model trained on imagenet. Is there an easy way to do this?
I have tried playing around with it and concatenating the outputs of 3 of these networks, putting them in a dense layer, and completing the model. But, I'm being thrown a error.
model1 = load_model(save_path1)
model2 = load_model(save_path2)
model1_out = model1.output
model2_out = model2.output
concat = tf.keras.layers.Concatenate()([model1_out, model2_out])
concatout = Dense(10, activation='softmax')(concat)
combined = models.Model(inputs=[model1.input, model2.input], outputs=[concatout])
combined.compile(optimizer='adam',
loss="categorical_crossentropy",
metrics=['accuracy'])
test_scores = combined.evaluate(x_test, y_test, verbose=2)
print('Loss for test dataset:', test_scores[0])
print('Accuracy for test dataset:', test_scores[1])
test_scores = combined.evaluate(x_test, y_test, verbose=2)
print('Loss for test dataset:', test_scores[0])
print('Accuracy for test dataset:', test_scores[1])
Error: File
ile "C:\\pythonProject1\backup.py", line 72, in <module>
test_scores = combined.evaluate(x_test, y_test, verbose=2)
File "C:\Users\\miniconda3\envs\tf1\lib\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler
raise e.with_traceback(filtered_tb) from None
File "C:\Users\\AppData\Local\Temp\__autograph_generated_file1poeqzyb.py", line 15, in tf__test_function
retval_ = ag__.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
ValueError: in user code:
File "C:\Users\\miniconda3\envs\tf1\lib\site-packages\keras\engine\training.py", line 1557, in test_function *
return step_function(self, iterator)
File "C:\Users\\miniconda3\envs\tf1\lib\site-packages\keras\engine\training.py", line 1546, in step_function **
outputs = model.distribute_strategy.run(run_step, args=(data,))
File "C:\Users\\miniconda3\envs\tf1\lib\site-packages\keras\engine\training.py", line 1535, in run_step **
outputs = model.test_step(data)
File "C:\Users\\miniconda3\envs\tf1\lib\site-packages\keras\engine\training.py", line 1499, in test_step
y_pred = self(x, training=False)
File "C:\Users\\miniconda3\envs\tf1\lib\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler
raise e.with_traceback(filtered_tb) from None
File "C:\Users\\miniconda3\envs\tf1\lib\site-packages\keras\engine\input_spec.py", line 200, in assert_input_compatibility
raise ValueError(f'Layer "{layer_name}" expects {len(input_spec)} input(s),'
ValueError: Layer "model" expects 2 input(s), but it received 1 input tensors. Inputs received: [<tf.Tensor 'IteratorGetNext:0' shape=(None, 48, 48, 3) dtype=float32>]