Cannot convert a symbolic Tensor (caps/strided_slice:0)

[Algabri]

I am trying to run this code using recent versions, but I got this error:

Traceback (most recent call last):
  File "/home/redhwan/2/HPE/FSA-Net/demo/demo_FSANET_ssd.py", line 222, in <module>
    main()
  File "/home/redhwan/2/HPE/FSA-Net/demo/demo_FSANET_ssd.py", line 128, in main
    model1 = FSA_net_Capsule(image_size, num_classes, stage_num, lambda_d, S_set)()
  File "/home/redhwan/2/HPE/FSA-Net/demo/../lib/FSANET_model.py", line 433, in __call__
    ssr_aggregation_model = self.ssr_aggregation_model_build((self.num_primcaps,64))
  File "/home/redhwan/2/HPE/FSA-Net/demo/../lib/FSANET_model.py", line 458, in ssr_aggregation_model_build
    capsule = CapsuleLayer(self.num_capsule, self.dim_capsule, routings=self.routings, name='caps')(input_primcaps)
  File "/home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py", line 925, in __call__
    return self._functional_construction_call(inputs, args, kwargs,
  File "/home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/keras/engine/base_layer.py", line 1117, in _functional_construction_call
    outputs = call_fn(cast_inputs, *args, **kwargs)
  File "/home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py", line 258, in wrapper
    raise e.ag_error_metadata.to_exception(e)
NotImplementedError: in user code:

    /home/redhwan/2/HPE/FSA-Net/demo/capsulelayers.py:189 call  *
        b = tf.zeros(shape=[K.shape(inputs_hat)[0],
    /home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/util/dispatch.py:201 wrapper  **
        return target(*args, **kwargs)
    /home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/ops/array_ops.py:2747 wrapped
        tensor = fun(*args, **kwargs)
    /home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/ops/array_ops.py:2794 zeros
        output = _constant_if_small(zero, shape, dtype, name)
    /home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/ops/array_ops.py:2732 _constant_if_small
        if np.prod(shape) < 1000:
    <__array_function__ internals>:180 prod
        
    /home/redhwan/.local/lib/python3.8/site-packages/numpy/core/fromnumeric.py:3045 prod
        return _wrapreduction(a, np.multiply, 'prod', axis, dtype, out,
    /home/redhwan/.local/lib/python3.8/site-packages/numpy/core/fromnumeric.py:86 _wrapreduction
        return ufunc.reduce(obj, axis, dtype, out, **passkwargs)
    /home/redhwan/.local/lib/python3.8/site-packages/tensorflow/python/framework/ops.py:845 __array__
        raise NotImplementedError(

    NotImplementedError: Cannot convert a symbolic Tensor (caps/strided_slice:0) to a numpy array. This error may indicate that you're trying to pass a Tensor to a NumPy call, which is not supported

I find a similar to this error here
tensorflow ==2.3

Note: if I use numpy == 1.19.5, it works fine with demo codes, but it is not working when I run the training code. So, I got this error:

   raise ImportError(
ImportError: this version of pandas is incompatible with numpy < 1.20.3
your numpy version is 1.19.5.
Please upgrade numpy to >= 1.20.3 to use this pandas version

Any help, please, I would like to run it using tensorflow ==2.3

[Algabri]

I fixed this issue following this