Zero equation model error - 'function' object has no attribute 'free_symbols

I am attempting a zero equation case of my own. My solver class is as follows.

class FlowOverBlockSolver(Solver):
  train_domain = FlowOverBlockTrain
  val_domain = FlowOverBlockVal
  #inference_domain = FlowOverBlockInference
  #monitor_domain = FlowOverBlockMonitor

  def __init__(self, **config):
    super(FlowOverBlockSolver, self).__init__(**config)

    self.equations = (NavierStokes(nu='nu', rho=rho, dim=3, time=False).make_node()
                        + IntegralContinuity(dim=3).make_node()
                        + ZeroEquation(nu=nu, dim=3, time=False, max_distance=0.75).make_node()
                        + [Node.from_sympy(geo.sdf, 'normal_distance')])
    flow_net = self.arch.make_node(name='flow_net',
                                   inputs=['x', 'y', 'z'],
                                   outputs=['u', 'v', 'w','p'])
    self.nets = [flow_net]

Like the LDC example I’ve ran my case with just the naiver stokes equation successfully before adjusting it for Zero Equation model. I added nu to the validation data and then made the necessary adjustments to the solver class detailed above.
However, when attempting to train, I get the following error:

(SimNetv21) [Usr@pc RTX_comparisons]$ CUDA_VISIBLE_DEVICES=0 python FOB3D_MF_RE50_slip_Zeq_25hz.py
WARNING:tensorflow:From /u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/site-packages/simnet-21.6-py3.7.egg/simnet/optimizer.py:353: The name tf.train.AdamOptimizer is deprecated. Please use tf.compat.v1.train.AdamOptimizer instead.

WARNING:tensorflow:From /u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/site-packages/simnet-21.6-py3.7.egg/simnet/optimizer.py:361: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.

/u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/importlib/_bootstrap.py:219: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  return f(*args, **kwds)


CONFIGS: FullyConnectedArch, /u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/site-packages/simnet-21.6-py3.7.egg/simnet/architecture/fully_connected.py
  activation_fn: swish
  layer_size: 512
  nr_layers: 6
  skip_connections: False
  weight_norm: True
  adaptive_activations: False
CONFIGS: ExponentialDecayLR, /u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/site-packages/simnet-21.6-py3.7.egg/simnet/learning_rate.py
  start_lr: 0.001
  end_lr: 0.0
  decay_steps: 7500
  decay_rate: 0.95
CONFIGS: AdamOptimizer, /u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/site-packages/simnet-21.6-py3.7.egg/simnet/optimizer.py
  beta1: 0.9
  beta2: 0.999
  epsilon: 1e-08
  amp: False
Traceback (most recent call last):
  File "FOB3D_MF_RE50_slip_Zeq_25hz.py", line 255, in <module>
    ctr.run()
  File "/u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/site-packages/simnet-21.6-py3.7.egg/simnet/controller.py", line 88, in run
    self.solver = self._solver(config=self.config)
  File "FOB3D_MF_RE50_slip_Zeq_25hz.py", line 235, in __init__
    + [Node.from_sympy(geo.sdf, 'normal_distance')])
  File "/u/Usr/anaconda3/envs/SimNetv21/lib/python3.7/site-packages/simnet-21.6-py3.7.egg/simnet/node.py", line 64, in from_sympy
    free_symbols = [x.name for x in tf_eq.free_symbols]
AttributeError: 'function' object has no attribute 'free_symbols'

My question is how should I resolve the following error?
The major difference between my case and the LDC case is that I meshed my geometry from an STL.

File ".../simnet-21.6-py3.7.egg/simnet/node.py", line 64, in from_sympy
    free_symbols = [x.name for x in tf_eq.free_symbols]
AttributeError: 'function' object has no attribute 'free_symbols'