UnboundLocalError: local variable 'eye_att' referenced before assignment

[G-force78]

Loading test data: 100% 21/21 [00:00<00:00, 3761.55it/s]
==> Start Test, save results to workspace/results
0% 0/21 [00:00<?, ?it/s]Traceback (most recent call last):
File "/content/SyncTalk/main.py", line 207, in
trainer.test(test_loader)
File "/content/SyncTalk/nerf_triplane/utils.py", line 1067, in test
preds, preds_depth = self.test_step(data)
File "/content/SyncTalk/nerf_triplane/utils.py", line 981, in test_step
outputs = self.model.render(rays_o, rays_d, auds, bg_coords, poses, eye=eye, index=index, staged=True, bg_color=bg_color, perturb=perturb, **vars(self.opt))
File "/content/SyncTalk/nerf_triplane/renderer.py", line 673, in render
results = _run(rays_o, rays_d, auds, bg_coords, poses, **kwargs)
File "/content/SyncTalk/nerf_triplane/renderer.py", line 259, in run_cuda
sigmas, rgbs, ambients_aud, ambients_eye, uncertainties = self(xyzs, dirs, enc_a, ind_code, eye)
File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 1511, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 1520, in _call_impl
return forward_call(*args, **kwargs)
File "/content/SyncTalk/nerf_triplane/network.py", line 337, in forward
sigma_result = self.density(x, enc_a, e, enc_x)
File "/content/SyncTalk/nerf_triplane/network.py", line 389, in density
'ambient_eye' : eye_att.norm(dim=-1, keepdim=True),
UnboundLocalError: local variable 'eye_att' referenced before assignment
0% 0/21 [00:00<?, ?it/s]