Error when run_cfgs in eval_nvs mode

[Realinxc]

python run_cf3dgs.py --source data/Tanks/Francis --mode eval_nvs --data_type tanks --model_path ./output/progressive/Tanks_Francis/chkpnt/ep00_init.pth
Using cache found in /data/linxincheng/.cache/torch/hub/intel-isl_MiDaS_master
Rotation type : 6d
Reading camera 150/150
Loading Training Cameras
Loading Test Cameras
Number of points at initialisation : 19361
Test images: 19
['000409', '000425', '000441', '000457', '000473', '000489', '000505', '000521', '000537', '000553', '000569', '000585', '000601', '000617', '000633', '000649', '000665', '000681', '000697']
Using cache found in /data/linxincheng/.cache/torch/hub/intel-isl_MiDaS_master
/data/linxincheng/anaconda3/envs/cf3dgs/lib/python3.7/site-packages/timm/models/_factory.py:121: UserWarning: Mapping deprecated model name vit_base_resnet50_384 to current vit_base_r50_s16_384.orig_in21k_ft_in1k.
**kwargs,
Using cache found in /data/linxincheng/.cache/torch/hub/intel-isl_MiDaS_master
Training progress: 0%| | 0/3800 [00:00<?, ?it/s]Traceback (most recent call last):
File "run_cf3dgs.py", line 53, in
trainer.eval_nvs()
File "/data/linxincheng/GaussianCodes/CF-3DGS/trainer/cf3dgs_trainer.py", line 604, in eval_nvs
reset=False,
File "/data/linxincheng/GaussianCodes/CF-3DGS/trainer/cf3dgs_trainer.py", line 105, in train_step
override_color=colors_precomp)
File "/data/linxincheng/GaussianCodes/CF-3DGS/scene/gaussian_model_cf.py", line 854, in render
camera_center = self.gaussians.get_RT(fidx).inverse()[
torch._C._LinAlgError: linalg.inv: The diagonal element 1 is zero, the inversion could not be completed because the input matrix is singular.
Training progress: 0%|

I did not change anything in the repo，I wonder if i should retrain the mode or do someting to continue？