can't run training command example

[yfq512]

(mvsnerf) root13456544@/work/mvsnerf# python train_mvs_nerf_pl.py --expname mvs-nerf-is-all-your-need --num_epochs 6 --use_viewdirs --dataset_name dtu --datadir dtu
Found ckpts []
GPU available: True, used: True
TPU available: False, using: 0 TPU cores
Traceback (most recent call last):
  File "train_mvs_nerf_pl.py", line 320, in <module>
    trainer.fit(system)
  File "/opt/conda/envs/mvsnerf/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 458, in fit
    self._run(model)
  File "/opt/conda/envs/mvsnerf/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 704, in _run
    self.data_connector.prepare_data(model)
  File "/opt/conda/envs/mvsnerf/lib/python3.8/site-packages/pytorch_lightning/trainer/connectors/data_connector.py", line 57, in prepare_data
    model.prepare_data()
  File "train_mvs_nerf_pl.py", line 80, in prepare_data
    self.train_dataset = dataset(root_dir=train_dir, split='train', max_len=-1 , downSample=args.imgScale_train)
  File "/cmdata/docker/yfq/mvsnerf/data/dtu.py", line 42, in __init__
    self.build_proj_mats()
  File "/cmdata/docker/yfq/mvsnerf/data/dtu.py", line 98, in build_proj_mats
    self.proj_mats, self.intrinsics = np.stack(proj_mats), np.stack(intrinsics)
  File "<__array_function__ internals>", line 200, in stack
  File "/opt/conda/envs/mvsnerf/lib/python3.8/site-packages/numpy/core/shape_base.py", line 458, in stack
    arrays = [asanyarray(arr) for arr in arrays]
  File "/opt/conda/envs/mvsnerf/lib/python3.8/site-packages/numpy/core/shape_base.py", line 458, in <listcomp>
    arrays = [asanyarray(arr) for arr in arrays]
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (2,) + inhomogeneous part.

can you help me!

[murphyl4]

I meet the same problem. Have you already solved it?

[Dongjiahua]

I think it may be a problem with the version of numpy. I solve it by simply modify the stack operation to:

self.proj_mats, self.intrinsics = proj_mats, intrinsics
self.world2cams, self.cam2worlds = world2cams, cam2worlds

Another thing to modify is the c2ws_all assignment in getitem. You can replace it by:

c2ws_all = np.array([self.cam2worlds[self.remap[i]] for i in view_ids_all])

Hope it can help.

[leo-frank]

I think it may be a problem with the version of numpy. I solve it by simply modify the stack operation to:

self.proj_mats, self.intrinsics = proj_mats, intrinsics
self.world2cams, self.cam2worlds = world2cams, cam2worlds

Another thing to modify is the c2ws_all assignment in getitem. You can replace it by:

c2ws_all = np.array([self.cam2worlds[self.remap[i]] for i in view_ids_all])

Hope it can help.

it works!