Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

custom dataset #60

Open
MaryneZa opened this issue Dec 2, 2024 · 0 comments
Open

custom dataset #60

MaryneZa opened this issue Dec 2, 2024 · 0 comments

Comments

@MaryneZa
Copy link

MaryneZa commented Dec 2, 2024

image
i have try this section for my custom data

`00:02:30 Runtime exception: arrays used as indices must be of integer (or boolean) type console_utils.py:393
╭─────────────────────────────────────────── Traceback (most recent call last) ───────────────────────────────────────────╮
│ /home/mare-4d/work/4K4D/easyvolcap/utils/console_utils.py:390 in inner │
│ │
│ ❱ 390 │ │ │ │ return func(*args, **kwargs) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/scripts/main.py:272 in main │
│ │
│ ❱ 272 │ else: globals()args.type # invoke this (call callable_from_cfg -> call_from_cfg) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:56 in inner │
│ │
│ ❱ 56 │ │ return call_from_cfg(func, cfg) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:47 in call_from_cfg │
│ │
│ ❱ 47 │ return func(**call_args) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/scripts/main.py:165 in test │
│ │
│ ❱ 165 │ val_dataloader: "VolumetricVideoDataloader" = DATALOADERS.build(val_dataloader_cfg) # reuse the validataion │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:300 in build │
│ │
│ ❱ 300 │ │ return self.build_func(*args, **kwargs, registry=self) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:131 in build_from_cfg │
│ │
│ ❱ 131 │ return call_from_cfg(obj_cls, args) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:47 in call_from_cfg │
│ │
│ ❱ 47 │ return func(**call_args) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/dataloaders/volumetric_video_dataloader.py:94 in init
│ │
│ ❱ 94 │ │ dataset: VolumetricVideoDataset = DATASETS.build(dataset_cfg) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:300 in build │
│ │
│ ❱ 300 │ │ return self.build_func(*args, **kwargs, registry=self) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:131 in build_from_cfg │
│ │
│ ❱ 131 │ return call_from_cfg(obj_cls, args) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:47 in call_from_cfg │
│ │
│ ❱ 47 │ return func(**call_args) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/dataloaders/datasets/image_based_dataset.py:56 in init
│ │
│ ❱ 56 │ │ call_from_cfg(super().init, kwargs, skip_loading_images=skip_loading_images) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/engine/registry.py:47 in call_from_cfg │
│ │
│ ❱ 47 │ return func(**call_args) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/dataloaders/datasets/volumetric_video_dataset.py:287 in init
│ │
│ ❱ 287 │ │ │ self.load_bytes() # load image bytes (also load vhulls) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/dataloaders/datasets/volumetric_video_dataset.py:472 in load_bytes │
│ │
│ ❱ 472 │ │ │ self.load_vhulls() # before cropping the mask (we need all the information we can get for visual hu │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/dataloaders/datasets/volumetric_video_dataset.py:618 in load_vhulls │
│ │
│ ❱ 618 │ │ │ │ vhulls, bounds, valid, inds = carve_using_bytes(H, W, K, R, T, latent_index) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/dataloaders/datasets/volumetric_video_dataset.py:575 in carve_using_bytes │
│ │
│ ❱ 575 │ │ │ vhulls, bounds, valid, inds = hierarchically_carve_vhull(*inputs, │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/utils/vhull_utils.py:173 in hierarchically_carve_vhull │
│ │
│ ❱ 173 │ │ │ return hierarchical_space_carving(*args, **kwargs, voxel_size=voxel_size) │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/utils/vhull_utils.py:238 in hierarchical_space_carving │
│ │
│ ❱ 238 │ vhull, bounds, valid, inds = carve_vhull(H, W, K, R, T, │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/utils/vhull_utils.py:156 in carve_vhull │
│ │
│ ❱ 156 │ │ vhull = fcds_utils.remove_outlier(vhull[None], K=5, std_ratio=5.0)[0] │
│ │
│ /home/mare-4d/work/4K4D/easyvolcap/utils/fcds_utils.py:306 in remove_outlier │
│ │
│ ❱ 306 │ return torch.as_tensor(np.array(o3d_pcd.points)[np.array(ind)]).to(pcd.device, pcd.dtype, non_blocking=True). │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
IndexError: arrays used as indices must be of integer (or boolean) type
00:02:30 Preparing vhulls data/pose1/vhulls VAL 0% ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0/300 0:00:04 < -:--:-- ? it/s v…
*** arrays used as indices must be of integer (or boolean) type

/home/mare-4d/work/4K4D/easyvolcap/utils/fcds_utils.py(306)remove_outlier()
304 if return_inds:
305 return torch.as_tensor(np.array(ind), device=pcd.device)[None] # N,
--> 306 return torch.as_tensor(np.array(o3d_pcd.points)[np.array(ind)]).to(pcd.device, pcd.dtype, non_blocking=True).vi
307
308 `

image

i found that "valid" eqauls to tensor([False, False, False, ..., False, False, False], device='cuda:0') all false, Do you have any suggestion ?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant