wzzheng / openocc Goto Github PK

Hello, will this repo provide support for more methods, such as VoxFormer, SurroundOcc? Thanks.

您好，请问下nuscenes_occ_infos_train.pkl可以从哪里下载？

Traceback (most recent call last):
File "/root/.vscode-server/extensions/ms-python.python-2023.6.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/pydevd.py", line 3489, in
main()
File "/root/.vscode-server/extensions/ms-python.python-2023.6.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/pydevd.py", line 3482, in main
globals = debugger.run(setup['file'], None, None, is_module)
File "/root/.vscode-server/extensions/ms-python.python-2023.6.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/pydevd.py", line 2510, in run
return self._exec(is_module, entry_point_fn, module_name, file, globals, locals)
File "/root/.vscode-server/extensions/ms-python.python-2023.6.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/pydevd.py", line 2517, in _exec
globals = pydevd_runpy.run_path(file, globals, 'main')
File "/root/.vscode-server/extensions/ms-python.python-2023.6.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 321, in run_path
return _run_module_code(code, init_globals, run_name,
File "/root/.vscode-server/extensions/ms-python.python-2023.6.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 135, in _run_module_code
_run_code(code, mod_globals, init_globals,
File "/root/.vscode-server/extensions/ms-python.python-2023.6.1/pythonFiles/lib/python/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 124, in _run_code
exec(code, run_globals)
File "./train.py", line 339, in
main(0, args)
File "./train.py", line 219, in main
model_outputs = my_model(**model_inputs)
File "/qcraft-vepfs-01/qcraft-vepfs-01/Perception/perception-users/lixiaodong/anaconda3/envs/OpenOcc/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/qcraft-vepfs-01/qcraft-vepfs-01/Perception/perception-users/lixiaodong/anaconda3/envs/OpenOcc/lib/python3.8/site-packages/mmcv/runner/fp16_utils.py", line 98, in new_func
return old_func(*args, **kwargs)
File "/qcraft-vepfs-01/Perception/perception-users/liwei/OpenOcc/model/segmentor/tpv_segmentor.py", line 36, in forward
outs = self.encoder(**results)
File "/qcraft-vepfs-01/qcraft-vepfs-01/Perception/perception-users/lixiaodong/anaconda3/envs/OpenOcc/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/qcraft-vepfs-01/Perception/perception-users/liwei/OpenOcc/model/encoder/tpvformer/tpvformer_encoder.py", line 201, in forward
# forward layers
File "/qcraft-vepfs-01/Perception/perception-users/liwei/OpenOcc/model/encoder/tpvformer/tpvformer_encoder.py", line 125, in forward_layers
reference_points_cam, tpv_mask = point_sampling(
File "/qcraft-vepfs-01/Perception/perception-users/liwei/OpenOcc/model/encoder/tpvformer/utils.py", line 143, in point_sampling
lidar2img = lidar2img.view(
RuntimeError: shape '[1, 1, 6, 1, 4, 4]' is invalid for input of size 192
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 1538708) of binary: /qcraft-vepfs-01/qcraft-vepfs-01/Perception/perception-users/lixiaodong/anaconda3/envs/OpenOcc/bin/python

any available pretrained models？Can I load the weights in TPVFormer?

When I ran the tpvformer_lidarseg_dim128_r50_800_3DOPC.py, the following bug happened.

File "/media/re/2384a6b4-4dae-400d-ad72-9b7044491b55/openmmlab/occ/OpenOcc/train.py", line 111, in main
label_str = train_dataset_loader.dataset.loader.nuScenes_label_name
AttributeError: 'NuScenes3DOPC' object has no attribute 'loader'

It seems that train_datasetloader.dataset doesn't have the attribute 'loader'.

What's wrong with that?

Hello, will this repo provide support for more methods, such as VoxFormer, SurroundOcc? Thanks.