Install ERROR: Command errored out with exit status 1

[sdlopezp]

when I run "pip install -ve my\path\to\LightConvPoint-0.2" I got an "ValueError: list.remove(x): x not in list" from setup.py file in the 29 line.

Any idea why is this happens?.

Thank you!

[yeminglang]

have you solved it？

[sdlopezp]

have you solved it？

Not yet.
The same thing happens to you?
Any idea?

[yeminglang]

have you solved it？

Not yet. The same thing happens to you? Any idea?

The same thing happens to me...
By the way ,Do you have any recommended repositories for point cloud semantic segmentation? (better if it can be trained fast and has higher mIOU)

[aboulch]

Hello,
Thanks for the issue.
I am planning to release a version without C++ module compilation by the end of the month (March).
Like that, it should rely only on PyTorch and maybe on TorchGeometric.
The code is almost ready, I have to make sure it is still compatible with the pre-trained networks I provided.
Best,

[sdlopezp]

Hello, Thanks for the issue. I am planning to release a version without C++ module compilation by the end of the month (March). Like that, it should rely only on PyTorch and maybe on TorchGeometric. The code is almost ready, I have to make sure it is still compatible with the pre-trained networks I provided. Best,

Great.
Thanks for all your work. I will expect the update.

[sdlopezp]

have you solved it？

Not yet. The same thing happens to you? Any idea?

The same thing happens to me... By the way ,Do you have any recommended repositories for point cloud semantic segmentation? (better if it can be trained fast and has higher mIOU)

I've been using the Open3D ML module (https://github.com/isl-org/Open3D-ML). They have different state-of-the-art architectures implemented and pretrained. It supposed to be easier to use than the original implementations.

[yeminglang]

have you solved it？

Not yet. The same thing happens to you? Any idea?

The same thing happens to me... By the way ,Do you have any recommended repositories for point cloud semantic segmentation? (better if it can be trained fast and has higher mIOU)

I've been using the Open3D ML module (https://github.com/isl-org/Open3D-ML). They have different state-of-the-art architectures implemented and pretrained. It supposed to be easier to use than the original implementations.

Nice, thank a lot!
Best wishes!

[tempted111]

Hello, can you tell me what the meaning of 2500000 in prepare_data.py is? thanks.

[tempted111]

I want use the code to run own datasets!