Xiaoyang Wu

Results 245 comments of Xiaoyang Wu

If your target dataset does not have normal information, please also disable it in other datasets.

> I meet the same problem. The memory usage continues to rise and the program is killed when it exceeds 100% usage. @DGCHAO @Novmaple Thanks for pointing out the reason....

> you have any chance to look at it, that would be really great! Hi, I received the email. I will check it soon.

Remove "Normal" in each `feat_key` list of `"Collect"` function, also don't forget to reduce the in_channel of the model config.

Good question! Actually, SpConv doesn't support depthwise submconv from now. And also, it is a huge challenge to accelerate a depthwise submconv (so no speed advantage). Therefore, we directly use...

> Hi @aldinorizaldy , I also met this problem, can you tell me how you solved it? I can perform normal testing in debug mode on Vscode, but when I...

Hi, you only contain 8 samples in your training set and the batch size is larger than the number. Consequently, the total steps number is 0.

Hi, save the pose here (https://github.com/Pointcept/Pointcept/blob/main/pointcept/datasets/preprocessing/s3dis/preprocess_s3dis.py#L116) is all you need. PS: I will delete your another duplicate question~

> te the "Area_1_alignmentAngle.txt" file when creating the dataset, I am also making the data set.. Thank you very much! Hi, do you mean that you want to know how...