Working with Voxel pooling in batch size greater than 1

Hello, I am trying to use voxel pooling in ml.torch to downsample point cloud in my network.
However, as most networks are trained with a batch size greater than 1, I wonder how I could use this function with a batch size greater than 1.
In the docs, it says the function takes the point cloud with shape [M,3], hence it gives an error if I pass my point cloud with shape
[batch size, M, 3].
Could you please let me know how to solve this problem?
Thank you.