The randint line is commented, but in the original implementation and the algorithm itself, it should be random? Maybe we can provide a fixed seed if we want to get the same result?
Based on the codes, we can see farthest is a temporal variable to record the maximal position latitude. In this case, I think zero initialization should be better than randomness.
I'm reading your codes.
https://github.com/jina-ai/executor-3d-encoder/blob/main/executor/models/modules.py#L75
The randint line is commented, but in the original implementation and the algorithm itself, it should be random? Maybe we can provide a fixed seed if we want to get the same result?