Where is the beta parameter getting updated?

[Saeedha-N]

Hello @ajhamdi just went through the paper and the concept is really cool! Thank you for your incredible contributions! You previously mentioned that the original submodules with the CUDA codes were left as it were in the latest version, which means the forward and backward propagation CUDA codes are same as 3DGS (shape (beta) = 2). However, I could not find many changes in the Python renderer also where the shape parameter was updated and incorporated in the forward and backward processes. Could you clarify whether the shape updates during optimization are performed in the __init__.py script or laplacian_model.py script, and if neither then which script?