kly

Request of implantmentation of lgamma function

5

I am trying to port an mlx implementation of the SCVI module of scvi-tools (a comprehensive de-batching approach using VAE models), which currently overcomes the problem of mlx.array not supporting...

Error when trying to feature native MLX accelerated scvi

4

**Description:** Before the recent update to PyTorch's MPS support, using PyTorch MPS to accelerate scVI would result in NaN values in the returned matrix. For more details, see: [Error when...