This should be fixed in a future release of PyTensor. It was fixed in: Remove restrictive broadcastability check in `local_blockwise_alloc` by ricardoV94 · Pull Request #751 · pymc-devs/pytensor · GitHub
This should be fixed in a future release of PyTensor. It was fixed in: Remove restrictive broadcastability check in `local_blockwise_alloc` by ricardoV94 · Pull Request #751 · pymc-devs/pytensor · GitHub