Weikang Qian

I had the same problem. About 980G of the memory allocation is required.

Honestly, I don't know in detail. According to the _[torch.linalg.lstsq](https://pytorch.org/docs/stable/generated/torch.linalg.lstsq.html)_ document, _gelsy_ is a general QR factorization to solve least-squares using CPU. _gels_ assumes the matrix is full rank. For...

Sure, please go ahead. I'm new to the GitHub collaboration. Glad this could help.

Update: I tried the codes in my local environment, and it works smoothly.

There is something wrong using _torch.linalg.lstsq_ in **spline.py**. it can lead to nan when we are not using CPU. I have added an alternative method to calculate the _coef_ in...

> > I tried running this on a couple of the examples in my CUDA set up, but I got the following error: > You may want to add a...

> I tried your code, but the following error occurred: NotImplementedError: The operator 'aten::linalg_lstsq.out' is not currently implemented for the MPS device. If you want this op to be added...

> I reset PYTORCH_ENABLE_MPS_FALLBACK=1 in the appropriate position, and it took effect. But unfortunately, it didn't help much. I tried your latest code. If coef_method='lstsq' is set, the training result...