YCY
> The `integrate` function looks also good to me. I have no idea what is wrong there but I found the order of the input point cloud affects the unwrapping...
Hi, if you were asking for the algorithm details, you can check the function here https://github.com/yoyololicon/duet-svs-diffusion/blob/69c8276ee91f030aa6686c3a5e18c2348737119c/eval.py#L19 We didn't have a simple script that could be used immediately for simple use...
Hi @insar-dev , thanks for stopping by. There are some discussions on this in previous issues #2 #9 . Basically, if simplices are given, `kamui` runs MCF using linear programming....
By `failed`, did you mean the results are incorrect, or did you run into memory issues?
@insar-dev, you can also try the graph cuts method (`method="gc"`). It can handle large data and is quite robust.
Hi @scotts, thanks for reporting the status of torchaudio and future plans. I don't understand the decision to drop the C++/CUDA extensions... They are implemented because of the super inefficiency...
https://github.com/IoSR-Surrey/DAFx24-Proceedings/blob/main/papers/DAFx24_paper_75.pdf The version on Surrey's website is updated.
Hi @ybourdin , glad to know you're interested in our work! Good suggestion! I suggest having a separate function like `sample_wise_lpc_zf` that implements what you describe and returns a tuple...
Hi @ybourdin , the feature has been added to the main branch f50b419ea58dd1dc26fab3e9602cc3bef1a0dbbb
Hi, happy to hear that you find my repo helpful to you :) I'll have a look when I'm available.