-
In order to support asymmetric PSF Kernels as well as e.g. pre computed PSF kernels, we should implement a `PSFKernelMap`.
-
continuing #7346 and merged PR #7364 for asymmetric and #7482 for discrete asymmetric kernels
Currently we only have asymmetric product kernels. In my copula experiments with Beta copula it looks …
-
another followup to supporting asymmetric kernels for support (0, 1) or R+
Chen 2000 uses Gasser Muller kernel regression, requires integral of kernel over neighborhood of a point.
That should be…
-
part of #7338
beta, gamma, invgauss and recipinvgauss kernels can be obtained through sccipy's distributions, with appropriate parameterization.
Birnbaum-Saunders (fatiguelife) should also be pos…
-
companion to general and specific issues #7454, ...
List of function and classes that we have or still want to get
("currently" means merged and what I include in PR #7254 )
- `distributions.be…
-
## Context
Current ZeroPointDomain is bound to the layout
https://github.com/pytorch/ao/blob/2ba1a61fe1244560325b5051b5d3c10044553be0/torchao/quantization/quant_api.py#L607-L615
Ideally, we shoul…
-
Hi,
I would like to clarify the following code that was used to implement geometric difference from the paper: "The power of data in quantum machine learning".
def calculate_geometric_difference…
-
(going a bit in circles)
- Bernstein polynomial density estimation is a kind of kernel estimation based on histogram data, or grouped probabilities. domain is interval [0,1] #7296
- in fast kde w…
-
Preliminary performance analysis of the coarray version of ICAR indicates that the algorithm exhibits significant load imbalance. We believe this is due to lopsided expense of evaluating a few physics…
-
Here is my understanding of the existing state of things and what I think we should be doing to make our lower-bit kernels more performant at both small and larger batch sizes. I'm making this an RFC …