-
As @maleadt mentioned in https://github.com/JuliaGPU/Metal.jl/issues/422.
I re-open a new issue here.
The current `LinearAlgebra.kron` only supports for `CuArray`, and the other `GPUArray` uses …
-
Do you have any idea how to make the backends (specifically, GLES or Vulkan) selectable at runtime? For exmaple, by making some object-safe traits? Alternatively, it would help if multiple backends ca…
-
### Description
Use case: backend only supports connections over TLS
We need to support https://gateway-api.sigs.k8s.io/api-types/backendtlspolicy/ or some subset of it
-
import torch
import torch.nn.functional as F
import droid_backends
anyone know what is the droid_backends?
-
I'd love to pass a comma separated list of `CE_BROKER_URL`s and have this spin up some multiprocessing under the hood, merging the results in the prometheus metrics endpoint. This is possible with [Mu…
-
Does plotnine have an abstraction level that would make it (relatively) easy to implement backends besides matplotlib? I'm asking because I really like the gg logic, but would prefer interactive graph…
-
Kotlin startup time is awfull
-
After seeing this tweet [1] by @tomaarsen in which he announced a set of new ONNX and OpenVINO models, I was hoping to run those models without having to depend on torch. Would it be possible to make …
-
### Description
Transpose convolutions are orders of magnitude slower than their complementary regular convolutions and their counterparts in torch (at least for the sizes in the example below). Thi…
-
The url generated by the backend editor should be copy-able (and view-able) by the user.
Also, the user should be able to paste in a previously saved url.