-
Hello. I'm trying to reuse the KV cache and currently need to implement the decoupling of the KV cache and RoPE.
* model: llama2
* executorch version: 0.2.0
At present, I have implemented that th…
-
Graph coloring problem is normally defined on the structurally symmetric graphs. Current kokkos-kernels implementation assumes the graph is symmetric, if it is not a preprocessing is required to symme…
-
Version: 1.0.522904+cdfa48b2ea1a27dfe0f545c42a34fd3ec7119074
is there something fundamental why the below doesn't work? Due to the fact that vars are shared but their types are not?
It would be a …
-
Good afternoon
I can't figure out how to draw graphs using spark scala kernels. The functionality works in pyspark. I would like to get something like this https://plotly.com/scala/user-guide/. Is it…
-
We should consider whether it is possible and desired to automatically combine kernels into CUDA graphs to reduce overhead of calling individual kernels.
Here is the relevant documentation:
- http…
-
### 🚀 Motivation and context
Is it possible to correlate kernel distribution with ranges annotated either through `torch.cuda.nvtx` or `torch.profiler.profile`?
The use case is model architectur…
-
I have started working on a package called [GraphKernels.jl](https://github.com/simonschoelly/GraphKernels.jl), that implements kernel functions between graphs. So far I have deliberately used similar…
-
I have setup a test environment with libeigen3-dev installed (/usr/include/eigen3)
The webassembly block
https://github.com/mil-tokyo/webdnn/blob/afe16593463b3ee9519d285baf693547fd5f25f9/src/graph…
-
## Improper use of CUDA Graph in TC-GNN
Hello,
I wanted to bring to your attention a potential issue regarding the usage of CUDA Graph in TC-GNN. Upon reviewing the torch [document](https://pyto…
-
Hi there! Thank you for your amazing work on implementing the faster components for transformer-based models! I've found that you have multiple gpu kernels in an encoder or decoder. Have you ever trie…