-
Hi,
I was trying out the compression library for ZeroQuant quantization (for GPT-J model). While I was able to compress the model, I didn't see any throughput/latency gain from the quantization dur…
-
I work in an organization with restrictive policies, but we are able to access pip.
Python and Pip are managed through a software center.
When trying to use Spyder with a managed python installatio…
rukie updated
1 month ago
-
As suggested by @wsmoses [1] I am adding here the code for executing my kernels where compilation takes a long time. there are two kernels - first takes around 15 minutes, second 54 minutes (earlier …
-
### What would you like to do?
Report an issue on quarto.org
### Description
The docs include guidance for adding Quarto support to a Jupyter Kernel [here](https://quarto.org/docs/advanced/jupyter/…
-
### System Info
`transformers==4.45.1`
`peft==0.13.0`
`liger-kernel==0.3.1`
So `isinstance(model.base_model.model, PreTrainedModel)` returns true but `isinstance(model, PreTrainedModel)` retur…
-
# Christian Mills - CUDA MODE Lecture 1: How to profile CUDA kernels in PyTorch
Lecture #1 provides a practical introduction to integrating and profiling custom CUDA kernels within PyTorch programs, …
-
### Some background and history
In August of 2023 Irwin Zaid (@izaid) made the following [comment](https://github.com/scipy/scipy/pull/19023#issuecomment-1689050946) on the pull request https://git…
-
DC(duplicated code) Score for onert-micro is 3.96 (with #13872 and #13870 and #13865)
Here is a DC list(first 20 DCs) :
Metric;BlockKey;Path;LineStart;LineEnd;LineCount;Token;Violation
DC;1;on…
-
there are parameters for kernel size and type and for most of the case the type is NONE, is there any recommendation for choosing the kernels size and types?
-
### Feature description
Throughout the kernel different conventions for comments and naming have been used, e.g. in [account.masm](https://github.com/0xPolygonMiden/miden-base/blob/next/miden-lib/asm…