-
Dear Zhi,
Thank you for sharing this amazing work! I have installed the repo following your instructions and was able to reproduce you showed in the YT video with the provided experimental data.
…
-
Some ideas, on top of my head.
## Simd modules
We are using `Simd` to accelerate some core image processing functions (e.g. color conversion, image resizing, ...) because:
- it is highly optimi…
-
1. By using command `CMAKE_ARGS="-DLLAMA_CUBLAS=on" FORCE_CMAKE=1 python setup.py bdist_wheel`, I can build out a wheel and have it installed as:
```console
llama_cpp:
total 3.8M
-rwxrwxr-x 1 …
-
Since glibc continues improving its performance a years-long, the latest version of glibc (2.35) delivered from ubuntu 22.04 seems having a better performance than the strings::memcpy_inlined function…
-
There is a [discussion](https://github.com/orgs/bitcraze/discussions/1018) indicating that there are issues running the classification example.
I did a quick test and found some (other) problems:
…
-
Hi, I encountered an error while running finetune.py, the program died, what is the reason?
python -m torch.distributed.run --nproc_per_node=8 finetune.py \
--model-type ram_plus \
--config r…
-
prompt time: 15761 ms / 17 tokens (927.114 ms/token)
output time: 60333.6 ms / 28 tokens (2154.77 ms/token)
total time: 76094.5 ms
-
i5-11400f
![Screenshot_242](https://user-images.githubusercontent.com/42765009/131723202-a2c98199-d7bf-4925-9088-80147421aebc.png)
-
**Introduction**:
SciPy's spatial distance computations are fundamental to various scientific and data science tasks. To accelerate these computations, it is imperative to ensure that the underlying …
-
Various ARM platforms have been doing it with big.LITTLE and DinamiQ for some time, with different cores, and Apple's M1 and Intel's Alder Lake made it into the mainstream now. It'd be useful to see w…