-
### Describe the bug
nvprof cannot start and run GPU profiling because it needs libraries `libaccinj64.so.12.2` and `libcuinj64.so.12.2` which can be found in the `cuda_packages.cuda_nvprof.lib` outp…
-
We're looking for someone with NVIDIA drivers to help test and debug the 'CUDA' mode in the lightmodel.py file. The code was originally written for older drivers, but currently, none of us have access…
-
### Description
I am encountering this problem on a fresh install of JAX.
I just followed the instructions to install JAX (pip, bundled with CUDA) as
```
pip install -U "jax[cuda12]"
```
b…
-
Spinoff from https://github.com/cms-sw/cmssw/pull/45285#issuecomment-2214607246
Question
> would it be feasible to add the this dependency from the alpaka-backend libraries to the "main library" a…
-
hi, i tried to debug the code Entrance_1.py according to the idea in the this video
[https://www.bilibili.com/video/BV1934y147p2/?spm_id_from=333.788.videopod.sections&vd_source=c276bd4c8dbb36d81805c…
-
### 🐛 Describe the bug
In the following code we add a backend-specific autograd kernel(e.g. AutogradCUDA) for aten builtin operator "tanh". Originally, it does not have a kernel for "AutogradCUDA",…
-
### 🐛 Describe the bug
Really not sure if this is a CMake issue or Pytorch one but I will start here as I think it's related to Pytorch cmake scripts.
I ran into two errors trying to compile with …
-
I tried to use the pipeline command but got an error from the spectral counts component:
```
FATAL: The NSAF and dNSAF computation for spectral-counts requires that the --protein-database option s…
-
Building on 8 cores
-- Found Torch7 in /home/chao/torch/install
CMake Error at /usr/share/cmake-3.5/Modules/FindPackageHandleStandardArgs.cmake:148 (message):
Could NOT find CUDA (missing: CUDA_C…
-
Hi, just want to confirm if It's possible to fine tune Llama-3.2 3B without GPU.
From the source code (https://github.com/unslothai/unsloth/blob/main/unsloth/__init__.py#L47) and (https://github.co…