-
Here is my description
```
[GIN] 2024/05/11 - 15:36:03 | 200 | 34.248274ms | 38.104.153.43 | GET "/api/v1/computing/cp"
time="2024-05-11 15:36:03.554" level=info msg="receive ubi task rece…
-
The [CExA project](https://cexa-project.org/) has been launched by the CEA to
fund a 10 people team to contribute to the Kokkos library and its ecosystem,
to onboard codes developed at CEA, and to off…
-
I was wondering if get_distance_matrix could go faster by using a GPU, which seems to be a possibility with dask functions. What do you think?
-
**Which section will this topic belongs to**
- [ ] 1. Why use a Cluster?
- [ ] 2. Working on a remote HPC system
- [ ] 3. Scheduling jobs
- [ ] 4. Accessing software
- [ ] 5. Transferring…
-
Traceback (most recent call last):
File "e:\SiTH-main\.conda\lib\site-packages\torch\utils\cpp_extension.py", line 2100, in _run_ninja_build
subprocess.run(
File "e:\SiTH-main\.conda\lib\su…
-
So suppose I have 4 million audio files, and I want to extract the features using `Encodec`. I found that the extracting process is kind of slow, where `Computing features in batches: 183301it [19:36,…
-
Hey,
I think the tokens/s calculation might be incorrect. I can see that you are computing time by timing the CPU clock here: https://github.com/likejazz/llama3.cuda/blob/master/llama3.cu#L789
…
-
-
Environment: miniconda3
Python = 3.12.2
torch.__version__ = 2.3.0+cu121
pip install flash-attn
pip install flash-attn --no-build-isolation
both failed to build
------------------------------…
-
I add the 'cl.exe' to PATH, but still says can not found.
```
error: failed to run custom build command for `candle-kernels v0.5.1 (https://github.com/huggingface/candle.git#cd4d941e)`
Caused b…