mirror of
https://github.com/huggingface/candle.git
synced 2025-06-16 10:38:54 +00:00
Apply suggestions from code review
This commit is contained in:

committed by
GitHub

parent
c98d3cfd8b
commit
1f58bdbb1d
@ -3,7 +3,7 @@
|
||||
**With Cuda support**:
|
||||
|
||||
1. First, make sure that Cuda is correctly installed.
|
||||
- `nvcc --version` should print your information about your Cuda compiler driver.
|
||||
- `nvcc --version` should print information about your Cuda compiler driver.
|
||||
- `nvidia-smi --query-gpu=compute_cap --format=csv` should print your GPUs compute capability, e.g. something
|
||||
like:
|
||||
|
||||
@ -14,7 +14,7 @@ compute_cap
|
||||
|
||||
If any of the above commands errors out, please make sure to update your Cuda version.
|
||||
|
||||
2. Create a new app and add [`candle-core`](https://github.com/huggingface/candle/tree/main/candle-core) with Cuda support
|
||||
2. Create a new app and add [`candle-core`](https://github.com/huggingface/candle/tree/main/candle-core) with Cuda support.
|
||||
|
||||
Start by creating a new cargo:
|
||||
|
||||
|
Reference in New Issue
Block a user