1
0
mirror of https://github.com/fumiama/Retrieval-based-Voice-Conversion-WebUI.git synced 2026-06-05 01:10:22 +08:00

doc: update instructions for ROCM (#106)

This commit is contained in:
Jax Leo
2025-01-29 13:46:08 +08:00
committed by GitHub
parent b0552625f7
commit 78dfad6cba
5 changed files with 245 additions and 4 deletions

View File

@@ -73,6 +73,7 @@ sh ./run.sh
```bash
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
```
3. Install the corresponding dependencies according to your own graphics card.
- Nvidia GPU
```bash
@@ -91,6 +92,11 @@ sh ./run.sh
pip install -r requirements/ipex.txt
```
4.If you are using an ROCM-capable AMD Radeon GPU, then you need to choose ROCM version of PyTorch.
```bash
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm6.2
```
## Preparation of Other Files
### 1. Assets
> RVC requires some models located in the `assets` folder for inference and training.
@@ -156,8 +162,8 @@ pacman -S rocm-hip-sdk rocm-opencl-sdk
````
For some models of graphics cards, you may need to configure the following environment variables (such as: RX6700XT).
````
export ROCM_PATH=/opt/rocm
export HSA_OVERRIDE_GFX_VERSION=10.3.0
export ROCM_PATH=/opt/rocm #Set ROCM Executables Path
export HSA_OVERRIDE_GFX_VERSION=10.3.0 #Spoof GPU Model for ROCM
````
Also, make sure your current user is in the `render` and `video` user groups.
````