OPENGIT/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2025-04-12 15:59:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

Updated Install and Run on AMD GPUs (markdown)

Chris Braun 2022-09-21 15:02:53 +02:00
parent 2e1ce22bd2
commit 3ac539d8a0
1 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
Install-and-Run-on-AMD-GPUs.md → Install-and-Run-on-Linux-with-AMD-GPU.md

@ -1,4 +1,4 @@
# Installing and running on Linux with AMD GPUs # Install and run on Linux with AMD GPU
If your AMD GPU is compatible with ROCm, you can try running: If your AMD GPU is compatible with ROCm, you can try running:
```bash ```bash
@ -17,7 +17,7 @@ The next generations should work with regular performance. You can follow the li
to use the same operating system, follow the steps there to fix this issue. If there is no clear way to compile or to use the same operating system, follow the steps there to fix this issue. If there is no clear way to compile or
install the MIOpen kernels for your operating system, consider following the Docker guide below. install the MIOpen kernels for your operating system, consider following the Docker guide below.
# Installing and running on Linux with AMD GPUs (Docker) # Install and run on Linux with AMD GPU (Docker)
Pull the latest `rocm/pytorch` Docker image, start the image and attach to the container (taken from the `rocm/pytorch` Pull the latest `rocm/pytorch` Docker image, start the image and attach to the container (taken from the `rocm/pytorch`
documentation): `docker run -it --network=host --device=/dev/kfd --device=/dev/dri --group-add=video --ipc=host documentation): `docker run -it --network=host --device=/dev/kfd --device=/dev/dri --group-add=video --ipc=host
--cap-add=SYS_PTRACE --security-opt seccomp=unconfined -v $HOME/dockerx:/dockerx rocm/pytorch` --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -v $HOME/dockerx:/dockerx rocm/pytorch`
@ -53,7 +53,7 @@ The `/dockerx` folder inside the container should be accessible in your home dir
## Updating Python version inside Docker ## Updating Python version inside Docker
If the web UI becomes incompatible with the pre-installed Python 3.7 version inside the Docker image, here are If the web UI becomes incompatible with the pre-installed Python 3.7 version inside the Docker image, here are
instructions on how to update it (assuming you have successfully followed "Installing and running on Linux with AMD GPUs instructions on how to update it (assuming you have successfully followed "Install and run on Linux with AMD GPU
(Docker)"): (Docker)"):
Execute the following inside the container: Execute the following inside the container:
@ -66,7 +66,7 @@ echo 'PATH=/usr/local/bin:$PATH' >> ~/.bashrc
Then restart the container and attach again. If you check `python --version` it should now say `Python 3.9.5` or newer. Then restart the container and attach again. If you check `python --version` it should now say `Python 3.9.5` or newer.
Run `rm -rf /dockerx/stable-diffusion-webui/venv` inside the container and then follow the steps in "Installing and Run `rm -rf /dockerx/stable-diffusion-webui/venv` inside the container and then follow the steps in "Installing and
running on Linux with AMD GPUs (Docker)" again, skipping the running on Linux with AMD GPU (Docker)" again, skipping the
`git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui` and using the modified launch-command below instead: `git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui` and using the modified launch-command below instead:
```bash ```bash
@ -74,4 +74,4 @@ running on Linux with AMD GPUs (Docker)" again, skipping the
TORCH_COMMAND='pip install torch torchvision --extra-index-url https://download.pytorch.org/whl/rocm5.1.1' python launch.py --precision full --no-half TORCH_COMMAND='pip install torch torchvision --extra-index-url https://download.pytorch.org/whl/rocm5.1.1' python launch.py --precision full --no-half
``` ```
Always use this new launch-command from now on, also when restarting the web UI in following runs. Always use this new launch-command from now on, also when restarting the web UI in following runs.