首先，从 Dockerfile.rocm 构建一个 docker 镜像，并从该镜像启动一个 docker 容器。Dockerfile.rocm 默认使用 ROCm 6.1，但在较旧的 vLLM 分支中也支持 ROCm 5.7 和 6.0。方法非常灵活，可以使用以下参数自定义 Docker 镜像的构建：BASE_IMAGE：指定运行 docker build 时使用的基础镜像，特别是 ROCm 基础镜...
Dockerfiles for the various software layers defined in the ROCm software platform - ROCm/ROCm-docker
1.3 Docker镜像拉取并运行一个Docker容器 下面在ROCm5.4.2上使用Pytorch2.0。请根据您的目标 ROCm 和 Pytorch 版本使用合适的 docker 镜像:https://hub.docker.com/r/rocm/pytorch/tags docker pull rocm/pytorch:rocm5.4.2_ubuntu20.04_py3.8_pytorch_2.0.0_preview sudo docker run --device=/dev/kfd --devi...
ROCm Developer Hub Start developing AMD GPU-accelerated applications. Visit the ROCm Developer Hub to get access to the latest user guides, containers, training videos, webinars, and more. Get Started ML Development on AMD Radeon GPUs Researchers and developers working with ML models and algorithms...
Docker images in the ROCm ecosystem# TheROCm Docker repositoryhosts images useful for building your own containers, leveraging ROCm. The built images are available onDocker Hub. In particular: rocm/rocm-terminalis a small image with the prerequisites to build HIP applications, but does not include...
首先,从 Dockerfile.rocm 构建一个 docker 镜像,并从该镜像启动一个 docker 容器。 Dockerfile.rocm 默认使用 ROCm 6.1,但在较旧的 vLLM 分支中也支持 ROCm 5.7 和 6.0。方法非常灵活,可以使用以下参数自定义 Docker 镜像的构建: BASE_IMAGE:指定运行docker build时使用的基础镜像,特别是 ROCm 基础镜像上的 Py...
Megatron-LM Docker User guide How to Tutorials for AI Workloads Explore more topics to find relevant learning resources to help you stay updated with the latest technological advancements from AMD. AI Inference Fine-tuning Training AI Inference Deploy AI models faster with higher accuracy and achiev...
Use the following command to pull the Docker image from Docker Hub. docker pull rocm/megatron-lm:v25.4 Launch the Docker container. docker run -it --device /dev/dri --device /dev/kfd --device /dev/infiniband --network host --ipc host --group-add video --cap-add SYS_PTRACE --secur...
You can use PyTorch unit tests to validate your PyTorch installation. If you used a prebuilt PyTorch Docker image from AMD ROCm Docker Hub or installed an official wheels package, validation tests are not necessary.If you want to manually run unit tests to validate your PyTorch installation ...
docker build -f Dockerfile -t fanwu103/mi308_inference:rocm6.3.1_ubuntu22.04_py3.10_vllm_0.6.5 . 2. pull images from docker hub docker pull fanwu103/mi308_inference:rocm6.3.1_ubuntu22.04_py3.10_vllm_0.6.5 AI Software VERSION: Torch: 2.5.1+gitabbfe77 vLLM: branch: moe_final_...