Fudan-SMI-lab
Official@fudan-smi-lab
Offers specialized environment configuration and dependency management for deploying PyTorch workloads on Ascend NPU hardware architectures.
Agent Skills by Fudan-SMI-lab
Showing 4 vetted skills indexed across 1 GitHub repositories.
Fudan-SMI-labtorch-npu-venv-setup-cpu-base
Pin CPU-base PyTorch and install matching torch-npu and torchvision.
Fudan-SMI-labtorch-npu-pyyaml-preinstall
Preinstall PyYAML before torch-npu to prevent transitive dependency failures.
Fudan-SMI-labapex-cuda-extension-removal
Replace NVIDIA Apex imports and APIs with NPU-native PyTorch equivalents.
Fudan-SMI-labe2e-v2-0dc9b40e0da8-promoted
Create virtual environments inheriting system torch-npu stacks on Ascend hosts.
Frequently Asked Questions About Fudan-SMI-lab
FAQPage SchemaWhat specific tasks are enabled by these environment configurations?▼
These configurations enable the deployment of PyTorch-based compute tasks on Ascend NPU hardware. They facilitate the installation of matching torch-npu and torchvision versions, resolve transitive dependency failures, and replace legacy NVIDIA-specific imports with native NPU-compatible equivalents for seamless hardware execution.
Which target personas benefit from these environment setups?▼
These configurations are designed for machine learning engineers and infrastructure specialists working with Ascend-based compute clusters. They are particularly useful for researchers migrating existing model architectures from NVIDIA-based environments to NPU-native hardware stacks.
What are the prerequisites for implementing these environment setups?▼
Implementation requires an Ascend host environment with existing system-level torch-npu stacks. Users must have administrative access to manage virtual environments and ensure that the host architecture supports the specific PyTorch and torchvision versions defined in the setup manifests.