Optional Installation Scenarios
ascend-deployer provides the following installation scenarios.
In scenarios where frameworks need to be installed, install GCC with a version later than 7.3.0 (by specifying --install to gcc) and configure environment variables or soft links. For details, see Configure GCC environment variables.
Installation Scenario |
Component to Be Installed |
Description |
|---|---|---|
auto |
sys_pkg, Python, NPU, Toolkit, NNRT, NNAE, TFPlugin, kernels, Toolbox, PyTorch, TensorFlow, MindSpore |
Install the downloaded software packages on the left, including the driver, firmware, CANN, MindSpore, TensorFlow, and PyTorch. |
offline_dev |
sys_pkg, Python, NPU, Toolkit, kernels |
Offline inference in the development environment |
offline_run |
sys_pkg, Python, NPU, NNRT |
Offline inference in the operating environment |
mindspore |
sys_pkg, Python, NPU, Toolkit, MindSpore, kernels |
MindSpore Scenario |
tensorflow_dev |
sys_pkg, Python, NPU, Toolkit, TFPlugin, TensorFlow, kernels |
TensorFlow development scenario |
tensorflow_run |
sys_pkg, Python, NPU, NNAE, TFPlugin, TensorFlow, kernels |
TensorFlow operating scenario |
pytorch_dev |
sys_pkg, Python, NPU, Toolkit, PyTorch, kernels |
PyTorch development scenario |
pytorch_run |
sys_pkg, Python, NPU, NNAE, PyTorch, kernels |
PyTorch operating scenario |
vmhost |
sys_pkg, NPU, Toolbox |
VM scenario |
edge |
sys_pkg, AtlasEdge, HA |
Install the MindX middleware and HA. |
dl |
sys_pkg, NPU, Kubernetes, Ascend Device Plugin, Ascend Docker Runtime, HCCL-Controller, NodeD, NPU-Exporter, Volcano, Ascend Operator, Resilience-Controller |
MindX DL installation scenario |
mef |
Docker, Kubernetes, KubeEdge, MEF Center |
MEF Center installation scenario |
The configuration files of the preceding installation scenarios are stored in the scene directory, for example, ascend-deployer/ascend_deployer/playbooks/scene/scene_auto.yml (for auto scenario). To customize an installation scenario, refer to its corresponding configuration file.