Setting environment for ST Edge AI Core ¶
Overview ¶
The ST Edge AI Core package is mainly a complete self-contained application package. For a given target,
it is recommanded to install the associated core tool-chain and adding the directory to the
PATH
environment variable.
Note
In this article,
%STEDGEAI_CORE_DIR%
designates the root path where the
ST Edge AI Core components
are installed.
Example for STM32 target ¶
For the generation of a relocatable binary network (refer to
“STM32 Relocatable binary model support”
article) or to
evaluate the
runtime memory size, a
GNU ARM Embedded tool-chain
(
'arm-none-eabi-'
prefix) should be available
in the PATH.
Environment settings ¶
Windows 10 64-bit or newer ¶
Open a Windows command prompt
Update system
PATHvariableset PATH=%STEDGEAI_CORE_DIR%\Utilities\windows;%PATH%[ Alternative solution ] Create an alias with
doskeycommanddoskey stedgeai="%STEDGEAI_CORE_DIR%\Utilities\windows\stedgeai.exe" $*Verify the environment (displayed versions depend on installed components)
$ stedgeai --version ST Edge AI Core v2.0.0-A1 ISPU 1.1.0 MLC 1.1.0 StellarStudioAI 2.0.0 STM32CubeAI 10.0.0 STM32MP2 1.0
or to have also the version of the embedded Python modules (displayed modules depend on installed components)
$ stedgeai --tools-version stedgeai - ST Edge AI Core v1.0.0 - Python version : 3.9.13 - Numpy version : 1.23.5 - TF version : 2.15.1 - TF Keras version : 2.15.0 - ONNX version : 1.15.0 - ONNX RT version : 1.18.1
Ubuntu 20.4 and Ubuntu 22.4 (or derived) ¶
Similar setting of the environment is expected. A native GCC (X86_64) tool-chain should be accessible in the system path.
export PATH=$STEDGEAI_CORE_DIR/Utilities/linux:$PATH
macOS (x64) - Minimum version High Sierra ¶
Similar setting of the environment is expected. A native GCC (X86_64) tool-chain should be accessible in the system path.
export PATH=$STEDGEAI_CORE_DIR/Utilities/mac:$PATH