4.1.0
Setting environment for ST Edge AI Core


ST Edge AI Core

Setting environment for ST Edge AI Core


ST Edge AI Core Technology 4.1.0



r1.1

Overview

The ST Edge AI Core package is a complete, self-contained application package. For a given target, install the associated core toolchain and add 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.

External tools dependencies

Optionally, to generate a relocatable or runtime loadable model (refer to “STM32 Relocatable binary model support” article) or to evaluate the runtime memory size, an Arm GNU Toolchain ('arm-none-eabi-' prefix) should be available in the PATH.

Environment settings

Windows® 10 64-bit or newer

  1. Open a Windows command prompt

  2. Update system PATH variable

    set PATH=%STEDGEAI_CORE_DIR%\Utilities\windows;%PATH%
  3. [ Alternative solution ] Create an alias with doskey command

    doskey stedgeai="%STEDGEAI_CORE_DIR%\Utilities\windows\stedgeai.exe" $*
  4. Verify the environment (displayed versions depend on installed components)

    $  stedgeai --version
    ST Edge AI Core v4.1.0
      ISPU 2.1.0
      MLC 1.3.0
      StellarAI 4.1.0
      STM32CubeAI 12.1.0

    or to have also the version of the embedded Python modules (displayed modules depend on installed components)

    $ stedgeai --tools
    stedgeai - ST Edge AI Core v4.1.0
    - Python version   : 3.13.12
    - Numpy version    : 2.4.3
    - TF version       : 2.21.0
    - TF Keras version : 3.15.1
    - ONNX version     : 1.22.0
    - ONNX RT version  : 1.25.1

Ubuntu® 20.4 and Ubuntu® 22.4 (or derived)

A similar environment setup is expected. A native GCC host toolchain should be accessible in the system path to perform the validate command in host mode.

export PATH=$STEDGEAI_CORE_DIR/Utilities/linux:$PATH

macOS® arm64 / aarch64 - Minimum version Sonoma (macOS® 14) or newer

This version of Apple’s operating system built for 64-bit ARM-based Apple Silicon processors, such as the M1, M2, M3, M4, and M5 chips. A similar environment setup is expected. A native GCC host toolchain should be accessible in the system path to perform the validate command (host mode).

export PATH=$STEDGEAI_CORE_DIR/Utilities/mac:$PATH

Intel-based Mac (Macintel / x86_64 architecture) is no longer supported since version 4.1 of ST Edge AI Core.