Skip to content

Edge Impulse machine learning firmware for Alif AppKit gen2

License

Apache-2.0, Unknown licenses found

Licenses found

Apache-2.0
LICENSE
Unknown
License.txt
Notifications You must be signed in to change notification settings

edgeimpulse/firmware-alif-csolution

Repository files navigation

firmware-alif-csolution

Firmware for Alif E7 AI/ML AppKit This application is built on VSCode Getting Started Template (alif_vscode-template) The default hardware is Gen 2 Ensemble DevKit with camera module and display.

The required software setup consists of VSCode, Git, CMake, cmsis-toolbox, Arm GNU toolchain, Alif tools and JLink (for debugging). Please refer to the Getting Started Guide for more details.

Note

The firmware by default uses the MT9M114 camera sensor, if you have ARX3A0 camera, modify the firmware-alif.cproject.yaml

# - component: AlifSemiconductor::BSP:External peripherals:CAMERA Sensor MT9M114
- component: AlifSemiconductor::BSP:External peripherals:CAMERA Sensor ARX3A0

Prerequisites

  1. Create an edge impulse account at edgeimpulse.com

  2. Install the latest Alif Security Toolkit:

    • Navigate to the Alif Semiconductor Kit documentation page (you will need to register to create an account or log in). and download the latest App Security Toolkit.
    • Extract archive where you prefer and create an envirionmental variable SETOOLS_ROOT pointing to it.
    • Follow the instructions in local the Alif Security Toolkit Quickstart Guide to finalize the installation.
  3. Clone this repo and intitialize the submodules:

git submodule init
git submodule update

Build

Important

To build and manage the project, you need to follow this guide on how to setup the CMSIS-Toolbox

Using Arm CMSIS Solution extension

Go to the CMSIS extension tab, and click the hammer icon.

Using the script

Run the script specifying the target you want to build the project for, if not specified the default is HP.

ie to compile HE:

sh build.sh HE

Large models that don't fit into DTCM memory can be placed in the SRAM0 section using the HP_SRAM argument:

sh build.sh HP_SRAM

To clean:

sh build.sh clean

Note

On windows, use build_win.bat to compile the project.

Using Docker

TODO

Flash the board

Important

To flash the board you need to set the SETOOLS_ROOT environmental variable to point to the Alif Security Toolkit

Flash using VS code

Flash using Secure toolkit

Task -> Run Task -> Program with security toolkit. The actual config will be flashed.

Flash using JLink

You can use a Segger JLink to start a debug session and flash the board. Connect the 20-pin connector to J13 and press F5 to start.

Note

When you start a debug session for a specific core, make sure you already flashed the bin for the core using the Security Toolkit, otherwise JLink can't connect to the core.

Flash using script

Run the script specifying the target you want to be flashed, if not specified the default is HP.

ie to flash HE:

flash.sh HE

Note

On windows, use flash_win.bat to flash the target.

Connect to serial console

To connect to the board, use the following settings: J15 3-5 4-6 115200, 8, N, 1

Run a debug session

Connect the 20pin cable to J13 and press F5.

Update your model

To update your model, unizp the CMSIS pack deployment, install the packs using cpackget add <your_project_name>.pack and paste the model.yml in the model folder.

About

Edge Impulse machine learning firmware for Alif AppKit gen2

Resources

License

Apache-2.0, Unknown licenses found

Licenses found

Apache-2.0
LICENSE
Unknown
License.txt

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages