# AMB82-MINI on Edge Impulse

**URL:** <https://forum.amebaiot.com/t/amb82-mini-on-edge-impulse/4657>\
**Category:** Hardware\
**Created:** [January 15, 2026, 7:11am UTC](https://forum.amebaiot.com/t/amb82-mini-on-edge-impulse/4657 "2026-01-15T07:11:12Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Arif](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.amebaiot.com/arif/32/3037_2.png) [@Arif](https://forum.amebaiot.com/u/Arif)\
**Post date:** [January 15, 2026, 7:11am UTC](https://forum.amebaiot.com/t/amb82-mini-on-edge-impulse/4657/1 "2026-01-15T07:11:12Z")

</div>

Has anyone ever know how to use AMB82-MINI on Edge Impulse?

---

<div class="post-metadata">

**Author:** ![M-ichae-l](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.amebaiot.com/m-ichae-l/32/33_2.png) [@M-ichae-l](https://forum.amebaiot.com/u/M-ichae-l)\
**Post date:** [February 4, 2026, 3:04am UTC](https://forum.amebaiot.com/t/amb82-mini-on-edge-impulse/4657/2 "2026-02-04T03:04:48Z")

</div>

Hi, thanks for trying ameba for your projects.

Edge Impulse doe support 3rd party embedded boards that compatible with Arduino.

Let us follow the steps to make use of the edge impulse.

1, we can make use of edge impulse to train the model we wanted. Then export the model.

- note that the models need to convert before used by ameba, lets try study how to process the [using-customized-nn-model](https://ameba-doc-ai-video-analytics-doc.readthedocs-hosted.com/en/latest/user_manual/NN_deployment/NN_deployment.html#using-customized-nn-model) first
- At edge impulse, refer the following to export the model that have trained. In my case, I have trained mobilenetv2 for image classification. I prefer to use tensorflow lite file that supported by AMB82-mini lately. We just need to find the `.tflite file`.
- Remember to check the model classes order, in my case is `Cups, Keys, unknown`
- Also remember to download at least 1 image file from “image“ for future quantization, my case is “001.jpg“.
- ![001](https://us1.discourse-cdn.com/flex016/uploads/ameba/original/2X/9/9a2755de68fec41517df091f0b3b785c151700a8.png) ![002](https://us1.discourse-cdn.com/flex016/uploads/ameba/original/2X/a/a3123795137ae396925bb2667f40cfc78687833b.png) ![003](https://us1.discourse-cdn.com/flex016/uploads/ameba/original/2X/5/599409c2edbf5532d102b1b6d280fe51621c08eb.png)

2, refer to the guide to convert NN model, it can be done both online/offline

- [Amebapro2 AI Convert Model – Realtek IoT/Wi-Fi MCU Solutions](https://www.amebaiot.com/en/amebapro2-ai-convert-model/)

- [Realtek AmebaPro2 Model Conversion Tool — AmebaPro2's Documentation v0.1 documentation](https://ameba-doc-ai-video-analytics-doc.readthedocs-hosted.com/en/latest/user_manual/Acuity_tool/index.html)

- I am using the offline tool by docker setup at ubuntu env. [Acuity\_installation\_docker](https://ameba-doc-ai-video-analytics-doc.readthedocs-hosted.com/en/latest/user_manual/Acuity_tool/Acuity_installation_docker.html#)

3, use Arduino IDE to upload the model into board (recommended to use SD card)

- some useful [example guides](https://ameba-doc-arduino-sdk.readthedocs-hosted.com/en/latest/ameba_pro2/amb82-mini/Example_Guides/Neural%20Network/index.html)

- I tried this example [Image Classification](https://ameba-doc-arduino-sdk.readthedocs-hosted.com/en/latest/ameba_pro2/amb82-mini/Example_Guides/Neural%20Network/Image%20Classification.html), refer to following for my setups