# Using the tiny\_conv model for image classification tasks

**URL:** <https://discuss.tinyml.seas.harvard.edu/t/using-the-tiny-conv-model-for-image-classification-tasks/984>\
**Category:** Programming Q&A\
**Created:** [February 5, 2022, 3:49am UTC](https://discuss.tinyml.seas.harvard.edu/t/using-the-tiny-conv-model-for-image-classification-tasks/984 "2022-02-05T03:49:09Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![suburban-daredevil](https://avatars.discourse-cdn.com/v4/letter/s/f04885/32.png) [@suburban-daredevil](https://discuss.tinyml.seas.harvard.edu/u/suburban-daredevil)\
**Post date:** [February 5, 2022, 3:49am UTC](https://discuss.tinyml.seas.harvard.edu/t/using-the-tiny-conv-model-for-image-classification-tasks/984/1 "2022-02-05T03:49:09Z")

</div>

I am trying to build a image classification model and deploy it onto the MCU. I have tried building my custom model and deploying. When the serial monitor is opened, it says that “Invoke() failed after Initialization failed”. I think it is because of the memory constraints imposed by the MCU. So I was thinking of using the tiny\_conv model that we used for performing classification of spectograms in KWS. But I was not able to find any resouce using tiny\_conv model for performing image classification. Can you please share some resources that I can use ?

Thank you

---

<div class="post-metadata">

**Author:** ![brian\_plancher](https://yyz1.discourse-cdn.com/flex027/user_avatar/discuss.tinyml.seas.harvard.edu/brian_plancher/32/85_2.png) [@brian\_plancher](https://discuss.tinyml.seas.harvard.edu/u/brian_plancher)\
**Post date:** [April 16, 2022, 5:31pm UTC](https://discuss.tinyml.seas.harvard.edu/t/using-the-tiny-conv-model-for-image-classification-tasks/984/2 "2022-04-16T17:31:40Z")

</div>

You likely need to expand the tensor arena to get over that Invoke() failure. And I don’t have an example handy of using tiny\_conv for image classification but/and you can totally try it. Edge Impulse also has an object detection tutorial with a default model architecture you could try to use: [Dashboard - Tutorial: object detection - Edge Impulse](https://studio.edgeimpulse.com/public/25483/latest)