# Introduction to On-Device AI **URL:** https://community.deeplearning.ai/c/short-course-q-a/introduction-to-on-device-ai/450.md [Latest](https://community.deeplearning.ai/latest.md) · [Categories](https://community.deeplearning.ai/categories.md) · [Tags](https://community.deeplearning.ai/tags.md) --- ## [Running on device](https://community.deeplearning.ai/t/running-on-device/657738)
**Author:** [@taha\_fawzy\_anwar](https://community.deeplearning.ai/u/taha_fawzy_anwar)\ **Replies:** 7\ **Last updated:** [September 22, 2026, 8:09am UTC](https://community.deeplearning.ai/t/running-on-device/657738 "2026-09-22T08:09:48Z")
Hello : I have 2 questions first :-I see that package support only some devices ,which are expensive ones ,do it support those only ? second : I see it compile app for each device ,how to deploy app in this case ,it me… --- ## [Is Intro to On Device Ai just for Snapdragon](https://community.deeplearning.ai/t/is-intro-to-on-device-ai-just-for-snapdragon/789903)
**Author:** [@HowardHonig](https://community.deeplearning.ai/u/HowardHonig)\ **Replies:** 0\ **Last updated:** [March 23, 2025, 4:53pm UTC](https://community.deeplearning.ai/t/is-intro-to-on-device-ai-just-for-snapdragon/789903 "2025-03-23T16:53:29Z")
Is Intro to On Device AI specifically for platforms that have a Snapdragon chip? I started taking the course and it seems to have very heavy emphasis on the functions and libraries of this chip. The intro doesn’t say t… --- ## [Can I get a Transcript of this course](https://community.deeplearning.ai/t/can-i-get-a-transcript-of-this-course/632904)
**Author:** [@108](https://community.deeplearning.ai/u/108)\ **Replies:** 10\ **Last updated:** [January 18, 2025, 5:30am UTC](https://community.deeplearning.ai/t/can-i-get-a-transcript-of-this-course/632904 "2025-01-18T05:30:08Z")
I need a transcript of this course, is there any way to download the transcript. I need this to take notes. Any help is appreciated. Thank you. --- ## [Checking On-Device Accuracy](https://community.deeplearning.ai/t/checking-on-device-accuracy/738150)
**Author:** [@chuckm](https://community.deeplearning.ai/u/chuckm)\ **Replies:** 5\ **Last updated:** [December 22, 2024, 10:41pm UTC](https://community.deeplearning.ai/t/checking-on-device-accuracy/738150 "2024-12-22T22:41:25Z")
Hi, Hoping someone is still following this class. I’m running the course labs locally. In Lab3 I run into an issue while Checking device accuracy. torch\_inputs = torch.Tensor(sample\_inputs\[‘image’\]\[0\]) torch\_outputs =… --- ## [On device is everywhere?](https://community.deeplearning.ai/t/on-device-is-everywhere/675311)
**Author:** [@leejun0402](https://community.deeplearning.ai/u/leejun0402)\ **Replies:** 0\ **Last updated:** [August 9, 2024, 5:11am UTC](https://community.deeplearning.ai/t/on-device-is-everywhere/675311 "2024-08-09T05:11:01Z")
This might be a small question, but what does he mean by 20+ AI models running when you take a picture? What are the AI models? How does that work? Also here, he mentioned that when we write on our keyboards on ou… --- ## [Unable to Unlock Accomplishment for "Introduction to On-Device AI" short course after completing the short course](https://community.deeplearning.ai/t/unable-to-unlock-accomplishment-for-introduction-to-on-device-ai-short-course-after-completing-the-short-course/655982)
**Author:** [@Miratun](https://community.deeplearning.ai/u/Miratun)\ **Replies:** 8\ **Last updated:** [July 2, 2024, 1:42pm UTC](https://community.deeplearning.ai/t/unable-to-unlock-accomplishment-for-introduction-to-on-device-ai-short-course-after-completing-the-short-course/655982 "2024-07-02T13:42:56Z")
I have completed the “Introduction to On-Device AI” course, and my progress shows 100% completion. However, when I try to unlock my accomplishment, I receive the error message: “Something went wrong: user has not complet… --- ## [Problem to run "L2: Deploying Segmentation Models On-Device" locally](https://community.deeplearning.ai/t/problem-to-run-l2-deploying-segmentation-models-on-device-locally/638724)
**Author:** [@Rishan\_Tan](https://community.deeplearning.ai/u/Rishan_Tan)\ **Replies:** 0\ **Last updated:** [June 1, 2024, 3:28am UTC](https://community.deeplearning.ai/t/problem-to-run-l2-deploying-segmentation-models-on-device-locally/638724 "2024-06-01T03:28:22Z")
I have trouble to run “L2: Deploying Segmentation Models On-Device” locally. Is there an easy way to get the model and weight on my local device? Can someone provide the steps to do so? Loading pretrained model state… --- ## [Deploy on my own device without AI Hub cloud?](https://community.deeplearning.ai/t/deploy-on-my-own-device-without-ai-hub-cloud/637087)
**Author:** [@abitrolly](https://community.deeplearning.ai/u/abitrolly)\ **Replies:** 0\ **Last updated:** [May 29, 2024, 1:29pm UTC](https://community.deeplearning.ai/t/deploy-on-my-own-device-without-ai-hub-cloud/637087 "2024-05-29T13:29:17Z")
I’ve just started it. The course promises to run AI without clouds, but it requires AI Hub account for deployment. So is there any way to upload and run models to my phone (also on Raspberry Pi and Linux based home NAS)… --- ## [Pickle Problems](https://community.deeplearning.ai/t/pickle-problems/633043)
**Author:** [@Martin\_Pollard](https://community.deeplearning.ai/u/Martin_Pollard)\ **Replies:** 4\ **Last updated:** [May 26, 2024, 5:02am UTC](https://community.deeplearning.ai/t/pickle-problems/633043 "2024-05-26T05:02:25Z")
Hi, I’ve run into an error unfortunately. Has anyone come across this (‘Deploying Segmentation Models On-Device’ lesson): UnpicklingError: could not find MARK This is from trying to run this locally, by the way. --- ## [Confusion regarding PSNR and differences in inference performance](https://community.deeplearning.ai/t/confusion-regarding-psnr-and-differences-in-inference-performance/633105)
**Author:** [@Nevermnd](https://community.deeplearning.ai/u/Nevermnd)\ **Replies:** 3\ **Last updated:** [May 24, 2024, 6:40am UTC](https://community.deeplearning.ai/t/confusion-regarding-psnr-and-differences-in-inference-performance/633105 "2024-05-24T06:40:29Z")
Ok, So I am a little confused about the difference in inference between the cloud and on device as represented by PSNR. I mean presuming you are using the same fundamental types/bits in both cases (float, double, etc),… --- ## [H5py error: Resolution](https://community.deeplearning.ai/t/h5py-error-resolution/632593)
**Author:** [@Martin\_Pollard](https://community.deeplearning.ai/u/Martin_Pollard)\ **Replies:** 0\ **Last updated:** [May 22, 2024, 11:56am UTC](https://community.deeplearning.ai/t/h5py-error-resolution/632593 "2024-05-22T11:56:37Z")
Hi all, Quick tip: if you come across this error: ImportError: cannot import name '\_errors' from partially initialized module 'h5py' (most likely due to a circular import) Uninstalling and reinstalling the h5py packag…