{"users":[{"id":1,"username":"janjongboom","name":"Jan Jongboom","avatar_template":"/user_avatar/forum.edgeimpulse.com/janjongboom/{size}/3861_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","admin":true,"trust_level":2},{"id":5873,"username":"marcpous","name":"marcpous","avatar_template":"/letter_avatar_proxy/v4/letter/m/e79b87/{size}.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","admin":true,"moderator":true,"trust_level":2},{"id":7181,"username":"ATXXWelli","name":"ATXXWelli","avatar_template":"/user_avatar/forum.edgeimpulse.com/atxxwelli/{size}/5649_2.png","flair_name":null,"trust_level":0},{"id":2200,"username":"Eoin","name":"Eoin","avatar_template":"/user_avatar/forum.edgeimpulse.com/eoin/{size}/3673_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","admin":true,"moderator":true,"trust_level":2},{"id":4626,"username":"matsujirushi","name":"MATSUOKA Takashi","avatar_template":"/letter_avatar_proxy/v4/letter/m/bc8723/{size}.png","flair_name":null,"trust_level":0},{"id":6891,"username":"edwards09","name":"edwards09","avatar_template":"/letter_avatar_proxy/v4/letter/e/c37758/{size}.png","flair_name":null,"trust_level":1},{"id":7127,"username":"kotobuki","name":"kotobuki","avatar_template":"/letter_avatar_proxy/v4/letter/k/919ad9/{size}.png","flair_name":null,"trust_level":1},{"id":7174,"username":"ozanoner","name":"Ozan","avatar_template":"/letter_avatar_proxy/v4/letter/o/b5e925/{size}.png","flair_name":null,"trust_level":0},{"id":7051,"username":"AS4065U","name":"AS4065U","avatar_template":"/letter_avatar_proxy/v4/letter/a/5e9695/{size}.png","flair_name":null,"trust_level":0},{"id":2030,"username":"jimbruges","name":"Jim Bruges","avatar_template":"/letter_avatar_proxy/v4/letter/j/a5b964/{size}.png","primary_group_name":"expert","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","trust_level":2},{"id":7145,"username":"KCoats","name":"KCoats","avatar_template":"/letter_avatar_proxy/v4/letter/k/839c29/{size}.png","flair_name":null,"trust_level":0},{"id":963,"username":"mateusz","name":"Mateusz Majchrzycki","avatar_template":"/user_avatar/forum.edgeimpulse.com/mateusz/{size}/5021_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","moderator":true,"trust_level":2},{"id":7121,"username":"emonaldi","name":"emonaldi","avatar_template":"/letter_avatar_proxy/v4/letter/e/8797f3/{size}.png","flair_name":null,"trust_level":0},{"id":1849,"username":"ei_francesco","name":"Francesco","avatar_template":"/user_avatar/forum.edgeimpulse.com/ei_francesco/{size}/3708_2.png","primary_group_name":"expert","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","trust_level":2},{"id":7153,"username":"AMIR","name":"AMIR","avatar_template":"/letter_avatar_proxy/v4/letter/a/c37758/{size}.png","flair_name":null,"trust_level":0},{"id":7150,"username":"Smallbeck","name":"Smallbeck","avatar_template":"/letter_avatar_proxy/v4/letter/s/d07c76/{size}.png","flair_name":null,"trust_level":0},{"id":6631,"username":"FengChi","name":"FengChi","avatar_template":"/letter_avatar_proxy/v4/letter/f/d2c977/{size}.png","flair_name":null,"trust_level":1},{"id":391,"username":"AIWintermuteAI","name":"Dmitry","avatar_template":"/letter_avatar_proxy/v4/letter/a/e9a140/{size}.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","trust_level":2},{"id":292,"username":"salmanfarisvp","name":"Salman Faris","avatar_template":"/user_avatar/forum.edgeimpulse.com/salmanfarisvp/{size}/3862_2.png","primary_group_name":"expert","flair_name":"expert","flair_url":"/uploads/default/original/2X/2/28831612ff87df286df263b2e50ceeda25bad8ac.png","flair_color":"a3d008","trust_level":1},{"id":1692,"username":"MMarcial","name":"Developer Marcial","avatar_template":"/user_avatar/forum.edgeimpulse.com/mmarcial/{size}/3678_2.png","primary_group_name":"expert","flair_name":"expert","flair_url":"/uploads/default/original/2X/2/28831612ff87df286df263b2e50ceeda25bad8ac.png","flair_color":"a3d008","trust_level":2},{"id":1200,"username":"shawn_edgeimpulse","name":"Shawn","avatar_template":"/user_avatar/forum.edgeimpulse.com/shawn_edgeimpulse/{size}/3667_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","admin":true,"moderator":true,"trust_level":2},{"id":4989,"username":"nischalj10","name":"Nischal Jain","avatar_template":"/letter_avatar_proxy/v4/letter/n/3ab097/{size}.png","flair_name":null,"trust_level":0},{"id":6911,"username":"bobber","name":"Robert Gallup","avatar_template":"/letter_avatar_proxy/v4/letter/b/aca169/{size}.png","flair_name":null,"trust_level":1},{"id":7010,"username":"Paul99","name":"Paul99","avatar_template":"/letter_avatar_proxy/v4/letter/p/0ea827/{size}.png","flair_name":null,"trust_level":0},{"id":5602,"username":"angelo.cioffari.btc","name":"Angelo Cioffari (BTC AG)","avatar_template":"/user_avatar/forum.edgeimpulse.com/angelo.cioffari.btc/{size}/5212_2.png","flair_name":null,"trust_level":1},{"id":7117,"username":"Prat","name":"Prat","avatar_template":"/letter_avatar_proxy/v4/letter/p/7bcc69/{size}.png","flair_name":null,"trust_level":0},{"id":261,"username":"jenny","name":"Jenny Plunkett","avatar_template":"/user_avatar/forum.edgeimpulse.com/jenny/{size}/5083_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","moderator":true,"trust_level":2},{"id":7104,"username":"salmanedgeimpulse","name":"SalmanEI","avatar_template":"/letter_avatar_proxy/v4/letter/s/8dc957/{size}.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","trust_level":0},{"id":5549,"username":"eduardsik","name":"Eduard Sik","avatar_template":"/letter_avatar_proxy/v4/letter/e/6f9a4e/{size}.png","flair_name":null,"trust_level":0},{"id":7034,"username":"marzouguioussema","name":"marzouguioussema","avatar_template":"/letter_avatar_proxy/v4/letter/m/97f17d/{size}.png","flair_name":null,"trust_level":1},{"id":919,"username":"rjames","name":"Raul James","avatar_template":"/user_avatar/forum.edgeimpulse.com/rjames/{size}/3645_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","admin":true,"moderator":true,"trust_level":2},{"id":7091,"username":"RichardWOttawa","name":"RichardWOttawa","avatar_template":"/letter_avatar_proxy/v4/letter/r/ecd19e/{size}.png","flair_name":null,"trust_level":0},{"id":7050,"username":"Yeriel","name":"Yeriel","avatar_template":"/letter_avatar_proxy/v4/letter/y/8edcca/{size}.png","flair_name":null,"trust_level":0},{"id":7089,"username":"mos_26","name":"mos_26","avatar_template":"/letter_avatar_proxy/v4/letter/m/71e660/{size}.png","flair_name":null,"trust_level":0},{"id":7014,"username":"danielferreira17","name":"danielferreira17","avatar_template":"/letter_avatar_proxy/v4/letter/d/e9c0ed/{size}.png","flair_name":null,"trust_level":0},{"id":6944,"username":"jiahsong","name":"J.H. Song","avatar_template":"/letter_avatar_proxy/v4/letter/j/58956e/{size}.png","flair_name":null,"trust_level":1},{"id":2402,"username":"brianmcfadden","name":"Brian McFadden","avatar_template":"/user_avatar/forum.edgeimpulse.com/brianmcfadden/{size}/5353_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","admin":true,"moderator":true,"trust_level":2},{"id":7021,"username":"OutterKlan","name":"OutterKlan","avatar_template":"/letter_avatar_proxy/v4/letter/o/b9bd4f/{size}.png","flair_name":null,"trust_level":0},{"id":7013,"username":"Demanico","name":"Demanico","avatar_template":"/letter_avatar_proxy/v4/letter/d/bc8723/{size}.png","flair_name":null,"trust_level":0},{"id":7052,"username":"edbrodbelt","name":"edbrodbelt","avatar_template":"/letter_avatar_proxy/v4/letter/e/f19dbf/{size}.png","flair_name":null,"trust_level":0},{"id":1935,"username":"davidtischler_edgeim","name":"David Tischler","avatar_template":"/user_avatar/forum.edgeimpulse.com/davidtischler_edgeim/{size}/3837_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","trust_level":2},{"id":3279,"username":"dayberry","name":"Brennan Dayberry","avatar_template":"/user_avatar/forum.edgeimpulse.com/dayberry/{size}/4163_2.png","primary_group_name":"team","flair_name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","trust_level":1},{"id":6169,"username":"LucasMCabral","name":"LucasMCabral","avatar_template":"/letter_avatar_proxy/v4/letter/l/35a633/{size}.png","flair_name":null,"trust_level":1}],"primary_groups":[{"id":42,"name":"team"},{"id":41,"name":"expert"}],"flair_groups":[{"id":42,"name":"team","flair_url":"/uploads/default/original/1X/a886d3f017b9dbdd19c0dfc38b6f0af02d13ac9b.png","flair_bg_color":"","flair_color":""},{"id":41,"name":"expert","flair_url":"/uploads/default/original/2X/2/28831612ff87df286df263b2e50ceeda25bad8ac.png","flair_bg_color":"","flair_color":"a3d008"}],"topic_list":{"can_create_topic":false,"more_topics_url":"/c/need-help-with-your-project/8?page=1","per_page":30,"top_tags":["arduino","esp32","cli","rpi4","dsp","linux","stm32","esp32-s3","nicla","nvidia","openmv","syntiant","byom","nordic","rpi5","onnx","jetson","cpp","embedded","linux-sdk-python","ndp","wasm","-esp32s3","himax","sales","ti","university","windows","ml","cmsis","college"],"topics":[{"id":1536,"title":"About the Help category","fancy_title":"About the Help category","slug":"about-the-help-category","posts_count":2,"reply_count":0,"highest_post_number":5,"image_url":null,"created_at":"2021-03-25T17:39:53.586Z","last_posted_at":"2025-12-29T10:15:12.733Z","bumped":true,"bumped_at":"2025-03-04T13:17:19.469Z","archetype":"regular","unseen":false,"pinned":true,"unpinned":null,"excerpt":null,"visible":true,"closed":true,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":1811,"like_count":2,"has_summary":false,"last_poster_username":"marcpous","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":1,"primary_group_id":42,"flair_group_id":42},{"extras":"latest","description":"Most Recent Poster","user_id":5873,"primary_group_id":42,"flair_group_id":42}]},{"id":17979,"title":"Accuracy difference between web deployment and deployment on Arduino Uno Q","fancy_title":"Accuracy difference between web deployment and deployment on Arduino Uno Q","slug":"accuracy-difference-between-web-deployment-and-deployment-on-arduino-uno-q","posts_count":8,"reply_count":5,"highest_post_number":8,"image_url":null,"created_at":"2026-04-21T07:52:04.917Z","last_posted_at":"2026-04-22T08:52:20.278Z","bumped":true,"bumped_at":"2026-04-22T08:52:20.278Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nHey, I trained a FOMO and YOLO Model that work very well, when deployed for Web. But when I deploy the same model with the same camera on my arduino, the accuracy dopps by about 30-40% \nProject ID: \n[Pro…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino"],"tags_descriptions":{},"views":16,"like_count":0,"has_summary":false,"last_poster_username":"ATXXWelli","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7181,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":5873,"primary_group_id":42,"flair_group_id":42},{"extras":null,"description":"Frequent Poster","user_id":2200,"primary_group_id":42,"flair_group_id":42}]},{"id":17986,"title":"How to configure label for image classification","fancy_title":"How to configure label for image classification","slug":"how-to-configure-label-for-image-classification","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":"https://forum.edgeimpulse.com/uploads/default/optimized/2X/5/57d13cf901d98e1d31f34c6305dd5bb4b3fd2c68_2_1024x438.jpeg","created_at":"2026-04-22T02:03:36.605Z","last_posted_at":"2026-04-22T02:04:22.260Z","bumped":true,"bumped_at":"2026-04-22T02:04:22.260Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nI believe that to perform image classification, I need to upload images and assign labels to them. However, I’m not sure how to add or configure the labels. \nI set the labels when capturing images with t…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":4,"like_count":0,"has_summary":false,"last_poster_username":"matsujirushi","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster","user_id":4626,"primary_group_id":null,"flair_group_id":null}]},{"id":17453,"title":"Stm32s3 static buffer example for classification not working as expected","fancy_title":"Stm32s3 static buffer example for classification not working as expected","slug":"stm32s3-static-buffer-example-for-classification-not-working-as-expected","posts_count":4,"reply_count":1,"highest_post_number":4,"image_url":"https://forum.edgeimpulse.com/uploads/default/optimized/2X/7/7c08fb151dee9e865689baf82035bb6ef91f88e4_2_1024x570.png","created_at":"2026-01-07T23:51:44.166Z","last_posted_at":"2026-04-17T23:02:12.273Z","bumped":true,"bumped_at":"2026-04-17T23:02:12.273Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nI have seemingly successfully trained an audio classification model.  The Live Classification seems to return 100% for my 3 classes which is great and using the Live Classification from my test sample se…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","esp32","dsp"],"tags_descriptions":{},"views":297,"like_count":1,"has_summary":false,"last_poster_username":"ozanoner","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":6891,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":7127,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7174,"primary_group_id":null,"flair_group_id":null}]},{"id":17772,"title":"Improving accuracy on npd120 model","fancy_title":"Improving accuracy on npd120 model","slug":"improving-accuracy-on-npd120-model","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-03-04T15:33:02.790Z","last_posted_at":"2026-04-17T10:12:41.185Z","bumped":true,"bumped_at":"2026-04-17T10:12:41.185Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \n[Describe the question or issue in detail] \nEven after adding an external dataset my accuracy is below 40% \nProject ID: \n[Provide the project ID] \nAS4065U-project-1 \nContext/Use case: \n[Provide context o…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","cli","rpi4","nicla"],"tags_descriptions":{},"views":173,"like_count":0,"has_summary":false,"last_poster_username":"jimbruges","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7051,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":2030,"primary_group_id":41,"flair_group_id":42}]},{"id":17926,"title":"Problem Connecting SiLabs xG24 board","fancy_title":"Problem Connecting SiLabs xG24 board","slug":"problem-connecting-silabs-xg24-board","posts_count":7,"reply_count":3,"highest_post_number":7,"image_url":null,"created_at":"2026-04-08T03:03:42.360Z","last_posted_at":"2026-04-15T14:59:40.436Z","bumped":true,"bumped_at":"2026-04-15T14:59:40.436Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hello!  I just set up my Edge Impulse account today and I’m trying to create a gesture recognition project.  I’m unable to connect my SiLabs xG24 board to my project after following the steps outlined in the official doc…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["cli"],"tags_descriptions":{},"views":124,"like_count":0,"has_summary":false,"last_poster_username":"mateusz","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7145,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":963,"primary_group_id":42,"flair_group_id":42}]},{"id":17898,"title":"Error while connecting to Edge Impulse (Ambiq Apollo 5)","fancy_title":"Error while connecting to Edge Impulse (Ambiq Apollo 5)","slug":"error-while-connecting-to-edge-impulse-ambiq-apollo-5","posts_count":8,"reply_count":3,"highest_post_number":8,"image_url":null,"created_at":"2026-04-02T09:50:05.767Z","last_posted_at":"2026-04-15T12:41:28.171Z","bumped":true,"bumped_at":"2026-04-15T12:41:28.171Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"I followed all the instructions, and also run the flas_win.bat successfully. When I try to run the edge-impulse-daemon it’s not able to estabilish connection and the terminal shows me \n“[SER] Serial is connected, trying …","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":98,"like_count":0,"has_summary":false,"last_poster_username":"ei_francesco","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7121,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":2200,"primary_group_id":42,"flair_group_id":42},{"extras":"latest","description":"Most Recent Poster","user_id":1849,"primary_group_id":41,"flair_group_id":42}]},{"id":17940,"title":"Enterprise Plan Request For PCB Defect Detection System Project","fancy_title":"Enterprise Plan Request For PCB Defect Detection System Project","slug":"enterprise-plan-request-for-pcb-defect-detection-system-project","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-04-11T13:21:09.928Z","last_posted_at":"2026-04-13T14:51:44.002Z","bumped":true,"bumped_at":"2026-04-13T14:51:44.002Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hi everyone, \nI’m currently working on my Final Year Project involving PCB defect detection using FOMO deployed on a Raspberry Pi 5. \nI have successfully prepared my dataset and trained an initial model, but I am current…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["rpi4","rpi5"],"tags_descriptions":{},"views":71,"like_count":0,"has_summary":false,"last_poster_username":"jimbruges","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7153,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":2030,"primary_group_id":41,"flair_group_id":42}]},{"id":17934,"title":"Need help with first UNO Q model","fancy_title":"Need help with first UNO Q model","slug":"need-help-with-first-uno-q-model","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-04-09T22:03:53.304Z","last_posted_at":"2026-04-12T19:09:27.782Z","bumped":true,"bumped_at":"2026-04-12T19:09:27.782Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nCreated a new keyword spotting model for a custom wake word.  Test great in edge impulse.  Using it on example ‘Hey Arduino’ and wake word does not trigger. \nProject ID: \nSmallbeck project-1 \nContext/Use…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","cli","rpi4"],"tags_descriptions":{},"views":69,"like_count":0,"has_summary":false,"last_poster_username":"marcpous","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7150,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":5873,"primary_group_id":42,"flair_group_id":42}]},{"id":17935,"title":"Dual model deployment","fancy_title":"Dual model deployment","slug":"dual-model-deployment","posts_count":3,"reply_count":1,"highest_post_number":3,"image_url":null,"created_at":"2026-04-10T01:49:30.534Z","last_posted_at":"2026-04-10T11:31:44.000Z","bumped":true,"bumped_at":"2026-04-10T11:31:44.000Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Is this platform able to deploy two models on the same MCU at the same time, and then switch between them during inference?","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":57,"like_count":0,"has_summary":false,"last_poster_username":"FengChi","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":6631,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster, Accepted Answer","user_id":391,"primary_group_id":42,"flair_group_id":42}]},{"id":8401,"title":"Error \"Arena size is too small for all buffers\" while inferencing audio project on XIAO ESP32-S3","fancy_title":"Error “Arena size is too small for all buffers” while inferencing audio project on XIAO ESP32-S3","slug":"error-arena-size-is-too-small-for-all-buffers-while-inferencing-audio-project-on-xiao-esp32-s3","posts_count":17,"reply_count":8,"highest_post_number":17,"image_url":null,"created_at":"2023-08-30T15:09:01.931Z","last_posted_at":"2026-04-08T22:46:00.523Z","bumped":true,"bumped_at":"2026-04-08T22:46:00.523Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hi, I was trying to inference an audio KWS project on XIAO ESP-S3 with Sense HAT. The device was working fine on the previous audio project, only facing issues with this particular project. Thanks in Advance. \nERR: Faile…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":2094,"like_count":3,"has_summary":false,"last_poster_username":"kotobuki","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":292,"primary_group_id":41,"flair_group_id":41},{"extras":null,"description":"Frequent Poster","user_id":1692,"primary_group_id":41,"flair_group_id":41},{"extras":null,"description":"Frequent Poster","user_id":1200,"primary_group_id":42,"flair_group_id":42},{"extras":null,"description":"Frequent Poster","user_id":4989,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":7127,"primary_group_id":null,"flair_group_id":null}]},{"id":17919,"title":"Failed to initialize linux tool Error","fancy_title":"Failed to initialize linux tool Error","slug":"failed-to-initialize-linux-tool-error","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-04-06T17:00:01.349Z","last_posted_at":"2026-04-07T14:18:11.090Z","bumped":true,"bumped_at":"2026-04-07T14:48:50.370Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nI just tried connecting my development board (UNO Q) to a newly-created project using edge-impulse-linux --clean. I receive the following error: \nFailed to initialize linux tool Error: Failed to load dev…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","cli","linux"],"tags_descriptions":{},"views":69,"like_count":0,"has_summary":false,"last_poster_username":"jimbruges","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":6911,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":2030,"primary_group_id":41,"flair_group_id":42}]},{"id":17893,"title":"How do I add new keywords to the audio classification impulse that I generated by following the tutorial?","fancy_title":"How do I add new keywords to the audio classification impulse that I generated by following the tutorial?","slug":"how-do-i-add-new-keywords-to-the-audio-classification-impulse-that-i-generated-by-following-the-tutorial","posts_count":4,"reply_count":1,"highest_post_number":4,"image_url":null,"created_at":"2026-04-01T13:30:44.283Z","last_posted_at":"2026-04-02T12:29:43.304Z","bumped":true,"bumped_at":"2026-04-02T12:29:43.304Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nUnable to support more than one keyword in audio classification impulse. \nI have just started with Edge Impulse and am finding it difficult to use. The tutorial leads to a working impulse for one keyword…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":123,"like_count":0,"has_summary":false,"last_poster_username":"Paul99","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7010,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":2200,"primary_group_id":42,"flair_group_id":42}]},{"id":13919,"title":"Jetson Nano. Error installing Edge Impulse Linux Python SDK","fancy_title":"Jetson Nano. Error installing Edge Impulse Linux Python SDK","slug":"jetson-nano-error-installing-edge-impulse-linux-python-sdk","posts_count":7,"reply_count":3,"highest_post_number":7,"image_url":null,"created_at":"2025-04-27T12:58:37.460Z","last_posted_at":"2026-04-02T11:35:29.470Z","bumped":true,"bumped_at":"2026-04-02T11:35:29.470Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nI am trying to install the Edge Impulse Linux Python SDK on a Jetson Nano following this documentation \nLinux Python SDK - Edge Impulse Documentation \nBut I have a problem when I try to install edge_impu…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["linux","nvidia"],"tags_descriptions":{},"views":1042,"like_count":9,"has_summary":false,"last_poster_username":"Eoin","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"posters":[{"extras":null,"description":"Original Poster","user_id":5602,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster, Accepted Answer","user_id":2200,"primary_group_id":42,"flair_group_id":42}]},{"id":17890,"title":"Edge AI for middle school students","fancy_title":"Edge AI for middle school students","slug":"edge-ai-for-middle-school-students","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-03-31T14:53:53.507Z","last_posted_at":"2026-04-02T11:18:09.718Z","bumped":true,"bumped_at":"2026-04-02T11:18:09.718Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hi, Since AI is penetrating fast everywhere i recently bought Adruino UNO Q for my middle school kid. However we are missing a place where there is structured learning. Can you help us finding some best courses or tutori…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino"],"tags_descriptions":{},"views":96,"like_count":0,"has_summary":false,"last_poster_username":"jenny","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7117,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":261,"primary_group_id":42,"flair_group_id":42}]},{"id":17852,"title":"Audio Sample Splitter Glitches?","fancy_title":"Audio Sample Splitter Glitches?","slug":"audio-sample-splitter-glitches","posts_count":6,"reply_count":3,"highest_post_number":6,"image_url":null,"created_at":"2026-03-24T17:09:36.231Z","last_posted_at":"2026-04-01T09:56:19.552Z","bumped":true,"bumped_at":"2026-04-01T09:56:19.552Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"I’m using the sample splitter to separate a 30-second sample into individual keyword samples. The splitter generally works, but when I zoom in then out, it forgets the last sample window and moves it somewhere else in th…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":131,"like_count":1,"has_summary":false,"last_poster_username":"jimbruges","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":6911,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":7104,"primary_group_id":42,"flair_group_id":42},{"extras":"latest","description":"Most Recent Poster","user_id":2030,"primary_group_id":41,"flair_group_id":42}]},{"id":13833,"title":"M5Stack Core2 Keyword Recognition - Why Does It Work Perfectly on Phone But Fails on the Board?","fancy_title":"M5Stack Core2 Keyword Recognition - Why Does It Work Perfectly on Phone But Fails on the Board?","slug":"m5stack-core2-keyword-recognition-why-does-it-work-perfectly-on-phone-but-fails-on-the-board","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2025-04-14T12:26:20.459Z","last_posted_at":"2026-03-30T07:39:37.185Z","bumped":true,"bumped_at":"2026-03-30T07:39:37.185Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hello Community! \nI’m working on a keyword recognition project with an M5Stack Core2 board and I’ve hit a roadblock. I hope someone can help or provide advice. \n\nThe Project\n\nUsing M5Stack Core2 board for keyword recogni…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","esp32"],"tags_descriptions":{},"views":429,"like_count":1,"has_summary":false,"last_poster_username":"AIWintermuteAI","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":5549,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":391,"primary_group_id":42,"flair_group_id":42}]},{"id":17869,"title":"Can't Change Project Target","fancy_title":"Can’t Change Project Target","slug":"cant-change-project-target","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-03-28T22:46:28.854Z","last_posted_at":"2026-03-29T00:00:02.873Z","bumped":true,"bumped_at":"2026-03-29T00:00:02.873Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"The project target control (upper-right) doesn’t work for me any more. If I choose a new target (e.g. select Rubik Pi 3), there’s a message behind the pop-up target dialog that’s impossible to read before it disappears a…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["rpi4"],"tags_descriptions":{},"views":77,"like_count":2,"has_summary":false,"last_poster_username":"bobber","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"posters":[{"extras":"latest single","description":"Original Poster, Most Recent Poster, Accepted Answer","user_id":6911,"primary_group_id":null,"flair_group_id":null}]},{"id":17861,"title":"Hello everyone Im making a smar recycle bin with esp32 and inmp441","fancy_title":"Hello everyone Im making a smar recycle bin with esp32 and inmp441","slug":"hello-everyone-im-making-a-smar-recycle-bin-with-esp32-and-inmp441","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-03-26T22:33:11.652Z","last_posted_at":"2026-03-27T09:12:40.335Z","bumped":true,"bumped_at":"2026-03-27T09:12:40.335Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"I did collected droping sounds in format audio.wav then i aploaded in edge impulse and i started traning the dara and when i upoad the library in arduino ide and start real predection the system went all wrong i made 3 c…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","esp32"],"tags_descriptions":{},"views":101,"like_count":0,"has_summary":false,"last_poster_username":"AIWintermuteAI","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7034,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":391,"primary_group_id":42,"flair_group_id":42}]},{"id":17851,"title":"Edge-impulse-linux-runner --threshold option not parsing value?","fancy_title":"Edge-impulse-linux-runner –threshold option not parsing value?","slug":"edge-impulse-linux-runner-threshold-option-not-parsing-value","posts_count":3,"reply_count":1,"highest_post_number":3,"image_url":"https://forum.edgeimpulse.com/uploads/default/original/2X/1/1dc7dbcee88cc71e710c8912c6e42a2c0590e3a6.png","created_at":"2026-03-24T15:45:03.529Z","last_posted_at":"2026-03-24T17:03:45.416Z","bumped":true,"bumped_at":"2026-03-24T17:43:09.528Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"I’m experimenting with edge-impulse-linux-runner and have tried using the --threshold option. I get the following output which suggests the --threshold value isn’t being parsed: \n  \nI’ve tried several different formats f…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["linux"],"tags_descriptions":{},"views":95,"like_count":0,"has_summary":false,"last_poster_username":"bobber","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":6911,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":919,"primary_group_id":42,"flair_group_id":42}]},{"id":17845,"title":"OpenMV AE3 audio classification","fancy_title":"OpenMV AE3 audio classification","slug":"openmv-ae3-audio-classification","posts_count":3,"reply_count":0,"highest_post_number":3,"image_url":null,"created_at":"2026-03-22T15:13:55.954Z","last_posted_at":"2026-03-24T12:02:14.915Z","bumped":true,"bumped_at":"2026-03-24T12:08:46.467Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nI’ve trained an audio model using Edge Impulse - this was straightforward. I selected the “Alif DevKit Gen2 HE core” and Inference Engine “TensorFlow Lite”. These seemed the closed fit to my OpenMV AE3 -…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","cli","rpi4","openmv"],"tags_descriptions":{},"views":124,"like_count":0,"has_summary":false,"last_poster_username":"RichardWOttawa","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7091,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":1849,"primary_group_id":41,"flair_group_id":42}]},{"id":17822,"title":"FOMO Arduino library not generating bounding boxes (ESP32)","fancy_title":"FOMO Arduino library not generating bounding boxes (ESP32)","slug":"fomo-arduino-library-not-generating-bounding-boxes-esp32","posts_count":6,"reply_count":3,"highest_post_number":6,"image_url":null,"created_at":"2026-03-16T16:11:49.864Z","last_posted_at":"2026-03-23T17:12:43.026Z","bumped":true,"bumped_at":"2026-03-23T17:12:43.026Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"I’m using FOMO MobileNetV2 0.1 for object detection. The model works in \nEdge Impulse testing (shows centroids), but the deployed Arduino library \nreturns 0 bounding boxes despite successful 126ms inference. \n\nProject: Y…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","esp32-s3"],"tags_descriptions":{},"views":153,"like_count":0,"has_summary":false,"last_poster_username":"AIWintermuteAI","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7050,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":391,"primary_group_id":42,"flair_group_id":42}]},{"id":17842,"title":"Edge_impulse_linux module not found","fancy_title":"Edge_impulse_linux module not found","slug":"edge-impulse-linux-module-not-found","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-03-22T05:04:01.585Z","last_posted_at":"2026-03-23T08:48:07.278Z","bumped":true,"bumped_at":"2026-03-23T08:48:07.278Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hi, I am a beginner on edge impulse. I tried to run python code containing  the line from edge_impulse_linux.image import ImageImpulseRunner on R-pi 5 (pi os: trixie) but python return error: module edge_impulse_linux no…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["rpi4","linux"],"tags_descriptions":{},"views":99,"like_count":0,"has_summary":false,"last_poster_username":"AIWintermuteAI","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7089,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":391,"primary_group_id":42,"flair_group_id":42}]},{"id":17824,"title":"Need help capturing audio from X-NUCLEO-CCA02M2 on STM32N657 (Edge Impulse working with raw features only)","fancy_title":"Need help capturing audio from X-NUCLEO-CCA02M2 on STM32N657 (Edge Impulse working with raw features only)","slug":"need-help-capturing-audio-from-x-nucleo-cca02m2-on-stm32n657-edge-impulse-working-with-raw-features-only","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-03-16T16:12:04.150Z","last_posted_at":"2026-03-17T14:41:28.355Z","bumped":true,"bumped_at":"2026-03-17T14:41:28.355Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hi Edge Impulse Team and Community, \nI’m working on deploying an audio model on an STM32N6 using the X-NUCLEO-CCA02M2 microphone expansion board with the NUCLEO-N657X0-Q, and I’m currently stuck at the audio capture stag…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["stm32"],"tags_descriptions":{},"views":124,"like_count":0,"has_summary":false,"last_poster_username":"ei_francesco","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7014,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":1849,"primary_group_id":41,"flair_group_id":42}]},{"id":17751,"title":"Data directory structure for custom learning block & deployment block","fancy_title":"Data directory structure for custom learning block & deployment block","slug":"data-directory-structure-for-custom-learning-block-deployment-block","posts_count":3,"reply_count":0,"highest_post_number":3,"image_url":null,"created_at":"2026-02-27T20:18:09.952Z","last_posted_at":"2026-03-10T18:49:26.753Z","bumped":true,"bumped_at":"2026-03-10T18:49:26.753Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hello! \nWhen working with custom deployment blocks, the layout of the /data directory mentioned on docs-dot-edgeimpulse-dot-com/studio/organizations/custom-blocks/custom-learning-blocks does not seem to align with realit…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":160,"like_count":0,"has_summary":false,"last_poster_username":"brianmcfadden","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":6944,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":2402,"primary_group_id":42,"flair_group_id":42}]},{"id":17738,"title":"How to connect esp32 + inmp441 there is no firmware for this","fancy_title":"How to connect esp32 + inmp441 there is no firmware for this","slug":"how-to-connect-esp32-inmp441-there-is-no-firmware-for-this","posts_count":3,"reply_count":1,"highest_post_number":3,"image_url":null,"created_at":"2026-02-26T04:35:37.018Z","last_posted_at":"2026-03-10T18:34:07.496Z","bumped":true,"bumped_at":"2026-03-10T18:34:07.496Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"How to connect esp32 + inmp441 there is no firmware for this please help im building a smart recycling bin and i am sticking with this probleme","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":[],"tags_descriptions":{},"views":188,"like_count":0,"has_summary":false,"last_poster_username":"marzouguioussema","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7034,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":391,"primary_group_id":42,"flair_group_id":42}]},{"id":17707,"title":"Difficulty uploading 1,000+ small WAV files: Browser timeouts and API 415 errors","fancy_title":"Difficulty uploading 1,000+ small WAV files: Browser timeouts and API 415 errors","slug":"difficulty-uploading-1-000-small-wav-files-browser-timeouts-and-api-415-errors","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-02-22T03:32:35.995Z","last_posted_at":"2026-03-10T15:34:49.961Z","bumped":true,"bumped_at":"2026-03-10T15:34:49.961Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: Difficulty uploading a batch of 1,121 small WAV files via the web interface and Ingestion API. \nProject Details: \n\n\nTarget hardware: ESP32-S3 (M5Stack CoreS3).\n\nDataset: Audio anomaly detection (ToyADMOS/…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["cli","esp32-s3"],"tags_descriptions":{},"views":198,"like_count":0,"has_summary":false,"last_poster_username":"brianmcfadden","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7021,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":2402,"primary_group_id":42,"flair_group_id":42}]},{"id":17683,"title":"I can not scan the Qr code in the connect device phone section","fancy_title":"I can not scan the Qr code in the connect device phone section","slug":"i-can-not-scan-the-qr-code-in-the-connect-device-phone-section","posts_count":2,"reply_count":0,"highest_post_number":2,"image_url":null,"created_at":"2026-02-18T07:55:33.273Z","last_posted_at":"2026-03-10T15:12:17.468Z","bumped":true,"bumped_at":"2026-03-10T15:12:17.468Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \n[* can not scan the Qr code in the connect device phone section*] \nProject ID: \n[Provide the project ID] \nContext/Use case: \n[Provide context or use case where the issue is encountered] \nSteps Taken: \n\n[…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino","cli","rpi4"],"tags_descriptions":{},"views":174,"like_count":0,"has_summary":false,"last_poster_username":"brianmcfadden","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":null,"description":"Original Poster","user_id":7013,"primary_group_id":null,"flair_group_id":null},{"extras":"latest","description":"Most Recent Poster","user_id":2402,"primary_group_id":42,"flair_group_id":42}]},{"id":17773,"title":"QCS6490 Tria Vision AI Kit Edge Impulse Can't Recognise Web Cams (MIPI)","fancy_title":"QCS6490 Tria Vision AI Kit Edge Impulse Can’t Recognise Web Cams (MIPI)","slug":"qcs6490-tria-vision-ai-kit-edge-impulse-cant-recognise-web-cams-mipi","posts_count":4,"reply_count":1,"highest_post_number":4,"image_url":"https://forum.edgeimpulse.com/uploads/default/original/2X/9/96a4a0153345729f2f93cb362a925df7610c759b.jpeg","created_at":"2026-03-04T16:05:53.224Z","last_posted_at":"2026-03-09T16:14:58.974Z","bumped":true,"bumped_at":"2026-03-09T16:14:58.974Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Question/Issue: \nTrying to follow the Tria Vision AI Kit Deployment, get to running the Edge Impulse tool on the Vision AI Kit, but it doesn’t recognise any webcams. I have a MIPI Camera plugged into MIPI interface 1, ho…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["linux"],"tags_descriptions":{},"views":166,"like_count":0,"has_summary":false,"last_poster_username":"edbrodbelt","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":false,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":7052,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster","user_id":1935,"primary_group_id":42,"flair_group_id":42},{"extras":null,"description":"Frequent Poster","user_id":3279,"primary_group_id":42,"flair_group_id":42}]},{"id":17780,"title":"Discrepancy between results shown on the Edge Impulse platform and those returned by an Arduino Nano 33 BLE Sense using an LSTM layer","fancy_title":"Discrepancy between results shown on the Edge Impulse platform and those returned by an Arduino Nano 33 BLE Sense using an LSTM layer","slug":"discrepancy-between-results-shown-on-the-edge-impulse-platform-and-those-returned-by-an-arduino-nano-33-ble-sense-using-an-lstm-layer","posts_count":3,"reply_count":1,"highest_post_number":3,"image_url":null,"created_at":"2026-03-05T20:04:05.199Z","last_posted_at":"2026-03-06T19:24:48.401Z","bumped":true,"bumped_at":"2026-03-06T19:24:48.401Z","archetype":"regular","unseen":false,"pinned":false,"unpinned":null,"excerpt":"Hello, \nI have been testing various architectures to classify ECG signals that are fed to my network as time series. When working with models composed only of convolutional and dense layers I had no problems deploying th…","visible":true,"closed":false,"archived":false,"bookmarked":null,"liked":null,"tags":["arduino"],"tags_descriptions":{},"views":177,"like_count":0,"has_summary":false,"last_poster_username":"LucasMCabral","category_id":8,"pinned_globally":false,"featured_link":null,"has_accepted_answer":true,"posters":[{"extras":"latest","description":"Original Poster, Most Recent Poster","user_id":6169,"primary_group_id":null,"flair_group_id":null},{"extras":null,"description":"Frequent Poster, Accepted Answer","user_id":391,"primary_group_id":42,"flair_group_id":42}]}]}}