# Seeed Ranger Training - Session 1 Source: https://www.youtube.com/watch?v=v6ZATvU2dLs Recap page: https://rapidrecap.app/video/v6ZATvU2dLs Generated: 2026-02-13T08:33:57.418+00:00 --- ## Quick Overview The Seeed Studio reComputer Jetson Edge AI Technology Guide AI training concludes with a review of the four-step process for building a custom defect detection solution using YOLO on Jetson devices, covering L1 (Awareness/Selection), L2 (Integration/Development), L3 (Debugging/After-sales), and a final Q&A section. **Key Points:** - The training objective was to guide attendees on how to become an AI Engineer (AE) using Seeed Studio's reComputer Jetson platform. - The process for custom AI deployment involves four main steps: Algorithm Selection (recommending YOLO), Data Collection (10 sample images), Data Annotation (using Roboflow with 'defect' and 'normal' labels), and Model Training/Deployment. - The demonstration showed that the pre-trained YOLO model failed to recognize custom stickers, necessitating custom training using the collected and annotated data. - Key prerequisites for the training included basic knowledge of Linux and Python, and proficiency in using the Linux system. - The reComputer J4012 is positioned as the most powerful computing platform among edge devices, offering better performance than SoC+NPU alternatives, despite higher power consumption. - System flashing requires an x86 Linux host (like Ubuntu) and specific commands, explicitly warning against using 'sudo apt upgrade' on the kernel. - The final Q&A confirmed that power supplies operate on a range (12-19V) to accommodate varying power requirements and that serial logs are crucial for diagnosing boot failures. ![Screenshot at 13:44: The start of the demonstration phase showing the hardware setup \(camera connected via USB to the reComputer\) and the question posed on how AI replaces manual inspection, leading into the custom training workflow.](https://ss.rapidrecap.app/screens/v6ZATvU2dLs/00-13-44.jpg) **Context:** This video serves as an advanced technical training session, specifically Level 1 (L1) through Level 3 (L3), detailing the end-to-end process for deploying custom Edge AI solutions using Seeed Studio's reComputer Jetson hardware, focusing on a practical example of building a defect detection system for custom stickers. ## Detailed Analysis The training covered three levels: L1 focused on Awareness and Selection, comparing Edge AI needs (inference vs. training) and positioning the reComputer Jetson (a GPU platform) as superior for high-performance edge devices over generic CPUs or SoC+NPU solutions due to its unmatched computing power and ecosystem. L2 detailed the practical development workflow using a custom defect detection scenario: 1. Algorithm Selection (recommending YOLO, guided by GPT), 2. Data Collection (10 sample images), 3. Data Annotation (using Roboflow to label 'defect' and 'normal'), and 4. Model Training/Deployment. The demonstration showed the pre-trained model failing, necessitating custom training. L3 covered system maintenance, including using system logs (like dmesg) for fault diagnosis and warning users not to use 'apt upgrade' on the kernel to avoid compatibility issues with customized drivers. The final quiz confirmed that flashing requires an x86 Linux host and that power supplies accommodate variable voltage needs. ### L1 - Awareness and Selection: The objective is to become an AE by understanding the need for Edge AI (proximal computing, low latency) over Cloud (uncontrollable latency, privacy risks); reComputer Jetson is highlighted as the most powerful edge GPU platform, offering superior performance compared to SoC+NPU and general-purpose CPUs. ### L1 Prerequisites - Basic knowledge of Linux and Python, and proficiency in using the Linux system are required. ### L2 - Integration and Development: The process is broken into four steps: 1. Algorithm Selection (using GPT to recommend YOLO for fast deployment); 2. Data Collection (gathering ~10 sample images); 3. Data Annotation (using Roboflow to label 'defect' and 'normal'); 4. Model Training and Deployment (training locally, then deploying the best.pt weights to the Jetson). ### L2 Workflow Demonstration - The initial generic YOLO model failed to recognize the custom stickers, confirming the need for custom training data (data/domain mismatch). The steps involve collecting images, annotating them, and training a custom YOLO model. ### L3 - Debugging and After-sales: Focuses on troubleshooting, emphasizing that system logs (dmseg, journalctl) are critical for diagnosis. Issues are triaged into Usage, Quality, Software Malfunction, or Hardware Failure categories. Flashing requires specific steps (Prepare image, REC Mode, Execute command) and users must not use 'sudo apt upgrade' to avoid breaking custom drivers. ### System Upgrade & Flashing - Flashing requires an x86 Linux PC and specific commands, while customizing firmware allows for environment cloning for mass deployment. ### Testing Time Q&A - Q7 established that flashing requires a Ubuntu x86 computer (D); Q8 confirms System boot log (Serial Log) is the first piece of information for boot failures (D); Q9 prohibits global apt upgrade because it causes incompatibility with customized drivers (A); Q10 states REC (Recovery) mode should be checked first for flashing timeouts (A). ![Screenshot at 00:00: Title slide outlining the training agenda \(L1, L2, L3\) and prerequisites for becoming an AE.](https://ss.rapidrecap.app/screens/v6ZATvU2dLs/00-00-00.jpg) ![Screenshot at 01:04: Slide explaining why Edge AI is needed, contrasting Training \(Cloud-based\) vs. Inference \(Edge-based\), and summarizing Cloud vs. Edge trade-offs \(latency/resources\).](https://ss.rapidrecap.app/screens/v6ZATvU2dLs/00-01-04.jpg) ![Screenshot at 03:38: Slide comparing General-purpose CPU, SoC+NPU, and GPU Platform, positioning Jetson as the most powerful edge computing platform.](https://ss.rapidrecap.app/screens/v6ZATvU2dLs/00-03-38.jpg) ![Screenshot at 07:04: Slide presenting the testing time questions \(Q1 and Q2\) covering AI Inference definition and the core advantage of Edge computing.](https://ss.rapidrecap.app/screens/v6ZATvU2dLs/00-07-04.jpg) ![Screenshot at 13:44: Demonstration of the defect detection scenario on the assembly line, showing the basic hardware setup \(Camera -\> reComputer -\> HDMI Display\).](https://ss.rapidrecap.app/screens/v6ZATvU2dLs/00-13-44.jpg)