|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/yolo-v5-label-train-and-test/
课程评论：没有评论
**课程名称：** YOLO v5：标注、训练与测试 **课程概述：** 这门完全实践性的课程将指导您使用最先进的 YOLO v5 算法训练您自己的目标检测器。课程从快速上手开始，让您测试已训练好的 YOLO v5 模型，在图像、视频及实时摄像头画面中检测目标。随后，您将学习如何标注自己的数据集（YOLO 格式），并从海量现有数据集中创建自定义数据集。接着，您将在本地和云端机器上训练 YOLO v5 模型。之后，您将测试在您自定义数据上训练得到的 YOLO v5 检测器。课程还包含一个奖励部分，您将进行练习测试并规划未来的学习方向。课程提供的所有代码模板均可修改并应用于您未来的工作中。本课程可以作为您个人项目的补充，帮助您向导师展示成果，在同学面前进行演示，甚至写入您的简历。 **内容组织：** 每章包含： * 视频讲座 * 代码模板及编码活动 * 测验 * 可下载的说明 * 讨论机会 **SMART 讲座：** 课程的视频讲座均设定了 SMART 目标： * **S - 具体 (Specific)：** 讲座有明确的学习目标。 * **M - 可衡量 (Measurable)：** 学习成果合理且可量化。 * **A - 可实现 (Attainable)：** 讲座包含清晰的步骤来实现目标。 * **R - 结果导向 (Result-oriented)：** 学习者能在讲座结束后获得实际成果。 * **T - 时限性 (Time-oriented)：** 学习成果能在可见的时间框架内达成。 **核心问题：** 该课程旨在解决以下痛点：希望使用 YOLO v5 算法处理自定义数据进行目标检测，但不知道如何开始的学生。 **先修知识：** 学生需要有扎实的 Python 编程基础。对目标检测算法有一定的实践经验（有则更佳，但非强制）。 **课程面向人群：** * 学习计算机视觉的学生； * 希望使用 YOLO v5 进行目标检测的学生； * 希望使用全新数据训练 YOLO v5 的学生； * 希望以 YOLO 格式标注自己数据的学生； * 希望将现有数据转换为 YOLO 格式的学生； * 希望在图像、视频及摄像头画面上测试 YOLO v5 的学生。 **学习目标与期望：** * 构建一个完整的基于 YOLO v5 的目标检测应用程序； * 撰写关于不同目标检测方法的研究论文； * 完成当前的关于目标检测的毕业设计项目； * 在下一次实习或理想工作面试前，提升在 YOLO v5 目标检测方面的硬技能。 **课程结束时您将能够：** * 在图像、视频及实时摄像头画面中应用已训练的 YOLO v5 检测目标； * 标注自己的数据集，并按 YOLO 格式整理文件； * 创建 YOLO 格式的自定义数据集； * 将现有的交通标志数据集转换为 YOLO 格式； * 通过几行代码，使用自定义数据训练 YOLO v5 检测器； * 在本地和云端机器上进行训练和测试。 **课程大纲：** 无
In this completely practical course, you'll train your own object detector by YOLO v5 as the state-of-the-art algorithm.As for the quick start, you'll test already trained YOLO v5 to detect objects on image, video and in real time by camera.After that, you'll label your own dataset in YOLO format and create custom dataset from huge existing one.Next, you'll train YOLO v5 in local machine as well as in cloud machine.Then, you'll test YOLO v5 detector that was trained on your own data.As for the bonus part, you'll pass practice test and plan your next steps.All the code templates can be modified and applied in your future work. The course can supplement your own project that you can represent as the results to your supervisor, or to make a presentation in front of classmates, or even mention it in your resume.Content OrganizationEach Section of the course contains:Video lecturesCode templates and coding activitiesQuizzesDownloadable instructionsDiscussion opportunitiesSMART lecturesVideo lectures of the course have SMART objectives:S - specific (the lecture has specific objectives)M - measurable (results are reasonable and can be quantified)A - attainable (the lecture has clear steps to achieve the objectives)R - result-oriented (results can be obtained by the end of the lecture)T - time-oriented (results can be obtained within the visible time frame)Principle questionsWhat pain point, need, or desire is addressed in the course?The course solves the student's pain point who want to use YOLO v5 algorithm with his/her custom data for object detection but don't know where to start.What is the prior knowledge that student has to have before starting the course?The student has written good amount of the code in Python. May or may not already have some practice of implementing object detection algorithms (good to have but not obligatory).Who is the course for?Student who studies computer vision and:wants to use YOLO v5 for object detection;wants to train YOLO v5 with completely new data;wants to label own data in YOLO format;wants to convert existing data in YOLO format;wants to test YOLO v5 on image, video and by camera.What are the aspirations for taking the course?The student's aspirations are:to build complete application for object detection with YOLO v5;to write scientific paper about different approaches for object detection;to accomplish final project about object detection that he/she might doing now;to improve his/her hard skills in object detection with YOLO v5 before the next interview for the internship or dream job.What will I be able to do at the end of the course?At the end of the course, you will be able to:apply trained YOLO v5 to detect objects on image, video and in real time by camera;label own dataset and structure files in YOLO format;create custom dataset in YOLO format;convert existing dataset of traffic signs in YOLO format;train YOLO v5 detector with custom data and few lines of the code;train and test both: in local machine and in cloud machine.
