19 Group - Sprint Log 5ed (12.12-12.14)

Killer-Queen 2022-12-23 14:48:31
The Link Your ClassFZU-SE-EE308-2022秋社区-CSDN社区云
The Link of Requirements of This Assignment

LAB 3-2 Project defense & alpha test - sprint-CSDN社区

TeamNameFittime
Sprint Summary LinkGroup 19 Sprint Summary-CSDN社区
Video demo linkEE308 GROUP19 Fittime_哔哩哔哩_bilibili
GitHub linkhttps://github.com/xingwang133/Lab3.git

 

 

Fundamental state

  • Team ID:19
  • Team Name:Fittime
  • team member:

    Team members
    Xiao Ma (Leader)
    Mingjun Cao
    Honghao Chen
    Xing Wang
    Xuan Chen

1.The progress of task

1.The progress of this sprint

team member

Tasks completed

    Time of consumption

Problems encountered in the process

The way to deal with the problemProportion
Xiao.MaThe function of "like", "favorite" and "consultation" was created, which solved the problem that some users would flash back when clicking the motion record12hWhen logging in to different accounts, clicking on my exercise record will appear a flash back. Different accounts of the same device will flash back, but the same account of different models will not flash back.After searching the Internet, I found that it was an Android problem. After adjustment, we successfully solved the port problem.110%
MingJun.CaoSearch github for open source projects with similar features to learn from6hSome projects are so old that some libraries cannot be called nowDo many searches on github95%
HongHao.ChenSearch github for open source projects with similar features to learn from6hSome projects are so old that some libraries cannot be called nowDo many searches on github100%
Xuan.Chenfunctional test4hWhen some functions are clicked, the APP will blink back; Some phones work just fine with a tapAfter searching the relevant information, it is found that there are some problems with the AS setting. The software can be adapted to different Android models95%
Xing.Wangfunctional test4hWhen some functions are clicked, the APP will blink back; Some phones work just fine with a tapAfter searching the relevant information, it is found that there are some problems with the AS setting. The software can be adapted to different Android models90%

2. Minutes of Meeting

2.1 Contents of the Meeting

  1. Basically complete the page design.
  2. Complete dynamic release, search function code
  3. Complete the question and answer function code
  4. Basically complete the community interface code
  5. Basically complete the mobile interface code

2.2 Pictures of the Meeting

4. Development Process

4.1 Check-in Records of Github

https://github.com/xingwang133/Lab3.git

5. Introduction to Progress

5.1 Progress of Development

The community back end has been basically written, the front end has completed the basic interface connection, the next plan is to complete the front end, and the page beautification.

 

5.2 Burnout Diagram

 

...全文
540 回复 打赏 收藏 举报
写回复
用AI写文章
回复
切换为时间正序
请发表友善的回复…
发表回复
详情可查看下方数据集可视化效果。 【数据集概况】 · 检测类别(中文):[中央处理器(CPU)] · 训练集:810 张 · 验证集:82 张 · 测试集:40 张 · 总计:932 张 该数据集聚焦于工业制造环境中中央处理器的精准识别,通过多背景、多光照条件下的图像采集,构建了覆盖典型生产场景的高质量样本库。数据集中中央处理器以标准封装形态呈现,背景涵盖金属台面、彩色渐变板及纹理表面,充分模拟实际生产线中的视觉干扰因素,为自动化质检与智能分拣系统提供了可靠的数据支撑。... 【训练曲线与评估图】 【模型训练配置】 参数 | 值 模型 | yolo26n 训练轮数 | 100 epochs 输入尺寸 | 640x640 批次大小 | 24 优化器 | auto 初始学习率 | 0.01 训练设备 【关键指标汇总】 训练了 37 个 epoch,最终轮指标: 指标 | 数值 mAP50 | **0.9650** mAP50-95 | 0.9173 Precision | 0.9967 Recall | 0.9647 train/box_loss | 0.4580 train/cls_loss | 0.2414 val/box_loss | 0.3658 val/cls_loss | 0.2987 【训练过程分析】 37 轮训练后 mAP50 达到 0.9650,模型收敛良好。Loss 曲线前段快速下降,后段趋于平稳,val_loss 无反弹,没有明显过拟合。mAP50-95 为 0.9173,和 mAP50 差距仅 0.05,框的定位精度也很扎实。 【模型性能评估】 Precision 0.9967、Recall 0.9647,精召双高,模型对中央处理器的检测能力强。 【预测效果展示】 验证集预测效果较好,检测框基本准确覆盖中央处理器,置信度整体偏高。 【改进建议】 1...

286

社区成员

发帖
与我相关
我的任务
社区描述
福州大学 梅努斯国际工程学院 软件工程(2022秋) 教学
软件工程 高校
社区管理员
  • LinQF39
加入社区
  • 近7日
  • 近30日
  • 至今
社区公告
暂无公告

试试用AI创作助手写篇文章吧