Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
yolov5m-icecream-detect – AI Model by binbinshen | AlphaNeural AI
You can deploy this model and start earning money today!
binbinshen
/
yolov5m-icecream-detect
like
0
ultralytics
object-detection
yolov5
yolov5m
icecream
empty_cup
no_cup
make_success
pytorch
en
other
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
YOLOv5m 冰激凌制作场景目标检测模型
模型介绍
模型架构:YOLOv5m(中等规模模型,平衡精度与速度)
训练参数:输入尺寸 640x640,AdamW 优化器,cosine LR 调度
支持类别(3类):
empty_cup(空杯子)
no_cup(无杯子)
make_success(制作完成)
性能指标(训练集验证结果)
整体精度(P):0.985
整体召回率(R):0.978
mAP50:0.993
mAP50-95:0.747
使用说明
置信度阈值:推荐 0.5(过滤低可信度标注)
IOU 阈值:0.45(解决重叠标注)
适配场景:冰激凌制作过程中的杯子状态检测
依赖库
txt
1
ultralytics>=8.0 # YOLOv5 依赖(实际按你的版本填写)
2
torch>=1.17.0
3
numpy>=1.24.0