:::tips
在这里将了解到最新的参数,对你接下来的调试具有参考作用
同时将在这里更新最新的调试过程,最新的代码
每天开始训练的时候请及时更新为最优参数
仿github团队协作模式,达到每天开始能回到同一个枝干上的目的
:::
Top最优参数
更新时间为: 2021-10-05 20:27
:::info box 由0.05调整到0.03,得到了目前最好的效果,map稳定在0.8以上 :::
lr0: 0.01 # initial learning rate (SGD=1E-2, Adam=1E-3)
lrf: 0.2 # final OneCycleLR learning rate (lr0 * lrf)
momentum: 0.937 # SGD momentum/Adam beta1
weight_decay: 0.0005 # optimizer weight decay 5e-4
warmup_epochs: 3.0 # warmup epochs (fractions ok)
warmup_momentum: 0.8 # warmup initial momentum
warmup_bias_lr: 0.1 # warmup initial bias lr
box: 0.03 # box loss gain
cls: 0.4 # cls loss gain
cls_pw: 1.0 # cls BCELoss positive_weight
obj: 1.0 # obj loss gain (scale with pixels)
obj_pw: 1.0 # obj BCELoss positive_weight
iou_t: 0.20 # IoU training threshold
anchor_t: 4.0 # anchor-multiple threshold
# anchors: 3 # anchors per output layer (0 to ignore)
fl_eiou_gamma: 0.0 #focal eiou loss gamma
iou_aware: 0.0
fl_gamma: 0.0 # focal loss gamma (efficientDet default gamma=1.5)
hsv_h: 0.015 # image HSV-Hue augmentation (fraction)
hsv_s: 0.7 # image HSV-Saturation augmentation (fraction)
hsv_v: 0.4 # image HSV-Value augmentation (fraction)
degrees: 0.0 # image rotation (+/- deg)
translate: 0.1 # image translation (+/- fraction)
scale: 0.5 # image scale (+/- gain)
shear: 0.0 # image shear (+/- deg)
perspective: 0.0 # image perspective (+/- fraction), range 0-0.001
flipud: 0.0 # image flip up-down (probability)
fliplr: 0.5 # image flip left-right (probability)
mosaic: 1.0 # image mosaic (probability)
mixup: 0.0 # image mixup (probability)
copy_paste: 0.0 # segment copy-paste (probability)
原始模型
lr0: 0.01 # initial learning rate (SGD=1E-2, Adam=1E-3)
lrf: 0.2 # final OneCycleLR learning rate (lr0 * lrf)
momentum: 0.937 # SGD momentum/Adam beta1
weight_decay: 0.0005 # optimizer weight decay 5e-4
warmup_epochs: 3.0 # warmup epochs (fractions ok)
warmup_momentum: 0.8 # warmup initial momentum
warmup_bias_lr: 0.1 # warmup initial bias lr
box: 0.05 # box loss gain
cls: 0.5 # cls loss gain
cls_pw: 1.0 # cls BCELoss positive_weight
obj: 1.0 # obj loss gain (scale with pixels)
obj_pw: 1.0 # obj BCELoss positive_weight
iou_t: 0.20 # IoU training threshold
anchor_t: 4.0 # anchor-multiple threshold
# anchors: 3 # anchors per output layer (0 to ignore)
fl_eiou_gamma: 0.0 #focal eiou loss gamma
iou_aware: 0.0
fl_gamma: 0.0 # focal loss gamma (efficientDet default gamma=1.5)
hsv_h: 0.015 # image HSV-Hue augmentation (fraction)
hsv_s: 0.7 # image HSV-Saturation augmentation (fraction)
hsv_v: 0.4 # image HSV-Value augmentation (fraction)
degrees: 0.0 # image rotation (+/- deg)
translate: 0.1 # image translation (+/- fraction)
scale: 0.5 # image scale (+/- gain)
shear: 0.0 # image shear (+/- deg)
perspective: 0.0 # image perspective (+/- fraction), range 0-0.001
flipud: 0.0 # image flip up-down (probability)
fliplr: 0.5 # image flip left-right (probability)
mosaic: 1.0 # image mosaic (probability)
mixup: 0.0 # image mixup (probability)
copy_paste: 0.0 # segment copy-paste (probability)
权重文件:
yolov5s6.pt
下载地址:
https://yun.hengyimonster.top/s/3zhE
操作使用:
这里你将学会简单的几步:方便大家同步前进 你的付出也就具有参考价值,快速学习
:::info
- 第一步:
找到当天的对应文档:打出##XXX(你的名字)并敲回车为二级标题
- 第二步:
相同的操作但是是3个###开始3级标题 为:你的操作日志
- 第三步:
请保持相同的格式:
- 输入 :::info 回车
- 数据:xxxx
- 命令行:xxxx
- 复制截图直接粘贴
- 复制参数并且粘贴同时将更改的参数标注为红色
- 每次训练一次请重复第二和第三步
- 小总结:
当你今天不想继续训练的时候: 请输入:::tips 回车
写一下今天调了哪一个 觉得那个最好
并在最后附上觉得最好的那张图 方便李师兄总结并更新小报
:::
快报
10月05日~10月06日
:::info 今日一共有 2人,提交了6份数据 :::
- 陈唯彬:
box 由0.05调整到0.03,得到了目前最好的效果,map稳定在0.8以上
得出的最优模型图: