Table 2 Values of various parameters in YOLOv5s.
From: Detection and identification of centipedes based on deep learning
Hyperparameters | Value |
|---|---|
lr0 (initial learning rate) | 0.01 |
lrf (final OneCycleLR learning rate) | 0.01 |
momentum (SGD momentum/Adam beta1) | 0.937 |
box (box loss gain) | 005 |
cls (cls loss gain) | 0.5 |
cls_pw (cls BCELoss) | 1.0 |
obj (obj loss gain) | 1.0 |
obj_pw (obj BCELoss) | 1.0 |
iou_t (IoU training threshold) | 0.20 |
anchor_t (anchor-multiple threshold) | 3 |
hsv_h (image HSV-Hue augmentation) | 0.015 |
hsv_s (image HSV-Saturation augmentation) | 0.7 |
hsv_v (image HSV-Value augmentation) | 0.4 |
translate (image translation) | 0.1 |
scale (image scale) | 0.5 |
fliplr (image flip left-right) | 0.5 |
mosaic (image mixup) | 1.0 |
