# Ultralytics YOLO 🚀, AGPL-3.0 license
# RT-DETR-l object detection model with P3-P5 outputs. For details see https://docs.ultralytics.com/models/rtdetr
# Parameters
nc: 80 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov8n-cls.yaml' will call yolov8-cls.yaml with scale 'n'
# [depth, width, max_channels]
l: [1.00, 1.00, 1024]
# From BiliBili 魔鬼面具
backbone:
# [from, repeats, module, args]
- [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
- [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
- [-1, 1, C2f_PFDConv, [128]]
- [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
- [-1, 1, C2f_PFDConv, [256]]
- [-1, 1, Conv, [384, 3, 2]] # 5-P4/16
- [-1, 1, C2f_PFDConv, [384]]
- [-1, 1, Conv, [384, 3, 2]] # 7-P5/32
- [-1, 3, C2f_PFDConv, [384]]
head:
- [-1, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 9 input_proj.2
- [-1, 1, AIFI, [1024, 8]] # 10
- [-1, 1, Conv, [256, 1, 1]] # 11, Y5, lateral_convs.0
- [-1, 1, nn.Upsample, [None, 2, 'nearest']] # 12
- [6, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 13 input_proj.1
- [[-2, -1], 1, Concat, [1]] # 14
- [-1, 1, CSP_PAC, [256]] # 15, fpn_blocks.0
- [-1, 1, Conv, [256, 1, 1]] # 16, Y4, lateral_convs.1
- [-1, 1, AttentionUpsample, []] # 17
- [4, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 18 input_proj.0
- [[-2, -1], 1, Concat, [1]] # 19 cat backbone P4
- [-1, 1, CSP_PAC, [256]] # X3 (20), fpn_blocks.1
- [-1, 1, AttentionDownsample, []] # 21, downsample_convs.0
- [[-1, 16], 1, Concat, [1]] # 22 cat Y4
- [-1, 3, CSP_PAC, [256]] # F4 (23), pan_blocks.0
- [-1, 1, AttentionDownsample, []] # 24, downsample_convs.1
- [[-1, 11], 1, Concat, [1]] # 25 cat Y5
- [-1, 3,CSP_PAC, [256]] # F5 (26), pan_blocks.1
- [[20, 23, 26], 1, RTDETRDecoder, [nc, 256, 300, 4, 8, 3]] # Detect(P3, P4, P5)检查错误
最新发布