Add YOLO11 P2 and P6 configs (#73)

* Merge pull request #57 from Y-T-G/scale-validation

Add support for multiple scales in YAML

* Handle invalid `flops` and `params`

* Support parsing list of FLOPs and parameters

* Add YOLO11-P2 Detect config (#56)

* Add YOLO11-P2 Detect config

* Update output shapes

* Update FLOPs and parameters

* Add discussion link to detect/yolo11-p2.yaml

* Add YOLO11-P2 Segment config (#58)

* Add YOLO11-P2 Segment config

* Update yolo11-p2.yaml

* Update FLOPs and parameters

* Add YOLO11-OBB P2 config (#59)

* Add discussion link to obb/yolo11-p2.yaml

* Add discussion link to segment/yolo11-p2.yaml

* Add YOLO11-Pose P2 config (#63)

* Add discussion link to pose/yolo11-p2.yaml

* Add YOLO11-Detect P6 config (#65)

* Add discussion link to detect/yolo11-p6.yaml

* Add YOLO11-Segment P6 config (#66)

* Add discussion link to segment/yolo11-p6.yaml

* Add YOLO11-Pose P6 config (#68)

* Add YOLO11-OBB P6 config (#67)

* Add discussion link to obb/yolo11-p6.yaml

* Add discussion link to pose/yolo11-p6.yaml

---------

Co-authored-by: GitHub Action <action@github.com>

Author: Y-T-G

cfg/detect/yolo11-p2.yaml

@@ -0,0 +1,87 @@
+# Metadata
+author: Y-T-G
+task: detect
+keywords: [cnn, yolo11, small-object]
+description: YOLO11 object detection model with an extra P2 scale for small object detecton.
+flops:
+  n: 11.4
+  s: 29.7
+  m: 88.9
+  l: 113.6
+  x: 245.1
+parameters:
+  n: 2740032
+  s: 9625968
+  m: 20591120
+  l: 26116368
+  x: 57849456
+min_version: 8.3.0
+discussion: https://github.com/Y-T-G/community/discussions/60
+
+# Model
+nc: 80
+strides: [4, 8, 16, 32]
+scales:
+  n: [0.50, 0.25, 1024]
+  s: [0.50, 0.50, 1024]
+  m: [0.50, 1.00, 512]
+  l: [1.00, 1.00, 512] 
+  x: [1.00, 1.50, 512] 
+
+backbone:
+  # Stem
+  - [-1, 1, Conv, [64, 3, 2]]                   # (1, 64, 320, 320) - 0
+
+  # P2 Backbone
+  - [-1, 1, Conv, [128, 3, 2]]                  # (1, 128, 160, 160) - 1
+  - [-1, 2, C3k2, [256, False, 0.25]]           # (1, 256, 160, 160) - 2
+
+  # P3 Backbone
+  - [-1, 1, Conv, [256, 3, 2]]                  # (1, 256, 80, 80) - 3
+  - [-1, 2, C3k2, [512, False, 0.25]]           # (1, 512, 80, 80) - 4
+
+  # P4 Backbone
+  - [-1, 1, Conv, [512, 3, 2]]                  # (1, 512, 40, 40) - 5
+  - [-1, 2, C3k2, [512, True]]                  # (1, 512, 40, 40) - 6
+
+  # P5 Backbone
+  - [-1, 1, Conv, [1024, 3, 2]]                 # (1, 512, 20, 20) - 7
+  - [-1, 2, C3k2, [1024, True]]                 # (1, 512, 20, 20) - 8
+  - [-1, 1, SPPF, [1024, 5]]                    # (1, 512, 20, 20) - 9
+  - [-1, 2, C2PSA, [1024]]                      # (1, 512, 20, 20) - 10
+
+head:
+  # Top-Down Path (Feature Pyramid)
+  # P4 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]    # (1, 512, 40, 40) - 11
+  - [[-1, 6], 1, Concat, [1]]                   # (1, 1024, 40, 40) - 12
+  - [-1, 2, C3k2, [512, False]]                 # (1, 512, 40, 40) - 13
+
+  # P3 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]    # (1, 512, 80, 80) - 14
+  - [[-1, 4], 1, Concat, [1]]                   # (1, 1024, 80, 80) - 15
+  - [-1, 2, C3k2, [256, False]]                 # (1, 256, 80, 80) - 16
+
+  # P2 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]    # (1, 256, 160, 160) - 17
+  - [[-1, 2], 1, Concat, [1]]                   # (1, 512, 160, 160) - 18
+  - [-1, 2, C3k2, [128, False]]                 # (1, 128, 160, 160) - 19
+
+  # Bottom-Up Path (PAN-FPN)
+  # P3
+  - [-1, 1, Conv, [128, 3, 2]]                  # (1, 128, 80, 80) - 20
+  - [[-1, 16], 1, Concat, [1]]                  # (1, 384, 80, 80) - 21
+  - [-1, 2, C3k2, [256, False]]                 # (1, 256, 80, 80) - 22
+
+  # P4
+  - [-1, 1, Conv, [256, 3, 2]]                  # (1, 256, 40, 40) - 23
+  - [[-1, 13], 1, Concat, [1]]                  # (1, 768, 40, 40) - 24
+  - [-1, 2, C3k2, [512, False]]                 # (1, 512, 40, 40) - 25
+
+  # P5
+  - [-1, 1, Conv, [512, 3, 2]]                  # (1, 512, 20, 20) - 26
+  - [[-1, 10], 1, Concat, [1]]                  # (1, 1024, 20, 20) - 27
+  - [-1, 2, C3k2, [1024, True]]                 # (1, 512, 20, 20) - 28
+
+  # Detection Head (P2, P3, P4, P5)
+  - [[19, 22, 25, 28], 1, Detect, [nc]]         # 29

cfg/detect/yolo11-p6.yaml

@@ -0,0 +1,91 @@
+# Metadata
+author: Y-T-G
+task: detect
+keywords: [cnn, yolo11, large-object]
+description: YOLO11 detection model with an extra P6 scale for large object detecton.
+flops:
+  n: 7.0
+  s: 23.2
+  m: 70.5
+  l: 90.3
+  x: 202.1
+parameters:
+  n: 4391920
+  s: 16130064
+  m: 30063504
+  l: 37883024
+  x: 85074352
+min_version: 8.3.0
+discussion: https://github.com/Y-T-G/community/discussions/69
+
+# Model
+nc: 80
+strides: [8, 16, 32, 64]
+scales:
+  n: [0.50, 0.25, 1024]
+  s: [0.50, 0.50, 1024]
+  m: [0.50, 1.00, 512]
+  l: [1.00, 1.00, 512] 
+  x: [1.00, 1.50, 512] 
+
+backbone:
+  # Stem
+  - [-1, 1, Conv, [64, 3, 2]]                        # (1, 64, 320, 320) - 0
+
+  # P2 Backbone
+  - [-1, 1, Conv, [128, 3, 2]]                       # (1, 128, 160, 160) - 1
+  - [-1, 2, C3k2, [256, False, 0.25]]                # (1, 256, 160, 160) - 2
+
+  # P3 Backbone
+  - [-1, 1, Conv, [256, 3, 2]]                       # (1, 256, 80, 80) - 3
+  - [-1, 2, C3k2, [512, False, 0.25]]                # (1, 512, 80, 80) - 4
+
+  # P4 Backbone
+  - [-1, 1, Conv, [512, 3, 2]]                       # (1, 512, 40, 40) - 5
+  - [-1, 2, C3k2, [768, True]]                       # (1, 512, 40, 40) - 6
+
+  # P5 Backbone
+  - [-1, 1, Conv, [768, 3, 2]]                       # (1, 512, 20, 20) - 7
+  - [-1, 2, C3k2, [768, True]]                       # (1, 512, 20, 20) - 8
+
+  # P6 Backbone
+  - [-1, 1, Conv, [1024, 3, 2]]                      # (1, 512, 10, 10) - 9
+  - [-1, 2, C3k2, [1024, True]]                      # (1, 512, 10, 10) - 10
+  - [-1, 1, SPPF, [1024, 5]]                         # (1, 512, 10, 10) - 11
+  - [-1, 2, C2PSA, [1024]]                           # (1, 512, 10, 10) - 12
+
+head:
+  # Top-Down Path (Feature Pyramid)
+  # P5 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 20, 20) - 13
+  - [[-1, 8], 1, Concat, [1]]                        # (1, 1024, 20, 20) - 14
+  - [-1, 2, C3k2, [768, False]]                      # (1, 512, 20, 20) - 15
+
+  # P4 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 40, 40) - 16
+  - [[-1, 6], 1, Concat, [1]]                        # (1, 1024, 40, 40) - 17
+  - [-1, 2, C3k2, [512, False]]                      # (1, 512, 40, 40) - 18
+
+  # P3 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 80, 80) - 19
+  - [[-1, 4], 1, Concat, [1]]                        # (1, 1024, 80, 80) - 20
+  - [-1, 2, C3k2, [256, False]]                      # (1, 256, 80, 80) - 21
+
+  # Bottom-Up Path (PAN-FPN)
+  # P4
+  - [-1, 1, Conv, [256, 3, 2]]                       # (1, 256, 40, 40) - 22
+  - [[-1, 16], 1, Concat, [1]]                       # (1, 768, 40, 40) - 23
+  - [-1, 2, C3k2, [512, False]]                      # (1, 512, 40, 40) - 24
+
+  # P5
+  - [-1, 1, Conv, [512, 3, 2]]                       # (1, 512, 20, 20) - 25
+  - [[-1, 13], 1, Concat, [1]]                       # (1, 1024, 20, 20) - 26
+  - [-1, 2, C3k2, [768, True]]                       # (1, 512, 20, 20) - 27
+
+  # P6
+  - [-1, 1, Conv, [768, 3, 2]]                       # (1, 512, 10, 10) - 28
+  - [[-1, 10], 1, Concat, [1]]                       # (1, 1024, 10, 10) - 29
+  - [-1, 2, C3k2, [1024, True]]                      # (1, 512, 10, 10) - 30
+
+  # Detect Head (P3, P4, P5, P6)
+  - [[21, 24, 27, 30], 1, Detect, [nc]]              # 31

cfg/obb/yolo11-p2.yaml

@@ -0,0 +1,87 @@
+# Metadata
+author: Y-T-G
+task: obb
+keywords: [cnn, yolo11, small-object]
+description: YOLO11 rotated bounding box model with an extra P2 scale for small object detecton.
+flops:
+  n: 11.7
+  s: 30.7
+  m: 93.0
+  l: 117.6
+  x: 254.3
+parameters:
+  n: 2777060 
+  s: 9773748
+  m: 21034132
+  l: 26559380
+  x: 58845748
+min_version: 8.3.0
+discussion: https://github.com/Y-T-G/community/discussions/62
+
+# Model
+nc: 80
+strides: [4, 8, 16, 32, ]
+scales:
+  n: [0.50, 0.25, 1024]
+  s: [0.50, 0.50, 1024]
+  m: [0.50, 1.00, 512]
+  l: [1.00, 1.00, 512] 
+  x: [1.00, 1.50, 512] 
+
+backbone:
+  # Stem
+  - [-1, 1, Conv, [64, 3, 2]]                        # (1, 64, 320, 320) - 0
+
+  # P2 Backbone
+  - [-1, 1, Conv, [128, 3, 2]]                       # (1, 128, 160, 160) - 1
+  - [-1, 2, C3k2, [256, False, 0.25]]                # (1, 256, 160, 160) - 2
+
+  # P3 Backbone
+  - [-1, 1, Conv, [256, 3, 2]]                       # (1, 256, 80, 80) - 3
+  - [-1, 2, C3k2, [512, False, 0.25]]                # (1, 512, 80, 80) - 4
+
+  # P4 Backbone
+  - [-1, 1, Conv, [512, 3, 2]]                       # (1, 512, 40, 40) - 5
+  - [-1, 2, C3k2, [512, True]]                       # (1, 512, 40, 40) - 6
+
+  # P5 Backbone
+  - [-1, 1, Conv, [1024, 3, 2]]                      # (1, 512, 20, 20) - 7
+  - [-1, 2, C3k2, [1024, True]]                      # (1, 512, 20, 20) - 8
+  - [-1, 1, SPPF, [1024, 5]]                         # (1, 512, 20, 20) - 9
+  - [-1, 2, C2PSA, [1024]]                           # (1, 512, 20, 20) - 10
+
+head:
+  # Top-Down Path (Feature Pyramid)
+  # P4 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 40, 40) - 11
+  - [[-1, 6], 1, Concat, [1]]                        # (1, 1024, 40, 40) - 12
+  - [-1, 2, C3k2, [512, False]]                      # (1, 512, 40, 40) - 13
+
+  # P3 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 80, 80) - 14
+  - [[-1, 4], 1, Concat, [1]]                        # (1, 1024, 80, 80) - 15
+  - [-1, 2, C3k2, [256, False]]                      # (1, 256, 80, 80) - 16
+
+  # P2 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 256, 160, 160) - 17
+  - [[-1, 2], 1, Concat, [1]]                        # (1, 512, 160, 160) - 18
+  - [-1, 2, C3k2, [128, False]]                      # (1, 128, 160, 160) - 19
+
+  # Bottom-Up Path (PAN-FPN)
+  # P3
+  - [-1, 1, Conv, [128, 3, 2]]                       # (1, 128, 80, 80) - 20
+  - [[-1, 16], 1, Concat, [1]]                       # (1, 384, 80, 80) - 21
+  - [-1, 2, C3k2, [256, False]]                      # (1, 256, 80, 80) - 22
+
+  # P4
+  - [-1, 1, Conv, [256, 3, 2]]                       # (1, 256, 40, 40) - 23
+  - [[-1, 13], 1, Concat, [1]]                       # (1, 768, 40, 40) - 24
+  - [-1, 2, C3k2, [512, False]]                      # (1, 512, 40, 40) - 25
+
+  # P5
+  - [-1, 1, Conv, [512, 3, 2]]                       # (1, 512, 20, 20) - 26
+  - [[-1, 10], 1, Concat, [1]]                       # (1, 1024, 20, 20) - 27
+  - [-1, 2, C3k2, [1024, True]]                      # (1, 512, 20, 20) - 28
+
+  # OBB Head (P2, P3, P4, P5)
+  - [[19, 22, 25, 28], 1, OBB, [nc, 1]]              # 29

cfg/obb/yolo11-p6.yaml

@@ -0,0 +1,91 @@
+# Metadata
+author: Y-T-G
+task: obb
+keywords: [cnn, yolo11, large-object]
+description: YOLO11 rotated bounding box model with an extra P6 scale for large object detecton.
+flops:
+  n: 7.2
+  s: 24.2
+  m: 74.3
+  l: 94.1
+  x: 210.6
+parameters:
+  n: 4493620
+  s: 16536212
+  m: 31244436
+  l: 39063956
+  x: 87730484
+min_version: 8.3.0
+discussion: https://github.com/Y-T-G/community/discussions/72
+
+# Model
+nc: 80
+strides: [8, 16, 32, 64]
+scales:
+  n: [0.50, 0.25, 1024]
+  s: [0.50, 0.50, 1024]
+  m: [0.50, 1.00, 512]
+  l: [1.00, 1.00, 512] 
+  x: [1.00, 1.50, 512] 
+
+backbone:
+  # Stem
+  - [-1, 1, Conv, [64, 3, 2]]                        # (1, 64, 320, 320) - 0
+
+  # P2 Backbone
+  - [-1, 1, Conv, [128, 3, 2]]                       # (1, 128, 160, 160) - 1
+  - [-1, 2, C3k2, [256, False, 0.25]]                # (1, 256, 160, 160) - 2
+
+  # P3 Backbone
+  - [-1, 1, Conv, [256, 3, 2]]                       # (1, 256, 80, 80) - 3
+  - [-1, 2, C3k2, [512, False, 0.25]]                # (1, 512, 80, 80) - 4
+
+  # P4 Backbone
+  - [-1, 1, Conv, [512, 3, 2]]                       # (1, 512, 40, 40) - 5
+  - [-1, 2, C3k2, [768, True]]                       # (1, 512, 40, 40) - 6
+
+  # P5 Backbone
+  - [-1, 1, Conv, [768, 3, 2]]                       # (1, 512, 20, 20) - 7
+  - [-1, 2, C3k2, [768, True]]                       # (1, 512, 20, 20) - 8
+
+  # P6 Backbone
+  - [-1, 1, Conv, [1024, 3, 2]]                      # (1, 512, 10, 10) - 9
+  - [-1, 2, C3k2, [1024, True]]                      # (1, 512, 10, 10) - 10
+  - [-1, 1, SPPF, [1024, 5]]                         # (1, 512, 10, 10) - 11
+  - [-1, 2, C2PSA, [1024]]                           # (1, 512, 10, 10) - 12
+
+head:
+  # Top-Down Path (Feature Pyramid)
+  # P5 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 20, 20) - 13
+  - [[-1, 8], 1, Concat, [1]]                        # (1, 1024, 20, 20) - 14
+  - [-1, 2, C3k2, [768, False]]                      # (1, 512, 20, 20) - 15
+
+  # P4 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 40, 40) - 16
+  - [[-1, 6], 1, Concat, [1]]                        # (1, 1024, 40, 40) - 17
+  - [-1, 2, C3k2, [512, False]]                      # (1, 512, 40, 40) - 18
+
+  # P3 Head
+  - [-1, 1, nn.Upsample, [None, 2, nearest]]         # (1, 512, 80, 80) - 19
+  - [[-1, 4], 1, Concat, [1]]                        # (1, 1024, 80, 80) - 20
+  - [-1, 2, C3k2, [256, False]]                      # (1, 256, 80, 80) - 21
+
+  # Bottom-Up Path (PAN-FPN)
+  # P4
+  - [-1, 1, Conv, [256, 3, 2]]                       # (1, 256, 40, 40) - 22
+  - [[-1, 16], 1, Concat, [1]]                       # (1, 768, 40, 40) - 23
+  - [-1, 2, C3k2, [512, False]]                      # (1, 512, 40, 40) - 24
+
+  # P5
+  - [-1, 1, Conv, [512, 3, 2]]                       # (1, 512, 20, 20) - 25
+  - [[-1, 13], 1, Concat, [1]]                       # (1, 1024, 20, 20) - 26
+  - [-1, 2, C3k2, [768, True]]                       # (1, 512, 20, 20) - 27
+
+  # P6
+  - [-1, 1, Conv, [768, 3, 2]]                       # (1, 512, 10, 10) - 28
+  - [[-1, 10], 1, Concat, [1]]                       # (1, 1024, 10, 10) - 29
+  - [-1, 2, C3k2, [1024, True]]                      # (1, 512, 10, 10) - 30
+
+  # OBB Head (P3, P4, P5, P6)
+  - [[21, 24, 27, 30], 1, OBB, [nc, 1]]              # 31