upload main files

ultralytics/cfg/datasets/Argoverse.yaml

@@ -0,0 +1,75 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Argoverse-HD dataset (ring-front-center camera) https://www.cs.cmu.edu/~mengtial/proj/streaming/ by Argo AI
+# Documentation: https://docs.ultralytics.com/datasets/detect/argoverse/
+# Example usage: yolo train data=Argoverse.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── Argoverse  ← downloads here (31.5 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/Argoverse # dataset root dir
+train: Argoverse-1.1/images/train/ # train images (relative to 'path') 39384 images
+val: Argoverse-1.1/images/val/ # val images (relative to 'path') 15062 images
+test: Argoverse-1.1/images/test/ # test images (optional) https://eval.ai/web/challenges/challenge-page/800/overview
+
+# Classes
+names:
+  0: person
+  1: bicycle
+  2: car
+  3: motorcycle
+  4: bus
+  5: truck
+  6: traffic_light
+  7: stop_sign
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import json
+  from tqdm import tqdm
+  from ultralytics.utils.downloads import download
+  from pathlib import Path
+
+  def argoverse2yolo(set):
+      labels = {}
+      a = json.load(open(set, "rb"))
+      for annot in tqdm(a['annotations'], desc=f"Converting {set} to YOLOv5 format..."):
+          img_id = annot['image_id']
+          img_name = a['images'][img_id]['name']
+          img_label_name = f'{img_name[:-3]}txt'
+
+          cls = annot['category_id']  # instance class id
+          x_center, y_center, width, height = annot['bbox']
+          x_center = (x_center + width / 2) / 1920.0  # offset and scale
+          y_center = (y_center + height / 2) / 1200.0  # offset and scale
+          width /= 1920.0  # scale
+          height /= 1200.0  # scale
+
+          img_dir = set.parents[2] / 'Argoverse-1.1' / 'labels' / a['seq_dirs'][a['images'][annot['image_id']]['sid']]
+          if not img_dir.exists():
+              img_dir.mkdir(parents=True, exist_ok=True)
+
+          k = str(img_dir / img_label_name)
+          if k not in labels:
+              labels[k] = []
+          labels[k].append(f"{cls} {x_center} {y_center} {width} {height}\n")
+
+      for k in labels:
+          with open(k, "w") as f:
+              f.writelines(labels[k])
+
+
+  # Download 'https://argoverse-hd.s3.us-east-2.amazonaws.com/Argoverse-HD-Full.zip' (deprecated S3 link)
+  dir = Path(yaml['path'])  # dataset root dir
+  urls = ['https://drive.google.com/file/d/1st9qW3BeIwQsnR0t8mRpvbsSWIo16ACi/view?usp=drive_link']
+  print("\n\nWARNING: Argoverse dataset MUST be downloaded manually, autodownload will NOT work.")
+  print(f"WARNING: Manually download Argoverse dataset '{urls[0]}' to '{dir}' and re-run your command.\n\n")
+  # download(urls, dir=dir)
+
+  # Convert
+  annotations_dir = 'Argoverse-HD/annotations/'
+  (dir / 'Argoverse-1.1' / 'tracking').rename(dir / 'Argoverse-1.1' / 'images')  # rename 'tracking' to 'images'
+  for d in "train.json", "val.json":
+      argoverse2yolo(dir / annotations_dir / d)  # convert Argoverse annotations to YOLO labels

ultralytics/cfg/datasets/DOTAv1.5.yaml

@@ -0,0 +1,37 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# DOTA 1.5 dataset https://captain-whu.github.io/DOTA/index.html for object detection in aerial images by Wuhan University
+# Documentation: https://docs.ultralytics.com/datasets/obb/dota-v2/
+# Example usage: yolo train model=yolov8n-obb.pt data=DOTAv1.5.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── dota1.5  ← downloads here (2GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/DOTAv1.5 # dataset root dir
+train: images/train # train images (relative to 'path') 1411 images
+val: images/val # val images (relative to 'path') 458 images
+test: images/test # test images (optional) 937 images
+
+# Classes for DOTA 1.5
+names:
+  0: plane
+  1: ship
+  2: storage tank
+  3: baseball diamond
+  4: tennis court
+  5: basketball court
+  6: ground track field
+  7: harbor
+  8: bridge
+  9: large vehicle
+  10: small vehicle
+  11: helicopter
+  12: roundabout
+  13: soccer ball field
+  14: swimming pool
+  15: container crane
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/DOTAv1.5.zip

ultralytics/cfg/datasets/DOTAv1.yaml

@@ -0,0 +1,36 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# DOTA 1.0 dataset https://captain-whu.github.io/DOTA/index.html for object detection in aerial images by Wuhan University
+# Documentation: https://docs.ultralytics.com/datasets/obb/dota-v2/
+# Example usage: yolo train model=yolov8n-obb.pt data=DOTAv1.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── dota1  ← downloads here (2GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/DOTAv1 # dataset root dir
+train: images/train # train images (relative to 'path') 1411 images
+val: images/val # val images (relative to 'path') 458 images
+test: images/test # test images (optional) 937 images
+
+# Classes for DOTA 1.0
+names:
+  0: plane
+  1: ship
+  2: storage tank
+  3: baseball diamond
+  4: tennis court
+  5: basketball court
+  6: ground track field
+  7: harbor
+  8: bridge
+  9: large vehicle
+  10: small vehicle
+  11: helicopter
+  12: roundabout
+  13: soccer ball field
+  14: swimming pool
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/DOTAv1.zip

ultralytics/cfg/datasets/GlobalWheat2020.yaml

@@ -0,0 +1,54 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Global Wheat 2020 dataset https://www.global-wheat.com/ by University of Saskatchewan
+# Documentation: https://docs.ultralytics.com/datasets/detect/globalwheat2020/
+# Example usage: yolo train data=GlobalWheat2020.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── GlobalWheat2020  ← downloads here (7.0 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/GlobalWheat2020 # dataset root dir
+train: # train images (relative to 'path') 3422 images
+  - images/arvalis_1
+  - images/arvalis_2
+  - images/arvalis_3
+  - images/ethz_1
+  - images/rres_1
+  - images/inrae_1
+  - images/usask_1
+val: # val images (relative to 'path') 748 images (WARNING: train set contains ethz_1)
+  - images/ethz_1
+test: # test images (optional) 1276 images
+  - images/utokyo_1
+  - images/utokyo_2
+  - images/nau_1
+  - images/uq_1
+
+# Classes
+names:
+  0: wheat_head
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  from ultralytics.utils.downloads import download
+  from pathlib import Path
+
+  # Download
+  dir = Path(yaml['path'])  # dataset root dir
+  urls = ['https://zenodo.org/record/4298502/files/global-wheat-codalab-official.zip',
+          'https://github.com/ultralytics/assets/releases/download/v0.0.0/GlobalWheat2020_labels.zip']
+  download(urls, dir=dir)
+
+  # Make Directories
+  for p in 'annotations', 'images', 'labels':
+      (dir / p).mkdir(parents=True, exist_ok=True)
+
+  # Move
+  for p in 'arvalis_1', 'arvalis_2', 'arvalis_3', 'ethz_1', 'rres_1', 'inrae_1', 'usask_1', \
+           'utokyo_1', 'utokyo_2', 'nau_1', 'uq_1':
+      (dir / 'global-wheat-codalab-official' / p).rename(dir / 'images' / p)  # move to /images
+      f = (dir / 'global-wheat-codalab-official' / p).with_suffix('.json')  # json file
+      if f.exists():
+          f.rename((dir / 'annotations' / p).with_suffix('.json'))  # move to /annotations

ultralytics/cfg/datasets/Objects365.yaml

@@ -0,0 +1,443 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Objects365 dataset https://www.objects365.org/ by Megvii
+# Documentation: https://docs.ultralytics.com/datasets/detect/objects365/
+# Example usage: yolo train data=Objects365.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── Objects365  ← downloads here (712 GB = 367G data + 345G zips)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/Objects365 # dataset root dir
+train: images/train # train images (relative to 'path') 1742289 images
+val: images/val # val images (relative to 'path') 80000 images
+test: # test images (optional)
+
+# Classes
+names:
+  0: Person
+  1: Sneakers
+  2: Chair
+  3: Other Shoes
+  4: Hat
+  5: Car
+  6: Lamp
+  7: Glasses
+  8: Bottle
+  9: Desk
+  10: Cup
+  11: Street Lights
+  12: Cabinet/shelf
+  13: Handbag/Satchel
+  14: Bracelet
+  15: Plate
+  16: Picture/Frame
+  17: Helmet
+  18: Book
+  19: Gloves
+  20: Storage box
+  21: Boat
+  22: Leather Shoes
+  23: Flower
+  24: Bench
+  25: Potted Plant
+  26: Bowl/Basin
+  27: Flag
+  28: Pillow
+  29: Boots
+  30: Vase
+  31: Microphone
+  32: Necklace
+  33: Ring
+  34: SUV
+  35: Wine Glass
+  36: Belt
+  37: Monitor/TV
+  38: Backpack
+  39: Umbrella
+  40: Traffic Light
+  41: Speaker
+  42: Watch
+  43: Tie
+  44: Trash bin Can
+  45: Slippers
+  46: Bicycle
+  47: Stool
+  48: Barrel/bucket
+  49: Van
+  50: Couch
+  51: Sandals
+  52: Basket
+  53: Drum
+  54: Pen/Pencil
+  55: Bus
+  56: Wild Bird
+  57: High Heels
+  58: Motorcycle
+  59: Guitar
+  60: Carpet
+  61: Cell Phone
+  62: Bread
+  63: Camera
+  64: Canned
+  65: Truck
+  66: Traffic cone
+  67: Cymbal
+  68: Lifesaver
+  69: Towel
+  70: Stuffed Toy
+  71: Candle
+  72: Sailboat
+  73: Laptop
+  74: Awning
+  75: Bed
+  76: Faucet
+  77: Tent
+  78: Horse
+  79: Mirror
+  80: Power outlet
+  81: Sink
+  82: Apple
+  83: Air Conditioner
+  84: Knife
+  85: Hockey Stick
+  86: Paddle
+  87: Pickup Truck
+  88: Fork
+  89: Traffic Sign
+  90: Balloon
+  91: Tripod
+  92: Dog
+  93: Spoon
+  94: Clock
+  95: Pot
+  96: Cow
+  97: Cake
+  98: Dining Table
+  99: Sheep
+  100: Hanger
+  101: Blackboard/Whiteboard
+  102: Napkin
+  103: Other Fish
+  104: Orange/Tangerine
+  105: Toiletry
+  106: Keyboard
+  107: Tomato
+  108: Lantern
+  109: Machinery Vehicle
+  110: Fan
+  111: Green Vegetables
+  112: Banana
+  113: Baseball Glove
+  114: Airplane
+  115: Mouse
+  116: Train
+  117: Pumpkin
+  118: Soccer
+  119: Skiboard
+  120: Luggage
+  121: Nightstand
+  122: Tea pot
+  123: Telephone
+  124: Trolley
+  125: Head Phone
+  126: Sports Car
+  127: Stop Sign
+  128: Dessert
+  129: Scooter
+  130: Stroller
+  131: Crane
+  132: Remote
+  133: Refrigerator
+  134: Oven
+  135: Lemon
+  136: Duck
+  137: Baseball Bat
+  138: Surveillance Camera
+  139: Cat
+  140: Jug
+  141: Broccoli
+  142: Piano
+  143: Pizza
+  144: Elephant
+  145: Skateboard
+  146: Surfboard
+  147: Gun
+  148: Skating and Skiing shoes
+  149: Gas stove
+  150: Donut
+  151: Bow Tie
+  152: Carrot
+  153: Toilet
+  154: Kite
+  155: Strawberry
+  156: Other Balls
+  157: Shovel
+  158: Pepper
+  159: Computer Box
+  160: Toilet Paper
+  161: Cleaning Products
+  162: Chopsticks
+  163: Microwave
+  164: Pigeon
+  165: Baseball
+  166: Cutting/chopping Board
+  167: Coffee Table
+  168: Side Table
+  169: Scissors
+  170: Marker
+  171: Pie
+  172: Ladder
+  173: Snowboard
+  174: Cookies
+  175: Radiator
+  176: Fire Hydrant
+  177: Basketball
+  178: Zebra
+  179: Grape
+  180: Giraffe
+  181: Potato
+  182: Sausage
+  183: Tricycle
+  184: Violin
+  185: Egg
+  186: Fire Extinguisher
+  187: Candy
+  188: Fire Truck
+  189: Billiards
+  190: Converter
+  191: Bathtub
+  192: Wheelchair
+  193: Golf Club
+  194: Briefcase
+  195: Cucumber
+  196: Cigar/Cigarette
+  197: Paint Brush
+  198: Pear
+  199: Heavy Truck
+  200: Hamburger
+  201: Extractor
+  202: Extension Cord
+  203: Tong
+  204: Tennis Racket
+  205: Folder
+  206: American Football
+  207: earphone
+  208: Mask
+  209: Kettle
+  210: Tennis
+  211: Ship
+  212: Swing
+  213: Coffee Machine
+  214: Slide
+  215: Carriage
+  216: Onion
+  217: Green beans
+  218: Projector
+  219: Frisbee
+  220: Washing Machine/Drying Machine
+  221: Chicken
+  222: Printer
+  223: Watermelon
+  224: Saxophone
+  225: Tissue
+  226: Toothbrush
+  227: Ice cream
+  228: Hot-air balloon
+  229: Cello
+  230: French Fries
+  231: Scale
+  232: Trophy
+  233: Cabbage
+  234: Hot dog
+  235: Blender
+  236: Peach
+  237: Rice
+  238: Wallet/Purse
+  239: Volleyball
+  240: Deer
+  241: Goose
+  242: Tape
+  243: Tablet
+  244: Cosmetics
+  245: Trumpet
+  246: Pineapple
+  247: Golf Ball
+  248: Ambulance
+  249: Parking meter
+  250: Mango
+  251: Key
+  252: Hurdle
+  253: Fishing Rod
+  254: Medal
+  255: Flute
+  256: Brush
+  257: Penguin
+  258: Megaphone
+  259: Corn
+  260: Lettuce
+  261: Garlic
+  262: Swan
+  263: Helicopter
+  264: Green Onion
+  265: Sandwich
+  266: Nuts
+  267: Speed Limit Sign
+  268: Induction Cooker
+  269: Broom
+  270: Trombone
+  271: Plum
+  272: Rickshaw
+  273: Goldfish
+  274: Kiwi fruit
+  275: Router/modem
+  276: Poker Card
+  277: Toaster
+  278: Shrimp
+  279: Sushi
+  280: Cheese
+  281: Notepaper
+  282: Cherry
+  283: Pliers
+  284: CD
+  285: Pasta
+  286: Hammer
+  287: Cue
+  288: Avocado
+  289: Hami melon
+  290: Flask
+  291: Mushroom
+  292: Screwdriver
+  293: Soap
+  294: Recorder
+  295: Bear
+  296: Eggplant
+  297: Board Eraser
+  298: Coconut
+  299: Tape Measure/Ruler
+  300: Pig
+  301: Showerhead
+  302: Globe
+  303: Chips
+  304: Steak
+  305: Crosswalk Sign
+  306: Stapler
+  307: Camel
+  308: Formula 1
+  309: Pomegranate
+  310: Dishwasher
+  311: Crab
+  312: Hoverboard
+  313: Meatball
+  314: Rice Cooker
+  315: Tuba
+  316: Calculator
+  317: Papaya
+  318: Antelope
+  319: Parrot
+  320: Seal
+  321: Butterfly
+  322: Dumbbell
+  323: Donkey
+  324: Lion
+  325: Urinal
+  326: Dolphin
+  327: Electric Drill
+  328: Hair Dryer
+  329: Egg tart
+  330: Jellyfish
+  331: Treadmill
+  332: Lighter
+  333: Grapefruit
+  334: Game board
+  335: Mop
+  336: Radish
+  337: Baozi
+  338: Target
+  339: French
+  340: Spring Rolls
+  341: Monkey
+  342: Rabbit
+  343: Pencil Case
+  344: Yak
+  345: Red Cabbage
+  346: Binoculars
+  347: Asparagus
+  348: Barbell
+  349: Scallop
+  350: Noddles
+  351: Comb
+  352: Dumpling
+  353: Oyster
+  354: Table Tennis paddle
+  355: Cosmetics Brush/Eyeliner Pencil
+  356: Chainsaw
+  357: Eraser
+  358: Lobster
+  359: Durian
+  360: Okra
+  361: Lipstick
+  362: Cosmetics Mirror
+  363: Curling
+  364: Table Tennis
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  from tqdm import tqdm
+
+  from ultralytics.utils.checks import check_requirements
+  from ultralytics.utils.downloads import download
+  from ultralytics.utils.ops import xyxy2xywhn
+
+  import numpy as np
+  from pathlib import Path
+
+  check_requirements(('pycocotools>=2.0',))
+  from pycocotools.coco import COCO
+
+  # Make Directories
+  dir = Path(yaml['path'])  # dataset root dir
+  for p in 'images', 'labels':
+      (dir / p).mkdir(parents=True, exist_ok=True)
+      for q in 'train', 'val':
+          (dir / p / q).mkdir(parents=True, exist_ok=True)
+
+  # Train, Val Splits
+  for split, patches in [('train', 50 + 1), ('val', 43 + 1)]:
+      print(f"Processing {split} in {patches} patches ...")
+      images, labels = dir / 'images' / split, dir / 'labels' / split
+
+      # Download
+      url = f"https://dorc.ks3-cn-beijing.ksyun.com/data-set/2020Objects365%E6%95%B0%E6%8D%AE%E9%9B%86/{split}/"
+      if split == 'train':
+          download([f'{url}zhiyuan_objv2_{split}.tar.gz'], dir=dir)  # annotations json
+          download([f'{url}patch{i}.tar.gz' for i in range(patches)], dir=images, curl=True, threads=8)
+      elif split == 'val':
+          download([f'{url}zhiyuan_objv2_{split}.json'], dir=dir)  # annotations json
+          download([f'{url}images/v1/patch{i}.tar.gz' for i in range(15 + 1)], dir=images, curl=True, threads=8)
+          download([f'{url}images/v2/patch{i}.tar.gz' for i in range(16, patches)], dir=images, curl=True, threads=8)
+
+      # Move
+      for f in tqdm(images.rglob('*.jpg'), desc=f'Moving {split} images'):
+          f.rename(images / f.name)  # move to /images/{split}
+
+      # Labels
+      coco = COCO(dir / f'zhiyuan_objv2_{split}.json')
+      names = [x["name"] for x in coco.loadCats(coco.getCatIds())]
+      for cid, cat in enumerate(names):
+          catIds = coco.getCatIds(catNms=[cat])
+          imgIds = coco.getImgIds(catIds=catIds)
+          for im in tqdm(coco.loadImgs(imgIds), desc=f'Class {cid + 1}/{len(names)} {cat}'):
+              width, height = im["width"], im["height"]
+              path = Path(im["file_name"])  # image filename
+              try:
+                  with open(labels / path.with_suffix('.txt').name, 'a') as file:
+                      annIds = coco.getAnnIds(imgIds=im["id"], catIds=catIds, iscrowd=None)
+                      for a in coco.loadAnns(annIds):
+                          x, y, w, h = a['bbox']  # bounding box in xywh (xy top-left corner)
+                          xyxy = np.array([x, y, x + w, y + h])[None]  # pixels(1,4)
+                          x, y, w, h = xyxy2xywhn(xyxy, w=width, h=height, clip=True)[0]  # normalized and clipped
+                          file.write(f"{cid} {x:.5f} {y:.5f} {w:.5f} {h:.5f}\n")
+              except Exception as e:
+                  print(e)

ultralytics/cfg/datasets/SKU-110K.yaml

@@ -0,0 +1,58 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# SKU-110K retail items dataset https://github.com/eg4000/SKU110K_CVPR19 by Trax Retail
+# Documentation: https://docs.ultralytics.com/datasets/detect/sku-110k/
+# Example usage: yolo train data=SKU-110K.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── SKU-110K  ← downloads here (13.6 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/SKU-110K # dataset root dir
+train: train.txt # train images (relative to 'path')  8219 images
+val: val.txt # val images (relative to 'path')  588 images
+test: test.txt # test images (optional)  2936 images
+
+# Classes
+names:
+  0: object
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import shutil
+  from pathlib import Path
+
+  import numpy as np
+  import pandas as pd
+  from tqdm import tqdm
+
+  from ultralytics.utils.downloads import download
+  from ultralytics.utils.ops import xyxy2xywh
+
+  # Download
+  dir = Path(yaml['path'])  # dataset root dir
+  parent = Path(dir.parent)  # download dir
+  urls = ['http://trax-geometry.s3.amazonaws.com/cvpr_challenge/SKU110K_fixed.tar.gz']
+  download(urls, dir=parent)
+
+  # Rename directories
+  if dir.exists():
+      shutil.rmtree(dir)
+  (parent / 'SKU110K_fixed').rename(dir)  # rename dir
+  (dir / 'labels').mkdir(parents=True, exist_ok=True)  # create labels dir
+
+  # Convert labels
+  names = 'image', 'x1', 'y1', 'x2', 'y2', 'class', 'image_width', 'image_height'  # column names
+  for d in 'annotations_train.csv', 'annotations_val.csv', 'annotations_test.csv':
+      x = pd.read_csv(dir / 'annotations' / d, names=names).values  # annotations
+      images, unique_images = x[:, 0], np.unique(x[:, 0])
+      with open((dir / d).with_suffix('.txt').__str__().replace('annotations_', ''), 'w') as f:
+          f.writelines(f'./images/{s}\n' for s in unique_images)
+      for im in tqdm(unique_images, desc=f'Converting {dir / d}'):
+          cls = 0  # single-class dataset
+          with open((dir / 'labels' / im).with_suffix('.txt'), 'a') as f:
+              for r in x[images == im]:
+                  w, h = r[6], r[7]  # image width, height
+                  xywh = xyxy2xywh(np.array([[r[1] / w, r[2] / h, r[3] / w, r[4] / h]]))[0]  # instance
+                  f.write(f"{cls} {xywh[0]:.5f} {xywh[1]:.5f} {xywh[2]:.5f} {xywh[3]:.5f}\n")  # write label

ultralytics/cfg/datasets/VOC.yaml

@@ -0,0 +1,100 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# PASCAL VOC dataset http://host.robots.ox.ac.uk/pascal/VOC by University of Oxford
+# Documentation: # Documentation: https://docs.ultralytics.com/datasets/detect/voc/
+# Example usage: yolo train data=VOC.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── VOC  ← downloads here (2.8 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/VOC
+train: # train images (relative to 'path')  16551 images
+  - images/train2012
+  - images/train2007
+  - images/val2012
+  - images/val2007
+val: # val images (relative to 'path')  4952 images
+  - images/test2007
+test: # test images (optional)
+  - images/test2007
+
+# Classes
+names:
+  0: aeroplane
+  1: bicycle
+  2: bird
+  3: boat
+  4: bottle
+  5: bus
+  6: car
+  7: cat
+  8: chair
+  9: cow
+  10: diningtable
+  11: dog
+  12: horse
+  13: motorbike
+  14: person
+  15: pottedplant
+  16: sheep
+  17: sofa
+  18: train
+  19: tvmonitor
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import xml.etree.ElementTree as ET
+
+  from tqdm import tqdm
+  from ultralytics.utils.downloads import download
+  from pathlib import Path
+
+  def convert_label(path, lb_path, year, image_id):
+      def convert_box(size, box):
+          dw, dh = 1. / size[0], 1. / size[1]
+          x, y, w, h = (box[0] + box[1]) / 2.0 - 1, (box[2] + box[3]) / 2.0 - 1, box[1] - box[0], box[3] - box[2]
+          return x * dw, y * dh, w * dw, h * dh
+
+      in_file = open(path / f'VOC{year}/Annotations/{image_id}.xml')
+      out_file = open(lb_path, 'w')
+      tree = ET.parse(in_file)
+      root = tree.getroot()
+      size = root.find('size')
+      w = int(size.find('width').text)
+      h = int(size.find('height').text)
+
+      names = list(yaml['names'].values())  # names list
+      for obj in root.iter('object'):
+          cls = obj.find('name').text
+          if cls in names and int(obj.find('difficult').text) != 1:
+              xmlbox = obj.find('bndbox')
+              bb = convert_box((w, h), [float(xmlbox.find(x).text) for x in ('xmin', 'xmax', 'ymin', 'ymax')])
+              cls_id = names.index(cls)  # class id
+              out_file.write(" ".join(str(a) for a in (cls_id, *bb)) + '\n')
+
+
+  # Download
+  dir = Path(yaml['path'])  # dataset root dir
+  url = 'https://github.com/ultralytics/assets/releases/download/v0.0.0/'
+  urls = [f'{url}VOCtrainval_06-Nov-2007.zip',  # 446MB, 5012 images
+          f'{url}VOCtest_06-Nov-2007.zip',  # 438MB, 4953 images
+          f'{url}VOCtrainval_11-May-2012.zip']  # 1.95GB, 17126 images
+  download(urls, dir=dir / 'images', curl=True, threads=3, exist_ok=True)  # download and unzip over existing paths (required)
+
+  # Convert
+  path = dir / 'images/VOCdevkit'
+  for year, image_set in ('2012', 'train'), ('2012', 'val'), ('2007', 'train'), ('2007', 'val'), ('2007', 'test'):
+      imgs_path = dir / 'images' / f'{image_set}{year}'
+      lbs_path = dir / 'labels' / f'{image_set}{year}'
+      imgs_path.mkdir(exist_ok=True, parents=True)
+      lbs_path.mkdir(exist_ok=True, parents=True)
+
+      with open(path / f'VOC{year}/ImageSets/Main/{image_set}.txt') as f:
+          image_ids = f.read().strip().split()
+      for id in tqdm(image_ids, desc=f'{image_set}{year}'):
+          f = path / f'VOC{year}/JPEGImages/{id}.jpg'  # old img path
+          lb_path = (lbs_path / f.name).with_suffix('.txt')  # new label path
+          f.rename(imgs_path / f.name)  # move image
+          convert_label(path, lb_path, year, id)  # convert labels to YOLO format

ultralytics/cfg/datasets/VisDrone.yaml

@@ -0,0 +1,73 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# VisDrone2019-DET dataset https://github.com/VisDrone/VisDrone-Dataset by Tianjin University
+# Documentation: https://docs.ultralytics.com/datasets/detect/visdrone/
+# Example usage: yolo train data=VisDrone.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── VisDrone  ← downloads here (2.3 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/VisDrone # dataset root dir
+train: VisDrone2019-DET-train/images # train images (relative to 'path')  6471 images
+val: VisDrone2019-DET-val/images # val images (relative to 'path')  548 images
+test: VisDrone2019-DET-test-dev/images # test images (optional)  1610 images
+
+# Classes
+names:
+  0: pedestrian
+  1: people
+  2: bicycle
+  3: car
+  4: van
+  5: truck
+  6: tricycle
+  7: awning-tricycle
+  8: bus
+  9: motor
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import os
+  from pathlib import Path
+
+  from ultralytics.utils.downloads import download
+
+  def visdrone2yolo(dir):
+      from PIL import Image
+      from tqdm import tqdm
+
+      def convert_box(size, box):
+          # Convert VisDrone box to YOLO xywh box
+          dw = 1. / size[0]
+          dh = 1. / size[1]
+          return (box[0] + box[2] / 2) * dw, (box[1] + box[3] / 2) * dh, box[2] * dw, box[3] * dh
+
+      (dir / 'labels').mkdir(parents=True, exist_ok=True)  # make labels directory
+      pbar = tqdm((dir / 'annotations').glob('*.txt'), desc=f'Converting {dir}')
+      for f in pbar:
+          img_size = Image.open((dir / 'images' / f.name).with_suffix('.jpg')).size
+          lines = []
+          with open(f, 'r') as file:  # read annotation.txt
+              for row in [x.split(',') for x in file.read().strip().splitlines()]:
+                  if row[4] == '0':  # VisDrone 'ignored regions' class 0
+                      continue
+                  cls = int(row[5]) - 1
+                  box = convert_box(img_size, tuple(map(int, row[:4])))
+                  lines.append(f"{cls} {' '.join(f'{x:.6f}' for x in box)}\n")
+                  with open(str(f).replace(f'{os.sep}annotations{os.sep}', f'{os.sep}labels{os.sep}'), 'w') as fl:
+                      fl.writelines(lines)  # write label.txt
+
+
+  # Download
+  dir = Path(yaml['path'])  # dataset root dir
+  urls = ['https://github.com/ultralytics/assets/releases/download/v0.0.0/VisDrone2019-DET-train.zip',
+          'https://github.com/ultralytics/assets/releases/download/v0.0.0/VisDrone2019-DET-val.zip',
+          'https://github.com/ultralytics/assets/releases/download/v0.0.0/VisDrone2019-DET-test-dev.zip',
+          'https://github.com/ultralytics/assets/releases/download/v0.0.0/VisDrone2019-DET-test-challenge.zip']
+  download(urls, dir=dir, curl=True, threads=4)
+
+  # Convert
+  for d in 'VisDrone2019-DET-train', 'VisDrone2019-DET-val', 'VisDrone2019-DET-test-dev':
+      visdrone2yolo(dir / d)  # convert VisDrone annotations to YOLO labels

ultralytics/cfg/datasets/african-wildlife.yaml

@@ -0,0 +1,25 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# African-wildlife dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/detect/african-wildlife/
+# Example usage: yolo train data=african-wildlife.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── african-wildlife  ← downloads here (100 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/african-wildlife # dataset root dir
+train: train/images # train images (relative to 'path') 1052 images
+val: valid/images # val images (relative to 'path') 225 images
+test: test/images # test images (relative to 'path') 227 images
+
+# Classes
+names:
+  0: buffalo
+  1: elephant
+  2: rhino
+  3: zebra
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/african-wildlife.zip

ultralytics/cfg/datasets/brain-tumor.yaml

@@ -0,0 +1,23 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Brain-tumor dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/detect/brain-tumor/
+# Example usage: yolo train data=brain-tumor.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── brain-tumor  ← downloads here (4.05 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/brain-tumor # dataset root dir
+train: train/images # train images (relative to 'path') 893 images
+val: valid/images # val images (relative to 'path') 223 images
+test: # test images (relative to 'path')
+
+# Classes
+names:
+  0: negative
+  1: positive
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/brain-tumor.zip

ultralytics/cfg/datasets/carparts-seg.yaml

@@ -0,0 +1,44 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Carparts-seg dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/segment/carparts-seg/
+# Example usage: yolo train data=carparts-seg.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── carparts-seg  ← downloads here (132 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/carparts-seg # dataset root dir
+train: train/images # train images (relative to 'path') 3516 images
+val: valid/images # val images (relative to 'path') 276 images
+test: test/images # test images (relative to 'path') 401 images
+
+# Classes
+names:
+  0: back_bumper
+  1: back_door
+  2: back_glass
+  3: back_left_door
+  4: back_left_light
+  5: back_light
+  6: back_right_door
+  7: back_right_light
+  8: front_bumper
+  9: front_door
+  10: front_glass
+  11: front_left_door
+  12: front_left_light
+  13: front_light
+  14: front_right_door
+  15: front_right_light
+  16: hood
+  17: left_mirror
+  18: object
+  19: right_mirror
+  20: tailgate
+  21: trunk
+  22: wheel
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/carparts-seg.zip

ultralytics/cfg/datasets/coco-pose.yaml

@@ -0,0 +1,39 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# COCO 2017 Keypoints dataset https://cocodataset.org by Microsoft
+# Documentation: https://docs.ultralytics.com/datasets/pose/coco/
+# Example usage: yolo train data=coco-pose.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco-pose  ← downloads here (20.1 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/coco-pose # dataset root dir
+train: train2017.txt # train images (relative to 'path') 56599 images
+val: val2017.txt # val images (relative to 'path') 2346 images
+test: test-dev2017.txt # 20288 of 40670 images, submit to https://codalab.lisn.upsaclay.fr/competitions/7403
+
+# Keypoints
+kpt_shape: [17, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+flip_idx: [0, 2, 1, 4, 3, 6, 5, 8, 7, 10, 9, 12, 11, 14, 13, 16, 15]
+
+# Classes
+names:
+  0: person
+
+# Download script/URL (optional)
+download: |
+  from ultralytics.utils.downloads import download
+  from pathlib import Path
+
+  # Download labels
+  dir = Path(yaml['path'])  # dataset root dir
+  url = 'https://github.com/ultralytics/assets/releases/download/v0.0.0/'
+  urls = [url + 'coco2017labels-pose.zip']  # labels
+  download(urls, dir=dir.parent)
+  # Download data
+  urls = ['http://images.cocodataset.org/zips/train2017.zip',  # 19G, 118k images
+          'http://images.cocodataset.org/zips/val2017.zip',  # 1G, 5k images
+          'http://images.cocodataset.org/zips/test2017.zip']  # 7G, 41k images (optional)
+  download(urls, dir=dir / 'images', threads=3)

ultralytics/cfg/datasets/coco.yaml

@@ -0,0 +1,115 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# COCO 2017 dataset https://cocodataset.org by Microsoft
+# Documentation: https://docs.ultralytics.com/datasets/detect/coco/
+# Example usage: yolo train data=coco.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco  ← downloads here (20.1 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: /data_local2/tianyunjie/Projects/YOLOV11/datasets/coco # dataset root dir
+train: train2017.txt # train images (relative to 'path') 118287 images
+val: val2017.txt # val images (relative to 'path') 5000 images
+test: test-dev2017.txt # 20288 of 40670 images, submit to https://competitions.codalab.org/competitions/20794
+
+# Classes
+names:
+  0: person
+  1: bicycle
+  2: car
+  3: motorcycle
+  4: airplane
+  5: bus
+  6: train
+  7: truck
+  8: boat
+  9: traffic light
+  10: fire hydrant
+  11: stop sign
+  12: parking meter
+  13: bench
+  14: bird
+  15: cat
+  16: dog
+  17: horse
+  18: sheep
+  19: cow
+  20: elephant
+  21: bear
+  22: zebra
+  23: giraffe
+  24: backpack
+  25: umbrella
+  26: handbag
+  27: tie
+  28: suitcase
+  29: frisbee
+  30: skis
+  31: snowboard
+  32: sports ball
+  33: kite
+  34: baseball bat
+  35: baseball glove
+  36: skateboard
+  37: surfboard
+  38: tennis racket
+  39: bottle
+  40: wine glass
+  41: cup
+  42: fork
+  43: knife
+  44: spoon
+  45: bowl
+  46: banana
+  47: apple
+  48: sandwich
+  49: orange
+  50: broccoli
+  51: carrot
+  52: hot dog
+  53: pizza
+  54: donut
+  55: cake
+  56: chair
+  57: couch
+  58: potted plant
+  59: bed
+  60: dining table
+  61: toilet
+  62: tv
+  63: laptop
+  64: mouse
+  65: remote
+  66: keyboard
+  67: cell phone
+  68: microwave
+  69: oven
+  70: toaster
+  71: sink
+  72: refrigerator
+  73: book
+  74: clock
+  75: vase
+  76: scissors
+  77: teddy bear
+  78: hair drier
+  79: toothbrush
+
+# Download script/URL (optional)
+download: |
+  from ultralytics.utils.downloads import download
+  from pathlib import Path
+
+  # Download labels
+  segments = True  # segment or box labels
+  dir = Path(yaml['path'])  # dataset root dir
+  url = 'https://github.com/ultralytics/assets/releases/download/v0.0.0/'
+  urls = [url + ('coco2017labels-segments.zip' if segments else 'coco2017labels.zip')]  # labels
+  download(urls, dir=dir.parent)
+  # Download data
+  urls = ['http://images.cocodataset.org/zips/train2017.zip',  # 19G, 118k images
+          'http://images.cocodataset.org/zips/val2017.zip',  # 1G, 5k images
+          'http://images.cocodataset.org/zips/test2017.zip']  # 7G, 41k images (optional)
+  download(urls, dir=dir / 'images', threads=3)

ultralytics/cfg/datasets/coco128-seg.yaml

@@ -0,0 +1,101 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# COCO128-seg dataset https://www.kaggle.com/datasets/ultralytics/coco128 (first 128 images from COCO train2017) by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/segment/coco/
+# Example usage: yolo train data=coco128.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco128-seg  ← downloads here (7 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/coco128-seg # dataset root dir
+train: images/train2017 # train images (relative to 'path') 128 images
+val: images/train2017 # val images (relative to 'path') 128 images
+test: # test images (optional)
+
+# Classes
+names:
+  0: person
+  1: bicycle
+  2: car
+  3: motorcycle
+  4: airplane
+  5: bus
+  6: train
+  7: truck
+  8: boat
+  9: traffic light
+  10: fire hydrant
+  11: stop sign
+  12: parking meter
+  13: bench
+  14: bird
+  15: cat
+  16: dog
+  17: horse
+  18: sheep
+  19: cow
+  20: elephant
+  21: bear
+  22: zebra
+  23: giraffe
+  24: backpack
+  25: umbrella
+  26: handbag
+  27: tie
+  28: suitcase
+  29: frisbee
+  30: skis
+  31: snowboard
+  32: sports ball
+  33: kite
+  34: baseball bat
+  35: baseball glove
+  36: skateboard
+  37: surfboard
+  38: tennis racket
+  39: bottle
+  40: wine glass
+  41: cup
+  42: fork
+  43: knife
+  44: spoon
+  45: bowl
+  46: banana
+  47: apple
+  48: sandwich
+  49: orange
+  50: broccoli
+  51: carrot
+  52: hot dog
+  53: pizza
+  54: donut
+  55: cake
+  56: chair
+  57: couch
+  58: potted plant
+  59: bed
+  60: dining table
+  61: toilet
+  62: tv
+  63: laptop
+  64: mouse
+  65: remote
+  66: keyboard
+  67: cell phone
+  68: microwave
+  69: oven
+  70: toaster
+  71: sink
+  72: refrigerator
+  73: book
+  74: clock
+  75: vase
+  76: scissors
+  77: teddy bear
+  78: hair drier
+  79: toothbrush
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/coco128-seg.zip

ultralytics/cfg/datasets/coco128.yaml

@@ -0,0 +1,101 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# COCO128 dataset https://www.kaggle.com/datasets/ultralytics/coco128 (first 128 images from COCO train2017) by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/detect/coco/
+# Example usage: yolo train data=coco128.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco128  ← downloads here (7 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/coco128 # dataset root dir
+train: images/train2017 # train images (relative to 'path') 128 images
+val: images/train2017 # val images (relative to 'path') 128 images
+test: # test images (optional)
+
+# Classes
+names:
+  0: person
+  1: bicycle
+  2: car
+  3: motorcycle
+  4: airplane
+  5: bus
+  6: train
+  7: truck
+  8: boat
+  9: traffic light
+  10: fire hydrant
+  11: stop sign
+  12: parking meter
+  13: bench
+  14: bird
+  15: cat
+  16: dog
+  17: horse
+  18: sheep
+  19: cow
+  20: elephant
+  21: bear
+  22: zebra
+  23: giraffe
+  24: backpack
+  25: umbrella
+  26: handbag
+  27: tie
+  28: suitcase
+  29: frisbee
+  30: skis
+  31: snowboard
+  32: sports ball
+  33: kite
+  34: baseball bat
+  35: baseball glove
+  36: skateboard
+  37: surfboard
+  38: tennis racket
+  39: bottle
+  40: wine glass
+  41: cup
+  42: fork
+  43: knife
+  44: spoon
+  45: bowl
+  46: banana
+  47: apple
+  48: sandwich
+  49: orange
+  50: broccoli
+  51: carrot
+  52: hot dog
+  53: pizza
+  54: donut
+  55: cake
+  56: chair
+  57: couch
+  58: potted plant
+  59: bed
+  60: dining table
+  61: toilet
+  62: tv
+  63: laptop
+  64: mouse
+  65: remote
+  66: keyboard
+  67: cell phone
+  68: microwave
+  69: oven
+  70: toaster
+  71: sink
+  72: refrigerator
+  73: book
+  74: clock
+  75: vase
+  76: scissors
+  77: teddy bear
+  78: hair drier
+  79: toothbrush
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/coco128.zip

ultralytics/cfg/datasets/coco8-pose.yaml

@@ -0,0 +1,26 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# COCO8-pose dataset (first 8 images from COCO train2017) by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/pose/coco8-pose/
+# Example usage: yolo train data=coco8-pose.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco8-pose  ← downloads here (1 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/coco8-pose # dataset root dir
+train: images/train # train images (relative to 'path') 4 images
+val: images/val # val images (relative to 'path') 4 images
+test: # test images (optional)
+
+# Keypoints
+kpt_shape: [17, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+flip_idx: [0, 2, 1, 4, 3, 6, 5, 8, 7, 10, 9, 12, 11, 14, 13, 16, 15]
+
+# Classes
+names:
+  0: person
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/coco8-pose.zip

ultralytics/cfg/datasets/coco8-seg.yaml

@@ -0,0 +1,101 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# COCO8-seg dataset (first 8 images from COCO train2017) by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/segment/coco8-seg/
+# Example usage: yolo train data=coco8-seg.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco8-seg  ← downloads here (1 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/coco8-seg # dataset root dir
+train: images/train # train images (relative to 'path') 4 images
+val: images/val # val images (relative to 'path') 4 images
+test: # test images (optional)
+
+# Classes
+names:
+  0: person
+  1: bicycle
+  2: car
+  3: motorcycle
+  4: airplane
+  5: bus
+  6: train
+  7: truck
+  8: boat
+  9: traffic light
+  10: fire hydrant
+  11: stop sign
+  12: parking meter
+  13: bench
+  14: bird
+  15: cat
+  16: dog
+  17: horse
+  18: sheep
+  19: cow
+  20: elephant
+  21: bear
+  22: zebra
+  23: giraffe
+  24: backpack
+  25: umbrella
+  26: handbag
+  27: tie
+  28: suitcase
+  29: frisbee
+  30: skis
+  31: snowboard
+  32: sports ball
+  33: kite
+  34: baseball bat
+  35: baseball glove
+  36: skateboard
+  37: surfboard
+  38: tennis racket
+  39: bottle
+  40: wine glass
+  41: cup
+  42: fork
+  43: knife
+  44: spoon
+  45: bowl
+  46: banana
+  47: apple
+  48: sandwich
+  49: orange
+  50: broccoli
+  51: carrot
+  52: hot dog
+  53: pizza
+  54: donut
+  55: cake
+  56: chair
+  57: couch
+  58: potted plant
+  59: bed
+  60: dining table
+  61: toilet
+  62: tv
+  63: laptop
+  64: mouse
+  65: remote
+  66: keyboard
+  67: cell phone
+  68: microwave
+  69: oven
+  70: toaster
+  71: sink
+  72: refrigerator
+  73: book
+  74: clock
+  75: vase
+  76: scissors
+  77: teddy bear
+  78: hair drier
+  79: toothbrush
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/coco8-seg.zip

ultralytics/cfg/datasets/coco8.yaml

@@ -0,0 +1,101 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# COCO8 dataset (first 8 images from COCO train2017) by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/detect/coco8/
+# Example usage: yolo train data=coco8.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco8  ← downloads here (1 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/coco8 # dataset root dir
+train: images/train # train images (relative to 'path') 4 images
+val: images/val # val images (relative to 'path') 4 images
+test: # test images (optional)
+
+# Classes
+names:
+  0: person
+  1: bicycle
+  2: car
+  3: motorcycle
+  4: airplane
+  5: bus
+  6: train
+  7: truck
+  8: boat
+  9: traffic light
+  10: fire hydrant
+  11: stop sign
+  12: parking meter
+  13: bench
+  14: bird
+  15: cat
+  16: dog
+  17: horse
+  18: sheep
+  19: cow
+  20: elephant
+  21: bear
+  22: zebra
+  23: giraffe
+  24: backpack
+  25: umbrella
+  26: handbag
+  27: tie
+  28: suitcase
+  29: frisbee
+  30: skis
+  31: snowboard
+  32: sports ball
+  33: kite
+  34: baseball bat
+  35: baseball glove
+  36: skateboard
+  37: surfboard
+  38: tennis racket
+  39: bottle
+  40: wine glass
+  41: cup
+  42: fork
+  43: knife
+  44: spoon
+  45: bowl
+  46: banana
+  47: apple
+  48: sandwich
+  49: orange
+  50: broccoli
+  51: carrot
+  52: hot dog
+  53: pizza
+  54: donut
+  55: cake
+  56: chair
+  57: couch
+  58: potted plant
+  59: bed
+  60: dining table
+  61: toilet
+  62: tv
+  63: laptop
+  64: mouse
+  65: remote
+  66: keyboard
+  67: cell phone
+  68: microwave
+  69: oven
+  70: toaster
+  71: sink
+  72: refrigerator
+  73: book
+  74: clock
+  75: vase
+  76: scissors
+  77: teddy bear
+  78: hair drier
+  79: toothbrush
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/coco8.zip

ultralytics/cfg/datasets/crack-seg.yaml

@@ -0,0 +1,22 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Crack-seg dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/segment/crack-seg/
+# Example usage: yolo train data=crack-seg.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── crack-seg  ← downloads here (91.2 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/crack-seg # dataset root dir
+train: train/images # train images (relative to 'path') 3717 images
+val: valid/images # val images (relative to 'path') 112 images
+test: test/images # test images (relative to 'path') 200 images
+
+# Classes
+names:
+  0: crack
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/crack-seg.zip

ultralytics/cfg/datasets/dog-pose.yaml

@@ -0,0 +1,24 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Dogs dataset http://vision.stanford.edu/aditya86/ImageNetDogs/ by Stanford
+# Documentation: https://docs.ultralytics.com/datasets/pose/dog-pose/
+# Example usage: yolo train data=dog-pose.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── dog-pose  ← downloads here (337 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/dog-pose # dataset root dir
+train: train # train images (relative to 'path') 6773 images
+val: val # val images (relative to 'path') 1703 images
+
+# Keypoints
+kpt_shape: [24, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+
+# Classes
+names:
+  0: dog
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/dog-pose.zip

ultralytics/cfg/datasets/dota8.yaml

@@ -0,0 +1,35 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# DOTA8 dataset 8 images from split DOTAv1 dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/obb/dota8/
+# Example usage: yolo train model=yolov8n-obb.pt data=dota8.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── dota8  ← downloads here (1MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/dota8 # dataset root dir
+train: images/train # train images (relative to 'path') 4 images
+val: images/val # val images (relative to 'path') 4 images
+
+# Classes for DOTA 1.0
+names:
+  0: plane
+  1: ship
+  2: storage tank
+  3: baseball diamond
+  4: tennis court
+  5: basketball court
+  6: ground track field
+  7: harbor
+  8: bridge
+  9: large vehicle
+  10: small vehicle
+  11: helicopter
+  12: roundabout
+  13: soccer ball field
+  14: swimming pool
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/dota8.zip

ultralytics/cfg/datasets/hand-keypoints.yaml

@@ -0,0 +1,26 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Hand Keypoints dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/pose/hand-keypoints/
+# Example usage: yolo train data=hand-keypoints.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── hand-keypoints  ← downloads here (369 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/hand-keypoints # dataset root dir
+train: train # train images (relative to 'path') 18776 images
+val: val # val images (relative to 'path') 7992 images
+
+# Keypoints
+kpt_shape: [21, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+flip_idx:
+  [0, 1, 2, 4, 3, 10, 11, 12, 13, 14, 5, 6, 7, 8, 9, 15, 16, 17, 18, 19, 20]
+
+# Classes
+names:
+  0: hand
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/hand-keypoints.zip

ultralytics/cfg/datasets/medical-pills.yaml

@@ -0,0 +1,22 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Medical-pills dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/detect/medical-pills/
+# Example usage: yolo train data=medical-pills.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── medical-pills  ← downloads here (8.19 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/medical-pills # dataset root dir
+train: train/images # train images (relative to 'path') 92 images
+val: valid/images # val images (relative to 'path') 23 images
+test: # test images (relative to 'path')
+
+# Classes
+names:
+  0: pill
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/medical-pills.zip

ultralytics/cfg/datasets/open-images-v7.yaml

@@ -0,0 +1,661 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Open Images v7 dataset https://storage.googleapis.com/openimages/web/index.html by Google
+# Documentation: https://docs.ultralytics.com/datasets/detect/open-images-v7/
+# Example usage: yolo train data=open-images-v7.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── open-images-v7  ← downloads here (561 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/open-images-v7 # dataset root dir
+train: images/train # train images (relative to 'path') 1743042 images
+val: images/val # val images (relative to 'path') 41620 images
+test: # test images (optional)
+
+# Classes
+names:
+  0: Accordion
+  1: Adhesive tape
+  2: Aircraft
+  3: Airplane
+  4: Alarm clock
+  5: Alpaca
+  6: Ambulance
+  7: Animal
+  8: Ant
+  9: Antelope
+  10: Apple
+  11: Armadillo
+  12: Artichoke
+  13: Auto part
+  14: Axe
+  15: Backpack
+  16: Bagel
+  17: Baked goods
+  18: Balance beam
+  19: Ball
+  20: Balloon
+  21: Banana
+  22: Band-aid
+  23: Banjo
+  24: Barge
+  25: Barrel
+  26: Baseball bat
+  27: Baseball glove
+  28: Bat (Animal)
+  29: Bathroom accessory
+  30: Bathroom cabinet
+  31: Bathtub
+  32: Beaker
+  33: Bear
+  34: Bed
+  35: Bee
+  36: Beehive
+  37: Beer
+  38: Beetle
+  39: Bell pepper
+  40: Belt
+  41: Bench
+  42: Bicycle
+  43: Bicycle helmet
+  44: Bicycle wheel
+  45: Bidet
+  46: Billboard
+  47: Billiard table
+  48: Binoculars
+  49: Bird
+  50: Blender
+  51: Blue jay
+  52: Boat
+  53: Bomb
+  54: Book
+  55: Bookcase
+  56: Boot
+  57: Bottle
+  58: Bottle opener
+  59: Bow and arrow
+  60: Bowl
+  61: Bowling equipment
+  62: Box
+  63: Boy
+  64: Brassiere
+  65: Bread
+  66: Briefcase
+  67: Broccoli
+  68: Bronze sculpture
+  69: Brown bear
+  70: Building
+  71: Bull
+  72: Burrito
+  73: Bus
+  74: Bust
+  75: Butterfly
+  76: Cabbage
+  77: Cabinetry
+  78: Cake
+  79: Cake stand
+  80: Calculator
+  81: Camel
+  82: Camera
+  83: Can opener
+  84: Canary
+  85: Candle
+  86: Candy
+  87: Cannon
+  88: Canoe
+  89: Cantaloupe
+  90: Car
+  91: Carnivore
+  92: Carrot
+  93: Cart
+  94: Cassette deck
+  95: Castle
+  96: Cat
+  97: Cat furniture
+  98: Caterpillar
+  99: Cattle
+  100: Ceiling fan
+  101: Cello
+  102: Centipede
+  103: Chainsaw
+  104: Chair
+  105: Cheese
+  106: Cheetah
+  107: Chest of drawers
+  108: Chicken
+  109: Chime
+  110: Chisel
+  111: Chopsticks
+  112: Christmas tree
+  113: Clock
+  114: Closet
+  115: Clothing
+  116: Coat
+  117: Cocktail
+  118: Cocktail shaker
+  119: Coconut
+  120: Coffee
+  121: Coffee cup
+  122: Coffee table
+  123: Coffeemaker
+  124: Coin
+  125: Common fig
+  126: Common sunflower
+  127: Computer keyboard
+  128: Computer monitor
+  129: Computer mouse
+  130: Container
+  131: Convenience store
+  132: Cookie
+  133: Cooking spray
+  134: Corded phone
+  135: Cosmetics
+  136: Couch
+  137: Countertop
+  138: Cowboy hat
+  139: Crab
+  140: Cream
+  141: Cricket ball
+  142: Crocodile
+  143: Croissant
+  144: Crown
+  145: Crutch
+  146: Cucumber
+  147: Cupboard
+  148: Curtain
+  149: Cutting board
+  150: Dagger
+  151: Dairy Product
+  152: Deer
+  153: Desk
+  154: Dessert
+  155: Diaper
+  156: Dice
+  157: Digital clock
+  158: Dinosaur
+  159: Dishwasher
+  160: Dog
+  161: Dog bed
+  162: Doll
+  163: Dolphin
+  164: Door
+  165: Door handle
+  166: Doughnut
+  167: Dragonfly
+  168: Drawer
+  169: Dress
+  170: Drill (Tool)
+  171: Drink
+  172: Drinking straw
+  173: Drum
+  174: Duck
+  175: Dumbbell
+  176: Eagle
+  177: Earrings
+  178: Egg (Food)
+  179: Elephant
+  180: Envelope
+  181: Eraser
+  182: Face powder
+  183: Facial tissue holder
+  184: Falcon
+  185: Fashion accessory
+  186: Fast food
+  187: Fax
+  188: Fedora
+  189: Filing cabinet
+  190: Fire hydrant
+  191: Fireplace
+  192: Fish
+  193: Flag
+  194: Flashlight
+  195: Flower
+  196: Flowerpot
+  197: Flute
+  198: Flying disc
+  199: Food
+  200: Food processor
+  201: Football
+  202: Football helmet
+  203: Footwear
+  204: Fork
+  205: Fountain
+  206: Fox
+  207: French fries
+  208: French horn
+  209: Frog
+  210: Fruit
+  211: Frying pan
+  212: Furniture
+  213: Garden Asparagus
+  214: Gas stove
+  215: Giraffe
+  216: Girl
+  217: Glasses
+  218: Glove
+  219: Goat
+  220: Goggles
+  221: Goldfish
+  222: Golf ball
+  223: Golf cart
+  224: Gondola
+  225: Goose
+  226: Grape
+  227: Grapefruit
+  228: Grinder
+  229: Guacamole
+  230: Guitar
+  231: Hair dryer
+  232: Hair spray
+  233: Hamburger
+  234: Hammer
+  235: Hamster
+  236: Hand dryer
+  237: Handbag
+  238: Handgun
+  239: Harbor seal
+  240: Harmonica
+  241: Harp
+  242: Harpsichord
+  243: Hat
+  244: Headphones
+  245: Heater
+  246: Hedgehog
+  247: Helicopter
+  248: Helmet
+  249: High heels
+  250: Hiking equipment
+  251: Hippopotamus
+  252: Home appliance
+  253: Honeycomb
+  254: Horizontal bar
+  255: Horse
+  256: Hot dog
+  257: House
+  258: Houseplant
+  259: Human arm
+  260: Human beard
+  261: Human body
+  262: Human ear
+  263: Human eye
+  264: Human face
+  265: Human foot
+  266: Human hair
+  267: Human hand
+  268: Human head
+  269: Human leg
+  270: Human mouth
+  271: Human nose
+  272: Humidifier
+  273: Ice cream
+  274: Indoor rower
+  275: Infant bed
+  276: Insect
+  277: Invertebrate
+  278: Ipod
+  279: Isopod
+  280: Jacket
+  281: Jacuzzi
+  282: Jaguar (Animal)
+  283: Jeans
+  284: Jellyfish
+  285: Jet ski
+  286: Jug
+  287: Juice
+  288: Kangaroo
+  289: Kettle
+  290: Kitchen & dining room table
+  291: Kitchen appliance
+  292: Kitchen knife
+  293: Kitchen utensil
+  294: Kitchenware
+  295: Kite
+  296: Knife
+  297: Koala
+  298: Ladder
+  299: Ladle
+  300: Ladybug
+  301: Lamp
+  302: Land vehicle
+  303: Lantern
+  304: Laptop
+  305: Lavender (Plant)
+  306: Lemon
+  307: Leopard
+  308: Light bulb
+  309: Light switch
+  310: Lighthouse
+  311: Lily
+  312: Limousine
+  313: Lion
+  314: Lipstick
+  315: Lizard
+  316: Lobster
+  317: Loveseat
+  318: Luggage and bags
+  319: Lynx
+  320: Magpie
+  321: Mammal
+  322: Man
+  323: Mango
+  324: Maple
+  325: Maracas
+  326: Marine invertebrates
+  327: Marine mammal
+  328: Measuring cup
+  329: Mechanical fan
+  330: Medical equipment
+  331: Microphone
+  332: Microwave oven
+  333: Milk
+  334: Miniskirt
+  335: Mirror
+  336: Missile
+  337: Mixer
+  338: Mixing bowl
+  339: Mobile phone
+  340: Monkey
+  341: Moths and butterflies
+  342: Motorcycle
+  343: Mouse
+  344: Muffin
+  345: Mug
+  346: Mule
+  347: Mushroom
+  348: Musical instrument
+  349: Musical keyboard
+  350: Nail (Construction)
+  351: Necklace
+  352: Nightstand
+  353: Oboe
+  354: Office building
+  355: Office supplies
+  356: Orange
+  357: Organ (Musical Instrument)
+  358: Ostrich
+  359: Otter
+  360: Oven
+  361: Owl
+  362: Oyster
+  363: Paddle
+  364: Palm tree
+  365: Pancake
+  366: Panda
+  367: Paper cutter
+  368: Paper towel
+  369: Parachute
+  370: Parking meter
+  371: Parrot
+  372: Pasta
+  373: Pastry
+  374: Peach
+  375: Pear
+  376: Pen
+  377: Pencil case
+  378: Pencil sharpener
+  379: Penguin
+  380: Perfume
+  381: Person
+  382: Personal care
+  383: Personal flotation device
+  384: Piano
+  385: Picnic basket
+  386: Picture frame
+  387: Pig
+  388: Pillow
+  389: Pineapple
+  390: Pitcher (Container)
+  391: Pizza
+  392: Pizza cutter
+  393: Plant
+  394: Plastic bag
+  395: Plate
+  396: Platter
+  397: Plumbing fixture
+  398: Polar bear
+  399: Pomegranate
+  400: Popcorn
+  401: Porch
+  402: Porcupine
+  403: Poster
+  404: Potato
+  405: Power plugs and sockets
+  406: Pressure cooker
+  407: Pretzel
+  408: Printer
+  409: Pumpkin
+  410: Punching bag
+  411: Rabbit
+  412: Raccoon
+  413: Racket
+  414: Radish
+  415: Ratchet (Device)
+  416: Raven
+  417: Rays and skates
+  418: Red panda
+  419: Refrigerator
+  420: Remote control
+  421: Reptile
+  422: Rhinoceros
+  423: Rifle
+  424: Ring binder
+  425: Rocket
+  426: Roller skates
+  427: Rose
+  428: Rugby ball
+  429: Ruler
+  430: Salad
+  431: Salt and pepper shakers
+  432: Sandal
+  433: Sandwich
+  434: Saucer
+  435: Saxophone
+  436: Scale
+  437: Scarf
+  438: Scissors
+  439: Scoreboard
+  440: Scorpion
+  441: Screwdriver
+  442: Sculpture
+  443: Sea lion
+  444: Sea turtle
+  445: Seafood
+  446: Seahorse
+  447: Seat belt
+  448: Segway
+  449: Serving tray
+  450: Sewing machine
+  451: Shark
+  452: Sheep
+  453: Shelf
+  454: Shellfish
+  455: Shirt
+  456: Shorts
+  457: Shotgun
+  458: Shower
+  459: Shrimp
+  460: Sink
+  461: Skateboard
+  462: Ski
+  463: Skirt
+  464: Skull
+  465: Skunk
+  466: Skyscraper
+  467: Slow cooker
+  468: Snack
+  469: Snail
+  470: Snake
+  471: Snowboard
+  472: Snowman
+  473: Snowmobile
+  474: Snowplow
+  475: Soap dispenser
+  476: Sock
+  477: Sofa bed
+  478: Sombrero
+  479: Sparrow
+  480: Spatula
+  481: Spice rack
+  482: Spider
+  483: Spoon
+  484: Sports equipment
+  485: Sports uniform
+  486: Squash (Plant)
+  487: Squid
+  488: Squirrel
+  489: Stairs
+  490: Stapler
+  491: Starfish
+  492: Stationary bicycle
+  493: Stethoscope
+  494: Stool
+  495: Stop sign
+  496: Strawberry
+  497: Street light
+  498: Stretcher
+  499: Studio couch
+  500: Submarine
+  501: Submarine sandwich
+  502: Suit
+  503: Suitcase
+  504: Sun hat
+  505: Sunglasses
+  506: Surfboard
+  507: Sushi
+  508: Swan
+  509: Swim cap
+  510: Swimming pool
+  511: Swimwear
+  512: Sword
+  513: Syringe
+  514: Table
+  515: Table tennis racket
+  516: Tablet computer
+  517: Tableware
+  518: Taco
+  519: Tank
+  520: Tap
+  521: Tart
+  522: Taxi
+  523: Tea
+  524: Teapot
+  525: Teddy bear
+  526: Telephone
+  527: Television
+  528: Tennis ball
+  529: Tennis racket
+  530: Tent
+  531: Tiara
+  532: Tick
+  533: Tie
+  534: Tiger
+  535: Tin can
+  536: Tire
+  537: Toaster
+  538: Toilet
+  539: Toilet paper
+  540: Tomato
+  541: Tool
+  542: Toothbrush
+  543: Torch
+  544: Tortoise
+  545: Towel
+  546: Tower
+  547: Toy
+  548: Traffic light
+  549: Traffic sign
+  550: Train
+  551: Training bench
+  552: Treadmill
+  553: Tree
+  554: Tree house
+  555: Tripod
+  556: Trombone
+  557: Trousers
+  558: Truck
+  559: Trumpet
+  560: Turkey
+  561: Turtle
+  562: Umbrella
+  563: Unicycle
+  564: Van
+  565: Vase
+  566: Vegetable
+  567: Vehicle
+  568: Vehicle registration plate
+  569: Violin
+  570: Volleyball (Ball)
+  571: Waffle
+  572: Waffle iron
+  573: Wall clock
+  574: Wardrobe
+  575: Washing machine
+  576: Waste container
+  577: Watch
+  578: Watercraft
+  579: Watermelon
+  580: Weapon
+  581: Whale
+  582: Wheel
+  583: Wheelchair
+  584: Whisk
+  585: Whiteboard
+  586: Willow
+  587: Window
+  588: Window blind
+  589: Wine
+  590: Wine glass
+  591: Wine rack
+  592: Winter melon
+  593: Wok
+  594: Woman
+  595: Wood-burning stove
+  596: Woodpecker
+  597: Worm
+  598: Wrench
+  599: Zebra
+  600: Zucchini
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  from ultralytics.utils import LOGGER, SETTINGS, Path, is_ubuntu, get_ubuntu_version
+  from ultralytics.utils.checks import check_requirements, check_version
+
+  check_requirements('fiftyone')
+  if is_ubuntu() and check_version(get_ubuntu_version(), '>=22.04'):
+      # Ubuntu>=22.04 patch https://github.com/voxel51/fiftyone/issues/2961#issuecomment-1666519347
+      check_requirements('fiftyone-db-ubuntu2204')
+
+  import fiftyone as fo
+  import fiftyone.zoo as foz
+  import warnings
+
+  name = 'open-images-v7'
+  fraction = 1.0  # fraction of full dataset to use
+  LOGGER.warning('WARNING ⚠️ Open Images V7 dataset requires at least **561 GB of free space. Starting download...')
+  for split in 'train', 'validation':  # 1743042 train, 41620 val images
+      train = split == 'train'
+
+      # Load Open Images dataset
+      dataset = foz.load_zoo_dataset(name,
+                                     split=split,
+                                     label_types=['detections'],
+                                     dataset_dir=Path(SETTINGS['datasets_dir']) / 'fiftyone' / name,
+                                     max_samples=round((1743042 if train else 41620) * fraction))
+
+      # Define classes
+      if train:
+          classes = dataset.default_classes  # all classes
+          # classes = dataset.distinct('ground_truth.detections.label')  # only observed classes
+
+      # Export to YOLO format
+      with warnings.catch_warnings():
+          warnings.filterwarnings("ignore", category=UserWarning, module="fiftyone.utils.yolo")
+          dataset.export(export_dir=str(Path(SETTINGS['datasets_dir']) / name),
+                         dataset_type=fo.types.YOLOv5Dataset,
+                         label_field='ground_truth',
+                         split='val' if split == 'validation' else split,
+                         classes=classes,
+                         overwrite=train)

ultralytics/cfg/datasets/package-seg.yaml

@@ -0,0 +1,22 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Package-seg dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/segment/package-seg/
+# Example usage: yolo train data=package-seg.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── package-seg  ← downloads here (102 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/package-seg # dataset root dir
+train: train/images # train images (relative to 'path') 1920 images
+val: valid/images # val images (relative to 'path') 89 images
+test: test/images # test images (relative to 'path') 188 images
+
+# Classes
+names:
+  0: package
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/package-seg.zip

ultralytics/cfg/datasets/signature.yaml

@@ -0,0 +1,21 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Signature dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/detect/signature/
+# Example usage: yolo train data=signature.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── signature  ← downloads here (11.2 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/signature # dataset root dir
+train: train/images # train images (relative to 'path') 143 images
+val: valid/images # val images (relative to 'path') 35 images
+
+# Classes
+names:
+  0: signature
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/signature.zip

ultralytics/cfg/datasets/tiger-pose.yaml

@@ -0,0 +1,25 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Tiger Pose dataset by Ultralytics
+# Documentation: https://docs.ultralytics.com/datasets/pose/tiger-pose/
+# Example usage: yolo train data=tiger-pose.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── tiger-pose  ← downloads here (75.3 MB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/tiger-pose # dataset root dir
+train: train # train images (relative to 'path') 210 images
+val: val # val images (relative to 'path') 53 images
+
+# Keypoints
+kpt_shape: [12, 2] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+flip_idx: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
+
+# Classes
+names:
+  0: tiger
+
+# Download script/URL (optional)
+download: https://github.com/ultralytics/assets/releases/download/v0.0.0/tiger-pose.zip

ultralytics/cfg/datasets/xView.yaml

@@ -0,0 +1,153 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# DIUx xView 2018 Challenge https://challenge.xviewdataset.org by U.S. National Geospatial-Intelligence Agency (NGA)
+# --------  DOWNLOAD DATA MANUALLY and jar xf val_images.zip to 'datasets/xView' before running train command!  --------
+# Documentation: https://docs.ultralytics.com/datasets/detect/xview/
+# Example usage: yolo train data=xView.yaml
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── xView  ← downloads here (20.7 GB)
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/xView # dataset root dir
+train: images/autosplit_train.txt # train images (relative to 'path') 90% of 847 train images
+val: images/autosplit_val.txt # train images (relative to 'path') 10% of 847 train images
+
+# Classes
+names:
+  0: Fixed-wing Aircraft
+  1: Small Aircraft
+  2: Cargo Plane
+  3: Helicopter
+  4: Passenger Vehicle
+  5: Small Car
+  6: Bus
+  7: Pickup Truck
+  8: Utility Truck
+  9: Truck
+  10: Cargo Truck
+  11: Truck w/Box
+  12: Truck Tractor
+  13: Trailer
+  14: Truck w/Flatbed
+  15: Truck w/Liquid
+  16: Crane Truck
+  17: Railway Vehicle
+  18: Passenger Car
+  19: Cargo Car
+  20: Flat Car
+  21: Tank car
+  22: Locomotive
+  23: Maritime Vessel
+  24: Motorboat
+  25: Sailboat
+  26: Tugboat
+  27: Barge
+  28: Fishing Vessel
+  29: Ferry
+  30: Yacht
+  31: Container Ship
+  32: Oil Tanker
+  33: Engineering Vehicle
+  34: Tower crane
+  35: Container Crane
+  36: Reach Stacker
+  37: Straddle Carrier
+  38: Mobile Crane
+  39: Dump Truck
+  40: Haul Truck
+  41: Scraper/Tractor
+  42: Front loader/Bulldozer
+  43: Excavator
+  44: Cement Mixer
+  45: Ground Grader
+  46: Hut/Tent
+  47: Shed
+  48: Building
+  49: Aircraft Hangar
+  50: Damaged Building
+  51: Facility
+  52: Construction Site
+  53: Vehicle Lot
+  54: Helipad
+  55: Storage Tank
+  56: Shipping container lot
+  57: Shipping Container
+  58: Pylon
+  59: Tower
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import json
+  import os
+  from pathlib import Path
+
+  import numpy as np
+  from PIL import Image
+  from tqdm import tqdm
+
+  from ultralytics.data.utils import autosplit
+  from ultralytics.utils.ops import xyxy2xywhn
+
+
+  def convert_labels(fname=Path('xView/xView_train.geojson')):
+      # Convert xView geoJSON labels to YOLO format
+      path = fname.parent
+      with open(fname) as f:
+          print(f'Loading {fname}...')
+          data = json.load(f)
+
+      # Make dirs
+      labels = Path(path / 'labels' / 'train')
+      os.system(f'rm -rf {labels}')
+      labels.mkdir(parents=True, exist_ok=True)
+
+      # xView classes 11-94 to 0-59
+      xview_class2index = [-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, 0, 1, 2, -1, 3, -1, 4, 5, 6, 7, 8, -1, 9, 10, 11,
+                           12, 13, 14, 15, -1, -1, 16, 17, 18, 19, 20, 21, 22, -1, 23, 24, 25, -1, 26, 27, -1, 28, -1,
+                           29, 30, 31, 32, 33, 34, 35, 36, 37, -1, 38, 39, 40, 41, 42, 43, 44, 45, -1, -1, -1, -1, 46,
+                           47, 48, 49, -1, 50, 51, -1, 52, -1, -1, -1, 53, 54, -1, 55, -1, -1, 56, -1, 57, -1, 58, 59]
+
+      shapes = {}
+      for feature in tqdm(data['features'], desc=f'Converting {fname}'):
+          p = feature['properties']
+          if p['bounds_imcoords']:
+              id = p['image_id']
+              file = path / 'train_images' / id
+              if file.exists():  # 1395.tif missing
+                  try:
+                      box = np.array([int(num) for num in p['bounds_imcoords'].split(",")])
+                      assert box.shape[0] == 4, f'incorrect box shape {box.shape[0]}'
+                      cls = p['type_id']
+                      cls = xview_class2index[int(cls)]  # xView class to 0-60
+                      assert 59 >= cls >= 0, f'incorrect class index {cls}'
+
+                      # Write YOLO label
+                      if id not in shapes:
+                          shapes[id] = Image.open(file).size
+                      box = xyxy2xywhn(box[None].astype(np.float), w=shapes[id][0], h=shapes[id][1], clip=True)
+                      with open((labels / id).with_suffix('.txt'), 'a') as f:
+                          f.write(f"{cls} {' '.join(f'{x:.6f}' for x in box[0])}\n")  # write label.txt
+                  except Exception as e:
+                      print(f'WARNING: skipping one label for {file}: {e}')
+
+
+  # Download manually from https://challenge.xviewdataset.org
+  dir = Path(yaml['path'])  # dataset root dir
+  # urls = ['https://d307kc0mrhucc3.cloudfront.net/train_labels.zip',  # train labels
+  #         'https://d307kc0mrhucc3.cloudfront.net/train_images.zip',  # 15G, 847 train images
+  #         'https://d307kc0mrhucc3.cloudfront.net/val_images.zip']  # 5G, 282 val images (no labels)
+  # download(urls, dir=dir)
+
+  # Convert labels
+  convert_labels(dir / 'xView_train.geojson')
+
+  # Move images
+  images = Path(dir / 'images')
+  images.mkdir(parents=True, exist_ok=True)
+  Path(dir / 'train_images').rename(dir / 'images' / 'train')
+  Path(dir / 'val_images').rename(dir / 'images' / 'val')
+
+  # Split
+  autosplit(dir / 'images' / 'train')

ultralytics/cfg/default.yaml

@@ -0,0 +1,130 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Global configuration YAML with settings and hyperparameters for YOLO training, validation, prediction and export
+# For documentation see https://docs.ultralytics.com/usage/cfg/
+
+task: detect # (str) YOLO task, i.e. detect, segment, classify, pose, obb
+mode: train # (str) YOLO mode, i.e. train, val, predict, export, track, benchmark
+
+# Train settings -------------------------------------------------------------------------------------------------------
+model: # (str, optional) path to model file, i.e. yolov8n.pt, yolov8n.yaml
+data: # (str, optional) path to data file, i.e. coco8.yaml
+epochs: 100 # (int) number of epochs to train for
+time: # (float, optional) number of hours to train for, overrides epochs if supplied
+patience: 100 # (int) epochs to wait for no observable improvement for early stopping of training
+batch: 16 # (int) number of images per batch (-1 for AutoBatch)
+imgsz: 640 # (int | list) input images size as int for train and val modes, or list[h,w] for predict and export modes
+save: True # (bool) save train checkpoints and predict results
+save_period: -1 # (int) Save checkpoint every x epochs (disabled if < 1)
+cache: False # (bool) True/ram, disk or False. Use cache for data loading
+device: # (int | str | list, optional) device to run on, i.e. cuda device=0 or device=0,1,2,3 or device=cpu
+workers: 8 # (int) number of worker threads for data loading (per RANK if DDP)
+project: # (str, optional) project name
+name: # (str, optional) experiment name, results saved to 'project/name' directory
+exist_ok: False # (bool) whether to overwrite existing experiment
+pretrained: True # (bool | str) whether to use a pretrained model (bool) or a model to load weights from (str)
+optimizer: auto # (str) optimizer to use, choices=[SGD, Adam, Adamax, AdamW, NAdam, RAdam, RMSProp, auto]
+verbose: True # (bool) whether to print verbose output
+seed: 0 # (int) random seed for reproducibility
+deterministic: True # (bool) whether to enable deterministic mode
+single_cls: False # (bool) train multi-class data as single-class
+rect: False # (bool) rectangular training if mode='train' or rectangular validation if mode='val'
+cos_lr: False # (bool) use cosine learning rate scheduler
+close_mosaic: 10 # (int) disable mosaic augmentation for final epochs (0 to disable)
+resume: False # (bool) resume training from last checkpoint
+amp: True # (bool) Automatic Mixed Precision (AMP) training, choices=[True, False], True runs AMP check
+fraction: 1.0 # (float) dataset fraction to train on (default is 1.0, all images in train set)
+profile: False # (bool) profile ONNX and TensorRT speeds during training for loggers
+freeze: None # (int | list, optional) freeze first n layers, or freeze list of layer indices during training
+multi_scale: False # (bool) Whether to use multiscale during training
+# Segmentation
+overlap_mask: True # (bool) merge object masks into a single image mask during training (segment train only)
+mask_ratio: 4 # (int) mask downsample ratio (segment train only)
+# Classification
+dropout: 0.0 # (float) use dropout regularization (classify train only)
+
+# Val/Test settings ----------------------------------------------------------------------------------------------------
+val: True # (bool) validate/test during training
+split: val # (str) dataset split to use for validation, i.e. 'val', 'test' or 'train'
+save_json: False # (bool) save results to JSON file
+save_hybrid: False # (bool) save hybrid version of labels (labels + additional predictions)
+conf: # (float, optional) object confidence threshold for detection (default 0.25 predict, 0.001 val)
+iou: 0.7 # (float) intersection over union (IoU) threshold for NMS
+max_det: 300 # (int) maximum number of detections per image
+half: False # (bool) use half precision (FP16)
+dnn: False # (bool) use OpenCV DNN for ONNX inference
+plots: True # (bool) save plots and images during train/val
+
+# Predict settings -----------------------------------------------------------------------------------------------------
+source: # (str, optional) source directory for images or videos
+vid_stride: 1 # (int) video frame-rate stride
+stream_buffer: False # (bool) buffer all streaming frames (True) or return the most recent frame (False)
+visualize: False # (bool) visualize model features
+augment: False # (bool) apply image augmentation to prediction sources
+agnostic_nms: False # (bool) class-agnostic NMS
+classes: # (int | list[int], optional) filter results by class, i.e. classes=0, or classes=[0,2,3]
+retina_masks: False # (bool) use high-resolution segmentation masks
+embed: # (list[int], optional) return feature vectors/embeddings from given layers
+
+# Visualize settings ---------------------------------------------------------------------------------------------------
+show: False # (bool) show predicted images and videos if environment allows
+save_frames: False # (bool) save predicted individual video frames
+save_txt: False # (bool) save results as .txt file
+save_conf: False # (bool) save results with confidence scores
+save_crop: False # (bool) save cropped images with results
+show_labels: True # (bool) show prediction labels, i.e. 'person'
+show_conf: True # (bool) show prediction confidence, i.e. '0.99'
+show_boxes: True # (bool) show prediction boxes
+line_width: # (int, optional) line width of the bounding boxes. Scaled to image size if None.
+
+# Export settings ------------------------------------------------------------------------------------------------------
+format: torchscript # (str) format to export to, choices at https://docs.ultralytics.com/modes/export/#export-formats
+keras: False # (bool) use Kera=s
+optimize: False # (bool) TorchScript: optimize for mobile
+int8: False # (bool) CoreML/TF INT8 quantization
+dynamic: False # (bool) ONNX/TF/TensorRT: dynamic axes
+simplify: True # (bool) ONNX: simplify model using `onnxslim`
+opset: # (int, optional) ONNX: opset version
+workspace: None # (float, optional) TensorRT: workspace size (GiB), `None` will let TensorRT auto-allocate memory
+nms: False # (bool) CoreML: add NMS
+
+# Hyperparameters ------------------------------------------------------------------------------------------------------
+lr0: 0.01 # (float) initial learning rate (i.e. SGD=1E-2, Adam=1E-3)
+lrf: 0.01 # (float) final learning rate (lr0 * lrf)
+momentum: 0.937 # (float) SGD momentum/Adam beta1
+weight_decay: 0.0005 # (float) optimizer weight decay 5e-4
+warmup_epochs: 3.0 # (float) warmup epochs (fractions ok)
+warmup_momentum: 0.8 # (float) warmup initial momentum
+warmup_bias_lr: 0.0 # 0.1 # (float) warmup initial bias lr
+box: 7.5 # (float) box loss gain
+cls: 0.5 # (float) cls loss gain (scale with pixels)
+dfl: 1.5 # (float) dfl loss gain
+pose: 12.0 # (float) pose loss gain
+kobj: 1.0 # (float) keypoint obj loss gain
+nbs: 64 # (int) nominal batch size
+hsv_h: 0.015 # (float) image HSV-Hue augmentation (fraction)
+hsv_s: 0.7 # (float) image HSV-Saturation augmentation (fraction)
+hsv_v: 0.4 # (float) image HSV-Value augmentation (fraction)
+degrees: 0.0 # (float) image rotation (+/- deg)
+translate: 0.1 # (float) image translation (+/- fraction)
+scale: 0.5 # (float) image scale (+/- gain)
+shear: 0.0 # (float) image shear (+/- deg)
+perspective: 0.0 # (float) image perspective (+/- fraction), range 0-0.001
+flipud: 0.0 # (float) image flip up-down (probability)
+fliplr: 0.5 # (float) image flip left-right (probability)
+bgr: 0.0 # (float) image channel BGR (probability)
+
+mosaic: 1.0 # (float) image mosaic (probability)
+mixup: 0.0 # (float) image mixup (probability)
+copy_paste: 0.1 # (float) segment copy-paste (probability)
+
+copy_paste_mode: "flip" # (str) the method to do copy_paste augmentation (flip, mixup)
+auto_augment: randaugment # (str) auto augmentation policy for classification (randaugment, autoaugment, augmix)
+erasing: 0.4 # (float) probability of random erasing during classification training (0-0.9), 0 means no erasing, must be less than 1.0.
+crop_fraction: 1.0 # (float) image crop fraction for classification (0.1-1), 1.0 means no crop, must be greater than 0.
+
+# Custom config.yaml ---------------------------------------------------------------------------------------------------
+cfg: # (str, optional) for overriding defaults.yaml
+
+# Tracker settings ------------------------------------------------------------------------------------------------------
+tracker: botsort.yaml # (str) tracker type, choices=[botsort.yaml, bytetrack.yaml]

ultralytics/cfg/models/11/yolo11-cls-resnet18.yaml

@@ -0,0 +1,24 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLO11-cls image classification model with ResNet18 backbone
+# Model docs: https://docs.ultralytics.com/models/yolo11
+# Task docs: https://docs.ultralytics.com/tasks/classify
+
+# Parameters
+nc: 10 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolo11n-cls.yaml' will call yolo11-cls.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 1024]
+  l: [1.00, 1.00, 1024]
+  x: [1.00, 1.25, 1024]
+
+# ResNet18 backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, TorchVision, [512, "resnet18", "DEFAULT", True, 2]] # truncate two layers from the end
+
+# YOLO11n head
+head:
+  - [-1, 1, Classify, [nc]] # Classify

ultralytics/cfg/models/11/yolo11-cls.yaml

@@ -0,0 +1,33 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLO11-cls image classification model
+# Model docs: https://docs.ultralytics.com/models/yolo11
+# Task docs: https://docs.ultralytics.com/tasks/classify
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolo11n-cls.yaml' will call yolo11-cls.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.50, 0.25, 1024] # summary: 151 layers, 1633584 parameters, 1633584 gradients, 3.3 GFLOPs
+  s: [0.50, 0.50, 1024] # summary: 151 layers, 5545488 parameters, 5545488 gradients, 12.2 GFLOPs
+  m: [0.50, 1.00, 512] # summary: 187 layers, 10455696 parameters, 10455696 gradients, 39.7 GFLOPs
+  l: [1.00, 1.00, 512] # summary: 309 layers, 12937104 parameters, 12937104 gradients, 49.9 GFLOPs
+  x: [1.00, 1.50, 512] # summary: 309 layers, 28458544 parameters, 28458544 gradients, 111.1 GFLOPs
+
+# YOLO11n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 2, C3k2, [256, False, 0.25]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 2, C3k2, [512, False, 0.25]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 2, C3k2, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 2, C3k2, [1024, True]]
+  - [-1, 2, C2PSA, [1024]] # 9
+
+# YOLO11n head
+head:
+  - [-1, 1, Classify, [nc]] # Classify

ultralytics/cfg/models/11/yolo11-obb.yaml

@@ -0,0 +1,50 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLO11-obb Oriented Bounding Boxes (OBB) model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolo11
+# Task docs: https://docs.ultralytics.com/tasks/obb
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolo11n-obb.yaml' will call yolo11-obb.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.50, 0.25, 1024] # summary: 344 layers, 2695747 parameters, 2695731 gradients, 6.9 GFLOPs
+  s: [0.50, 0.50, 1024] # summary: 344 layers, 9744931 parameters, 9744915 gradients, 22.7 GFLOPs
+  m: [0.50, 1.00, 512] # summary: 434 layers, 20963523 parameters, 20963507 gradients, 72.2 GFLOPs
+  l: [1.00, 1.00, 512] # summary: 656 layers, 26220995 parameters, 26220979 gradients, 91.3 GFLOPs
+  x: [1.00, 1.50, 512] # summary: 656 layers, 58875331 parameters, 58875315 gradients, 204.3 GFLOPs
+
+# YOLO11n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 2, C3k2, [256, False, 0.25]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 2, C3k2, [512, False, 0.25]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 2, C3k2, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 2, C3k2, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 2, C2PSA, [1024]] # 10
+
+# YOLO11n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 2, C3k2, [512, False]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 2, C3k2, [256, False]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 2, C3k2, [512, False]] # 19 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 2, C3k2, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, OBB, [nc, 1]] # Detect(P3, P4, P5)

ultralytics/cfg/models/11/yolo11-pose.yaml

@@ -0,0 +1,51 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLO11-pose keypoints/pose estimation model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolo11
+# Task docs: https://docs.ultralytics.com/tasks/pose
+
+# Parameters
+nc: 80 # number of classes
+kpt_shape: [17, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+scales: # model compound scaling constants, i.e. 'model=yolo11n-pose.yaml' will call yolo11.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.50, 0.25, 1024] # summary: 344 layers, 2908507 parameters, 2908491 gradients, 7.7 GFLOPs
+  s: [0.50, 0.50, 1024] # summary: 344 layers, 9948811 parameters, 9948795 gradients, 23.5 GFLOPs
+  m: [0.50, 1.00, 512] # summary: 434 layers, 20973273 parameters, 20973257 gradients, 72.3 GFLOPs
+  l: [1.00, 1.00, 512] # summary: 656 layers, 26230745 parameters, 26230729 gradients, 91.4 GFLOPs
+  x: [1.00, 1.50, 512] # summary: 656 layers, 58889881 parameters, 58889865 gradients, 204.3 GFLOPs
+
+# YOLO11n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 2, C3k2, [256, False, 0.25]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 2, C3k2, [512, False, 0.25]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 2, C3k2, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 2, C3k2, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 2, C2PSA, [1024]] # 10
+
+# YOLO11n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 2, C3k2, [512, False]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 2, C3k2, [256, False]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 2, C3k2, [512, False]] # 19 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 2, C3k2, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, Pose, [nc, kpt_shape]] # Detect(P3, P4, P5)

ultralytics/cfg/models/11/yolo11-seg.yaml

@@ -0,0 +1,50 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLO11-seg instance segmentation model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolo11
+# Task docs: https://docs.ultralytics.com/tasks/segment
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolo11n-seg.yaml' will call yolo11-seg.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.50, 0.25, 1024] # summary: 355 layers, 2876848 parameters, 2876832 gradients, 10.5 GFLOPs
+  s: [0.50, 0.50, 1024] # summary: 355 layers, 10113248 parameters, 10113232 gradients, 35.8 GFLOPs
+  m: [0.50, 1.00, 512] # summary: 445 layers, 22420896 parameters, 22420880 gradients, 123.9 GFLOPs
+  l: [1.00, 1.00, 512] # summary: 667 layers, 27678368 parameters, 27678352 gradients, 143.0 GFLOPs
+  x: [1.00, 1.50, 512] # summary: 667 layers, 62142656 parameters, 62142640 gradients, 320.2 GFLOPs
+
+# YOLO11n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 2, C3k2, [256, False, 0.25]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 2, C3k2, [512, False, 0.25]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 2, C3k2, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 2, C3k2, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 2, C2PSA, [1024]] # 10
+
+# YOLO11n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 2, C3k2, [512, False]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 2, C3k2, [256, False]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 2, C3k2, [512, False]] # 19 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 2, C3k2, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, Segment, [nc, 32, 256]] # Detect(P3, P4, P5)

ultralytics/cfg/models/11/yolo11.yaml

@@ -0,0 +1,50 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLO11 object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolo11
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolo11n.yaml' will call yolo11.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.50, 0.25, 1024] # summary: 319 layers, 2624080 parameters, 2624064 gradients, 6.6 GFLOPs
+  s: [0.50, 0.50, 1024] # summary: 319 layers, 9458752 parameters, 9458736 gradients, 21.7 GFLOPs
+  m: [0.50, 1.00, 512] # summary: 409 layers, 20114688 parameters, 20114672 gradients, 68.5 GFLOPs
+  l: [1.00, 1.00, 512] # summary: 631 layers, 25372160 parameters, 25372144 gradients, 87.6 GFLOPs
+  x: [1.00, 1.50, 512] # summary: 631 layers, 56966176 parameters, 56966160 gradients, 196.0 GFLOPs
+
+# YOLO11n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 2, C3k2, [256, False, 0.25]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 2, C3k2, [512, False, 0.25]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 2, C3k2, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 2, C3k2, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 2, C2PSA, [1024]] # 10
+
+# YOLO11n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 2, C3k2, [512, False]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 2, C3k2, [256, False]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 2, C3k2, [512, False]] # 19 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 2, C3k2, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/README.md

@@ -0,0 +1,48 @@
+## Models
+
+Welcome to the [Ultralytics](https://www.ultralytics.com/) Models directory! Here you will find a wide variety of pre-configured model configuration files (`*.yaml`s) that can be used to create custom YOLO models. The models in this directory have been expertly crafted and fine-tuned by the Ultralytics team to provide the best performance for a wide range of object detection and image segmentation tasks.
+
+These model configurations cover a wide range of scenarios, from simple object detection to more complex tasks like instance segmentation and object tracking. They are also designed to run efficiently on a variety of hardware platforms, from CPUs to GPUs. Whether you are a seasoned machine learning practitioner or just getting started with YOLO, this directory provides a great starting point for your custom model development needs.
+
+To get started, simply browse through the models in this directory and find one that best suits your needs. Once you've selected a model, you can use the provided `*.yaml` file to train and deploy your custom YOLO model with ease. See full details at the Ultralytics [Docs](https://docs.ultralytics.com/models/), and if you need help or have any questions, feel free to reach out to the Ultralytics team for support. So, don't wait, start creating your custom YOLO model now!
+
+### Usage
+
+Model `*.yaml` files may be used directly in the [Command Line Interface (CLI)](https://docs.ultralytics.com/usage/cli/) with a `yolo` command:
+
+```bash
+# Train a YOLO11n model using the coco8 dataset for 100 epochs
+yolo task=detect mode=train model=yolo11n.yaml data=coco8.yaml epochs=100
+```
+
+They may also be used directly in a Python environment, and accept the same [arguments](https://docs.ultralytics.com/usage/cfg/) as in the CLI example above:
+
+```python
+from ultralytics import YOLO
+
+# Initialize a YOLO11n model from a YAML configuration file
+model = YOLO("model.yaml")
+
+# If a pre-trained model is available, use it instead
+# model = YOLO("model.pt")
+
+# Display model information
+model.info()
+
+# Train the model using the COCO8 dataset for 100 epochs
+model.train(data="coco8.yaml", epochs=100)
+```
+
+## Pre-trained Model Architectures
+
+Ultralytics supports many model architectures. Visit [Ultralytics Models](https://docs.ultralytics.com/models/) to view detailed information and usage. Any of these models can be used by loading their configurations or pretrained checkpoints if available.
+
+## Contribute New Models
+
+Have you trained a new YOLO variant or achieved state-of-the-art performance with specific tuning? We'd love to showcase your work in our Models section! Contributions from the community in the form of new models, architectures, or optimizations are highly valued and can significantly enrich our repository.
+
+By contributing to this section, you're helping us offer a wider array of model choices and configurations to the community. It's a fantastic way to share your knowledge and expertise while making the Ultralytics YOLO ecosystem even more versatile.
+
+To get started, please consult our [Contributing Guide](https://docs.ultralytics.com/help/contributing/) for step-by-step instructions on how to submit a Pull Request (PR) 🛠️. Your contributions are eagerly awaited!
+
+Let's join hands to extend the range and capabilities of the Ultralytics YOLO models 🙏!

ultralytics/cfg/models/rt-detr/rtdetr-l.yaml

@@ -0,0 +1,53 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics RT-DETR-l hybrid object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/rtdetr
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# 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]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, HGStem, [32, 48]] # 0-P2/4
+  - [-1, 6, HGBlock, [48, 128, 3]] # stage 1
+
+  - [-1, 1, DWConv, [128, 3, 2, 1, False]] # 2-P3/8
+  - [-1, 6, HGBlock, [96, 512, 3]] # stage 2
+
+  - [-1, 1, DWConv, [512, 3, 2, 1, False]] # 4-P3/16
+  - [-1, 6, HGBlock, [192, 1024, 5, True, False]] # cm, c2, k, light, shortcut
+  - [-1, 6, HGBlock, [192, 1024, 5, True, True]]
+  - [-1, 6, HGBlock, [192, 1024, 5, True, True]] # stage 3
+
+  - [-1, 1, DWConv, [1024, 3, 2, 1, False]] # 8-P4/32
+  - [-1, 6, HGBlock, [384, 2048, 5, True, False]] # stage 4
+
+head:
+  - [-1, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 10 input_proj.2
+  - [-1, 1, AIFI, [1024, 8]]
+  - [-1, 1, Conv, [256, 1, 1]] # 12, Y5, lateral_convs.0
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [7, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 14 input_proj.1
+  - [[-2, -1], 1, Concat, [1]]
+  - [-1, 3, RepC3, [256]] # 16, fpn_blocks.0
+  - [-1, 1, Conv, [256, 1, 1]] # 17, Y4, lateral_convs.1
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [3, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 19 input_proj.0
+  - [[-2, -1], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, RepC3, [256]] # X3 (21), fpn_blocks.1
+
+  - [-1, 1, Conv, [256, 3, 2]] # 22, downsample_convs.0
+  - [[-1, 17], 1, Concat, [1]] # cat Y4
+  - [-1, 3, RepC3, [256]] # F4 (24), pan_blocks.0
+
+  - [-1, 1, Conv, [256, 3, 2]] # 25, downsample_convs.1
+  - [[-1, 12], 1, Concat, [1]] # cat Y5
+  - [-1, 3, RepC3, [256]] # F5 (27), pan_blocks.1
+
+  - [[21, 24, 27], 1, RTDETRDecoder, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/rt-detr/rtdetr-resnet101.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics RT-DETR-ResNet101 hybrid object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/rtdetr
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# 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]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, ResNetLayer, [3, 64, 1, True, 1]] # 0
+  - [-1, 1, ResNetLayer, [64, 64, 1, False, 3]] # 1
+  - [-1, 1, ResNetLayer, [256, 128, 2, False, 4]] # 2
+  - [-1, 1, ResNetLayer, [512, 256, 2, False, 23]] # 3
+  - [-1, 1, ResNetLayer, [1024, 512, 2, False, 3]] # 4
+
+head:
+  - [-1, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 5
+  - [-1, 1, AIFI, [1024, 8]]
+  - [-1, 1, Conv, [256, 1, 1]] # 7
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [3, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 9
+  - [[-2, -1], 1, Concat, [1]]
+  - [-1, 3, RepC3, [256]] # 11
+  - [-1, 1, Conv, [256, 1, 1]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [2, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 14
+  - [[-2, -1], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, RepC3, [256]] # X3 (16), fpn_blocks.1
+
+  - [-1, 1, Conv, [256, 3, 2]] # 17, downsample_convs.0
+  - [[-1, 12], 1, Concat, [1]] # cat Y4
+  - [-1, 3, RepC3, [256]] # F4 (19), pan_blocks.0
+
+  - [-1, 1, Conv, [256, 3, 2]] # 20, downsample_convs.1
+  - [[-1, 7], 1, Concat, [1]] # cat Y5
+  - [-1, 3, RepC3, [256]] # F5 (22), pan_blocks.1
+
+  - [[16, 19, 22], 1, RTDETRDecoder, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/rt-detr/rtdetr-resnet50.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics RT-DETR-ResNet50 hybrid object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/rtdetr
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# 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]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, ResNetLayer, [3, 64, 1, True, 1]] # 0
+  - [-1, 1, ResNetLayer, [64, 64, 1, False, 3]] # 1
+  - [-1, 1, ResNetLayer, [256, 128, 2, False, 4]] # 2
+  - [-1, 1, ResNetLayer, [512, 256, 2, False, 6]] # 3
+  - [-1, 1, ResNetLayer, [1024, 512, 2, False, 3]] # 4
+
+head:
+  - [-1, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 5
+  - [-1, 1, AIFI, [1024, 8]]
+  - [-1, 1, Conv, [256, 1, 1]] # 7
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [3, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 9
+  - [[-2, -1], 1, Concat, [1]]
+  - [-1, 3, RepC3, [256]] # 11
+  - [-1, 1, Conv, [256, 1, 1]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [2, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 14
+  - [[-2, -1], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, RepC3, [256]] # X3 (16), fpn_blocks.1
+
+  - [-1, 1, Conv, [256, 3, 2]] # 17, downsample_convs.0
+  - [[-1, 12], 1, Concat, [1]] # cat Y4
+  - [-1, 3, RepC3, [256]] # F4 (19), pan_blocks.0
+
+  - [-1, 1, Conv, [256, 3, 2]] # 20, downsample_convs.1
+  - [[-1, 7], 1, Concat, [1]] # cat Y5
+  - [-1, 3, RepC3, [256]] # F5 (22), pan_blocks.1
+
+  - [[16, 19, 22], 1, RTDETRDecoder, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/rt-detr/rtdetr-x.yaml

@@ -0,0 +1,57 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics RT-DETR-x hybrid object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/rtdetr
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# 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]
+  x: [1.00, 1.00, 2048]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, HGStem, [32, 64]] # 0-P2/4
+  - [-1, 6, HGBlock, [64, 128, 3]] # stage 1
+
+  - [-1, 1, DWConv, [128, 3, 2, 1, False]] # 2-P3/8
+  - [-1, 6, HGBlock, [128, 512, 3]]
+  - [-1, 6, HGBlock, [128, 512, 3, False, True]] # 4-stage 2
+
+  - [-1, 1, DWConv, [512, 3, 2, 1, False]] # 5-P3/16
+  - [-1, 6, HGBlock, [256, 1024, 5, True, False]] # cm, c2, k, light, shortcut
+  - [-1, 6, HGBlock, [256, 1024, 5, True, True]]
+  - [-1, 6, HGBlock, [256, 1024, 5, True, True]]
+  - [-1, 6, HGBlock, [256, 1024, 5, True, True]]
+  - [-1, 6, HGBlock, [256, 1024, 5, True, True]] # 10-stage 3
+
+  - [-1, 1, DWConv, [1024, 3, 2, 1, False]] # 11-P4/32
+  - [-1, 6, HGBlock, [512, 2048, 5, True, False]]
+  - [-1, 6, HGBlock, [512, 2048, 5, True, True]] # 13-stage 4
+
+head:
+  - [-1, 1, Conv, [384, 1, 1, None, 1, 1, False]] # 14 input_proj.2
+  - [-1, 1, AIFI, [2048, 8]]
+  - [-1, 1, Conv, [384, 1, 1]] # 16, Y5, lateral_convs.0
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [10, 1, Conv, [384, 1, 1, None, 1, 1, False]] # 18 input_proj.1
+  - [[-2, -1], 1, Concat, [1]]
+  - [-1, 3, RepC3, [384]] # 20, fpn_blocks.0
+  - [-1, 1, Conv, [384, 1, 1]] # 21, Y4, lateral_convs.1
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [4, 1, Conv, [384, 1, 1, None, 1, 1, False]] # 23 input_proj.0
+  - [[-2, -1], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, RepC3, [384]] # X3 (25), fpn_blocks.1
+
+  - [-1, 1, Conv, [384, 3, 2]] # 26, downsample_convs.0
+  - [[-1, 21], 1, Concat, [1]] # cat Y4
+  - [-1, 3, RepC3, [384]] # F4 (28), pan_blocks.0
+
+  - [-1, 1, Conv, [384, 3, 2]] # 29, downsample_convs.1
+  - [[-1, 16], 1, Concat, [1]] # cat Y5
+  - [-1, 3, RepC3, [384]] # F5 (31), pan_blocks.1
+
+  - [[25, 28, 31], 1, RTDETRDecoder, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v10/yolov10b.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv10b object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov10
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov10n.yaml' will call yolov10.yaml with scale 'n'
+  # [depth, width, max_channels]
+  b: [0.67, 1.00, 512]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, SCDown, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2fCIB, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 1, PSA, [1024]] # 10
+
+# YOLOv10.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2fCIB, [512, True]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2fCIB, [512, True]] # 19 (P4/16-medium)
+
+  - [-1, 1, SCDown, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fCIB, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v10/yolov10l.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv10l object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov10
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov10n.yaml' will call yolov10.yaml with scale 'n'
+  # [depth, width, max_channels]
+  l: [1.00, 1.00, 512]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, SCDown, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2fCIB, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 1, PSA, [1024]] # 10
+
+# YOLOv10.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2fCIB, [512, True]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2fCIB, [512, True]] # 19 (P4/16-medium)
+
+  - [-1, 1, SCDown, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fCIB, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v10/yolov10m.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv10m object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov10
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov10n.yaml' will call yolov10.yaml with scale 'n'
+  # [depth, width, max_channels]
+  m: [0.67, 0.75, 768]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, SCDown, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2fCIB, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 1, PSA, [1024]] # 10
+
+# YOLOv10.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2fCIB, [512, True]] # 19 (P4/16-medium)
+
+  - [-1, 1, SCDown, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fCIB, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v10/yolov10n.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv10n object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov10
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov10n.yaml' will call yolov10.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, SCDown, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 1, PSA, [1024]] # 10
+
+# YOLOv10.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 19 (P4/16-medium)
+
+  - [-1, 1, SCDown, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fCIB, [1024, True, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v10/yolov10s.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv10s object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov10
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov10n.yaml' will call yolov10.yaml with scale 'n'
+  # [depth, width, max_channels]
+  s: [0.33, 0.50, 1024]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, SCDown, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2fCIB, [1024, True, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 1, PSA, [1024]] # 10
+
+# YOLOv10.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 19 (P4/16-medium)
+
+  - [-1, 1, SCDown, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fCIB, [1024, True, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v10/yolov10x.yaml

@@ -0,0 +1,45 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv10x object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov10
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov10n.yaml' will call yolov10.yaml with scale 'n'
+  # [depth, width, max_channels]
+  x: [1.00, 1.25, 512]
+
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, SCDown, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2fCIB, [512, True]]
+  - [-1, 1, SCDown, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2fCIB, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+  - [-1, 1, PSA, [1024]] # 10
+
+# YOLOv10.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2fCIB, [512, True]] # 13
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 16 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 13], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2fCIB, [512, True]] # 19 (P4/16-medium)
+
+  - [-1, 1, SCDown, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fCIB, [1024, True]] # 22 (P5/32-large)
+
+  - [[16, 19, 22], 1, v10Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v12/yolov12.yaml

@@ -0,0 +1,47 @@
+# YOLOv12 🚀, AGPL-3.0 license
+# YOLOv12 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov12n.yaml' will call yolov12.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.50, 0.25, 1024] # summary: 411 layers, 2,538,872 parameters, 2,538,856 gradients, 6.7 GFLOPs
+  s: [0.50, 0.50, 1024] # summary: 411 layers, 8,986,272 parameters, 8,986,256 gradients, 22.0 GFLOPs
+  m: [0.50, 1.00, 512] # summary: 541 layers, 19,918,024 parameters, 19,918,008 gradients, 69.7 GFLOPs
+  l: [1.00, 1.00, 512] # summary: 917 layers, 28,329,872 parameters, 28,329,856 gradients, 97.2 GFLOPs
+  x: [1.00, 1.50, 512] # summary: 917 layers, 63,190,624 parameters, 63,190,608 gradients, 216.5 GFLOPs
+
+
+# YOLO12n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv,  [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv,  [128, 3, 2]] # 1-P2/4
+  - [-1, 2, C3k2,  [256, False, 0.25]]
+  - [-1, 1, Conv,  [256, 3, 2]] # 3-P3/8
+  - [-1, 2, C3k2,  [512, False, 0.25]]
+  - [-1, 1, Conv,  [512, 3, 2]] # 5-P4/16
+  - [-1, 4, A2C2f, [512, True, 4]]
+  - [-1, 1, Conv,  [1024, 3, 2]] # 7-P5/32
+  - [-1, 4, A2C2f, [1024, True, 1]]  # 8
+
+
+# YOLO12n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 2, A2C2f, [512, False, 4, True]] # 11
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 2, C3k2, [256, False]] # 14 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 11], 1, Concat, [1]] # cat head P4
+  - [-1, 2, A2C2f, [512, False, 4, True]] # 17 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 8], 1, Concat, [1]] # cat head P5
+  - [-1, 2, C3k2, [1024, True]] # 20 (P5/32-large)
+
+  - [[14, 17, 20], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v3/yolov3-spp.yaml

@@ -0,0 +1,49 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv3-SPP object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov3
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+depth_multiple: 1.0 # model depth multiple
+width_multiple: 1.0 # layer channel multiple
+
+# darknet53 backbone
+backbone:
+  # [from, number, module, args]
+  - [-1, 1, Conv, [32, 3, 1]] # 0
+  - [-1, 1, Conv, [64, 3, 2]] # 1-P1/2
+  - [-1, 1, Bottleneck, [64]]
+  - [-1, 1, Conv, [128, 3, 2]] # 3-P2/4
+  - [-1, 2, Bottleneck, [128]]
+  - [-1, 1, Conv, [256, 3, 2]] # 5-P3/8
+  - [-1, 8, Bottleneck, [256]]
+  - [-1, 1, Conv, [512, 3, 2]] # 7-P4/16
+  - [-1, 8, Bottleneck, [512]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 9-P5/32
+  - [-1, 4, Bottleneck, [1024]] # 10
+
+# YOLOv3-SPP head
+head:
+  - [-1, 1, Bottleneck, [1024, False]]
+  - [-1, 1, SPP, [512, [5, 9, 13]]]
+  - [-1, 1, Conv, [1024, 3, 1]]
+  - [-1, 1, Conv, [512, 1, 1]]
+  - [-1, 1, Conv, [1024, 3, 1]] # 15 (P5/32-large)
+
+  - [-2, 1, Conv, [256, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, Bottleneck, [512, False]]
+  - [-1, 1, Bottleneck, [512, False]]
+  - [-1, 1, Conv, [256, 1, 1]]
+  - [-1, 1, Conv, [512, 3, 1]] # 22 (P4/16-medium)
+
+  - [-2, 1, Conv, [128, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, Bottleneck, [256, False]]
+  - [-1, 2, Bottleneck, [256, False]] # 27 (P3/8-small)
+
+  - [[27, 22, 15], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v3/yolov3-tiny.yaml

@@ -0,0 +1,40 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv3-tiiny object detection model with P4/16 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov3
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+depth_multiple: 1.0 # model depth multiple
+width_multiple: 1.0 # layer channel multiple
+
+# YOLOv3-tiny backbone
+backbone:
+  # [from, number, module, args]
+  - [-1, 1, Conv, [16, 3, 1]] # 0
+  - [-1, 1, nn.MaxPool2d, [2, 2, 0]] # 1-P1/2
+  - [-1, 1, Conv, [32, 3, 1]]
+  - [-1, 1, nn.MaxPool2d, [2, 2, 0]] # 3-P2/4
+  - [-1, 1, Conv, [64, 3, 1]]
+  - [-1, 1, nn.MaxPool2d, [2, 2, 0]] # 5-P3/8
+  - [-1, 1, Conv, [128, 3, 1]]
+  - [-1, 1, nn.MaxPool2d, [2, 2, 0]] # 7-P4/16
+  - [-1, 1, Conv, [256, 3, 1]]
+  - [-1, 1, nn.MaxPool2d, [2, 2, 0]] # 9-P5/32
+  - [-1, 1, Conv, [512, 3, 1]]
+  - [-1, 1, nn.ZeroPad2d, [[0, 1, 0, 1]]] # 11
+  - [-1, 1, nn.MaxPool2d, [2, 1, 0]] # 12
+
+# YOLOv3-tiny head
+head:
+  - [-1, 1, Conv, [1024, 3, 1]]
+  - [-1, 1, Conv, [256, 1, 1]]
+  - [-1, 1, Conv, [512, 3, 1]] # 15 (P5/32-large)
+
+  - [-2, 1, Conv, [128, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, Conv, [256, 3, 1]] # 19 (P4/16-medium)
+
+  - [[19, 15], 1, Detect, [nc]] # Detect(P4, P5)

ultralytics/cfg/models/v3/yolov3.yaml

@@ -0,0 +1,49 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv3 object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov3
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+depth_multiple: 1.0 # model depth multiple
+width_multiple: 1.0 # layer channel multiple
+
+# darknet53 backbone
+backbone:
+  # [from, number, module, args]
+  - [-1, 1, Conv, [32, 3, 1]] # 0
+  - [-1, 1, Conv, [64, 3, 2]] # 1-P1/2
+  - [-1, 1, Bottleneck, [64]]
+  - [-1, 1, Conv, [128, 3, 2]] # 3-P2/4
+  - [-1, 2, Bottleneck, [128]]
+  - [-1, 1, Conv, [256, 3, 2]] # 5-P3/8
+  - [-1, 8, Bottleneck, [256]]
+  - [-1, 1, Conv, [512, 3, 2]] # 7-P4/16
+  - [-1, 8, Bottleneck, [512]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 9-P5/32
+  - [-1, 4, Bottleneck, [1024]] # 10
+
+# YOLOv3 head
+head:
+  - [-1, 1, Bottleneck, [1024, False]]
+  - [-1, 1, Conv, [512, 1, 1]]
+  - [-1, 1, Conv, [1024, 3, 1]]
+  - [-1, 1, Conv, [512, 1, 1]]
+  - [-1, 1, Conv, [1024, 3, 1]] # 15 (P5/32-large)
+
+  - [-2, 1, Conv, [256, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, Bottleneck, [512, False]]
+  - [-1, 1, Bottleneck, [512, False]]
+  - [-1, 1, Conv, [256, 1, 1]]
+  - [-1, 1, Conv, [512, 3, 1]] # 22 (P4/16-medium)
+
+  - [-2, 1, Conv, [128, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, Bottleneck, [256, False]]
+  - [-1, 2, Bottleneck, [256, False]] # 27 (P3/8-small)
+
+  - [[27, 22, 15], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v5/yolov5-p6.yaml

@@ -0,0 +1,62 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv5 object detection model with P3/8 - P6/64 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov5
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov5n-p6.yaml' will call yolov5-p6.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 1024]
+  l: [1.00, 1.00, 1024]
+  x: [1.33, 1.25, 1024]
+
+# YOLOv5 v6.0 backbone
+backbone:
+  # [from, number, module, args]
+  - [-1, 1, Conv, [64, 6, 2, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C3, [128]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C3, [256]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 9, C3, [512]]
+  - [-1, 1, Conv, [768, 3, 2]] # 7-P5/32
+  - [-1, 3, C3, [768]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 9-P6/64
+  - [-1, 3, C3, [1024]]
+  - [-1, 1, SPPF, [1024, 5]] # 11
+
+# YOLOv5 v6.0 head
+head:
+  - [-1, 1, Conv, [768, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P5
+  - [-1, 3, C3, [768, False]] # 15
+
+  - [-1, 1, Conv, [512, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C3, [512, False]] # 19
+
+  - [-1, 1, Conv, [256, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C3, [256, False]] # 23 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 20], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C3, [512, False]] # 26 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 16], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C3, [768, False]] # 29 (P5/32-large)
+
+  - [-1, 1, Conv, [768, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P6
+  - [-1, 3, C3, [1024, False]] # 32 (P6/64-xlarge)
+
+  - [[23, 26, 29, 32], 1, Detect, [nc]] # Detect(P3, P4, P5, P6)

ultralytics/cfg/models/v5/yolov5.yaml

@@ -0,0 +1,51 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv5 object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov5
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov5n.yaml' will call yolov5.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 1024]
+  l: [1.00, 1.00, 1024]
+  x: [1.33, 1.25, 1024]
+
+# YOLOv5 v6.0 backbone
+backbone:
+  # [from, number, module, args]
+  - [-1, 1, Conv, [64, 6, 2, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C3, [128]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C3, [256]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 9, C3, [512]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C3, [1024]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv5 v6.0 head
+head:
+  - [-1, 1, Conv, [512, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C3, [512, False]] # 13
+
+  - [-1, 1, Conv, [256, 1, 1]]
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C3, [256, False]] # 17 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 14], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C3, [512, False]] # 20 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C3, [1024, False]] # 23 (P5/32-large)
+
+  - [[17, 20, 23], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v6/yolov6.yaml

@@ -0,0 +1,56 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Meituan YOLOv6 object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov6
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+activation: nn.ReLU() # (optional) model default activation function
+scales: # model compound scaling constants, i.e. 'model=yolov6n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 768]
+  l: [1.00, 1.00, 512]
+  x: [1.00, 1.25, 512]
+
+# YOLOv6-3.0s backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 6, Conv, [128, 3, 1]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 12, Conv, [256, 3, 1]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 18, Conv, [512, 3, 1]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 6, Conv, [1024, 3, 1]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv6-3.0s head
+head:
+  - [-1, 1, Conv, [256, 1, 1]]
+  - [-1, 1, nn.ConvTranspose2d, [256, 2, 2, 0]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, Conv, [256, 3, 1]]
+  - [-1, 9, Conv, [256, 3, 1]] # 14
+
+  - [-1, 1, Conv, [128, 1, 1]]
+  - [-1, 1, nn.ConvTranspose2d, [128, 2, 2, 0]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, Conv, [128, 3, 1]]
+  - [-1, 9, Conv, [128, 3, 1]] # 19
+
+  - [-1, 1, Conv, [128, 3, 2]]
+  - [[-1, 15], 1, Concat, [1]] # cat head P4
+  - [-1, 1, Conv, [256, 3, 1]]
+  - [-1, 9, Conv, [256, 3, 1]] # 23
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 10], 1, Concat, [1]] # cat head P5
+  - [-1, 1, Conv, [512, 3, 1]]
+  - [-1, 9, Conv, [512, 3, 1]] # 27
+
+  - [[19, 23, 27], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-cls-resnet101.yaml

@@ -0,0 +1,28 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-cls image classification model with ResNet101 backbone
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/classify
+
+# Parameters
+nc: 1000 # 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]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 1024]
+  l: [1.00, 1.00, 1024]
+  x: [1.00, 1.25, 1024]
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, ResNetLayer, [3, 64, 1, True, 1]] # 0-P1/2
+  - [-1, 1, ResNetLayer, [64, 64, 1, False, 3]] # 1-P2/4
+  - [-1, 1, ResNetLayer, [256, 128, 2, False, 4]] # 2-P3/8
+  - [-1, 1, ResNetLayer, [512, 256, 2, False, 23]] # 3-P4/16
+  - [-1, 1, ResNetLayer, [1024, 512, 2, False, 3]] # 4-P5/32
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, Classify, [nc]] # Classify

ultralytics/cfg/models/v8/yolov8-cls-resnet50.yaml

@@ -0,0 +1,28 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-cls image classification model with ResNet50 backbone
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/classify
+
+# Parameters
+nc: 1000 # 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]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 1024]
+  l: [1.00, 1.00, 1024]
+  x: [1.00, 1.25, 1024]
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, ResNetLayer, [3, 64, 1, True, 1]] # 0-P1/2
+  - [-1, 1, ResNetLayer, [64, 64, 1, False, 3]] # 1-P2/4
+  - [-1, 1, ResNetLayer, [256, 128, 2, False, 4]] # 2-P3/8
+  - [-1, 1, ResNetLayer, [512, 256, 2, False, 6]] # 3-P4/16
+  - [-1, 1, ResNetLayer, [1024, 512, 2, False, 3]] # 4-P5/32
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, Classify, [nc]] # Classify

ultralytics/cfg/models/v8/yolov8-cls.yaml

@@ -0,0 +1,32 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-cls image classification model with YOLO backbone
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/classify
+
+# Parameters
+nc: 1000 # 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]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 1024]
+  l: [1.00, 1.00, 1024]
+  x: [1.00, 1.25, 1024]
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, Classify, [nc]] # Classify

ultralytics/cfg/models/v8/yolov8-ghost-p2.yaml

@@ -0,0 +1,58 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8 object detection model with P2/4 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# Employs Ghost convolutions and modules proposed in Huawei's GhostNet in https://arxiv.org/abs/1911.11907v2
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n-ghost-p2 summary: 491 layers, 2033944 parameters,   2033928 gradients,  13.8 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s-ghost-p2 summary: 491 layers, 5562080 parameters,   5562064 gradients,  25.1 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m-ghost-p2 summary: 731 layers, 9031728 parameters,   9031712 gradients,  42.8 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l-ghost-p2 summary: 971 layers, 12214448 parameters, 12214432 gradients,  69.1 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x-ghost-p2 summary: 971 layers, 18664776 parameters, 18664760 gradients, 103.3 GFLOPs
+
+# YOLOv8.0-ghost backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, GhostConv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C3Ghost, [128, True]]
+  - [-1, 1, GhostConv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C3Ghost, [256, True]]
+  - [-1, 1, GhostConv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C3Ghost, [512, True]]
+  - [-1, 1, GhostConv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C3Ghost, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0-ghost-p2 head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C3Ghost, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C3Ghost, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 2], 1, Concat, [1]] # cat backbone P2
+  - [-1, 3, C3Ghost, [128]] # 18 (P2/4-xsmall)
+
+  - [-1, 1, GhostConv, [128, 3, 2]]
+  - [[-1, 15], 1, Concat, [1]] # cat head P3
+  - [-1, 3, C3Ghost, [256]] # 21 (P3/8-small)
+
+  - [-1, 1, GhostConv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C3Ghost, [512]] # 24 (P4/16-medium)
+
+  - [-1, 1, GhostConv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C3Ghost, [1024]] # 27 (P5/32-large)
+
+  - [[18, 21, 24, 27], 1, Detect, [nc]] # Detect(P2, P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-ghost-p6.yaml

@@ -0,0 +1,60 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8 object detection model with P3/8 - P6/64 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# Employs Ghost convolutions and modules proposed in Huawei's GhostNet in https://arxiv.org/abs/1911.11907v2
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n-p6.yaml' will call yolov8-p6.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n-ghost-p6 summary: 529 layers, 2901100 parameters, 2901084 gradients, 5.8 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s-ghost-p6 summary: 529 layers, 9520008 parameters, 9519992 gradients, 16.4 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m-ghost-p6 summary: 789 layers, 18002904 parameters, 18002888 gradients, 34.4 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l-ghost-p6 summary: 1049 layers, 21227584 parameters, 21227568 gradients, 55.3 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x-ghost-p6 summary: 1049 layers, 33057852 parameters, 33057836 gradients, 85.7 GFLOPs
+
+# YOLOv8.0-ghost backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, GhostConv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C3Ghost, [128, True]]
+  - [-1, 1, GhostConv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C3Ghost, [256, True]]
+  - [-1, 1, GhostConv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C3Ghost, [512, True]]
+  - [-1, 1, GhostConv, [768, 3, 2]] # 7-P5/32
+  - [-1, 3, C3Ghost, [768, True]]
+  - [-1, 1, GhostConv, [1024, 3, 2]] # 9-P6/64
+  - [-1, 3, C3Ghost, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 11
+
+# YOLOv8.0-ghost-p6 head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P5
+  - [-1, 3, C3Ghost, [768]] # 14
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C3Ghost, [512]] # 17
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C3Ghost, [256]] # 20 (P3/8-small)
+
+  - [-1, 1, GhostConv, [256, 3, 2]]
+  - [[-1, 17], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C3Ghost, [512]] # 23 (P4/16-medium)
+
+  - [-1, 1, GhostConv, [512, 3, 2]]
+  - [[-1, 14], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C3Ghost, [768]] # 26 (P5/32-large)
+
+  - [-1, 1, GhostConv, [768, 3, 2]]
+  - [[-1, 11], 1, Concat, [1]] # cat head P6
+  - [-1, 3, C3Ghost, [1024]] # 29 (P6/64-xlarge)
+
+  - [[20, 23, 26, 29], 1, Detect, [nc]] # Detect(P3, P4, P5, P6)

ultralytics/cfg/models/v8/yolov8-ghost.yaml

@@ -0,0 +1,50 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8 object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# Employs Ghost convolutions and modules proposed in Huawei's GhostNet in https://arxiv.org/abs/1911.11907v2
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n-ghost summary: 403 layers,  1865316 parameters,  1865300 gradients,   5.8 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s-ghost summary: 403 layers,  5960072 parameters,  5960056 gradients,  16.4 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m-ghost summary: 603 layers, 10336312 parameters, 10336296 gradients,  32.7 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l-ghost summary: 803 layers, 14277872 parameters, 14277856 gradients,  53.7 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x-ghost summary: 803 layers, 22229308 parameters, 22229292 gradients,  83.3 GFLOPs
+
+# YOLOv8.0n-ghost backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, GhostConv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C3Ghost, [128, True]]
+  - [-1, 1, GhostConv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C3Ghost, [256, True]]
+  - [-1, 1, GhostConv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C3Ghost, [512, True]]
+  - [-1, 1, GhostConv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C3Ghost, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C3Ghost, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C3Ghost, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, GhostConv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C3Ghost, [512]] # 18 (P4/16-medium)
+
+  - [-1, 1, GhostConv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C3Ghost, [1024]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-obb.yaml

@@ -0,0 +1,49 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-obb Oriented Bounding Boxes (OBB) model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/obb
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n summary: 225 layers,  3157200 parameters,  3157184 gradients,   8.9 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s summary: 225 layers, 11166560 parameters, 11166544 gradients,  28.8 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients,  79.3 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l summary: 365 layers, 43691520 parameters, 43691504 gradients, 165.7 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x summary: 365 layers, 68229648 parameters, 68229632 gradients, 258.5 GFLOPs
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 18 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2f, [1024]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, OBB, [nc, 1]] # OBB(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-p2.yaml

@@ -0,0 +1,57 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8 object detection model with P2/4 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 768]
+  l: [1.00, 1.00, 512]
+  x: [1.00, 1.25, 512]
+
+# YOLOv8.0 backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0-p2 head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 2], 1, Concat, [1]] # cat backbone P2
+  - [-1, 3, C2f, [128]] # 18 (P2/4-xsmall)
+
+  - [-1, 1, Conv, [128, 3, 2]]
+  - [[-1, 15], 1, Concat, [1]] # cat head P3
+  - [-1, 3, C2f, [256]] # 21 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 24 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2f, [1024]] # 27 (P5/32-large)
+
+  - [[18, 21, 24, 27], 1, Detect, [nc]] # Detect(P2, P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-p6.yaml

@@ -0,0 +1,59 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8 object detection model with P3/8 - P6/64 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n-p6.yaml' will call yolov8-p6.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n-p6 summary (fused): 220 layers, 4976656 parameters, 42560 gradients, 8.7 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s-p6 summary (fused): 220 layers, 17897168 parameters, 57920 gradients, 28.5 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m-p6 summary (fused): 285 layers, 44862352 parameters, 78400 gradients, 83.1 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l-p6 summary (fused): 350 layers, 62351440 parameters, 98880 gradients, 167.3 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x-p6 summary (fused): 350 layers, 97382352 parameters, 123456 gradients, 261.1 GFLOPs
+
+# YOLOv8.0x6 backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [768, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [768, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 9-P6/64
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 11
+
+# YOLOv8.0x6 head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P5
+  - [-1, 3, C2, [768, False]] # 14
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2, [512, False]] # 17
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2, [256, False]] # 20 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 17], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2, [512, False]] # 23 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 14], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2, [768, False]] # 26 (P5/32-large)
+
+  - [-1, 1, Conv, [768, 3, 2]]
+  - [[-1, 11], 1, Concat, [1]] # cat head P6
+  - [-1, 3, C2, [1024, False]] # 29 (P6/64-xlarge)
+
+  - [[20, 23, 26, 29], 1, Detect, [nc]] # Detect(P3, P4, P5, P6)

ultralytics/cfg/models/v8/yolov8-pose-p6.yaml

@@ -0,0 +1,60 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-pose keypoints/pose estimation model with P3/8 - P6/64 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/pose
+
+# Parameters
+nc: 1 # number of classes
+kpt_shape: [17, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+scales: # model compound scaling constants, i.e. 'model=yolov8n-p6.yaml' will call yolov8-p6.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 768]
+  l: [1.00, 1.00, 512]
+  x: [1.00, 1.25, 512]
+
+# YOLOv8.0x6 backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [768, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [768, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 9-P6/64
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 11
+
+# YOLOv8.0x6 head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P5
+  - [-1, 3, C2, [768, False]] # 14
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2, [512, False]] # 17
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2, [256, False]] # 20 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 17], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2, [512, False]] # 23 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 14], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2, [768, False]] # 26 (P5/32-large)
+
+  - [-1, 1, Conv, [768, 3, 2]]
+  - [[-1, 11], 1, Concat, [1]] # cat head P6
+  - [-1, 3, C2, [1024, False]] # 29 (P6/64-xlarge)
+
+  - [[20, 23, 26, 29], 1, Pose, [nc, kpt_shape]] # Pose(P3, P4, P5, P6)

ultralytics/cfg/models/v8/yolov8-pose.yaml

@@ -0,0 +1,50 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-pose keypoints/pose estimation model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/pose
+
+# Parameters
+nc: 1 # number of classes
+kpt_shape: [17, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
+scales: # model compound scaling constants, i.e. 'model=yolov8n-pose.yaml' will call yolov8-pose.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 768]
+  l: [1.00, 1.00, 512]
+  x: [1.00, 1.25, 512]
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 18 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2f, [1024]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Pose, [nc, kpt_shape]] # Pose(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-rtdetr.yaml

@@ -0,0 +1,49 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-RTDETR hybrid object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/rtdetr
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n summary: 225 layers,  3157200 parameters,  3157184 gradients,   8.9 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s summary: 225 layers, 11166560 parameters, 11166544 gradients,  28.8 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients,  79.3 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l summary: 365 layers, 43691520 parameters, 43691504 gradients, 165.7 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x summary: 365 layers, 68229648 parameters, 68229632 gradients, 258.5 GFLOPs
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 18 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2f, [1024]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, RTDETRDecoder, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-seg-p6.yaml

@@ -0,0 +1,59 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-seg instance segmentation model with P3/8 - P6/64 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/segment
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n-seg-p6.yaml' will call yolov8-seg-p6.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 768]
+  l: [1.00, 1.00, 512]
+  x: [1.00, 1.25, 512]
+
+# YOLOv8.0x6 backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [768, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [768, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 9-P6/64
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 11
+
+# YOLOv8.0x6 head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 8], 1, Concat, [1]] # cat backbone P5
+  - [-1, 3, C2, [768, False]] # 14
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2, [512, False]] # 17
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2, [256, False]] # 20 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 17], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2, [512, False]] # 23 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 14], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2, [768, False]] # 26 (P5/32-large)
+
+  - [-1, 1, Conv, [768, 3, 2]]
+  - [[-1, 11], 1, Concat, [1]] # cat head P6
+  - [-1, 3, C2, [1024, False]] # 29 (P6/64-xlarge)
+
+  - [[20, 23, 26, 29], 1, Segment, [nc, 32, 256]] # Pose(P3, P4, P5, P6)

ultralytics/cfg/models/v8/yolov8-seg.yaml

@@ -0,0 +1,49 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-seg instance segmentation model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/segment
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n-seg.yaml' will call yolov8-seg.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024]
+  s: [0.33, 0.50, 1024]
+  m: [0.67, 0.75, 768]
+  l: [1.00, 1.00, 512]
+  x: [1.00, 1.25, 512]
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 18 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2f, [1024]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Segment, [nc, 32, 256]] # Segment(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-world.yaml

@@ -0,0 +1,51 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-World hybrid object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolo-world
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n summary: 225 layers,  3157200 parameters,  3157184 gradients,   8.9 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s summary: 225 layers, 11166560 parameters, 11166544 gradients,  28.8 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients,  79.3 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l summary: 365 layers, 43691520 parameters, 43691504 gradients, 165.7 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x summary: 365 layers, 68229648 parameters, 68229632 gradients, 258.5 GFLOPs
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2fAttn, [512, 256, 8]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2fAttn, [256, 128, 4]] # 15 (P3/8-small)
+
+  - [[15, 12, 9], 1, ImagePoolingAttn, [256]] # 16 (P3/8-small)
+
+  - [15, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2fAttn, [512, 256, 8]] # 19 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fAttn, [1024, 512, 16]] # 22 (P5/32-large)
+
+  - [[15, 19, 22], 1, WorldDetect, [nc, 512, False]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8-worldv2.yaml

@@ -0,0 +1,49 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8-Worldv2 hybrid object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolo-world
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n summary: 225 layers,  3157200 parameters,  3157184 gradients,   8.9 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s summary: 225 layers, 11166560 parameters, 11166544 gradients,  28.8 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients,  79.3 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l summary: 365 layers, 43691520 parameters, 43691504 gradients, 165.7 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x summary: 365 layers, 68229648 parameters, 68229632 gradients, 258.5 GFLOPs
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2fAttn, [512, 256, 8]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2fAttn, [256, 128, 4]] # 15 (P3/8-small)
+
+  - [15, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2fAttn, [512, 256, 8]] # 18 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2fAttn, [1024, 512, 16]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, WorldDetect, [nc, 512, True]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v8/yolov8.yaml

@@ -0,0 +1,49 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Ultralytics YOLOv8 object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov8
+# Task docs: https://docs.ultralytics.com/tasks/detect
+
+# Parameters
+nc: 80 # number of classes
+scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'
+  # [depth, width, max_channels]
+  n: [0.33, 0.25, 1024] # YOLOv8n summary: 225 layers,  3157200 parameters,  3157184 gradients,   8.9 GFLOPs
+  s: [0.33, 0.50, 1024] # YOLOv8s summary: 225 layers, 11166560 parameters, 11166544 gradients,  28.8 GFLOPs
+  m: [0.67, 0.75, 768] # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients,  79.3 GFLOPs
+  l: [1.00, 1.00, 512] # YOLOv8l summary: 365 layers, 43691520 parameters, 43691504 gradients, 165.7 GFLOPs
+  x: [1.00, 1.25, 512] # YOLOv8x summary: 365 layers, 68229648 parameters, 68229632 gradients, 258.5 GFLOPs
+
+# YOLOv8.0n backbone
+backbone:
+  # [from, repeats, module, args]
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 3, C2f, [128, True]]
+  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
+  - [-1, 6, C2f, [256, True]]
+  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
+  - [-1, 6, C2f, [512, True]]
+  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
+  - [-1, 3, C2f, [1024, True]]
+  - [-1, 1, SPPF, [1024, 5]] # 9
+
+# YOLOv8.0n head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 3, C2f, [512]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 3, C2f, [256]] # 15 (P3/8-small)
+
+  - [-1, 1, Conv, [256, 3, 2]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 3, C2f, [512]] # 18 (P4/16-medium)
+
+  - [-1, 1, Conv, [512, 3, 2]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 3, C2f, [1024]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v9/yolov9c-seg.yaml

@@ -0,0 +1,41 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv9c-seg instance segmentation model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov9
+# Task docs: https://docs.ultralytics.com/tasks/segment
+# 654 layers, 27897120 parameters, 159.4 GFLOPs
+
+# Parameters
+nc: 80 # number of classes
+
+# GELAN backbone
+backbone:
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 1, RepNCSPELAN4, [256, 128, 64, 1]] # 2
+  - [-1, 1, ADown, [256]] # 3-P3/8
+  - [-1, 1, RepNCSPELAN4, [512, 256, 128, 1]] # 4
+  - [-1, 1, ADown, [512]] # 5-P4/16
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 6
+  - [-1, 1, ADown, [512]] # 7-P5/32
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 8
+  - [-1, 1, SPPELAN, [512, 256]] # 9
+
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, RepNCSPELAN4, [256, 256, 128, 1]] # 15 (P3/8-small)
+
+  - [-1, 1, ADown, [256]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 18 (P4/16-medium)
+
+  - [-1, 1, ADown, [512]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Segment, [nc, 32, 256]] # Segment(P3, P4, P5)

ultralytics/cfg/models/v9/yolov9c.yaml

@@ -0,0 +1,41 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv9c object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov9
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# 618 layers, 25590912 parameters, 104.0 GFLOPs
+
+# Parameters
+nc: 80 # number of classes
+
+# GELAN backbone
+backbone:
+  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
+  - [-1, 1, RepNCSPELAN4, [256, 128, 64, 1]] # 2
+  - [-1, 1, ADown, [256]] # 3-P3/8
+  - [-1, 1, RepNCSPELAN4, [512, 256, 128, 1]] # 4
+  - [-1, 1, ADown, [512]] # 5-P4/16
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 6
+  - [-1, 1, ADown, [512]] # 7-P5/32
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 8
+  - [-1, 1, SPPELAN, [512, 256]] # 9
+
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, RepNCSPELAN4, [256, 256, 128, 1]] # 15 (P3/8-small)
+
+  - [-1, 1, ADown, [256]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 18 (P4/16-medium)
+
+  - [-1, 1, ADown, [512]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 1]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v9/yolov9e-seg.yaml

@@ -0,0 +1,64 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv9e-seg instance segmentation model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov9
+# Task docs: https://docs.ultralytics.com/tasks/segment
+# 1261 layers, 60512800 parameters, 248.4 GFLOPs
+
+# Parameters
+nc: 80 # number of classes
+
+# GELAN backbone
+backbone:
+  - [-1, 1, nn.Identity, []]
+  - [-1, 1, Conv, [64, 3, 2]] # 1-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 2-P2/4
+  - [-1, 1, RepNCSPELAN4, [256, 128, 64, 2]] # 3
+  - [-1, 1, ADown, [256]] # 4-P3/8
+  - [-1, 1, RepNCSPELAN4, [512, 256, 128, 2]] # 5
+  - [-1, 1, ADown, [512]] # 6-P4/16
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 7
+  - [-1, 1, ADown, [1024]] # 8-P5/32
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 9
+
+  - [1, 1, CBLinear, [[64]]] # 10
+  - [3, 1, CBLinear, [[64, 128]]] # 11
+  - [5, 1, CBLinear, [[64, 128, 256]]] # 12
+  - [7, 1, CBLinear, [[64, 128, 256, 512]]] # 13
+  - [9, 1, CBLinear, [[64, 128, 256, 512, 1024]]] # 14
+
+  - [0, 1, Conv, [64, 3, 2]] # 15-P1/2
+  - [[10, 11, 12, 13, 14, -1], 1, CBFuse, [[0, 0, 0, 0, 0]]] # 16
+  - [-1, 1, Conv, [128, 3, 2]] # 17-P2/4
+  - [[11, 12, 13, 14, -1], 1, CBFuse, [[1, 1, 1, 1]]] # 18
+  - [-1, 1, RepNCSPELAN4, [256, 128, 64, 2]] # 19
+  - [-1, 1, ADown, [256]] # 20-P3/8
+  - [[12, 13, 14, -1], 1, CBFuse, [[2, 2, 2]]] # 21
+  - [-1, 1, RepNCSPELAN4, [512, 256, 128, 2]] # 22
+  - [-1, 1, ADown, [512]] # 23-P4/16
+  - [[13, 14, -1], 1, CBFuse, [[3, 3]]] # 24
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 25
+  - [-1, 1, ADown, [1024]] # 26-P5/32
+  - [[14, -1], 1, CBFuse, [[4]]] # 27
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 28
+  - [-1, 1, SPPELAN, [512, 256]] # 29
+
+# GELAN head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 25], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 2]] # 32
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 22], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, RepNCSPELAN4, [256, 256, 128, 2]] # 35 (P3/8-small)
+
+  - [-1, 1, ADown, [256]]
+  - [[-1, 32], 1, Concat, [1]] # cat head P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 2]] # 38 (P4/16-medium)
+
+  - [-1, 1, ADown, [512]]
+  - [[-1, 29], 1, Concat, [1]] # cat head P5
+  - [-1, 1, RepNCSPELAN4, [512, 1024, 512, 2]] # 41 (P5/32-large)
+
+  - [[35, 38, 41], 1, Segment, [nc, 32, 256]] # Segment (P3, P4, P5)

ultralytics/cfg/models/v9/yolov9e.yaml

@@ -0,0 +1,64 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv9e object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov9
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# 1225 layers, 58206592 parameters, 193.0 GFLOPs
+
+# Parameters
+nc: 80 # number of classes
+
+# GELAN backbone
+backbone:
+  - [-1, 1, nn.Identity, []]
+  - [-1, 1, Conv, [64, 3, 2]] # 1-P1/2
+  - [-1, 1, Conv, [128, 3, 2]] # 2-P2/4
+  - [-1, 1, RepNCSPELAN4, [256, 128, 64, 2]] # 3
+  - [-1, 1, ADown, [256]] # 4-P3/8
+  - [-1, 1, RepNCSPELAN4, [512, 256, 128, 2]] # 5
+  - [-1, 1, ADown, [512]] # 6-P4/16
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 7
+  - [-1, 1, ADown, [1024]] # 8-P5/32
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 9
+
+  - [1, 1, CBLinear, [[64]]] # 10
+  - [3, 1, CBLinear, [[64, 128]]] # 11
+  - [5, 1, CBLinear, [[64, 128, 256]]] # 12
+  - [7, 1, CBLinear, [[64, 128, 256, 512]]] # 13
+  - [9, 1, CBLinear, [[64, 128, 256, 512, 1024]]] # 14
+
+  - [0, 1, Conv, [64, 3, 2]] # 15-P1/2
+  - [[10, 11, 12, 13, 14, -1], 1, CBFuse, [[0, 0, 0, 0, 0]]] # 16
+  - [-1, 1, Conv, [128, 3, 2]] # 17-P2/4
+  - [[11, 12, 13, 14, -1], 1, CBFuse, [[1, 1, 1, 1]]] # 18
+  - [-1, 1, RepNCSPELAN4, [256, 128, 64, 2]] # 19
+  - [-1, 1, ADown, [256]] # 20-P3/8
+  - [[12, 13, 14, -1], 1, CBFuse, [[2, 2, 2]]] # 21
+  - [-1, 1, RepNCSPELAN4, [512, 256, 128, 2]] # 22
+  - [-1, 1, ADown, [512]] # 23-P4/16
+  - [[13, 14, -1], 1, CBFuse, [[3, 3]]] # 24
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 25
+  - [-1, 1, ADown, [1024]] # 26-P5/32
+  - [[14, -1], 1, CBFuse, [[4]]] # 27
+  - [-1, 1, RepNCSPELAN4, [1024, 512, 256, 2]] # 28
+  - [-1, 1, SPPELAN, [512, 256]] # 29
+
+# GELAN head
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 25], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 2]] # 32
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 22], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, RepNCSPELAN4, [256, 256, 128, 2]] # 35 (P3/8-small)
+
+  - [-1, 1, ADown, [256]]
+  - [[-1, 32], 1, Concat, [1]] # cat head P4
+  - [-1, 1, RepNCSPELAN4, [512, 512, 256, 2]] # 38 (P4/16-medium)
+
+  - [-1, 1, ADown, [512]]
+  - [[-1, 29], 1, Concat, [1]] # cat head P5
+  - [-1, 1, RepNCSPELAN4, [512, 1024, 512, 2]] # 41 (P5/32-large)
+
+  - [[35, 38, 41], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v9/yolov9m.yaml

@@ -0,0 +1,41 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv9m object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov9
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# 603 layers, 20216160 parameters, 77.9 GFLOPs
+
+# Parameters
+nc: 80 # number of classes
+
+# GELAN backbone
+backbone:
+  - [-1, 1, Conv, [32, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [64, 3, 2]] # 1-P2/4
+  - [-1, 1, RepNCSPELAN4, [128, 128, 64, 1]] # 2
+  - [-1, 1, AConv, [240]] # 3-P3/8
+  - [-1, 1, RepNCSPELAN4, [240, 240, 120, 1]] # 4
+  - [-1, 1, AConv, [360]] # 5-P4/16
+  - [-1, 1, RepNCSPELAN4, [360, 360, 180, 1]] # 6
+  - [-1, 1, AConv, [480]] # 7-P5/32
+  - [-1, 1, RepNCSPELAN4, [480, 480, 240, 1]] # 8
+  - [-1, 1, SPPELAN, [480, 240]] # 9
+
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, RepNCSPELAN4, [360, 360, 180, 1]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, RepNCSPELAN4, [240, 240, 120, 1]] # 15
+
+  - [-1, 1, AConv, [180]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 1, RepNCSPELAN4, [360, 360, 180, 1]] # 18 (P4/16-medium)
+
+  - [-1, 1, AConv, [240]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 1, RepNCSPELAN4, [480, 480, 240, 1]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/models/v9/yolov9s.yaml

@@ -0,0 +1,41 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv9s object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov9
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# 917 layers, 7318368 parameters, 27.6 GFLOPs
+
+# Parameters
+nc: 80 # number of classes
+
+# GELAN backbone
+backbone:
+  - [-1, 1, Conv, [32, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [64, 3, 2]] # 1-P2/4
+  - [-1, 1, ELAN1, [64, 64, 32]] # 2
+  - [-1, 1, AConv, [128]] # 3-P3/8
+  - [-1, 1, RepNCSPELAN4, [128, 128, 64, 3]] # 4
+  - [-1, 1, AConv, [192]] # 5-P4/16
+  - [-1, 1, RepNCSPELAN4, [192, 192, 96, 3]] # 6
+  - [-1, 1, AConv, [256]] # 7-P5/32
+  - [-1, 1, RepNCSPELAN4, [256, 256, 128, 3]] # 8
+  - [-1, 1, SPPELAN, [256, 128]] # 9
+
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, RepNCSPELAN4, [192, 192, 96, 3]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, RepNCSPELAN4, [128, 128, 64, 3]] # 15
+
+  - [-1, 1, AConv, [96]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 1, RepNCSPELAN4, [192, 192, 96, 3]] # 18 (P4/16-medium)
+
+  - [-1, 1, AConv, [128]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 1, RepNCSPELAN4, [256, 256, 128, 3]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Detect, [nc]] # Detect(P3, P4 P5)

ultralytics/cfg/models/v9/yolov9t.yaml

@@ -0,0 +1,41 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# YOLOv9t object detection model with P3/8 - P5/32 outputs
+# Model docs: https://docs.ultralytics.com/models/yolov9
+# Task docs: https://docs.ultralytics.com/tasks/detect
+# 917 layers, 2128720 parameters, 8.5 GFLOPs
+
+# Parameters
+nc: 80 # number of classes
+
+# GELAN backbone
+backbone:
+  - [-1, 1, Conv, [16, 3, 2]] # 0-P1/2
+  - [-1, 1, Conv, [32, 3, 2]] # 1-P2/4
+  - [-1, 1, ELAN1, [32, 32, 16]] # 2
+  - [-1, 1, AConv, [64]] # 3-P3/8
+  - [-1, 1, RepNCSPELAN4, [64, 64, 32, 3]] # 4
+  - [-1, 1, AConv, [96]] # 5-P4/16
+  - [-1, 1, RepNCSPELAN4, [96, 96, 48, 3]] # 6
+  - [-1, 1, AConv, [128]] # 7-P5/32
+  - [-1, 1, RepNCSPELAN4, [128, 128, 64, 3]] # 8
+  - [-1, 1, SPPELAN, [128, 64]] # 9
+
+head:
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
+  - [-1, 1, RepNCSPELAN4, [96, 96, 48, 3]] # 12
+
+  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
+  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
+  - [-1, 1, RepNCSPELAN4, [64, 64, 32, 3]] # 15
+
+  - [-1, 1, AConv, [48]]
+  - [[-1, 12], 1, Concat, [1]] # cat head P4
+  - [-1, 1, RepNCSPELAN4, [96, 96, 48, 3]] # 18 (P4/16-medium)
+
+  - [-1, 1, AConv, [64]]
+  - [[-1, 9], 1, Concat, [1]] # cat head P5
+  - [-1, 1, RepNCSPELAN4, [128, 128, 64, 3]] # 21 (P5/32-large)
+
+  - [[15, 18, 21], 1, Detect, [nc]] # Detect(P3, P4, P5)

ultralytics/cfg/solutions/default.yaml

@@ -0,0 +1,24 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Global configuration YAML with settings and arguments for Ultralytics Solutions
+# For documentation see https://docs.ultralytics.com/solutions/
+
+# Object counting settings  --------------------------------------------------------------------------------------------
+region: # list[tuple[int, int]] object counting, queue or speed estimation region points.
+show_in: True # (bool) flag to display objects moving *into* the defined region
+show_out: True # (bool) flag to display objects moving *out of* the defined region
+
+# Heatmaps settings ----------------------------------------------------------------------------------------------------
+colormap: #  (int | str) colormap for heatmap, Only OPENCV supported colormaps can be used.
+
+# Workouts monitoring settings -----------------------------------------------------------------------------------------
+up_angle: 145.0 # (float) Workouts up_angle for counts, 145.0 is default value.
+down_angle: 90 # (float) Workouts down_angle for counts, 90 is default value. Y
+kpts: [6, 8, 10] # (list[int]) keypoints for workouts monitoring, i.e. for push-ups kpts have values of [6, 8, 10].
+
+# Analytics settings ---------------------------------------------------------------------------------------------------
+analytics_type: "line" # (str) analytics type i.e "line", "pie", "bar" or "area" charts.
+json_file: # (str) parking system regions file path.
+
+# Security alarm system settings ---------------------------------------------------------------------------------------
+records: 5 # (int) Total detections count to send an email about security

ultralytics/cfg/trackers/botsort.yaml

@@ -0,0 +1,21 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Default Ultralytics settings for BoT-SORT tracker when using mode="track"
+# For documentation and examples see https://docs.ultralytics.com/modes/track/
+# For BoT-SORT source code see https://github.com/NirAharon/BoT-SORT
+
+tracker_type: botsort # tracker type, ['botsort', 'bytetrack']
+track_high_thresh: 0.25 # threshold for the first association
+track_low_thresh: 0.1 # threshold for the second association
+new_track_thresh: 0.25 # threshold for init new track if the detection does not match any tracks
+track_buffer: 30 # buffer to calculate the time when to remove tracks
+match_thresh: 0.8 # threshold for matching tracks
+fuse_score: True # Whether to fuse confidence scores with the iou distances before matching
+# min_box_area: 10  # threshold for min box areas(for tracker evaluation, not used for now)
+
+# BoT-SORT settings
+gmc_method: sparseOptFlow # method of global motion compensation
+# ReID model related thresh (not supported yet)
+proximity_thresh: 0.5
+appearance_thresh: 0.25
+with_reid: False

ultralytics/cfg/trackers/bytetrack.yaml

@@ -0,0 +1,14 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+# Default Ultralytics settings for ByteTrack tracker when using mode="track"
+# For documentation and examples see https://docs.ultralytics.com/modes/track/
+# For ByteTrack source code see https://github.com/ifzhang/ByteTrack
+
+tracker_type: bytetrack # tracker type, ['botsort', 'bytetrack']
+track_high_thresh: 0.25 # threshold for the first association
+track_low_thresh: 0.1 # threshold for the second association
+new_track_thresh: 0.25 # threshold for init new track if the detection does not match any tracks
+track_buffer: 30 # buffer to calculate the time when to remove tracks
+match_thresh: 0.8 # threshold for matching tracks
+fuse_score: True # Whether to fuse confidence scores with the iou distances before matching
+# min_box_area: 10  # threshold for min box areas(for tracker evaluation, not used for now)

ultralytics/data/__init__.py

@@ -0,0 +1,26 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+from .base import BaseDataset
+from .build import build_dataloader, build_grounding, build_yolo_dataset, load_inference_source
+from .dataset import (
+    ClassificationDataset,
+    GroundingDataset,
+    SemanticDataset,
+    YOLOConcatDataset,
+    YOLODataset,
+    YOLOMultiModalDataset,
+)
+
+__all__ = (
+    "BaseDataset",
+    "ClassificationDataset",
+    "SemanticDataset",
+    "YOLODataset",
+    "YOLOMultiModalDataset",
+    "YOLOConcatDataset",
+    "GroundingDataset",
+    "build_yolo_dataset",
+    "build_grounding",
+    "build_dataloader",
+    "load_inference_source",
+)

ultralytics/data/annotator.py

@@ -0,0 +1,72 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+from pathlib import Path
+
+from ultralytics import SAM, YOLO
+
+
+def auto_annotate(
+    data,
+    det_model="yolo11x.pt",
+    sam_model="sam_b.pt",
+    device="",
+    conf=0.25,
+    iou=0.45,
+    imgsz=640,
+    max_det=300,
+    classes=None,
+    output_dir=None,
+):
+    """
+    Automatically annotates images using a YOLO object detection model and a SAM segmentation model.
+
+    This function processes images in a specified directory, detects objects using a YOLO model, and then generates
+    segmentation masks using a SAM model. The resulting annotations are saved as text files.
+
+    Args:
+        data (str): Path to a folder containing images to be annotated.
+        det_model (str): Path or name of the pre-trained YOLO detection model.
+        sam_model (str): Path or name of the pre-trained SAM segmentation model.
+        device (str): Device to run the models on (e.g., 'cpu', 'cuda', '0').
+        conf (float): Confidence threshold for detection model; default is 0.25.
+        iou (float): IoU threshold for filtering overlapping boxes in detection results; default is 0.45.
+        imgsz (int): Input image resize dimension; default is 640.
+        max_det (int): Limits detections per image to control outputs in dense scenes.
+        classes (list): Filters predictions to specified class IDs, returning only relevant detections.
+        output_dir (str | None): Directory to save the annotated results. If None, a default directory is created.
+
+    Examples:
+        >>> from ultralytics.data.annotator import auto_annotate
+        >>> auto_annotate(data="ultralytics/assets", det_model="yolo11n.pt", sam_model="mobile_sam.pt")
+
+    Notes:
+        - The function creates a new directory for output if not specified.
+        - Annotation results are saved as text files with the same names as the input images.
+        - Each line in the output text file represents a detected object with its class ID and segmentation points.
+    """
+    det_model = YOLO(det_model)
+    sam_model = SAM(sam_model)
+
+    data = Path(data)
+    if not output_dir:
+        output_dir = data.parent / f"{data.stem}_auto_annotate_labels"
+    Path(output_dir).mkdir(exist_ok=True, parents=True)
+
+    det_results = det_model(
+        data, stream=True, device=device, conf=conf, iou=iou, imgsz=imgsz, max_det=max_det, classes=classes
+    )
+
+    for result in det_results:
+        class_ids = result.boxes.cls.int().tolist()  # noqa
+        if len(class_ids):
+            boxes = result.boxes.xyxy  # Boxes object for bbox outputs
+            sam_results = sam_model(result.orig_img, bboxes=boxes, verbose=False, save=False, device=device)
+            segments = sam_results[0].masks.xyn  # noqa
+
+            with open(f"{Path(output_dir) / Path(result.path).stem}.txt", "w") as f:
+                for i in range(len(segments)):
+                    s = segments[i]
+                    if len(s) == 0:
+                        continue
+                    segment = map(str, segments[i].reshape(-1).tolist())
+                    f.write(f"{class_ids[i]} " + " ".join(segment) + "\n")

ultralytics/data/base.py

@@ -0,0 +1,346 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+import glob
+import math
+import os
+import random
+from copy import deepcopy
+from multiprocessing.pool import ThreadPool
+from pathlib import Path
+from typing import Optional
+
+import cv2
+import numpy as np
+import psutil
+from torch.utils.data import Dataset
+
+from ultralytics.data.utils import FORMATS_HELP_MSG, HELP_URL, IMG_FORMATS
+from ultralytics.utils import DEFAULT_CFG, LOCAL_RANK, LOGGER, NUM_THREADS, TQDM
+
+
+class BaseDataset(Dataset):
+    """
+    Base dataset class for loading and processing image data.
+
+    Args:
+        img_path (str): Path to the folder containing images.
+        imgsz (int, optional): Image size. Defaults to 640.
+        cache (bool, optional): Cache images to RAM or disk during training. Defaults to False.
+        augment (bool, optional): If True, data augmentation is applied. Defaults to True.
+        hyp (dict, optional): Hyperparameters to apply data augmentation. Defaults to None.
+        prefix (str, optional): Prefix to print in log messages. Defaults to ''.
+        rect (bool, optional): If True, rectangular training is used. Defaults to False.
+        batch_size (int, optional): Size of batches. Defaults to None.
+        stride (int, optional): Stride. Defaults to 32.
+        pad (float, optional): Padding. Defaults to 0.0.
+        single_cls (bool, optional): If True, single class training is used. Defaults to False.
+        classes (list): List of included classes. Default is None.
+        fraction (float): Fraction of dataset to utilize. Default is 1.0 (use all data).
+
+    Attributes:
+        im_files (list): List of image file paths.
+        labels (list): List of label data dictionaries.
+        ni (int): Number of images in the dataset.
+        ims (list): List of loaded images.
+        npy_files (list): List of numpy file paths.
+        transforms (callable): Image transformation function.
+    """
+
+    def __init__(
+        self,
+        img_path,
+        imgsz=640,
+        cache=False,
+        augment=True,
+        hyp=DEFAULT_CFG,
+        prefix="",
+        rect=False,
+        batch_size=16,
+        stride=32,
+        pad=0.5,
+        single_cls=False,
+        classes=None,
+        fraction=1.0,
+    ):
+        """Initialize BaseDataset with given configuration and options."""
+        super().__init__()
+        self.img_path = img_path
+        self.imgsz = imgsz
+        self.augment = augment
+        self.single_cls = single_cls
+        self.prefix = prefix
+        self.fraction = fraction
+        self.im_files = self.get_img_files(self.img_path)
+        self.labels = self.get_labels()
+        self.update_labels(include_class=classes)  # single_cls and include_class
+        self.ni = len(self.labels)  # number of images
+        self.rect = rect
+        self.batch_size = batch_size
+        self.stride = stride
+        self.pad = pad
+        if self.rect:
+            assert self.batch_size is not None
+            self.set_rectangle()
+
+        # Buffer thread for mosaic images
+        self.buffer = []  # buffer size = batch size
+        self.max_buffer_length = min((self.ni, self.batch_size * 8, 1000)) if self.augment else 0
+
+        # Cache images (options are cache = True, False, None, "ram", "disk")
+        self.ims, self.im_hw0, self.im_hw = [None] * self.ni, [None] * self.ni, [None] * self.ni
+        self.npy_files = [Path(f).with_suffix(".npy") for f in self.im_files]
+        self.cache = cache.lower() if isinstance(cache, str) else "ram" if cache is True else None
+        if self.cache == "ram" and self.check_cache_ram():
+            if hyp.deterministic:
+                LOGGER.warning(
+                    "WARNING ⚠️ cache='ram' may produce non-deterministic training results. "
+                    "Consider cache='disk' as a deterministic alternative if your disk space allows."
+                )
+            self.cache_images()
+        elif self.cache == "disk" and self.check_cache_disk():
+            self.cache_images()
+
+        # Transforms
+        self.transforms = self.build_transforms(hyp=hyp)
+
+    def get_img_files(self, img_path):
+        """Read image files."""
+        try:
+            f = []  # image files
+            for p in img_path if isinstance(img_path, list) else [img_path]:
+                p = Path(p)  # os-agnostic
+                if p.is_dir():  # dir
+                    f += glob.glob(str(p / "**" / "*.*"), recursive=True)
+                    # F = list(p.rglob('*.*'))  # pathlib
+                elif p.is_file():  # file
+                    with open(p) as t:
+                        t = t.read().strip().splitlines()
+                        parent = str(p.parent) + os.sep
+                        f += [x.replace("./", parent) if x.startswith("./") else x for x in t]  # local to global path
+                        # F += [p.parent / x.lstrip(os.sep) for x in t]  # local to global path (pathlib)
+                else:
+                    raise FileNotFoundError(f"{self.prefix}{p} does not exist")
+            im_files = sorted(x.replace("/", os.sep) for x in f if x.split(".")[-1].lower() in IMG_FORMATS)
+            # self.img_files = sorted([x for x in f if x.suffix[1:].lower() in IMG_FORMATS])  # pathlib
+            assert im_files, f"{self.prefix}No images found in {img_path}. {FORMATS_HELP_MSG}"
+        except Exception as e:
+            raise FileNotFoundError(f"{self.prefix}Error loading data from {img_path}\n{HELP_URL}") from e
+        if self.fraction < 1:
+            im_files = im_files[: round(len(im_files) * self.fraction)]  # retain a fraction of the dataset
+        return im_files
+
+    def update_labels(self, include_class: Optional[list]):
+        """Update labels to include only these classes (optional)."""
+        include_class_array = np.array(include_class).reshape(1, -1)
+        for i in range(len(self.labels)):
+            if include_class is not None:
+                cls = self.labels[i]["cls"]
+                bboxes = self.labels[i]["bboxes"]
+                segments = self.labels[i]["segments"]
+                keypoints = self.labels[i]["keypoints"]
+                j = (cls == include_class_array).any(1)
+                self.labels[i]["cls"] = cls[j]
+                self.labels[i]["bboxes"] = bboxes[j]
+                if segments:
+                    self.labels[i]["segments"] = [segments[si] for si, idx in enumerate(j) if idx]
+                if keypoints is not None:
+                    self.labels[i]["keypoints"] = keypoints[j]
+            if self.single_cls:
+                self.labels[i]["cls"][:, 0] = 0
+
+    def load_image(self, i, rect_mode=True):
+        """Loads 1 image from dataset index 'i', returns (im, resized hw)."""
+        im, f, fn = self.ims[i], self.im_files[i], self.npy_files[i]
+        if im is None:  # not cached in RAM
+            if fn.exists():  # load npy
+                try:
+                    im = np.load(fn)
+                except Exception as e:
+                    LOGGER.warning(f"{self.prefix}WARNING ⚠️ Removing corrupt *.npy image file {fn} due to: {e}")
+                    Path(fn).unlink(missing_ok=True)
+                    im = cv2.imread(f)  # BGR
+            else:  # read image
+                im = cv2.imread(f)  # BGR
+            if im is None:
+                raise FileNotFoundError(f"Image Not Found {f}")
+
+            h0, w0 = im.shape[:2]  # orig hw
+            if rect_mode:  # resize long side to imgsz while maintaining aspect ratio
+                r = self.imgsz / max(h0, w0)  # ratio
+                if r != 1:  # if sizes are not equal
+                    w, h = (min(math.ceil(w0 * r), self.imgsz), min(math.ceil(h0 * r), self.imgsz))
+                    im = cv2.resize(im, (w, h), interpolation=cv2.INTER_LINEAR)
+            elif not (h0 == w0 == self.imgsz):  # resize by stretching image to square imgsz
+                im = cv2.resize(im, (self.imgsz, self.imgsz), interpolation=cv2.INTER_LINEAR)
+
+            # Add to buffer if training with augmentations
+            if self.augment:
+                self.ims[i], self.im_hw0[i], self.im_hw[i] = im, (h0, w0), im.shape[:2]  # im, hw_original, hw_resized
+                self.buffer.append(i)
+                if 1 < len(self.buffer) >= self.max_buffer_length:  # prevent empty buffer
+                    j = self.buffer.pop(0)
+                    if self.cache != "ram":
+                        self.ims[j], self.im_hw0[j], self.im_hw[j] = None, None, None
+
+            return im, (h0, w0), im.shape[:2]
+
+        return self.ims[i], self.im_hw0[i], self.im_hw[i]
+
+    def cache_images(self):
+        """Cache images to memory or disk."""
+        b, gb = 0, 1 << 30  # bytes of cached images, bytes per gigabytes
+        fcn, storage = (self.cache_images_to_disk, "Disk") if self.cache == "disk" else (self.load_image, "RAM")
+        with ThreadPool(NUM_THREADS) as pool:
+            results = pool.imap(fcn, range(self.ni))
+            pbar = TQDM(enumerate(results), total=self.ni, disable=LOCAL_RANK > 0)
+            for i, x in pbar:
+                if self.cache == "disk":
+                    b += self.npy_files[i].stat().st_size
+                else:  # 'ram'
+                    self.ims[i], self.im_hw0[i], self.im_hw[i] = x  # im, hw_orig, hw_resized = load_image(self, i)
+                    b += self.ims[i].nbytes
+                pbar.desc = f"{self.prefix}Caching images ({b / gb:.1f}GB {storage})"
+            pbar.close()
+
+    def cache_images_to_disk(self, i):
+        """Saves an image as an *.npy file for faster loading."""
+        f = self.npy_files[i]
+        if not f.exists():
+            np.save(f.as_posix(), cv2.imread(self.im_files[i]), allow_pickle=False)
+
+    def check_cache_disk(self, safety_margin=0.5):
+        """Check image caching requirements vs available disk space."""
+        import shutil
+
+        b, gb = 0, 1 << 30  # bytes of cached images, bytes per gigabytes
+        n = min(self.ni, 30)  # extrapolate from 30 random images
+        for _ in range(n):
+            im_file = random.choice(self.im_files)
+            im = cv2.imread(im_file)
+            if im is None:
+                continue
+            b += im.nbytes
+            if not os.access(Path(im_file).parent, os.W_OK):
+                self.cache = None
+                LOGGER.info(f"{self.prefix}Skipping caching images to disk, directory not writeable ⚠️")
+                return False
+        disk_required = b * self.ni / n * (1 + safety_margin)  # bytes required to cache dataset to disk
+        total, used, free = shutil.disk_usage(Path(self.im_files[0]).parent)
+        if disk_required > free:
+            self.cache = None
+            LOGGER.info(
+                f"{self.prefix}{disk_required / gb:.1f}GB disk space required, "
+                f"with {int(safety_margin * 100)}% safety margin but only "
+                f"{free / gb:.1f}/{total / gb:.1f}GB free, not caching images to disk ⚠️"
+            )
+            return False
+        return True
+
+    def check_cache_ram(self, safety_margin=0.5):
+        """Check image caching requirements vs available memory."""
+        b, gb = 0, 1 << 30  # bytes of cached images, bytes per gigabytes
+        n = min(self.ni, 30)  # extrapolate from 30 random images
+        for _ in range(n):
+            im = cv2.imread(random.choice(self.im_files))  # sample image
+            if im is None:
+                continue
+            ratio = self.imgsz / max(im.shape[0], im.shape[1])  # max(h, w)  # ratio
+            b += im.nbytes * ratio**2
+        mem_required = b * self.ni / n * (1 + safety_margin)  # GB required to cache dataset into RAM
+        mem = psutil.virtual_memory()
+        if mem_required > mem.available:
+            self.cache = None
+            LOGGER.info(
+                f"{self.prefix}{mem_required / gb:.1f}GB RAM required to cache images "
+                f"with {int(safety_margin * 100)}% safety margin but only "
+                f"{mem.available / gb:.1f}/{mem.total / gb:.1f}GB available, not caching images ⚠️"
+            )
+            return False
+        return True
+
+    def set_rectangle(self):
+        """Sets the shape of bounding boxes for YOLO detections as rectangles."""
+        bi = np.floor(np.arange(self.ni) / self.batch_size).astype(int)  # batch index
+        nb = bi[-1] + 1  # number of batches
+
+        s = np.array([x.pop("shape") for x in self.labels])  # hw
+        ar = s[:, 0] / s[:, 1]  # aspect ratio
+        irect = ar.argsort()
+        self.im_files = [self.im_files[i] for i in irect]
+        self.labels = [self.labels[i] for i in irect]
+        ar = ar[irect]
+
+        # Set training image shapes
+        shapes = [[1, 1]] * nb
+        for i in range(nb):
+            ari = ar[bi == i]
+            mini, maxi = ari.min(), ari.max()
+            if maxi < 1:
+                shapes[i] = [maxi, 1]
+            elif mini > 1:
+                shapes[i] = [1, 1 / mini]
+
+        self.batch_shapes = np.ceil(np.array(shapes) * self.imgsz / self.stride + self.pad).astype(int) * self.stride
+        self.batch = bi  # batch index of image
+
+    def __getitem__(self, index):
+        """Returns transformed label information for given index."""
+        return self.transforms(self.get_image_and_label(index))
+
+    def get_image_and_label(self, index):
+        """Get and return label information from the dataset."""
+        label = deepcopy(self.labels[index])  # requires deepcopy() https://github.com/ultralytics/ultralytics/pull/1948
+        label.pop("shape", None)  # shape is for rect, remove it
+        label["img"], label["ori_shape"], label["resized_shape"] = self.load_image(index)
+        label["ratio_pad"] = (
+            label["resized_shape"][0] / label["ori_shape"][0],
+            label["resized_shape"][1] / label["ori_shape"][1],
+        )  # for evaluation
+        if self.rect:
+            label["rect_shape"] = self.batch_shapes[self.batch[index]]
+        return self.update_labels_info(label)
+
+    def __len__(self):
+        """Returns the length of the labels list for the dataset."""
+        return len(self.labels)
+
+    def update_labels_info(self, label):
+        """Custom your label format here."""
+        return label
+
+    def build_transforms(self, hyp=None):
+        """
+        Users can customize augmentations here.
+
+        Example:
+            ```python
+            if self.augment:
+                # Training transforms
+                return Compose([])
+            else:
+                # Val transforms
+                return Compose([])
+            ```
+        """
+        raise NotImplementedError
+
+    def get_labels(self):
+        """
+        Users can customize their own format here.
+
+        Note:
+            Ensure output is a dictionary with the following keys:
+            ```python
+            dict(
+                im_file=im_file,
+                shape=shape,  # format: (height, width)
+                cls=cls,
+                bboxes=bboxes,  # xywh
+                segments=segments,  # xy
+                keypoints=keypoints,  # xy
+                normalized=True,  # or False
+                bbox_format="xyxy",  # or xywh, ltwh
+            )
+            ```
+        """
+        raise NotImplementedError

ultralytics/data/build.py

@@ -0,0 +1,215 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+import os
+import random
+from pathlib import Path
+
+import numpy as np
+import torch
+from PIL import Image
+from torch.utils.data import dataloader, distributed
+
+from ultralytics.data.dataset import GroundingDataset, YOLODataset, YOLOMultiModalDataset
+from ultralytics.data.loaders import (
+    LOADERS,
+    LoadImagesAndVideos,
+    LoadPilAndNumpy,
+    LoadScreenshots,
+    LoadStreams,
+    LoadTensor,
+    SourceTypes,
+    autocast_list,
+)
+from ultralytics.data.utils import IMG_FORMATS, PIN_MEMORY, VID_FORMATS
+from ultralytics.utils import RANK, colorstr
+from ultralytics.utils.checks import check_file
+
+
+class InfiniteDataLoader(dataloader.DataLoader):
+    """
+    Dataloader that reuses workers.
+
+    Uses same syntax as vanilla DataLoader.
+    """
+
+    def __init__(self, *args, **kwargs):
+        """Dataloader that infinitely recycles workers, inherits from DataLoader."""
+        super().__init__(*args, **kwargs)
+        object.__setattr__(self, "batch_sampler", _RepeatSampler(self.batch_sampler))
+        self.iterator = super().__iter__()
+
+    def __len__(self):
+        """Returns the length of the batch sampler's sampler."""
+        return len(self.batch_sampler.sampler)
+
+    def __iter__(self):
+        """Creates a sampler that repeats indefinitely."""
+        for _ in range(len(self)):
+            yield next(self.iterator)
+
+    def __del__(self):
+        """Ensure that workers are terminated."""
+        if hasattr(self.iterator, "_workers"):
+            for w in self.iterator._workers:  # force terminate
+                if w.is_alive():
+                    w.terminate()
+            self.iterator._shutdown_workers()  # cleanup
+
+    def reset(self):
+        """
+        Reset iterator.
+
+        This is useful when we want to modify settings of dataset while training.
+        """
+        self.iterator = self._get_iterator()
+
+
+class _RepeatSampler:
+    """
+    Sampler that repeats forever.
+
+    Args:
+        sampler (Dataset.sampler): The sampler to repeat.
+    """
+
+    def __init__(self, sampler):
+        """Initializes an object that repeats a given sampler indefinitely."""
+        self.sampler = sampler
+
+    def __iter__(self):
+        """Iterates over the 'sampler' and yields its contents."""
+        while True:
+            yield from iter(self.sampler)
+
+
+def seed_worker(worker_id):  # noqa
+    """Set dataloader worker seed https://pytorch.org/docs/stable/notes/randomness.html#dataloader."""
+    worker_seed = torch.initial_seed() % 2**32
+    np.random.seed(worker_seed)
+    random.seed(worker_seed)
+
+
+def build_yolo_dataset(cfg, img_path, batch, data, mode="train", rect=False, stride=32, multi_modal=False):
+    """Build YOLO Dataset."""
+    dataset = YOLOMultiModalDataset if multi_modal else YOLODataset
+    return dataset(
+        img_path=img_path,
+        imgsz=cfg.imgsz,
+        batch_size=batch,
+        augment=mode == "train",  # augmentation
+        hyp=cfg,  # TODO: probably add a get_hyps_from_cfg function
+        rect=cfg.rect or rect,  # rectangular batches
+        cache=cfg.cache or None,
+        single_cls=cfg.single_cls or False,
+        stride=int(stride),
+        pad=0.0 if mode == "train" else 0.5,
+        prefix=colorstr(f"{mode}: "),
+        task=cfg.task,
+        classes=cfg.classes,
+        data=data,
+        fraction=cfg.fraction if mode == "train" else 1.0,
+    )
+
+
+def build_grounding(cfg, img_path, json_file, batch, mode="train", rect=False, stride=32):
+    """Build YOLO Dataset."""
+    return GroundingDataset(
+        img_path=img_path,
+        json_file=json_file,
+        imgsz=cfg.imgsz,
+        batch_size=batch,
+        augment=mode == "train",  # augmentation
+        hyp=cfg,  # TODO: probably add a get_hyps_from_cfg function
+        rect=cfg.rect or rect,  # rectangular batches
+        cache=cfg.cache or None,
+        single_cls=cfg.single_cls or False,
+        stride=int(stride),
+        pad=0.0 if mode == "train" else 0.5,
+        prefix=colorstr(f"{mode}: "),
+        task=cfg.task,
+        classes=cfg.classes,
+        fraction=cfg.fraction if mode == "train" else 1.0,
+    )
+
+
+def build_dataloader(dataset, batch, workers, shuffle=True, rank=-1):
+    """Return an InfiniteDataLoader or DataLoader for training or validation set."""
+    batch = min(batch, len(dataset))
+    nd = torch.cuda.device_count()  # number of CUDA devices
+    nw = min(os.cpu_count() // max(nd, 1), workers)  # number of workers
+    sampler = None if rank == -1 else distributed.DistributedSampler(dataset, shuffle=shuffle)
+    generator = torch.Generator()
+    generator.manual_seed(6148914691236517205 + RANK)
+    return InfiniteDataLoader(
+        dataset=dataset,
+        batch_size=batch,
+        shuffle=shuffle and sampler is None,
+        num_workers=nw,
+        sampler=sampler,
+        pin_memory=PIN_MEMORY,
+        collate_fn=getattr(dataset, "collate_fn", None),
+        worker_init_fn=seed_worker,
+        generator=generator,
+    )
+
+
+def check_source(source):
+    """Check source type and return corresponding flag values."""
+    webcam, screenshot, from_img, in_memory, tensor = False, False, False, False, False
+    if isinstance(source, (str, int, Path)):  # int for local usb camera
+        source = str(source)
+        is_file = Path(source).suffix[1:] in (IMG_FORMATS | VID_FORMATS)
+        is_url = source.lower().startswith(("https://", "http://", "rtsp://", "rtmp://", "tcp://"))
+        webcam = source.isnumeric() or source.endswith(".streams") or (is_url and not is_file)
+        screenshot = source.lower() == "screen"
+        if is_url and is_file:
+            source = check_file(source)  # download
+    elif isinstance(source, LOADERS):
+        in_memory = True
+    elif isinstance(source, (list, tuple)):
+        source = autocast_list(source)  # convert all list elements to PIL or np arrays
+        from_img = True
+    elif isinstance(source, (Image.Image, np.ndarray)):
+        from_img = True
+    elif isinstance(source, torch.Tensor):
+        tensor = True
+    else:
+        raise TypeError("Unsupported image type. For supported types see https://docs.ultralytics.com/modes/predict")
+
+    return source, webcam, screenshot, from_img, in_memory, tensor
+
+
+def load_inference_source(source=None, batch=1, vid_stride=1, buffer=False):
+    """
+    Loads an inference source for object detection and applies necessary transformations.
+
+    Args:
+        source (str, Path, Tensor, PIL.Image, np.ndarray): The input source for inference.
+        batch (int, optional): Batch size for dataloaders. Default is 1.
+        vid_stride (int, optional): The frame interval for video sources. Default is 1.
+        buffer (bool, optional): Determined whether stream frames will be buffered. Default is False.
+
+    Returns:
+        dataset (Dataset): A dataset object for the specified input source.
+    """
+    source, stream, screenshot, from_img, in_memory, tensor = check_source(source)
+    source_type = source.source_type if in_memory else SourceTypes(stream, screenshot, from_img, tensor)
+
+    # Dataloader
+    if tensor:
+        dataset = LoadTensor(source)
+    elif in_memory:
+        dataset = source
+    elif stream:
+        dataset = LoadStreams(source, vid_stride=vid_stride, buffer=buffer)
+    elif screenshot:
+        dataset = LoadScreenshots(source)
+    elif from_img:
+        dataset = LoadPilAndNumpy(source)
+    else:
+        dataset = LoadImagesAndVideos(source, batch=batch, vid_stride=vid_stride)
+
+    # Attach source types to the dataset
+    setattr(dataset, "source_type", source_type)
+
+    return dataset

ultralytics/data/converter.py

@@ -0,0 +1,702 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+import json
+import random
+import shutil
+from collections import defaultdict
+from concurrent.futures import ThreadPoolExecutor, as_completed
+from pathlib import Path
+
+import cv2
+import numpy as np
+from PIL import Image
+
+from ultralytics.utils import DATASETS_DIR, LOGGER, NUM_THREADS, TQDM
+from ultralytics.utils.downloads import download
+from ultralytics.utils.files import increment_path
+
+
+def coco91_to_coco80_class():
+    """
+    Converts 91-index COCO class IDs to 80-index COCO class IDs.
+
+    Returns:
+        (list): A list of 91 class IDs where the index represents the 80-index class ID and the value is the
+            corresponding 91-index class ID.
+    """
+    return [
+        0,
+        1,
+        2,
+        3,
+        4,
+        5,
+        6,
+        7,
+        8,
+        9,
+        10,
+        None,
+        11,
+        12,
+        13,
+        14,
+        15,
+        16,
+        17,
+        18,
+        19,
+        20,
+        21,
+        22,
+        23,
+        None,
+        24,
+        25,
+        None,
+        None,
+        26,
+        27,
+        28,
+        29,
+        30,
+        31,
+        32,
+        33,
+        34,
+        35,
+        36,
+        37,
+        38,
+        39,
+        None,
+        40,
+        41,
+        42,
+        43,
+        44,
+        45,
+        46,
+        47,
+        48,
+        49,
+        50,
+        51,
+        52,
+        53,
+        54,
+        55,
+        56,
+        57,
+        58,
+        59,
+        None,
+        60,
+        None,
+        None,
+        61,
+        None,
+        62,
+        63,
+        64,
+        65,
+        66,
+        67,
+        68,
+        69,
+        70,
+        71,
+        72,
+        None,
+        73,
+        74,
+        75,
+        76,
+        77,
+        78,
+        79,
+        None,
+    ]
+
+
+def coco80_to_coco91_class():
+    r"""
+    Converts 80-index (val2014) to 91-index (paper).
+    For details see https://tech.amikelive.com/node-718/what-object-categories-labels-are-in-coco-dataset/.
+
+    Example:
+        ```python
+        import numpy as np
+
+        a = np.loadtxt("data/coco.names", dtype="str", delimiter="\n")
+        b = np.loadtxt("data/coco_paper.names", dtype="str", delimiter="\n")
+        x1 = [list(a[i] == b).index(True) + 1 for i in range(80)]  # darknet to coco
+        x2 = [list(b[i] == a).index(True) if any(b[i] == a) else None for i in range(91)]  # coco to darknet
+        ```
+    """
+    return [
+        1,
+        2,
+        3,
+        4,
+        5,
+        6,
+        7,
+        8,
+        9,
+        10,
+        11,
+        13,
+        14,
+        15,
+        16,
+        17,
+        18,
+        19,
+        20,
+        21,
+        22,
+        23,
+        24,
+        25,
+        27,
+        28,
+        31,
+        32,
+        33,
+        34,
+        35,
+        36,
+        37,
+        38,
+        39,
+        40,
+        41,
+        42,
+        43,
+        44,
+        46,
+        47,
+        48,
+        49,
+        50,
+        51,
+        52,
+        53,
+        54,
+        55,
+        56,
+        57,
+        58,
+        59,
+        60,
+        61,
+        62,
+        63,
+        64,
+        65,
+        67,
+        70,
+        72,
+        73,
+        74,
+        75,
+        76,
+        77,
+        78,
+        79,
+        80,
+        81,
+        82,
+        84,
+        85,
+        86,
+        87,
+        88,
+        89,
+        90,
+    ]
+
+
+def convert_coco(
+    labels_dir="../coco/annotations/",
+    save_dir="coco_converted/",
+    use_segments=False,
+    use_keypoints=False,
+    cls91to80=True,
+    lvis=False,
+):
+    """
+    Converts COCO dataset annotations to a YOLO annotation format  suitable for training YOLO models.
+
+    Args:
+        labels_dir (str, optional): Path to directory containing COCO dataset annotation files.
+        save_dir (str, optional): Path to directory to save results to.
+        use_segments (bool, optional): Whether to include segmentation masks in the output.
+        use_keypoints (bool, optional): Whether to include keypoint annotations in the output.
+        cls91to80 (bool, optional): Whether to map 91 COCO class IDs to the corresponding 80 COCO class IDs.
+        lvis (bool, optional): Whether to convert data in lvis dataset way.
+
+    Example:
+        ```python
+        from ultralytics.data.converter import convert_coco
+
+        convert_coco("../datasets/coco/annotations/", use_segments=True, use_keypoints=False, cls91to80=False)
+        convert_coco(
+            "../datasets/lvis/annotations/", use_segments=True, use_keypoints=False, cls91to80=False, lvis=True
+        )
+        ```
+
+    Output:
+        Generates output files in the specified output directory.
+    """
+    # Create dataset directory
+    save_dir = increment_path(save_dir)  # increment if save directory already exists
+    for p in save_dir / "labels", save_dir / "images":
+        p.mkdir(parents=True, exist_ok=True)  # make dir
+
+    # Convert classes
+    coco80 = coco91_to_coco80_class()
+
+    # Import json
+    for json_file in sorted(Path(labels_dir).resolve().glob("*.json")):
+        lname = "" if lvis else json_file.stem.replace("instances_", "")
+        fn = Path(save_dir) / "labels" / lname  # folder name
+        fn.mkdir(parents=True, exist_ok=True)
+        if lvis:
+            # NOTE: create folders for both train and val in advance,
+            # since LVIS val set contains images from COCO 2017 train in addition to the COCO 2017 val split.
+            (fn / "train2017").mkdir(parents=True, exist_ok=True)
+            (fn / "val2017").mkdir(parents=True, exist_ok=True)
+        with open(json_file, encoding="utf-8") as f:
+            data = json.load(f)
+
+        # Create image dict
+        images = {f"{x['id']:d}": x for x in data["images"]}
+        # Create image-annotations dict
+        imgToAnns = defaultdict(list)
+        for ann in data["annotations"]:
+            imgToAnns[ann["image_id"]].append(ann)
+
+        image_txt = []
+        # Write labels file
+        for img_id, anns in TQDM(imgToAnns.items(), desc=f"Annotations {json_file}"):
+            img = images[f"{img_id:d}"]
+            h, w = img["height"], img["width"]
+            f = str(Path(img["coco_url"]).relative_to("http://images.cocodataset.org")) if lvis else img["file_name"]
+            if lvis:
+                image_txt.append(str(Path("./images") / f))
+
+            bboxes = []
+            segments = []
+            keypoints = []
+            for ann in anns:
+                if ann.get("iscrowd", False):
+                    continue
+                # The COCO box format is [top left x, top left y, width, height]
+                box = np.array(ann["bbox"], dtype=np.float64)
+                box[:2] += box[2:] / 2  # xy top-left corner to center
+                box[[0, 2]] /= w  # normalize x
+                box[[1, 3]] /= h  # normalize y
+                if box[2] <= 0 or box[3] <= 0:  # if w <= 0 and h <= 0
+                    continue
+
+                cls = coco80[ann["category_id"] - 1] if cls91to80 else ann["category_id"] - 1  # class
+                box = [cls] + box.tolist()
+                if box not in bboxes:
+                    bboxes.append(box)
+                    if use_segments and ann.get("segmentation") is not None:
+                        if len(ann["segmentation"]) == 0:
+                            segments.append([])
+                            continue
+                        elif len(ann["segmentation"]) > 1:
+                            s = merge_multi_segment(ann["segmentation"])
+                            s = (np.concatenate(s, axis=0) / np.array([w, h])).reshape(-1).tolist()
+                        else:
+                            s = [j for i in ann["segmentation"] for j in i]  # all segments concatenated
+                            s = (np.array(s).reshape(-1, 2) / np.array([w, h])).reshape(-1).tolist()
+                        s = [cls] + s
+                        segments.append(s)
+                    if use_keypoints and ann.get("keypoints") is not None:
+                        keypoints.append(
+                            box + (np.array(ann["keypoints"]).reshape(-1, 3) / np.array([w, h, 1])).reshape(-1).tolist()
+                        )
+
+            # Write
+            with open((fn / f).with_suffix(".txt"), "a") as file:
+                for i in range(len(bboxes)):
+                    if use_keypoints:
+                        line = (*(keypoints[i]),)  # cls, box, keypoints
+                    else:
+                        line = (
+                            *(segments[i] if use_segments and len(segments[i]) > 0 else bboxes[i]),
+                        )  # cls, box or segments
+                    file.write(("%g " * len(line)).rstrip() % line + "\n")
+
+        if lvis:
+            with open((Path(save_dir) / json_file.name.replace("lvis_v1_", "").replace(".json", ".txt")), "a") as f:
+                f.writelines(f"{line}\n" for line in image_txt)
+
+    LOGGER.info(f"{'LVIS' if lvis else 'COCO'} data converted successfully.\nResults saved to {save_dir.resolve()}")
+
+
+def convert_segment_masks_to_yolo_seg(masks_dir, output_dir, classes):
+    """
+    Converts a dataset of segmentation mask images to the YOLO segmentation format.
+
+    This function takes the directory containing the binary format mask images and converts them into YOLO segmentation format.
+    The converted masks are saved in the specified output directory.
+
+    Args:
+        masks_dir (str): The path to the directory where all mask images (png, jpg) are stored.
+        output_dir (str): The path to the directory where the converted YOLO segmentation masks will be stored.
+        classes (int): Total classes in the dataset i.e. for COCO classes=80
+
+    Example:
+        ```python
+        from ultralytics.data.converter import convert_segment_masks_to_yolo_seg
+
+        # The classes here is the total classes in the dataset, for COCO dataset we have 80 classes
+        convert_segment_masks_to_yolo_seg("path/to/masks_directory", "path/to/output/directory", classes=80)
+        ```
+
+    Notes:
+        The expected directory structure for the masks is:
+
+            - masks
+                ├─ mask_image_01.png or mask_image_01.jpg
+                ├─ mask_image_02.png or mask_image_02.jpg
+                ├─ mask_image_03.png or mask_image_03.jpg
+                └─ mask_image_04.png or mask_image_04.jpg
+
+        After execution, the labels will be organized in the following structure:
+
+            - output_dir
+                ├─ mask_yolo_01.txt
+                ├─ mask_yolo_02.txt
+                ├─ mask_yolo_03.txt
+                └─ mask_yolo_04.txt
+    """
+    pixel_to_class_mapping = {i + 1: i for i in range(classes)}
+    for mask_path in Path(masks_dir).iterdir():
+        if mask_path.suffix in {".png", ".jpg"}:
+            mask = cv2.imread(str(mask_path), cv2.IMREAD_GRAYSCALE)  # Read the mask image in grayscale
+            img_height, img_width = mask.shape  # Get image dimensions
+            LOGGER.info(f"Processing {mask_path} imgsz = {img_height} x {img_width}")
+
+            unique_values = np.unique(mask)  # Get unique pixel values representing different classes
+            yolo_format_data = []
+
+            for value in unique_values:
+                if value == 0:
+                    continue  # Skip background
+                class_index = pixel_to_class_mapping.get(value, -1)
+                if class_index == -1:
+                    LOGGER.warning(f"Unknown class for pixel value {value} in file {mask_path}, skipping.")
+                    continue
+
+                # Create a binary mask for the current class and find contours
+                contours, _ = cv2.findContours(
+                    (mask == value).astype(np.uint8), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE
+                )  # Find contours
+
+                for contour in contours:
+                    if len(contour) >= 3:  # YOLO requires at least 3 points for a valid segmentation
+                        contour = contour.squeeze()  # Remove single-dimensional entries
+                        yolo_format = [class_index]
+                        for point in contour:
+                            # Normalize the coordinates
+                            yolo_format.append(round(point[0] / img_width, 6))  # Rounding to 6 decimal places
+                            yolo_format.append(round(point[1] / img_height, 6))
+                        yolo_format_data.append(yolo_format)
+            # Save Ultralytics YOLO format data to file
+            output_path = Path(output_dir) / f"{mask_path.stem}.txt"
+            with open(output_path, "w") as file:
+                for item in yolo_format_data:
+                    line = " ".join(map(str, item))
+                    file.write(line + "\n")
+            LOGGER.info(f"Processed and stored at {output_path} imgsz = {img_height} x {img_width}")
+
+
+def convert_dota_to_yolo_obb(dota_root_path: str):
+    """
+    Converts DOTA dataset annotations to YOLO OBB (Oriented Bounding Box) format.
+
+    The function processes images in the 'train' and 'val' folders of the DOTA dataset. For each image, it reads the
+    associated label from the original labels directory and writes new labels in YOLO OBB format to a new directory.
+
+    Args:
+        dota_root_path (str): The root directory path of the DOTA dataset.
+
+    Example:
+        ```python
+        from ultralytics.data.converter import convert_dota_to_yolo_obb
+
+        convert_dota_to_yolo_obb("path/to/DOTA")
+        ```
+
+    Notes:
+        The directory structure assumed for the DOTA dataset:
+
+            - DOTA
+                ├─ images
+                │   ├─ train
+                │   └─ val
+                └─ labels
+                    ├─ train_original
+                    └─ val_original
+
+        After execution, the function will organize the labels into:
+
+            - DOTA
+                └─ labels
+                    ├─ train
+                    └─ val
+    """
+    dota_root_path = Path(dota_root_path)
+
+    # Class names to indices mapping
+    class_mapping = {
+        "plane": 0,
+        "ship": 1,
+        "storage-tank": 2,
+        "baseball-diamond": 3,
+        "tennis-court": 4,
+        "basketball-court": 5,
+        "ground-track-field": 6,
+        "harbor": 7,
+        "bridge": 8,
+        "large-vehicle": 9,
+        "small-vehicle": 10,
+        "helicopter": 11,
+        "roundabout": 12,
+        "soccer-ball-field": 13,
+        "swimming-pool": 14,
+        "container-crane": 15,
+        "airport": 16,
+        "helipad": 17,
+    }
+
+    def convert_label(image_name, image_width, image_height, orig_label_dir, save_dir):
+        """Converts a single image's DOTA annotation to YOLO OBB format and saves it to a specified directory."""
+        orig_label_path = orig_label_dir / f"{image_name}.txt"
+        save_path = save_dir / f"{image_name}.txt"
+
+        with orig_label_path.open("r") as f, save_path.open("w") as g:
+            lines = f.readlines()
+            for line in lines:
+                parts = line.strip().split()
+                if len(parts) < 9:
+                    continue
+                class_name = parts[8]
+                class_idx = class_mapping[class_name]
+                coords = [float(p) for p in parts[:8]]
+                normalized_coords = [
+                    coords[i] / image_width if i % 2 == 0 else coords[i] / image_height for i in range(8)
+                ]
+                formatted_coords = [f"{coord:.6g}" for coord in normalized_coords]
+                g.write(f"{class_idx} {' '.join(formatted_coords)}\n")
+
+    for phase in ["train", "val"]:
+        image_dir = dota_root_path / "images" / phase
+        orig_label_dir = dota_root_path / "labels" / f"{phase}_original"
+        save_dir = dota_root_path / "labels" / phase
+
+        save_dir.mkdir(parents=True, exist_ok=True)
+
+        image_paths = list(image_dir.iterdir())
+        for image_path in TQDM(image_paths, desc=f"Processing {phase} images"):
+            if image_path.suffix != ".png":
+                continue
+            image_name_without_ext = image_path.stem
+            img = cv2.imread(str(image_path))
+            h, w = img.shape[:2]
+            convert_label(image_name_without_ext, w, h, orig_label_dir, save_dir)
+
+
+def min_index(arr1, arr2):
+    """
+    Find a pair of indexes with the shortest distance between two arrays of 2D points.
+
+    Args:
+        arr1 (np.ndarray): A NumPy array of shape (N, 2) representing N 2D points.
+        arr2 (np.ndarray): A NumPy array of shape (M, 2) representing M 2D points.
+
+    Returns:
+        (tuple): A tuple containing the indexes of the points with the shortest distance in arr1 and arr2 respectively.
+    """
+    dis = ((arr1[:, None, :] - arr2[None, :, :]) ** 2).sum(-1)
+    return np.unravel_index(np.argmin(dis, axis=None), dis.shape)
+
+
+def merge_multi_segment(segments):
+    """
+    Merge multiple segments into one list by connecting the coordinates with the minimum distance between each segment.
+    This function connects these coordinates with a thin line to merge all segments into one.
+
+    Args:
+        segments (List[List]): Original segmentations in COCO's JSON file.
+                               Each element is a list of coordinates, like [segmentation1, segmentation2,...].
+
+    Returns:
+        s (List[np.ndarray]): A list of connected segments represented as NumPy arrays.
+    """
+    s = []
+    segments = [np.array(i).reshape(-1, 2) for i in segments]
+    idx_list = [[] for _ in range(len(segments))]
+
+    # Record the indexes with min distance between each segment
+    for i in range(1, len(segments)):
+        idx1, idx2 = min_index(segments[i - 1], segments[i])
+        idx_list[i - 1].append(idx1)
+        idx_list[i].append(idx2)
+
+    # Use two round to connect all the segments
+    for k in range(2):
+        # Forward connection
+        if k == 0:
+            for i, idx in enumerate(idx_list):
+                # Middle segments have two indexes, reverse the index of middle segments
+                if len(idx) == 2 and idx[0] > idx[1]:
+                    idx = idx[::-1]
+                    segments[i] = segments[i][::-1, :]
+
+                segments[i] = np.roll(segments[i], -idx[0], axis=0)
+                segments[i] = np.concatenate([segments[i], segments[i][:1]])
+                # Deal with the first segment and the last one
+                if i in {0, len(idx_list) - 1}:
+                    s.append(segments[i])
+                else:
+                    idx = [0, idx[1] - idx[0]]
+                    s.append(segments[i][idx[0] : idx[1] + 1])
+
+        else:
+            for i in range(len(idx_list) - 1, -1, -1):
+                if i not in {0, len(idx_list) - 1}:
+                    idx = idx_list[i]
+                    nidx = abs(idx[1] - idx[0])
+                    s.append(segments[i][nidx:])
+    return s
+
+
+def yolo_bbox2segment(im_dir, save_dir=None, sam_model="sam_b.pt", device=None):
+    """
+    Converts existing object detection dataset (bounding boxes) to segmentation dataset or oriented bounding box (OBB)
+    in YOLO format. Generates segmentation data using SAM auto-annotator as needed.
+
+    Args:
+        im_dir (str | Path): Path to image directory to convert.
+        save_dir (str | Path): Path to save the generated labels, labels will be saved
+            into `labels-segment` in the same directory level of `im_dir` if save_dir is None. Default: None.
+        sam_model (str): Segmentation model to use for intermediate segmentation data; optional.
+        device (int | str): The specific device to run SAM models. Default: None.
+
+    Notes:
+        The input directory structure assumed for dataset:
+
+            - im_dir
+                ├─ 001.jpg
+                ├─ ...
+                └─ NNN.jpg
+            - labels
+                ├─ 001.txt
+                ├─ ...
+                └─ NNN.txt
+    """
+    from ultralytics import SAM
+    from ultralytics.data import YOLODataset
+    from ultralytics.utils import LOGGER
+    from ultralytics.utils.ops import xywh2xyxy
+
+    # NOTE: add placeholder to pass class index check
+    dataset = YOLODataset(im_dir, data=dict(names=list(range(1000))))
+    if len(dataset.labels[0]["segments"]) > 0:  # if it's segment data
+        LOGGER.info("Segmentation labels detected, no need to generate new ones!")
+        return
+
+    LOGGER.info("Detection labels detected, generating segment labels by SAM model!")
+    sam_model = SAM(sam_model)
+    for label in TQDM(dataset.labels, total=len(dataset.labels), desc="Generating segment labels"):
+        h, w = label["shape"]
+        boxes = label["bboxes"]
+        if len(boxes) == 0:  # skip empty labels
+            continue
+        boxes[:, [0, 2]] *= w
+        boxes[:, [1, 3]] *= h
+        im = cv2.imread(label["im_file"])
+        sam_results = sam_model(im, bboxes=xywh2xyxy(boxes), verbose=False, save=False, device=device)
+        label["segments"] = sam_results[0].masks.xyn
+
+    save_dir = Path(save_dir) if save_dir else Path(im_dir).parent / "labels-segment"
+    save_dir.mkdir(parents=True, exist_ok=True)
+    for label in dataset.labels:
+        texts = []
+        lb_name = Path(label["im_file"]).with_suffix(".txt").name
+        txt_file = save_dir / lb_name
+        cls = label["cls"]
+        for i, s in enumerate(label["segments"]):
+            if len(s) == 0:
+                continue
+            line = (int(cls[i]), *s.reshape(-1))
+            texts.append(("%g " * len(line)).rstrip() % line)
+        with open(txt_file, "a") as f:
+            f.writelines(text + "\n" for text in texts)
+    LOGGER.info(f"Generated segment labels saved in {save_dir}")
+
+
+def create_synthetic_coco_dataset():
+    """
+    Creates a synthetic COCO dataset with random images based on filenames from label lists.
+
+    This function downloads COCO labels, reads image filenames from label list files,
+    creates synthetic images for train2017 and val2017 subsets, and organizes
+    them in the COCO dataset structure. It uses multithreading to generate images efficiently.
+
+    Examples:
+        >>> from ultralytics.data.converter import create_synthetic_coco_dataset
+        >>> create_synthetic_coco_dataset()
+
+    Notes:
+        - Requires internet connection to download label files.
+        - Generates random RGB images of varying sizes (480x480 to 640x640 pixels).
+        - Existing test2017 directory is removed as it's not needed.
+        - Reads image filenames from train2017.txt and val2017.txt files.
+    """
+
+    def create_synthetic_image(image_file):
+        """Generates synthetic images with random sizes and colors for dataset augmentation or testing purposes."""
+        if not image_file.exists():
+            size = (random.randint(480, 640), random.randint(480, 640))
+            Image.new(
+                "RGB",
+                size=size,
+                color=(random.randint(0, 255), random.randint(0, 255), random.randint(0, 255)),
+            ).save(image_file)
+
+    # Download labels
+    dir = DATASETS_DIR / "coco"
+    url = "https://github.com/ultralytics/assets/releases/download/v0.0.0/"
+    label_zip = "coco2017labels-segments.zip"
+    download([url + label_zip], dir=dir.parent)
+
+    # Create synthetic images
+    shutil.rmtree(dir / "labels" / "test2017", ignore_errors=True)  # Remove test2017 directory as not needed
+    with ThreadPoolExecutor(max_workers=NUM_THREADS) as executor:
+        for subset in ["train2017", "val2017"]:
+            subset_dir = dir / "images" / subset
+            subset_dir.mkdir(parents=True, exist_ok=True)
+
+            # Read image filenames from label list file
+            label_list_file = dir / f"{subset}.txt"
+            if label_list_file.exists():
+                with open(label_list_file) as f:
+                    image_files = [dir / line.strip() for line in f]
+
+                # Submit all tasks
+                futures = [executor.submit(create_synthetic_image, image_file) for image_file in image_files]
+                for _ in TQDM(as_completed(futures), total=len(futures), desc=f"Generating images for {subset}"):
+                    pass  # The actual work is done in the background
+            else:
+                print(f"Warning: Labels file {label_list_file} does not exist. Skipping image creation for {subset}.")
+
+    print("Synthetic COCO dataset created successfully.")

ultralytics/data/dataset.py

@@ -0,0 +1,521 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+import json
+from collections import defaultdict
+from itertools import repeat
+from multiprocessing.pool import ThreadPool
+from pathlib import Path
+
+import cv2
+import numpy as np
+import torch
+from PIL import Image
+from torch.utils.data import ConcatDataset
+
+from ultralytics.utils import LOCAL_RANK, NUM_THREADS, TQDM, colorstr
+from ultralytics.utils.ops import resample_segments
+from ultralytics.utils.torch_utils import TORCHVISION_0_18
+
+from .augment import (
+    Compose,
+    Format,
+    Instances,
+    LetterBox,
+    RandomLoadText,
+    classify_augmentations,
+    classify_transforms,
+    v8_transforms,
+)
+from .base import BaseDataset
+from .utils import (
+    HELP_URL,
+    LOGGER,
+    get_hash,
+    img2label_paths,
+    load_dataset_cache_file,
+    save_dataset_cache_file,
+    verify_image,
+    verify_image_label,
+)
+
+# Ultralytics dataset *.cache version, >= 1.0.0 for YOLOv8
+DATASET_CACHE_VERSION = "1.0.3"
+
+
+class YOLODataset(BaseDataset):
+    """
+    Dataset class for loading object detection and/or segmentation labels in YOLO format.
+
+    Args:
+        data (dict, optional): A dataset YAML dictionary. Defaults to None.
+        task (str): An explicit arg to point current task, Defaults to 'detect'.
+
+    Returns:
+        (torch.utils.data.Dataset): A PyTorch dataset object that can be used for training an object detection model.
+    """
+
+    def __init__(self, *args, data=None, task="detect", **kwargs):
+        """Initializes the YOLODataset with optional configurations for segments and keypoints."""
+        self.use_segments = task == "segment"
+        self.use_keypoints = task == "pose"
+        self.use_obb = task == "obb"
+        self.data = data
+        assert not (self.use_segments and self.use_keypoints), "Can not use both segments and keypoints."
+        super().__init__(*args, **kwargs)
+
+    def cache_labels(self, path=Path("./labels.cache")):
+        """
+        Cache dataset labels, check images and read shapes.
+
+        Args:
+            path (Path): Path where to save the cache file. Default is Path("./labels.cache").
+
+        Returns:
+            (dict): labels.
+        """
+        x = {"labels": []}
+        nm, nf, ne, nc, msgs = 0, 0, 0, 0, []  # number missing, found, empty, corrupt, messages
+        desc = f"{self.prefix}Scanning {path.parent / path.stem}..."
+        total = len(self.im_files)
+        nkpt, ndim = self.data.get("kpt_shape", (0, 0))
+        if self.use_keypoints and (nkpt <= 0 or ndim not in {2, 3}):
+            raise ValueError(
+                "'kpt_shape' in data.yaml missing or incorrect. Should be a list with [number of "
+                "keypoints, number of dims (2 for x,y or 3 for x,y,visible)], i.e. 'kpt_shape: [17, 3]'"
+            )
+        with ThreadPool(NUM_THREADS) as pool:
+            results = pool.imap(
+                func=verify_image_label,
+                iterable=zip(
+                    self.im_files,
+                    self.label_files,
+                    repeat(self.prefix),
+                    repeat(self.use_keypoints),
+                    repeat(len(self.data["names"])),
+                    repeat(nkpt),
+                    repeat(ndim),
+                ),
+            )
+            pbar = TQDM(results, desc=desc, total=total)
+            for im_file, lb, shape, segments, keypoint, nm_f, nf_f, ne_f, nc_f, msg in pbar:
+                nm += nm_f
+                nf += nf_f
+                ne += ne_f
+                nc += nc_f
+                if im_file:
+                    x["labels"].append(
+                        {
+                            "im_file": im_file,
+                            "shape": shape,
+                            "cls": lb[:, 0:1],  # n, 1
+                            "bboxes": lb[:, 1:],  # n, 4
+                            "segments": segments,
+                            "keypoints": keypoint,
+                            "normalized": True,
+                            "bbox_format": "xywh",
+                        }
+                    )
+                if msg:
+                    msgs.append(msg)
+                pbar.desc = f"{desc} {nf} images, {nm + ne} backgrounds, {nc} corrupt"
+            pbar.close()
+
+        if msgs:
+            LOGGER.info("\n".join(msgs))
+        if nf == 0:
+            LOGGER.warning(f"{self.prefix}WARNING ⚠️ No labels found in {path}. {HELP_URL}")
+        x["hash"] = get_hash(self.label_files + self.im_files)
+        x["results"] = nf, nm, ne, nc, len(self.im_files)
+        x["msgs"] = msgs  # warnings
+        save_dataset_cache_file(self.prefix, path, x, DATASET_CACHE_VERSION)
+        return x
+
+    def get_labels(self):
+        """Returns dictionary of labels for YOLO training."""
+        self.label_files = img2label_paths(self.im_files)
+        cache_path = Path(self.label_files[0]).parent.with_suffix(".cache")
+        try:
+            cache, exists = load_dataset_cache_file(cache_path), True  # attempt to load a *.cache file
+            assert cache["version"] == DATASET_CACHE_VERSION  # matches current version
+            assert cache["hash"] == get_hash(self.label_files + self.im_files)  # identical hash
+        except (FileNotFoundError, AssertionError, AttributeError):
+            cache, exists = self.cache_labels(cache_path), False  # run cache ops
+
+        # Display cache
+        nf, nm, ne, nc, n = cache.pop("results")  # found, missing, empty, corrupt, total
+        if exists and LOCAL_RANK in {-1, 0}:
+            d = f"Scanning {cache_path}... {nf} images, {nm + ne} backgrounds, {nc} corrupt"
+            TQDM(None, desc=self.prefix + d, total=n, initial=n)  # display results
+            if cache["msgs"]:
+                LOGGER.info("\n".join(cache["msgs"]))  # display warnings
+
+        # Read cache
+        [cache.pop(k) for k in ("hash", "version", "msgs")]  # remove items
+        labels = cache["labels"]
+        if not labels:
+            LOGGER.warning(f"WARNING ⚠️ No images found in {cache_path}, training may not work correctly. {HELP_URL}")
+        self.im_files = [lb["im_file"] for lb in labels]  # update im_files
+
+        # Check if the dataset is all boxes or all segments
+        lengths = ((len(lb["cls"]), len(lb["bboxes"]), len(lb["segments"])) for lb in labels)
+        len_cls, len_boxes, len_segments = (sum(x) for x in zip(*lengths))
+        if len_segments and len_boxes != len_segments:
+            LOGGER.warning(
+                f"WARNING ⚠️ Box and segment counts should be equal, but got len(segments) = {len_segments}, "
+                f"len(boxes) = {len_boxes}. To resolve this only boxes will be used and all segments will be removed. "
+                "To avoid this please supply either a detect or segment dataset, not a detect-segment mixed dataset."
+            )
+            for lb in labels:
+                lb["segments"] = []
+        if len_cls == 0:
+            LOGGER.warning(f"WARNING ⚠️ No labels found in {cache_path}, training may not work correctly. {HELP_URL}")
+        return labels
+
+    def build_transforms(self, hyp=None):
+        """Builds and appends transforms to the list."""
+        if self.augment:
+            hyp.mosaic = hyp.mosaic if self.augment and not self.rect else 0.0
+            hyp.mixup = hyp.mixup if self.augment and not self.rect else 0.0
+            transforms = v8_transforms(self, self.imgsz, hyp)
+        else:
+            transforms = Compose([LetterBox(new_shape=(self.imgsz, self.imgsz), scaleup=False)])
+        transforms.append(
+            Format(
+                bbox_format="xywh",
+                normalize=True,
+                return_mask=self.use_segments,
+                return_keypoint=self.use_keypoints,
+                return_obb=self.use_obb,
+                batch_idx=True,
+                mask_ratio=hyp.mask_ratio,
+                mask_overlap=hyp.overlap_mask,
+                bgr=hyp.bgr if self.augment else 0.0,  # only affect training.
+            )
+        )
+        return transforms
+
+    def close_mosaic(self, hyp):
+        """Sets mosaic, copy_paste and mixup options to 0.0 and builds transformations."""
+        hyp.mosaic = 0.0  # set mosaic ratio=0.0
+        hyp.copy_paste = 0.0  # keep the same behavior as previous v8 close-mosaic
+        hyp.mixup = 0.0  # keep the same behavior as previous v8 close-mosaic
+        self.transforms = self.build_transforms(hyp)
+
+    def update_labels_info(self, label):
+        """
+        Custom your label format here.
+
+        Note:
+            cls is not with bboxes now, classification and semantic segmentation need an independent cls label
+            Can also support classification and semantic segmentation by adding or removing dict keys there.
+        """
+        bboxes = label.pop("bboxes")
+        segments = label.pop("segments", [])
+        keypoints = label.pop("keypoints", None)
+        bbox_format = label.pop("bbox_format")
+        normalized = label.pop("normalized")
+
+        # NOTE: do NOT resample oriented boxes
+        segment_resamples = 100 if self.use_obb else 1000
+        if len(segments) > 0:
+            # make sure segments interpolate correctly if original length is greater than segment_resamples
+            max_len = max(len(s) for s in segments)
+            segment_resamples = (max_len + 1) if segment_resamples < max_len else segment_resamples
+            # list[np.array(segment_resamples, 2)] * num_samples
+            segments = np.stack(resample_segments(segments, n=segment_resamples), axis=0)
+        else:
+            segments = np.zeros((0, segment_resamples, 2), dtype=np.float32)
+        label["instances"] = Instances(bboxes, segments, keypoints, bbox_format=bbox_format, normalized=normalized)
+        return label
+
+    @staticmethod
+    def collate_fn(batch):
+        """Collates data samples into batches."""
+        new_batch = {}
+        keys = batch[0].keys()
+        values = list(zip(*[list(b.values()) for b in batch]))
+        for i, k in enumerate(keys):
+            value = values[i]
+            if k == "img":
+                value = torch.stack(value, 0)
+            if k in {"masks", "keypoints", "bboxes", "cls", "segments", "obb"}:
+                value = torch.cat(value, 0)
+            new_batch[k] = value
+        new_batch["batch_idx"] = list(new_batch["batch_idx"])
+        for i in range(len(new_batch["batch_idx"])):
+            new_batch["batch_idx"][i] += i  # add target image index for build_targets()
+        new_batch["batch_idx"] = torch.cat(new_batch["batch_idx"], 0)
+        return new_batch
+
+
+class YOLOMultiModalDataset(YOLODataset):
+    """
+    Dataset class for loading object detection and/or segmentation labels in YOLO format.
+
+    Args:
+        data (dict, optional): A dataset YAML dictionary. Defaults to None.
+        task (str): An explicit arg to point current task, Defaults to 'detect'.
+
+    Returns:
+        (torch.utils.data.Dataset): A PyTorch dataset object that can be used for training an object detection model.
+    """
+
+    def __init__(self, *args, data=None, task="detect", **kwargs):
+        """Initializes a dataset object for object detection tasks with optional specifications."""
+        super().__init__(*args, data=data, task=task, **kwargs)
+
+    def update_labels_info(self, label):
+        """Add texts information for multi-modal model training."""
+        labels = super().update_labels_info(label)
+        # NOTE: some categories are concatenated with its synonyms by `/`.
+        labels["texts"] = [v.split("/") for _, v in self.data["names"].items()]
+        return labels
+
+    def build_transforms(self, hyp=None):
+        """Enhances data transformations with optional text augmentation for multi-modal training."""
+        transforms = super().build_transforms(hyp)
+        if self.augment:
+            # NOTE: hard-coded the args for now.
+            transforms.insert(-1, RandomLoadText(max_samples=min(self.data["nc"], 80), padding=True))
+        return transforms
+
+
+class GroundingDataset(YOLODataset):
+    """Handles object detection tasks by loading annotations from a specified JSON file, supporting YOLO format."""
+
+    def __init__(self, *args, task="detect", json_file, **kwargs):
+        """Initializes a GroundingDataset for object detection, loading annotations from a specified JSON file."""
+        assert task == "detect", "`GroundingDataset` only support `detect` task for now!"
+        self.json_file = json_file
+        super().__init__(*args, task=task, data={}, **kwargs)
+
+    def get_img_files(self, img_path):
+        """The image files would be read in `get_labels` function, return empty list here."""
+        return []
+
+    def get_labels(self):
+        """Loads annotations from a JSON file, filters, and normalizes bounding boxes for each image."""
+        labels = []
+        LOGGER.info("Loading annotation file...")
+        with open(self.json_file) as f:
+            annotations = json.load(f)
+        images = {f"{x['id']:d}": x for x in annotations["images"]}
+        img_to_anns = defaultdict(list)
+        for ann in annotations["annotations"]:
+            img_to_anns[ann["image_id"]].append(ann)
+        for img_id, anns in TQDM(img_to_anns.items(), desc=f"Reading annotations {self.json_file}"):
+            img = images[f"{img_id:d}"]
+            h, w, f = img["height"], img["width"], img["file_name"]
+            im_file = Path(self.img_path) / f
+            if not im_file.exists():
+                continue
+            self.im_files.append(str(im_file))
+            bboxes = []
+            cat2id = {}
+            texts = []
+            for ann in anns:
+                if ann["iscrowd"]:
+                    continue
+                box = np.array(ann["bbox"], dtype=np.float32)
+                box[:2] += box[2:] / 2
+                box[[0, 2]] /= float(w)
+                box[[1, 3]] /= float(h)
+                if box[2] <= 0 or box[3] <= 0:
+                    continue
+
+                caption = img["caption"]
+                cat_name = " ".join([caption[t[0] : t[1]] for t in ann["tokens_positive"]])
+                if cat_name not in cat2id:
+                    cat2id[cat_name] = len(cat2id)
+                    texts.append([cat_name])
+                cls = cat2id[cat_name]  # class
+                box = [cls] + box.tolist()
+                if box not in bboxes:
+                    bboxes.append(box)
+            lb = np.array(bboxes, dtype=np.float32) if len(bboxes) else np.zeros((0, 5), dtype=np.float32)
+            labels.append(
+                {
+                    "im_file": im_file,
+                    "shape": (h, w),
+                    "cls": lb[:, 0:1],  # n, 1
+                    "bboxes": lb[:, 1:],  # n, 4
+                    "normalized": True,
+                    "bbox_format": "xywh",
+                    "texts": texts,
+                }
+            )
+        return labels
+
+    def build_transforms(self, hyp=None):
+        """Configures augmentations for training with optional text loading; `hyp` adjusts augmentation intensity."""
+        transforms = super().build_transforms(hyp)
+        if self.augment:
+            # NOTE: hard-coded the args for now.
+            transforms.insert(-1, RandomLoadText(max_samples=80, padding=True))
+        return transforms
+
+
+class YOLOConcatDataset(ConcatDataset):
+    """
+    Dataset as a concatenation of multiple datasets.
+
+    This class is useful to assemble different existing datasets.
+    """
+
+    @staticmethod
+    def collate_fn(batch):
+        """Collates data samples into batches."""
+        return YOLODataset.collate_fn(batch)
+
+
+# TODO: support semantic segmentation
+class SemanticDataset(BaseDataset):
+    """
+    Semantic Segmentation Dataset.
+
+    This class is responsible for handling datasets used for semantic segmentation tasks. It inherits functionalities
+    from the BaseDataset class.
+
+    Note:
+        This class is currently a placeholder and needs to be populated with methods and attributes for supporting
+        semantic segmentation tasks.
+    """
+
+    def __init__(self):
+        """Initialize a SemanticDataset object."""
+        super().__init__()
+
+
+class ClassificationDataset:
+    """
+    Extends torchvision ImageFolder to support YOLO classification tasks, offering functionalities like image
+    augmentation, caching, and verification. It's designed to efficiently handle large datasets for training deep
+    learning models, with optional image transformations and caching mechanisms to speed up training.
+
+    This class allows for augmentations using both torchvision and Albumentations libraries, and supports caching images
+    in RAM or on disk to reduce IO overhead during training. Additionally, it implements a robust verification process
+    to ensure data integrity and consistency.
+
+    Attributes:
+        cache_ram (bool): Indicates if caching in RAM is enabled.
+        cache_disk (bool): Indicates if caching on disk is enabled.
+        samples (list): A list of tuples, each containing the path to an image, its class index, path to its .npy cache
+                        file (if caching on disk), and optionally the loaded image array (if caching in RAM).
+        torch_transforms (callable): PyTorch transforms to be applied to the images.
+    """
+
+    def __init__(self, root, args, augment=False, prefix=""):
+        """
+        Initialize YOLO object with root, image size, augmentations, and cache settings.
+
+        Args:
+            root (str): Path to the dataset directory where images are stored in a class-specific folder structure.
+            args (Namespace): Configuration containing dataset-related settings such as image size, augmentation
+                parameters, and cache settings. It includes attributes like `imgsz` (image size), `fraction` (fraction
+                of data to use), `scale`, `fliplr`, `flipud`, `cache` (disk or RAM caching for faster training),
+                `auto_augment`, `hsv_h`, `hsv_s`, `hsv_v`, and `crop_fraction`.
+            augment (bool, optional): Whether to apply augmentations to the dataset. Default is False.
+            prefix (str, optional): Prefix for logging and cache filenames, aiding in dataset identification and
+                debugging. Default is an empty string.
+        """
+        import torchvision  # scope for faster 'import ultralytics'
+
+        # Base class assigned as attribute rather than used as base class to allow for scoping slow torchvision import
+        if TORCHVISION_0_18:  # 'allow_empty' argument first introduced in torchvision 0.18
+            self.base = torchvision.datasets.ImageFolder(root=root, allow_empty=True)
+        else:
+            self.base = torchvision.datasets.ImageFolder(root=root)
+        self.samples = self.base.samples
+        self.root = self.base.root
+
+        # Initialize attributes
+        if augment and args.fraction < 1.0:  # reduce training fraction
+            self.samples = self.samples[: round(len(self.samples) * args.fraction)]
+        self.prefix = colorstr(f"{prefix}: ") if prefix else ""
+        self.cache_ram = args.cache is True or str(args.cache).lower() == "ram"  # cache images into RAM
+        if self.cache_ram:
+            LOGGER.warning(
+                "WARNING ⚠️ Classification `cache_ram` training has known memory leak in "
+                "https://github.com/ultralytics/ultralytics/issues/9824, setting `cache_ram=False`."
+            )
+            self.cache_ram = False
+        self.cache_disk = str(args.cache).lower() == "disk"  # cache images on hard drive as uncompressed *.npy files
+        self.samples = self.verify_images()  # filter out bad images
+        self.samples = [list(x) + [Path(x[0]).with_suffix(".npy"), None] for x in self.samples]  # file, index, npy, im
+        scale = (1.0 - args.scale, 1.0)  # (0.08, 1.0)
+        self.torch_transforms = (
+            classify_augmentations(
+                size=args.imgsz,
+                scale=scale,
+                hflip=args.fliplr,
+                vflip=args.flipud,
+                erasing=args.erasing,
+                auto_augment=args.auto_augment,
+                hsv_h=args.hsv_h,
+                hsv_s=args.hsv_s,
+                hsv_v=args.hsv_v,
+            )
+            if augment
+            else classify_transforms(size=args.imgsz, crop_fraction=args.crop_fraction)
+        )
+
+    def __getitem__(self, i):
+        """Returns subset of data and targets corresponding to given indices."""
+        f, j, fn, im = self.samples[i]  # filename, index, filename.with_suffix('.npy'), image
+        if self.cache_ram:
+            if im is None:  # Warning: two separate if statements required here, do not combine this with previous line
+                im = self.samples[i][3] = cv2.imread(f)
+        elif self.cache_disk:
+            if not fn.exists():  # load npy
+                np.save(fn.as_posix(), cv2.imread(f), allow_pickle=False)
+            im = np.load(fn)
+        else:  # read image
+            im = cv2.imread(f)  # BGR
+        # Convert NumPy array to PIL image
+        im = Image.fromarray(cv2.cvtColor(im, cv2.COLOR_BGR2RGB))
+        sample = self.torch_transforms(im)
+        return {"img": sample, "cls": j}
+
+    def __len__(self) -> int:
+        """Return the total number of samples in the dataset."""
+        return len(self.samples)
+
+    def verify_images(self):
+        """Verify all images in dataset."""
+        desc = f"{self.prefix}Scanning {self.root}..."
+        path = Path(self.root).with_suffix(".cache")  # *.cache file path
+
+        try:
+            cache = load_dataset_cache_file(path)  # attempt to load a *.cache file
+            assert cache["version"] == DATASET_CACHE_VERSION  # matches current version
+            assert cache["hash"] == get_hash([x[0] for x in self.samples])  # identical hash
+            nf, nc, n, samples = cache.pop("results")  # found, missing, empty, corrupt, total
+            if LOCAL_RANK in {-1, 0}:
+                d = f"{desc} {nf} images, {nc} corrupt"
+                TQDM(None, desc=d, total=n, initial=n)
+                if cache["msgs"]:
+                    LOGGER.info("\n".join(cache["msgs"]))  # display warnings
+            return samples
+
+        except (FileNotFoundError, AssertionError, AttributeError):
+            # Run scan if *.cache retrieval failed
+            nf, nc, msgs, samples, x = 0, 0, [], [], {}
+            with ThreadPool(NUM_THREADS) as pool:
+                results = pool.imap(func=verify_image, iterable=zip(self.samples, repeat(self.prefix)))
+                pbar = TQDM(results, desc=desc, total=len(self.samples))
+                for sample, nf_f, nc_f, msg in pbar:
+                    if nf_f:
+                        samples.append(sample)
+                    if msg:
+                        msgs.append(msg)
+                    nf += nf_f
+                    nc += nc_f
+                    pbar.desc = f"{desc} {nf} images, {nc} corrupt"
+                pbar.close()
+            if msgs:
+                LOGGER.info("\n".join(msgs))
+            x["hash"] = get_hash([x[0] for x in self.samples])
+            x["results"] = nf, nc, len(samples), samples
+            x["msgs"] = msgs  # warnings
+            save_dataset_cache_file(self.prefix, path, x, DATASET_CACHE_VERSION)
+            return samples

ultralytics/data/loaders.py

@@ -0,0 +1,658 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+import glob
+import math
+import os
+import time
+from dataclasses import dataclass
+from pathlib import Path
+from threading import Thread
+from urllib.parse import urlparse
+
+import cv2
+import numpy as np
+import requests
+import torch
+from PIL import Image
+
+from ultralytics.data.utils import FORMATS_HELP_MSG, IMG_FORMATS, VID_FORMATS
+from ultralytics.utils import IS_COLAB, IS_KAGGLE, LOGGER, ops
+from ultralytics.utils.checks import check_requirements
+from ultralytics.utils.patches import imread
+
+
+@dataclass
+class SourceTypes:
+    """
+    Class to represent various types of input sources for predictions.
+
+    This class uses dataclass to define boolean flags for different types of input sources that can be used for
+    making predictions with YOLO models.
+
+    Attributes:
+        stream (bool): Flag indicating if the input source is a video stream.
+        screenshot (bool): Flag indicating if the input source is a screenshot.
+        from_img (bool): Flag indicating if the input source is an image file.
+
+    Examples:
+        >>> source_types = SourceTypes(stream=True, screenshot=False, from_img=False)
+        >>> print(source_types.stream)
+        True
+        >>> print(source_types.from_img)
+        False
+    """
+
+    stream: bool = False
+    screenshot: bool = False
+    from_img: bool = False
+    tensor: bool = False
+
+
+class LoadStreams:
+    """
+    Stream Loader for various types of video streams.
+
+    Supports RTSP, RTMP, HTTP, and TCP streams. This class handles the loading and processing of multiple video
+    streams simultaneously, making it suitable for real-time video analysis tasks.
+
+    Attributes:
+        sources (List[str]): The source input paths or URLs for the video streams.
+        vid_stride (int): Video frame-rate stride.
+        buffer (bool): Whether to buffer input streams.
+        running (bool): Flag to indicate if the streaming thread is running.
+        mode (str): Set to 'stream' indicating real-time capture.
+        imgs (List[List[np.ndarray]]): List of image frames for each stream.
+        fps (List[float]): List of FPS for each stream.
+        frames (List[int]): List of total frames for each stream.
+        threads (List[Thread]): List of threads for each stream.
+        shape (List[Tuple[int, int, int]]): List of shapes for each stream.
+        caps (List[cv2.VideoCapture]): List of cv2.VideoCapture objects for each stream.
+        bs (int): Batch size for processing.
+
+    Methods:
+        update: Read stream frames in daemon thread.
+        close: Close stream loader and release resources.
+        __iter__: Returns an iterator object for the class.
+        __next__: Returns source paths, transformed, and original images for processing.
+        __len__: Return the length of the sources object.
+
+    Examples:
+        >>> stream_loader = LoadStreams("rtsp://example.com/stream1.mp4")
+        >>> for sources, imgs, _ in stream_loader:
+        ...     # Process the images
+        ...     pass
+        >>> stream_loader.close()
+
+    Notes:
+        - The class uses threading to efficiently load frames from multiple streams simultaneously.
+        - It automatically handles YouTube links, converting them to the best available stream URL.
+        - The class implements a buffer system to manage frame storage and retrieval.
+    """
+
+    def __init__(self, sources="file.streams", vid_stride=1, buffer=False):
+        """Initialize stream loader for multiple video sources, supporting various stream types."""
+        torch.backends.cudnn.benchmark = True  # faster for fixed-size inference
+        self.buffer = buffer  # buffer input streams
+        self.running = True  # running flag for Thread
+        self.mode = "stream"
+        self.vid_stride = vid_stride  # video frame-rate stride
+
+        sources = Path(sources).read_text().rsplit() if os.path.isfile(sources) else [sources]
+        n = len(sources)
+        self.bs = n
+        self.fps = [0] * n  # frames per second
+        self.frames = [0] * n
+        self.threads = [None] * n
+        self.caps = [None] * n  # video capture objects
+        self.imgs = [[] for _ in range(n)]  # images
+        self.shape = [[] for _ in range(n)]  # image shapes
+        self.sources = [ops.clean_str(x) for x in sources]  # clean source names for later
+        for i, s in enumerate(sources):  # index, source
+            # Start thread to read frames from video stream
+            st = f"{i + 1}/{n}: {s}... "
+            if urlparse(s).hostname in {"www.youtube.com", "youtube.com", "youtu.be"}:  # if source is YouTube video
+                # YouTube format i.e. 'https://www.youtube.com/watch?v=Jsn8D3aC840' or 'https://youtu.be/Jsn8D3aC840'
+                s = get_best_youtube_url(s)
+            s = eval(s) if s.isnumeric() else s  # i.e. s = '0' local webcam
+            if s == 0 and (IS_COLAB or IS_KAGGLE):
+                raise NotImplementedError(
+                    "'source=0' webcam not supported in Colab and Kaggle notebooks. "
+                    "Try running 'source=0' in a local environment."
+                )
+            self.caps[i] = cv2.VideoCapture(s)  # store video capture object
+            if not self.caps[i].isOpened():
+                raise ConnectionError(f"{st}Failed to open {s}")
+            w = int(self.caps[i].get(cv2.CAP_PROP_FRAME_WIDTH))
+            h = int(self.caps[i].get(cv2.CAP_PROP_FRAME_HEIGHT))
+            fps = self.caps[i].get(cv2.CAP_PROP_FPS)  # warning: may return 0 or nan
+            self.frames[i] = max(int(self.caps[i].get(cv2.CAP_PROP_FRAME_COUNT)), 0) or float(
+                "inf"
+            )  # infinite stream fallback
+            self.fps[i] = max((fps if math.isfinite(fps) else 0) % 100, 0) or 30  # 30 FPS fallback
+
+            success, im = self.caps[i].read()  # guarantee first frame
+            if not success or im is None:
+                raise ConnectionError(f"{st}Failed to read images from {s}")
+            self.imgs[i].append(im)
+            self.shape[i] = im.shape
+            self.threads[i] = Thread(target=self.update, args=([i, self.caps[i], s]), daemon=True)
+            LOGGER.info(f"{st}Success ✅ ({self.frames[i]} frames of shape {w}x{h} at {self.fps[i]:.2f} FPS)")
+            self.threads[i].start()
+        LOGGER.info("")  # newline
+
+    def update(self, i, cap, stream):
+        """Read stream frames in daemon thread and update image buffer."""
+        n, f = 0, self.frames[i]  # frame number, frame array
+        while self.running and cap.isOpened() and n < (f - 1):
+            if len(self.imgs[i]) < 30:  # keep a <=30-image buffer
+                n += 1
+                cap.grab()  # .read() = .grab() followed by .retrieve()
+                if n % self.vid_stride == 0:
+                    success, im = cap.retrieve()
+                    if not success:
+                        im = np.zeros(self.shape[i], dtype=np.uint8)
+                        LOGGER.warning("WARNING ⚠️ Video stream unresponsive, please check your IP camera connection.")
+                        cap.open(stream)  # re-open stream if signal was lost
+                    if self.buffer:
+                        self.imgs[i].append(im)
+                    else:
+                        self.imgs[i] = [im]
+            else:
+                time.sleep(0.01)  # wait until the buffer is empty
+
+    def close(self):
+        """Terminates stream loader, stops threads, and releases video capture resources."""
+        self.running = False  # stop flag for Thread
+        for thread in self.threads:
+            if thread.is_alive():
+                thread.join(timeout=5)  # Add timeout
+        for cap in self.caps:  # Iterate through the stored VideoCapture objects
+            try:
+                cap.release()  # release video capture
+            except Exception as e:
+                LOGGER.warning(f"WARNING ⚠️ Could not release VideoCapture object: {e}")
+        cv2.destroyAllWindows()
+
+    def __iter__(self):
+        """Iterates through YOLO image feed and re-opens unresponsive streams."""
+        self.count = -1
+        return self
+
+    def __next__(self):
+        """Returns the next batch of frames from multiple video streams for processing."""
+        self.count += 1
+
+        images = []
+        for i, x in enumerate(self.imgs):
+            # Wait until a frame is available in each buffer
+            while not x:
+                if not self.threads[i].is_alive() or cv2.waitKey(1) == ord("q"):  # q to quit
+                    self.close()
+                    raise StopIteration
+                time.sleep(1 / min(self.fps))
+                x = self.imgs[i]
+                if not x:
+                    LOGGER.warning(f"WARNING ⚠️ Waiting for stream {i}")
+
+            # Get and remove the first frame from imgs buffer
+            if self.buffer:
+                images.append(x.pop(0))
+
+            # Get the last frame, and clear the rest from the imgs buffer
+            else:
+                images.append(x.pop(-1) if x else np.zeros(self.shape[i], dtype=np.uint8))
+                x.clear()
+
+        return self.sources, images, [""] * self.bs
+
+    def __len__(self):
+        """Return the number of video streams in the LoadStreams object."""
+        return self.bs  # 1E12 frames = 32 streams at 30 FPS for 30 years
+
+
+class LoadScreenshots:
+    """
+    Ultralytics screenshot dataloader for capturing and processing screen images.
+
+    This class manages the loading of screenshot images for processing with YOLO. It is suitable for use with
+    `yolo predict source=screen`.
+
+    Attributes:
+        source (str): The source input indicating which screen to capture.
+        screen (int): The screen number to capture.
+        left (int): The left coordinate for screen capture area.
+        top (int): The top coordinate for screen capture area.
+        width (int): The width of the screen capture area.
+        height (int): The height of the screen capture area.
+        mode (str): Set to 'stream' indicating real-time capture.
+        frame (int): Counter for captured frames.
+        sct (mss.mss): Screen capture object from `mss` library.
+        bs (int): Batch size, set to 1.
+        fps (int): Frames per second, set to 30.
+        monitor (Dict[str, int]): Monitor configuration details.
+
+    Methods:
+        __iter__: Returns an iterator object.
+        __next__: Captures the next screenshot and returns it.
+
+    Examples:
+        >>> loader = LoadScreenshots("0 100 100 640 480")  # screen 0, top-left (100,100), 640x480
+        >>> for source, im, im0s, vid_cap, s in loader:
+        ...     print(f"Captured frame: {im.shape}")
+    """
+
+    def __init__(self, source):
+        """Initialize screenshot capture with specified screen and region parameters."""
+        check_requirements("mss")
+        import mss  # noqa
+
+        source, *params = source.split()
+        self.screen, left, top, width, height = 0, None, None, None, None  # default to full screen 0
+        if len(params) == 1:
+            self.screen = int(params[0])
+        elif len(params) == 4:
+            left, top, width, height = (int(x) for x in params)
+        elif len(params) == 5:
+            self.screen, left, top, width, height = (int(x) for x in params)
+        self.mode = "stream"
+        self.frame = 0
+        self.sct = mss.mss()
+        self.bs = 1
+        self.fps = 30
+
+        # Parse monitor shape
+        monitor = self.sct.monitors[self.screen]
+        self.top = monitor["top"] if top is None else (monitor["top"] + top)
+        self.left = monitor["left"] if left is None else (monitor["left"] + left)
+        self.width = width or monitor["width"]
+        self.height = height or monitor["height"]
+        self.monitor = {"left": self.left, "top": self.top, "width": self.width, "height": self.height}
+
+    def __iter__(self):
+        """Yields the next screenshot image from the specified screen or region for processing."""
+        return self
+
+    def __next__(self):
+        """Captures and returns the next screenshot as a numpy array using the mss library."""
+        im0 = np.asarray(self.sct.grab(self.monitor))[:, :, :3]  # BGRA to BGR
+        s = f"screen {self.screen} (LTWH): {self.left},{self.top},{self.width},{self.height}: "
+
+        self.frame += 1
+        return [str(self.screen)], [im0], [s]  # screen, img, string
+
+
+class LoadImagesAndVideos:
+    """
+    A class for loading and processing images and videos for YOLO object detection.
+
+    This class manages the loading and pre-processing of image and video data from various sources, including
+    single image files, video files, and lists of image and video paths.
+
+    Attributes:
+        files (List[str]): List of image and video file paths.
+        nf (int): Total number of files (images and videos).
+        video_flag (List[bool]): Flags indicating whether a file is a video (True) or an image (False).
+        mode (str): Current mode, 'image' or 'video'.
+        vid_stride (int): Stride for video frame-rate.
+        bs (int): Batch size.
+        cap (cv2.VideoCapture): Video capture object for OpenCV.
+        frame (int): Frame counter for video.
+        frames (int): Total number of frames in the video.
+        count (int): Counter for iteration, initialized at 0 during __iter__().
+        ni (int): Number of images.
+
+    Methods:
+        __init__: Initialize the LoadImagesAndVideos object.
+        __iter__: Returns an iterator object for VideoStream or ImageFolder.
+        __next__: Returns the next batch of images or video frames along with their paths and metadata.
+        _new_video: Creates a new video capture object for the given path.
+        __len__: Returns the number of batches in the object.
+
+    Examples:
+        >>> loader = LoadImagesAndVideos("path/to/data", batch=32, vid_stride=1)
+        >>> for paths, imgs, info in loader:
+        ...     # Process batch of images or video frames
+        ...     pass
+
+    Notes:
+        - Supports various image formats including HEIC.
+        - Handles both local files and directories.
+        - Can read from a text file containing paths to images and videos.
+    """
+
+    def __init__(self, path, batch=1, vid_stride=1):
+        """Initialize dataloader for images and videos, supporting various input formats."""
+        parent = None
+        if isinstance(path, str) and Path(path).suffix == ".txt":  # *.txt file with img/vid/dir on each line
+            parent = Path(path).parent
+            path = Path(path).read_text().splitlines()  # list of sources
+        files = []
+        for p in sorted(path) if isinstance(path, (list, tuple)) else [path]:
+            a = str(Path(p).absolute())  # do not use .resolve() https://github.com/ultralytics/ultralytics/issues/2912
+            if "*" in a:
+                files.extend(sorted(glob.glob(a, recursive=True)))  # glob
+            elif os.path.isdir(a):
+                files.extend(sorted(glob.glob(os.path.join(a, "*.*"))))  # dir
+            elif os.path.isfile(a):
+                files.append(a)  # files (absolute or relative to CWD)
+            elif parent and (parent / p).is_file():
+                files.append(str((parent / p).absolute()))  # files (relative to *.txt file parent)
+            else:
+                raise FileNotFoundError(f"{p} does not exist")
+
+        # Define files as images or videos
+        images, videos = [], []
+        for f in files:
+            suffix = f.split(".")[-1].lower()  # Get file extension without the dot and lowercase
+            if suffix in IMG_FORMATS:
+                images.append(f)
+            elif suffix in VID_FORMATS:
+                videos.append(f)
+        ni, nv = len(images), len(videos)
+
+        self.files = images + videos
+        self.nf = ni + nv  # number of files
+        self.ni = ni  # number of images
+        self.video_flag = [False] * ni + [True] * nv
+        self.mode = "video" if ni == 0 else "image"  # default to video if no images
+        self.vid_stride = vid_stride  # video frame-rate stride
+        self.bs = batch
+        if any(videos):
+            self._new_video(videos[0])  # new video
+        else:
+            self.cap = None
+        if self.nf == 0:
+            raise FileNotFoundError(f"No images or videos found in {p}. {FORMATS_HELP_MSG}")
+
+    def __iter__(self):
+        """Iterates through image/video files, yielding source paths, images, and metadata."""
+        self.count = 0
+        return self
+
+    def __next__(self):
+        """Returns the next batch of images or video frames with their paths and metadata."""
+        paths, imgs, info = [], [], []
+        while len(imgs) < self.bs:
+            if self.count >= self.nf:  # end of file list
+                if imgs:
+                    return paths, imgs, info  # return last partial batch
+                else:
+                    raise StopIteration
+
+            path = self.files[self.count]
+            if self.video_flag[self.count]:
+                self.mode = "video"
+                if not self.cap or not self.cap.isOpened():
+                    self._new_video(path)
+
+                success = False
+                for _ in range(self.vid_stride):
+                    success = self.cap.grab()
+                    if not success:
+                        break  # end of video or failure
+
+                if success:
+                    success, im0 = self.cap.retrieve()
+                    if success:
+                        self.frame += 1
+                        paths.append(path)
+                        imgs.append(im0)
+                        info.append(f"video {self.count + 1}/{self.nf} (frame {self.frame}/{self.frames}) {path}: ")
+                        if self.frame == self.frames:  # end of video
+                            self.count += 1
+                            self.cap.release()
+                else:
+                    # Move to the next file if the current video ended or failed to open
+                    self.count += 1
+                    if self.cap:
+                        self.cap.release()
+                    if self.count < self.nf:
+                        self._new_video(self.files[self.count])
+            else:
+                # Handle image files (including HEIC)
+                self.mode = "image"
+                if path.split(".")[-1].lower() == "heic":
+                    # Load HEIC image using Pillow with pillow-heif
+                    check_requirements("pillow-heif")
+
+                    from pillow_heif import register_heif_opener
+
+                    register_heif_opener()  # Register HEIF opener with Pillow
+                    with Image.open(path) as img:
+                        im0 = cv2.cvtColor(np.asarray(img), cv2.COLOR_RGB2BGR)  # convert image to BGR nparray
+                else:
+                    im0 = imread(path)  # BGR
+                if im0 is None:
+                    LOGGER.warning(f"WARNING ⚠️ Image Read Error {path}")
+                else:
+                    paths.append(path)
+                    imgs.append(im0)
+                    info.append(f"image {self.count + 1}/{self.nf} {path}: ")
+                self.count += 1  # move to the next file
+                if self.count >= self.ni:  # end of image list
+                    break
+
+        return paths, imgs, info
+
+    def _new_video(self, path):
+        """Creates a new video capture object for the given path and initializes video-related attributes."""
+        self.frame = 0
+        self.cap = cv2.VideoCapture(path)
+        self.fps = int(self.cap.get(cv2.CAP_PROP_FPS))
+        if not self.cap.isOpened():
+            raise FileNotFoundError(f"Failed to open video {path}")
+        self.frames = int(self.cap.get(cv2.CAP_PROP_FRAME_COUNT) / self.vid_stride)
+
+    def __len__(self):
+        """Returns the number of files (images and videos) in the dataset."""
+        return math.ceil(self.nf / self.bs)  # number of batches
+
+
+class LoadPilAndNumpy:
+    """
+    Load images from PIL and Numpy arrays for batch processing.
+
+    This class manages loading and pre-processing of image data from both PIL and Numpy formats. It performs basic
+    validation and format conversion to ensure that the images are in the required format for downstream processing.
+
+    Attributes:
+        paths (List[str]): List of image paths or autogenerated filenames.
+        im0 (List[np.ndarray]): List of images stored as Numpy arrays.
+        mode (str): Type of data being processed, set to 'image'.
+        bs (int): Batch size, equivalent to the length of `im0`.
+
+    Methods:
+        _single_check: Validate and format a single image to a Numpy array.
+
+    Examples:
+        >>> from PIL import Image
+        >>> import numpy as np
+        >>> pil_img = Image.new("RGB", (100, 100))
+        >>> np_img = np.random.randint(0, 255, (100, 100, 3), dtype=np.uint8)
+        >>> loader = LoadPilAndNumpy([pil_img, np_img])
+        >>> paths, images, _ = next(iter(loader))
+        >>> print(f"Loaded {len(images)} images")
+        Loaded 2 images
+    """
+
+    def __init__(self, im0):
+        """Initializes a loader for PIL and Numpy images, converting inputs to a standardized format."""
+        if not isinstance(im0, list):
+            im0 = [im0]
+        # use `image{i}.jpg` when Image.filename returns an empty path.
+        self.paths = [getattr(im, "filename", "") or f"image{i}.jpg" for i, im in enumerate(im0)]
+        self.im0 = [self._single_check(im) for im in im0]
+        self.mode = "image"
+        self.bs = len(self.im0)
+
+    @staticmethod
+    def _single_check(im):
+        """Validate and format an image to numpy array, ensuring RGB order and contiguous memory."""
+        assert isinstance(im, (Image.Image, np.ndarray)), f"Expected PIL/np.ndarray image type, but got {type(im)}"
+        if isinstance(im, Image.Image):
+            if im.mode != "RGB":
+                im = im.convert("RGB")
+            im = np.asarray(im)[:, :, ::-1]
+            im = np.ascontiguousarray(im)  # contiguous
+        return im
+
+    def __len__(self):
+        """Returns the length of the 'im0' attribute, representing the number of loaded images."""
+        return len(self.im0)
+
+    def __next__(self):
+        """Returns the next batch of images, paths, and metadata for processing."""
+        if self.count == 1:  # loop only once as it's batch inference
+            raise StopIteration
+        self.count += 1
+        return self.paths, self.im0, [""] * self.bs
+
+    def __iter__(self):
+        """Iterates through PIL/numpy images, yielding paths, raw images, and metadata for processing."""
+        self.count = 0
+        return self
+
+
+class LoadTensor:
+    """
+    A class for loading and processing tensor data for object detection tasks.
+
+    This class handles the loading and pre-processing of image data from PyTorch tensors, preparing them for
+    further processing in object detection pipelines.
+
+    Attributes:
+        im0 (torch.Tensor): The input tensor containing the image(s) with shape (B, C, H, W).
+        bs (int): Batch size, inferred from the shape of `im0`.
+        mode (str): Current processing mode, set to 'image'.
+        paths (List[str]): List of image paths or auto-generated filenames.
+
+    Methods:
+        _single_check: Validates and formats an input tensor.
+
+    Examples:
+        >>> import torch
+        >>> tensor = torch.rand(1, 3, 640, 640)
+        >>> loader = LoadTensor(tensor)
+        >>> paths, images, info = next(iter(loader))
+        >>> print(f"Processed {len(images)} images")
+    """
+
+    def __init__(self, im0) -> None:
+        """Initialize LoadTensor object for processing torch.Tensor image data."""
+        self.im0 = self._single_check(im0)
+        self.bs = self.im0.shape[0]
+        self.mode = "image"
+        self.paths = [getattr(im, "filename", f"image{i}.jpg") for i, im in enumerate(im0)]
+
+    @staticmethod
+    def _single_check(im, stride=32):
+        """Validates and formats a single image tensor, ensuring correct shape and normalization."""
+        s = (
+            f"WARNING ⚠️ torch.Tensor inputs should be BCHW i.e. shape(1, 3, 640, 640) "
+            f"divisible by stride {stride}. Input shape{tuple(im.shape)} is incompatible."
+        )
+        if len(im.shape) != 4:
+            if len(im.shape) != 3:
+                raise ValueError(s)
+            LOGGER.warning(s)
+            im = im.unsqueeze(0)
+        if im.shape[2] % stride or im.shape[3] % stride:
+            raise ValueError(s)
+        if im.max() > 1.0 + torch.finfo(im.dtype).eps:  # torch.float32 eps is 1.2e-07
+            LOGGER.warning(
+                f"WARNING ⚠️ torch.Tensor inputs should be normalized 0.0-1.0 but max value is {im.max()}. "
+                f"Dividing input by 255."
+            )
+            im = im.float() / 255.0
+
+        return im
+
+    def __iter__(self):
+        """Yields an iterator object for iterating through tensor image data."""
+        self.count = 0
+        return self
+
+    def __next__(self):
+        """Yields the next batch of tensor images and metadata for processing."""
+        if self.count == 1:
+            raise StopIteration
+        self.count += 1
+        return self.paths, self.im0, [""] * self.bs
+
+    def __len__(self):
+        """Returns the batch size of the tensor input."""
+        return self.bs
+
+
+def autocast_list(source):
+    """Merges a list of sources into a list of numpy arrays or PIL images for Ultralytics prediction."""
+    files = []
+    for im in source:
+        if isinstance(im, (str, Path)):  # filename or uri
+            files.append(Image.open(requests.get(im, stream=True).raw if str(im).startswith("http") else im))
+        elif isinstance(im, (Image.Image, np.ndarray)):  # PIL or np Image
+            files.append(im)
+        else:
+            raise TypeError(
+                f"type {type(im).__name__} is not a supported Ultralytics prediction source type. \n"
+                f"See https://docs.ultralytics.com/modes/predict for supported source types."
+            )
+
+    return files
+
+
+def get_best_youtube_url(url, method="pytube"):
+    """
+    Retrieves the URL of the best quality MP4 video stream from a given YouTube video.
+
+    Args:
+        url (str): The URL of the YouTube video.
+        method (str): The method to use for extracting video info. Options are "pytube", "pafy", and "yt-dlp".
+            Defaults to "pytube".
+
+    Returns:
+        (str | None): The URL of the best quality MP4 video stream, or None if no suitable stream is found.
+
+    Examples:
+        >>> url = "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
+        >>> best_url = get_best_youtube_url(url)
+        >>> print(best_url)
+        https://rr4---sn-q4flrnek.googlevideo.com/videoplayback?expire=...
+
+    Notes:
+        - Requires additional libraries based on the chosen method: pytubefix, pafy, or yt-dlp.
+        - The function prioritizes streams with at least 1080p resolution when available.
+        - For the "yt-dlp" method, it looks for formats with video codec, no audio, and *.mp4 extension.
+    """
+    if method == "pytube":
+        # Switched from pytube to pytubefix to resolve https://github.com/pytube/pytube/issues/1954
+        check_requirements("pytubefix>=6.5.2")
+        from pytubefix import YouTube
+
+        streams = YouTube(url).streams.filter(file_extension="mp4", only_video=True)
+        streams = sorted(streams, key=lambda s: s.resolution, reverse=True)  # sort streams by resolution
+        for stream in streams:
+            if stream.resolution and int(stream.resolution[:-1]) >= 1080:  # check if resolution is at least 1080p
+                return stream.url
+
+    elif method == "pafy":
+        check_requirements(("pafy", "youtube_dl==2020.12.2"))
+        import pafy  # noqa
+
+        return pafy.new(url).getbestvideo(preftype="mp4").url
+
+    elif method == "yt-dlp":
+        check_requirements("yt-dlp")
+        import yt_dlp
+
+        with yt_dlp.YoutubeDL({"quiet": True}) as ydl:
+            info_dict = ydl.extract_info(url, download=False)  # extract info
+        for f in reversed(info_dict.get("formats", [])):  # reversed because best is usually last
+            # Find a format with video codec, no audio, *.mp4 extension at least 1920x1080 size
+            good_size = (f.get("width") or 0) >= 1920 or (f.get("height") or 0) >= 1080
+            if good_size and f["vcodec"] != "none" and f["acodec"] == "none" and f["ext"] == "mp4":
+                return f.get("url")
+
+
+# Define constants
+LOADERS = (LoadStreams, LoadPilAndNumpy, LoadImagesAndVideos, LoadScreenshots)

ultralytics/data/scripts/download_weights.sh

@@ -0,0 +1,18 @@
+#!/bin/bash
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+# Download latest models from https://github.com/ultralytics/assets/releases
+# Example usage: bash ultralytics/data/scripts/download_weights.sh
+# parent
+# └── weights
+#     ├── yolov8n.pt  ← downloads here
+#     ├── yolov8s.pt
+#     └── ...
+
+python - <<EOF
+from ultralytics.utils.downloads import attempt_download_asset
+
+assets = [f"yolov8{size}{suffix}.pt" for size in "nsmlx" for suffix in ("", "-cls", "-seg", "-pose")]
+for x in assets:
+    attempt_download_asset(f"weights/{x}")
+
+EOF

ultralytics/data/scripts/get_coco.sh

@@ -0,0 +1,60 @@
+#!/bin/bash
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+# Download COCO 2017 dataset https://cocodataset.org
+# Example usage: bash data/scripts/get_coco.sh
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco  ← downloads here
+
+# Arguments (optional) Usage: bash data/scripts/get_coco.sh --train --val --test --segments
+if [ "$#" -gt 0 ]; then
+  for opt in "$@"; do
+    case "${opt}" in
+    --train) train=true ;;
+    --val) val=true ;;
+    --test) test=true ;;
+    --segments) segments=true ;;
+    --sama) sama=true ;;
+    esac
+  done
+else
+  train=true
+  val=true
+  test=false
+  segments=false
+  sama=false
+fi
+
+# Download/unzip labels
+d='../datasets' # unzip directory
+url=https://github.com/ultralytics/assets/releases/download/v0.0.0/
+if [ "$segments" == "true" ]; then
+  f='coco2017labels-segments.zip' # 169 MB
+elif [ "$sama" == "true" ]; then
+  f='coco2017labels-segments-sama.zip' # 199 MB https://www.sama.com/sama-coco-dataset/
+else
+  f='coco2017labels.zip' # 46 MB
+fi
+echo 'Downloading' $url$f ' ...'
+curl -L $url$f -o $f -# && unzip -q $f -d $d && rm $f &
+
+# Download/unzip images
+d='../datasets/coco/images' # unzip directory
+url=http://images.cocodataset.org/zips/
+if [ "$train" == "true" ]; then
+  f='train2017.zip' # 19G, 118k images
+  echo 'Downloading' $url$f '...'
+  curl -L $url$f -o $f -# && unzip -q $f -d $d && rm $f &
+fi
+if [ "$val" == "true" ]; then
+  f='val2017.zip' # 1G, 5k images
+  echo 'Downloading' $url$f '...'
+  curl -L $url$f -o $f -# && unzip -q $f -d $d && rm $f &
+fi
+if [ "$test" == "true" ]; then
+  f='test2017.zip' # 7G, 41k images (optional)
+  echo 'Downloading' $url$f '...'
+  curl -L $url$f -o $f -# && unzip -q $f -d $d && rm $f &
+fi
+wait # finish background tasks

ultralytics/data/scripts/get_coco128.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+# Download COCO128 dataset https://www.kaggle.com/ultralytics/coco128 (first 128 images from COCO train2017)
+# Example usage: bash data/scripts/get_coco128.sh
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── coco128  ← downloads here
+
+# Download/unzip images and labels
+d='../datasets' # unzip directory
+url=https://github.com/ultralytics/assets/releases/download/v0.0.0/
+f='coco128.zip' # or 'coco128-segments.zip', 68 MB
+echo 'Downloading' $url$f ' ...'
+curl -L $url$f -o $f -# && unzip -q $f -d $d && rm $f &
+
+wait # finish background tasks

ultralytics/data/scripts/get_imagenet.sh

@@ -0,0 +1,51 @@
+#!/bin/bash
+# Ultralytics YOLO 🚀, AGPL-3.0 license
+# Download ILSVRC2012 ImageNet dataset https://image-net.org
+# Example usage: bash data/scripts/get_imagenet.sh
+# parent
+# ├── ultralytics
+# └── datasets
+#     └── imagenet  ← downloads here
+
+# Arguments (optional) Usage: bash data/scripts/get_imagenet.sh --train --val
+if [ "$#" -gt 0 ]; then
+  for opt in "$@"; do
+    case "${opt}" in
+    --train) train=true ;;
+    --val) val=true ;;
+    esac
+  done
+else
+  train=true
+  val=true
+fi
+
+# Make dir
+d='../datasets/imagenet' # unzip directory
+mkdir -p $d && cd $d
+
+# Download/unzip train
+if [ "$train" == "true" ]; then
+  wget https://image-net.org/data/ILSVRC/2012/ILSVRC2012_img_train.tar # download 138G, 1281167 images
+  mkdir train && mv ILSVRC2012_img_train.tar train/ && cd train
+  tar -xf ILSVRC2012_img_train.tar && rm -f ILSVRC2012_img_train.tar
+  find . -name "*.tar" | while read NAME; do
+    mkdir -p "${NAME%.tar}"
+    tar -xf "${NAME}" -C "${NAME%.tar}"
+    rm -f "${NAME}"
+  done
+  cd ..
+fi
+
+# Download/unzip val
+if [ "$val" == "true" ]; then
+  wget https://image-net.org/data/ILSVRC/2012/ILSVRC2012_img_val.tar # download 6.3G, 50000 images
+  mkdir val && mv ILSVRC2012_img_val.tar val/ && cd val && tar -xf ILSVRC2012_img_val.tar
+  wget -qO- https://raw.githubusercontent.com/soumith/imagenetloader.torch/master/valprep.sh | bash # move into subdirs
+fi
+
+# Delete corrupted image (optional: PNG under JPEG name that may cause dataloaders to fail)
+# rm train/n04266014/n04266014_10835.JPEG
+
+# TFRecords (optional)
+# wget https://raw.githubusercontent.com/tensorflow/models/master/research/slim/datasets/imagenet_lsvrc_2015_synsets.txt

ultralytics/data/split_dota.py

@@ -0,0 +1,298 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+import itertools
+from glob import glob
+from math import ceil
+from pathlib import Path
+
+import cv2
+import numpy as np
+from PIL import Image
+from tqdm import tqdm
+
+from ultralytics.data.utils import exif_size, img2label_paths
+from ultralytics.utils.checks import check_requirements
+
+
+def bbox_iof(polygon1, bbox2, eps=1e-6):
+    """
+    Calculate Intersection over Foreground (IoF) between polygons and bounding boxes.
+
+    Args:
+        polygon1 (np.ndarray): Polygon coordinates, shape (n, 8).
+        bbox2 (np.ndarray): Bounding boxes, shape (n, 4).
+        eps (float, optional): Small value to prevent division by zero. Defaults to 1e-6.
+
+    Returns:
+        (np.ndarray): IoF scores, shape (n, 1) or (n, m) if bbox2 is (m, 4).
+
+    Note:
+        Polygon format: [x1, y1, x2, y2, x3, y3, x4, y4].
+        Bounding box format: [x_min, y_min, x_max, y_max].
+    """
+    check_requirements("shapely")
+    from shapely.geometry import Polygon
+
+    polygon1 = polygon1.reshape(-1, 4, 2)
+    lt_point = np.min(polygon1, axis=-2)  # left-top
+    rb_point = np.max(polygon1, axis=-2)  # right-bottom
+    bbox1 = np.concatenate([lt_point, rb_point], axis=-1)
+
+    lt = np.maximum(bbox1[:, None, :2], bbox2[..., :2])
+    rb = np.minimum(bbox1[:, None, 2:], bbox2[..., 2:])
+    wh = np.clip(rb - lt, 0, np.inf)
+    h_overlaps = wh[..., 0] * wh[..., 1]
+
+    left, top, right, bottom = (bbox2[..., i] for i in range(4))
+    polygon2 = np.stack([left, top, right, top, right, bottom, left, bottom], axis=-1).reshape(-1, 4, 2)
+
+    sg_polys1 = [Polygon(p) for p in polygon1]
+    sg_polys2 = [Polygon(p) for p in polygon2]
+    overlaps = np.zeros(h_overlaps.shape)
+    for p in zip(*np.nonzero(h_overlaps)):
+        overlaps[p] = sg_polys1[p[0]].intersection(sg_polys2[p[-1]]).area
+    unions = np.array([p.area for p in sg_polys1], dtype=np.float32)
+    unions = unions[..., None]
+
+    unions = np.clip(unions, eps, np.inf)
+    outputs = overlaps / unions
+    if outputs.ndim == 1:
+        outputs = outputs[..., None]
+    return outputs
+
+
+def load_yolo_dota(data_root, split="train"):
+    """
+    Load DOTA dataset.
+
+    Args:
+        data_root (str): Data root.
+        split (str): The split data set, could be `train` or `val`.
+
+    Notes:
+        The directory structure assumed for the DOTA dataset:
+            - data_root
+                - images
+                    - train
+                    - val
+                - labels
+                    - train
+                    - val
+    """
+    assert split in {"train", "val"}, f"Split must be 'train' or 'val', not {split}."
+    im_dir = Path(data_root) / "images" / split
+    assert im_dir.exists(), f"Can't find {im_dir}, please check your data root."
+    im_files = glob(str(Path(data_root) / "images" / split / "*"))
+    lb_files = img2label_paths(im_files)
+    annos = []
+    for im_file, lb_file in zip(im_files, lb_files):
+        w, h = exif_size(Image.open(im_file))
+        with open(lb_file) as f:
+            lb = [x.split() for x in f.read().strip().splitlines() if len(x)]
+            lb = np.array(lb, dtype=np.float32)
+        annos.append(dict(ori_size=(h, w), label=lb, filepath=im_file))
+    return annos
+
+
+def get_windows(im_size, crop_sizes=(1024,), gaps=(200,), im_rate_thr=0.6, eps=0.01):
+    """
+    Get the coordinates of windows.
+
+    Args:
+        im_size (tuple): Original image size, (h, w).
+        crop_sizes (List(int)): Crop size of windows.
+        gaps (List(int)): Gap between crops.
+        im_rate_thr (float): Threshold of windows areas divided by image ares.
+        eps (float): Epsilon value for math operations.
+    """
+    h, w = im_size
+    windows = []
+    for crop_size, gap in zip(crop_sizes, gaps):
+        assert crop_size > gap, f"invalid crop_size gap pair [{crop_size} {gap}]"
+        step = crop_size - gap
+
+        xn = 1 if w <= crop_size else ceil((w - crop_size) / step + 1)
+        xs = [step * i for i in range(xn)]
+        if len(xs) > 1 and xs[-1] + crop_size > w:
+            xs[-1] = w - crop_size
+
+        yn = 1 if h <= crop_size else ceil((h - crop_size) / step + 1)
+        ys = [step * i for i in range(yn)]
+        if len(ys) > 1 and ys[-1] + crop_size > h:
+            ys[-1] = h - crop_size
+
+        start = np.array(list(itertools.product(xs, ys)), dtype=np.int64)
+        stop = start + crop_size
+        windows.append(np.concatenate([start, stop], axis=1))
+    windows = np.concatenate(windows, axis=0)
+
+    im_in_wins = windows.copy()
+    im_in_wins[:, 0::2] = np.clip(im_in_wins[:, 0::2], 0, w)
+    im_in_wins[:, 1::2] = np.clip(im_in_wins[:, 1::2], 0, h)
+    im_areas = (im_in_wins[:, 2] - im_in_wins[:, 0]) * (im_in_wins[:, 3] - im_in_wins[:, 1])
+    win_areas = (windows[:, 2] - windows[:, 0]) * (windows[:, 3] - windows[:, 1])
+    im_rates = im_areas / win_areas
+    if not (im_rates > im_rate_thr).any():
+        max_rate = im_rates.max()
+        im_rates[abs(im_rates - max_rate) < eps] = 1
+    return windows[im_rates > im_rate_thr]
+
+
+def get_window_obj(anno, windows, iof_thr=0.7):
+    """Get objects for each window."""
+    h, w = anno["ori_size"]
+    label = anno["label"]
+    if len(label):
+        label[:, 1::2] *= w
+        label[:, 2::2] *= h
+        iofs = bbox_iof(label[:, 1:], windows)
+        # Unnormalized and misaligned coordinates
+        return [(label[iofs[:, i] >= iof_thr]) for i in range(len(windows))]  # window_anns
+    else:
+        return [np.zeros((0, 9), dtype=np.float32) for _ in range(len(windows))]  # window_anns
+
+
+def crop_and_save(anno, windows, window_objs, im_dir, lb_dir, allow_background_images=True):
+    """
+    Crop images and save new labels.
+
+    Args:
+        anno (dict): Annotation dict, including `filepath`, `label`, `ori_size` as its keys.
+        windows (list): A list of windows coordinates.
+        window_objs (list): A list of labels inside each window.
+        im_dir (str): The output directory path of images.
+        lb_dir (str): The output directory path of labels.
+        allow_background_images (bool): Whether to include background images without labels.
+
+    Notes:
+        The directory structure assumed for the DOTA dataset:
+            - data_root
+                - images
+                    - train
+                    - val
+                - labels
+                    - train
+                    - val
+    """
+    im = cv2.imread(anno["filepath"])
+    name = Path(anno["filepath"]).stem
+    for i, window in enumerate(windows):
+        x_start, y_start, x_stop, y_stop = window.tolist()
+        new_name = f"{name}__{x_stop - x_start}__{x_start}___{y_start}"
+        patch_im = im[y_start:y_stop, x_start:x_stop]
+        ph, pw = patch_im.shape[:2]
+
+        label = window_objs[i]
+        if len(label) or allow_background_images:
+            cv2.imwrite(str(Path(im_dir) / f"{new_name}.jpg"), patch_im)
+        if len(label):
+            label[:, 1::2] -= x_start
+            label[:, 2::2] -= y_start
+            label[:, 1::2] /= pw
+            label[:, 2::2] /= ph
+
+            with open(Path(lb_dir) / f"{new_name}.txt", "w") as f:
+                for lb in label:
+                    formatted_coords = [f"{coord:.6g}" for coord in lb[1:]]
+                    f.write(f"{int(lb[0])} {' '.join(formatted_coords)}\n")
+
+
+def split_images_and_labels(data_root, save_dir, split="train", crop_sizes=(1024,), gaps=(200,)):
+    """
+    Split both images and labels.
+
+    Notes:
+        The directory structure assumed for the DOTA dataset:
+            - data_root
+                - images
+                    - split
+                - labels
+                    - split
+        and the output directory structure is:
+            - save_dir
+                - images
+                    - split
+                - labels
+                    - split
+    """
+    im_dir = Path(save_dir) / "images" / split
+    im_dir.mkdir(parents=True, exist_ok=True)
+    lb_dir = Path(save_dir) / "labels" / split
+    lb_dir.mkdir(parents=True, exist_ok=True)
+
+    annos = load_yolo_dota(data_root, split=split)
+    for anno in tqdm(annos, total=len(annos), desc=split):
+        windows = get_windows(anno["ori_size"], crop_sizes, gaps)
+        window_objs = get_window_obj(anno, windows)
+        crop_and_save(anno, windows, window_objs, str(im_dir), str(lb_dir))
+
+
+def split_trainval(data_root, save_dir, crop_size=1024, gap=200, rates=(1.0,)):
+    """
+    Split train and val set of DOTA.
+
+    Notes:
+        The directory structure assumed for the DOTA dataset:
+            - data_root
+                - images
+                    - train
+                    - val
+                - labels
+                    - train
+                    - val
+        and the output directory structure is:
+            - save_dir
+                - images
+                    - train
+                    - val
+                - labels
+                    - train
+                    - val
+    """
+    crop_sizes, gaps = [], []
+    for r in rates:
+        crop_sizes.append(int(crop_size / r))
+        gaps.append(int(gap / r))
+    for split in ["train", "val"]:
+        split_images_and_labels(data_root, save_dir, split, crop_sizes, gaps)
+
+
+def split_test(data_root, save_dir, crop_size=1024, gap=200, rates=(1.0,)):
+    """
+    Split test set of DOTA, labels are not included within this set.
+
+    Notes:
+        The directory structure assumed for the DOTA dataset:
+            - data_root
+                - images
+                    - test
+        and the output directory structure is:
+            - save_dir
+                - images
+                    - test
+    """
+    crop_sizes, gaps = [], []
+    for r in rates:
+        crop_sizes.append(int(crop_size / r))
+        gaps.append(int(gap / r))
+    save_dir = Path(save_dir) / "images" / "test"
+    save_dir.mkdir(parents=True, exist_ok=True)
+
+    im_dir = Path(data_root) / "images" / "test"
+    assert im_dir.exists(), f"Can't find {im_dir}, please check your data root."
+    im_files = glob(str(im_dir / "*"))
+    for im_file in tqdm(im_files, total=len(im_files), desc="test"):
+        w, h = exif_size(Image.open(im_file))
+        windows = get_windows((h, w), crop_sizes=crop_sizes, gaps=gaps)
+        im = cv2.imread(im_file)
+        name = Path(im_file).stem
+        for window in windows:
+            x_start, y_start, x_stop, y_stop = window.tolist()
+            new_name = f"{name}__{x_stop - x_start}__{x_start}___{y_start}"
+            patch_im = im[y_start:y_stop, x_start:x_stop]
+            cv2.imwrite(str(save_dir / f"{new_name}.jpg"), patch_im)
+
+
+if __name__ == "__main__":
+    split_trainval(data_root="DOTAv2", save_dir="DOTAv2-split")
+    split_test(data_root="DOTAv2", save_dir="DOTAv2-split")

ultralytics/data/utils.py

@@ -0,0 +1,721 @@
+# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
+
+import hashlib
+import json
+import os
+import random
+import subprocess
+import time
+import zipfile
+from multiprocessing.pool import ThreadPool
+from pathlib import Path
+from tarfile import is_tarfile
+
+import cv2
+import numpy as np
+from PIL import Image, ImageOps
+
+from ultralytics.nn.autobackend import check_class_names
+from ultralytics.utils import (
+    DATASETS_DIR,
+    LOGGER,
+    NUM_THREADS,
+    ROOT,
+    SETTINGS_FILE,
+    TQDM,
+    clean_url,
+    colorstr,
+    emojis,
+    is_dir_writeable,
+    yaml_load,
+    yaml_save,
+)
+from ultralytics.utils.checks import check_file, check_font, is_ascii
+from ultralytics.utils.downloads import download, safe_download, unzip_file
+from ultralytics.utils.ops import segments2boxes
+
+HELP_URL = "See https://docs.ultralytics.com/datasets for dataset formatting guidance."
+IMG_FORMATS = {"bmp", "dng", "jpeg", "jpg", "mpo", "png", "tif", "tiff", "webp", "pfm", "heic"}  # image suffixes
+VID_FORMATS = {"asf", "avi", "gif", "m4v", "mkv", "mov", "mp4", "mpeg", "mpg", "ts", "wmv", "webm"}  # video suffixes
+PIN_MEMORY = str(os.getenv("PIN_MEMORY", True)).lower() == "true"  # global pin_memory for dataloaders
+FORMATS_HELP_MSG = f"Supported formats are:\nimages: {IMG_FORMATS}\nvideos: {VID_FORMATS}"
+
+
+def img2label_paths(img_paths):
+    """Define label paths as a function of image paths."""
+    sa, sb = f"{os.sep}images{os.sep}", f"{os.sep}labels{os.sep}"  # /images/, /labels/ substrings
+    return [sb.join(x.rsplit(sa, 1)).rsplit(".", 1)[0] + ".txt" for x in img_paths]
+
+
+def get_hash(paths):
+    """Returns a single hash value of a list of paths (files or dirs)."""
+    size = sum(os.path.getsize(p) for p in paths if os.path.exists(p))  # sizes
+    h = hashlib.sha256(str(size).encode())  # hash sizes
+    h.update("".join(paths).encode())  # hash paths
+    return h.hexdigest()  # return hash
+
+
+def exif_size(img: Image.Image):
+    """Returns exif-corrected PIL size."""
+    s = img.size  # (width, height)
+    if img.format == "JPEG":  # only support JPEG images
+        try:
+            if exif := img.getexif():
+                rotation = exif.get(274, None)  # the EXIF key for the orientation tag is 274
+                if rotation in {6, 8}:  # rotation 270 or 90
+                    s = s[1], s[0]
+        except Exception:
+            pass
+    return s
+
+
+def verify_image(args):
+    """Verify one image."""
+    (im_file, cls), prefix = args
+    # Number (found, corrupt), message
+    nf, nc, msg = 0, 0, ""
+    try:
+        im = Image.open(im_file)
+        im.verify()  # PIL verify
+        shape = exif_size(im)  # image size
+        shape = (shape[1], shape[0])  # hw
+        assert (shape[0] > 9) & (shape[1] > 9), f"image size {shape} <10 pixels"
+        assert im.format.lower() in IMG_FORMATS, f"Invalid image format {im.format}. {FORMATS_HELP_MSG}"
+        if im.format.lower() in {"jpg", "jpeg"}:
+            with open(im_file, "rb") as f:
+                f.seek(-2, 2)
+                if f.read() != b"\xff\xd9":  # corrupt JPEG
+                    ImageOps.exif_transpose(Image.open(im_file)).save(im_file, "JPEG", subsampling=0, quality=100)
+                    msg = f"{prefix}WARNING ⚠️ {im_file}: corrupt JPEG restored and saved"
+        nf = 1
+    except Exception as e:
+        nc = 1
+        msg = f"{prefix}WARNING ⚠️ {im_file}: ignoring corrupt image/label: {e}"
+    return (im_file, cls), nf, nc, msg
+
+
+def verify_image_label(args):
+    """Verify one image-label pair."""
+    im_file, lb_file, prefix, keypoint, num_cls, nkpt, ndim = args
+    # Number (missing, found, empty, corrupt), message, segments, keypoints
+    nm, nf, ne, nc, msg, segments, keypoints = 0, 0, 0, 0, "", [], None
+    try:
+        # Verify images
+        im = Image.open(im_file)
+        im.verify()  # PIL verify
+        shape = exif_size(im)  # image size
+        shape = (shape[1], shape[0])  # hw
+        assert (shape[0] > 9) & (shape[1] > 9), f"image size {shape} <10 pixels"
+        assert im.format.lower() in IMG_FORMATS, f"invalid image format {im.format}. {FORMATS_HELP_MSG}"
+        if im.format.lower() in {"jpg", "jpeg"}:
+            with open(im_file, "rb") as f:
+                f.seek(-2, 2)
+                if f.read() != b"\xff\xd9":  # corrupt JPEG
+                    ImageOps.exif_transpose(Image.open(im_file)).save(im_file, "JPEG", subsampling=0, quality=100)
+                    msg = f"{prefix}WARNING ⚠️ {im_file}: corrupt JPEG restored and saved"
+
+        # Verify labels
+        if os.path.isfile(lb_file):
+            nf = 1  # label found
+            with open(lb_file) as f:
+                lb = [x.split() for x in f.read().strip().splitlines() if len(x)]
+                if any(len(x) > 6 for x in lb) and (not keypoint):  # is segment
+                    classes = np.array([x[0] for x in lb], dtype=np.float32)
+                    segments = [np.array(x[1:], dtype=np.float32).reshape(-1, 2) for x in lb]  # (cls, xy1...)
+                    lb = np.concatenate((classes.reshape(-1, 1), segments2boxes(segments)), 1)  # (cls, xywh)
+                lb = np.array(lb, dtype=np.float32)
+            if nl := len(lb):
+                if keypoint:
+                    assert lb.shape[1] == (5 + nkpt * ndim), f"labels require {(5 + nkpt * ndim)} columns each"
+                    points = lb[:, 5:].reshape(-1, ndim)[:, :2]
+                else:
+                    assert lb.shape[1] == 5, f"labels require 5 columns, {lb.shape[1]} columns detected"
+                    points = lb[:, 1:]
+                assert points.max() <= 1, f"non-normalized or out of bounds coordinates {points[points > 1]}"
+                assert lb.min() >= 0, f"negative label values {lb[lb < 0]}"
+
+                # All labels
+                max_cls = lb[:, 0].max()  # max label count
+                assert max_cls <= num_cls, (
+                    f"Label class {int(max_cls)} exceeds dataset class count {num_cls}. "
+                    f"Possible class labels are 0-{num_cls - 1}"
+                )
+                _, i = np.unique(lb, axis=0, return_index=True)
+                if len(i) < nl:  # duplicate row check
+                    lb = lb[i]  # remove duplicates
+                    if segments:
+                        segments = [segments[x] for x in i]
+                    msg = f"{prefix}WARNING ⚠️ {im_file}: {nl - len(i)} duplicate labels removed"
+            else:
+                ne = 1  # label empty
+                lb = np.zeros((0, (5 + nkpt * ndim) if keypoint else 5), dtype=np.float32)
+        else:
+            nm = 1  # label missing
+            lb = np.zeros((0, (5 + nkpt * ndim) if keypoints else 5), dtype=np.float32)
+        if keypoint:
+            keypoints = lb[:, 5:].reshape(-1, nkpt, ndim)
+            if ndim == 2:
+                kpt_mask = np.where((keypoints[..., 0] < 0) | (keypoints[..., 1] < 0), 0.0, 1.0).astype(np.float32)
+                keypoints = np.concatenate([keypoints, kpt_mask[..., None]], axis=-1)  # (nl, nkpt, 3)
+        lb = lb[:, :5]
+        return im_file, lb, shape, segments, keypoints, nm, nf, ne, nc, msg
+    except Exception as e:
+        nc = 1
+        msg = f"{prefix}WARNING ⚠️ {im_file}: ignoring corrupt image/label: {e}"
+        return [None, None, None, None, None, nm, nf, ne, nc, msg]
+
+
+def visualize_image_annotations(image_path, txt_path, label_map):
+    """
+    Visualizes YOLO annotations (bounding boxes and class labels) on an image.
+
+    This function reads an image and its corresponding annotation file in YOLO format, then
+    draws bounding boxes around detected objects and labels them with their respective class names.
+    The bounding box colors are assigned based on the class ID, and the text color is dynamically
+    adjusted for readability, depending on the background color's luminance.
+
+    Args:
+        image_path (str): The path to the image file to annotate, and it can be in formats supported by PIL (e.g., .jpg, .png).
+        txt_path (str): The path to the annotation file in YOLO format, that should contain one line per object with:
+                        - class_id (int): The class index.
+                        - x_center (float): The X center of the bounding box (relative to image width).
+                        - y_center (float): The Y center of the bounding box (relative to image height).
+                        - width (float): The width of the bounding box (relative to image width).
+                        - height (float): The height of the bounding box (relative to image height).
+        label_map (dict): A dictionary that maps class IDs (integers) to class labels (strings).
+
+    Example:
+        >>> label_map = {0: "cat", 1: "dog", 2: "bird"}  # It should include all annotated classes details
+        >>> visualize_image_annotations("path/to/image.jpg", "path/to/annotations.txt", label_map)
+    """
+    import matplotlib.pyplot as plt
+
+    from ultralytics.utils.plotting import colors
+
+    img = np.array(Image.open(image_path))
+    img_height, img_width = img.shape[:2]
+    annotations = []
+    with open(txt_path) as file:
+        for line in file:
+            class_id, x_center, y_center, width, height = map(float, line.split())
+            x = (x_center - width / 2) * img_width
+            y = (y_center - height / 2) * img_height
+            w = width * img_width
+            h = height * img_height
+            annotations.append((x, y, w, h, int(class_id)))
+    fig, ax = plt.subplots(1)  # Plot the image and annotations
+    for x, y, w, h, label in annotations:
+        color = tuple(c / 255 for c in colors(label, True))  # Get and normalize the RGB color
+        rect = plt.Rectangle((x, y), w, h, linewidth=2, edgecolor=color, facecolor="none")  # Create a rectangle
+        ax.add_patch(rect)
+        luminance = 0.2126 * color[0] + 0.7152 * color[1] + 0.0722 * color[2]  # Formula for luminance
+        ax.text(x, y - 5, label_map[label], color="white" if luminance < 0.5 else "black", backgroundcolor=color)
+    ax.imshow(img)
+    plt.show()
+
+
+def polygon2mask(imgsz, polygons, color=1, downsample_ratio=1):
+    """
+    Convert a list of polygons to a binary mask of the specified image size.
+
+    Args:
+        imgsz (tuple): The size of the image as (height, width).
+        polygons (list[np.ndarray]): A list of polygons. Each polygon is an array with shape [N, M], where
+                                     N is the number of polygons, and M is the number of points such that M % 2 = 0.
+        color (int, optional): The color value to fill in the polygons on the mask. Defaults to 1.
+        downsample_ratio (int, optional): Factor by which to downsample the mask. Defaults to 1.
+
+    Returns:
+        (np.ndarray): A binary mask of the specified image size with the polygons filled in.
+    """
+    mask = np.zeros(imgsz, dtype=np.uint8)
+    polygons = np.asarray(polygons, dtype=np.int32)
+    polygons = polygons.reshape((polygons.shape[0], -1, 2))
+    cv2.fillPoly(mask, polygons, color=color)
+    nh, nw = (imgsz[0] // downsample_ratio, imgsz[1] // downsample_ratio)
+    # Note: fillPoly first then resize is trying to keep the same loss calculation method when mask-ratio=1
+    return cv2.resize(mask, (nw, nh))
+
+
+def polygons2masks(imgsz, polygons, color, downsample_ratio=1):
+    """
+    Convert a list of polygons to a set of binary masks of the specified image size.
+
+    Args:
+        imgsz (tuple): The size of the image as (height, width).
+        polygons (list[np.ndarray]): A list of polygons. Each polygon is an array with shape [N, M], where
+                                     N is the number of polygons, and M is the number of points such that M % 2 = 0.
+        color (int): The color value to fill in the polygons on the masks.
+        downsample_ratio (int, optional): Factor by which to downsample each mask. Defaults to 1.
+
+    Returns:
+        (np.ndarray): A set of binary masks of the specified image size with the polygons filled in.
+    """
+    return np.array([polygon2mask(imgsz, [x.reshape(-1)], color, downsample_ratio) for x in polygons])
+
+
+def polygons2masks_overlap(imgsz, segments, downsample_ratio=1):
+    """Return a (640, 640) overlap mask."""
+    masks = np.zeros(
+        (imgsz[0] // downsample_ratio, imgsz[1] // downsample_ratio),
+        dtype=np.int32 if len(segments) > 255 else np.uint8,
+    )
+    areas = []
+    ms = []
+    for si in range(len(segments)):
+        mask = polygon2mask(imgsz, [segments[si].reshape(-1)], downsample_ratio=downsample_ratio, color=1)
+        ms.append(mask.astype(masks.dtype))
+        areas.append(mask.sum())
+    areas = np.asarray(areas)
+    index = np.argsort(-areas)
+    ms = np.array(ms)[index]
+    for i in range(len(segments)):
+        mask = ms[i] * (i + 1)
+        masks = masks + mask
+        masks = np.clip(masks, a_min=0, a_max=i + 1)
+    return masks, index
+
+
+def find_dataset_yaml(path: Path) -> Path:
+    """
+    Find and return the YAML file associated with a Detect, Segment or Pose dataset.
+
+    This function searches for a YAML file at the root level of the provided directory first, and if not found, it
+    performs a recursive search. It prefers YAML files that have the same stem as the provided path. An AssertionError
+    is raised if no YAML file is found or if multiple YAML files are found.
+
+    Args:
+        path (Path): The directory path to search for the YAML file.
+
+    Returns:
+        (Path): The path of the found YAML file.
+    """
+    files = list(path.glob("*.yaml")) or list(path.rglob("*.yaml"))  # try root level first and then recursive
+    assert files, f"No YAML file found in '{path.resolve()}'"
+    if len(files) > 1:
+        files = [f for f in files if f.stem == path.stem]  # prefer *.yaml files that match
+    assert len(files) == 1, f"Expected 1 YAML file in '{path.resolve()}', but found {len(files)}.\n{files}"
+    return files[0]
+
+
+def check_det_dataset(dataset, autodownload=True):
+    """
+    Download, verify, and/or unzip a dataset if not found locally.
+
+    This function checks the availability of a specified dataset, and if not found, it has the option to download and
+    unzip the dataset. It then reads and parses the accompanying YAML data, ensuring key requirements are met and also
+    resolves paths related to the dataset.
+
+    Args:
+        dataset (str): Path to the dataset or dataset descriptor (like a YAML file).
+        autodownload (bool, optional): Whether to automatically download the dataset if not found. Defaults to True.
+
+    Returns:
+        (dict): Parsed dataset information and paths.
+    """
+    file = check_file(dataset)
+
+    # Download (optional)
+    extract_dir = ""
+    if zipfile.is_zipfile(file) or is_tarfile(file):
+        new_dir = safe_download(file, dir=DATASETS_DIR, unzip=True, delete=False)
+        file = find_dataset_yaml(DATASETS_DIR / new_dir)
+        extract_dir, autodownload = file.parent, False
+
+    # Read YAML
+    data = yaml_load(file, append_filename=True)  # dictionary
+
+    # Checks
+    for k in "train", "val":
+        if k not in data:
+            if k != "val" or "validation" not in data:
+                raise SyntaxError(
+                    emojis(f"{dataset} '{k}:' key missing ❌.\n'train' and 'val' are required in all data YAMLs.")
+                )
+            LOGGER.info("WARNING ⚠️ renaming data YAML 'validation' key to 'val' to match YOLO format.")
+            data["val"] = data.pop("validation")  # replace 'validation' key with 'val' key
+    if "names" not in data and "nc" not in data:
+        raise SyntaxError(emojis(f"{dataset} key missing ❌.\n either 'names' or 'nc' are required in all data YAMLs."))
+    if "names" in data and "nc" in data and len(data["names"]) != data["nc"]:
+        raise SyntaxError(emojis(f"{dataset} 'names' length {len(data['names'])} and 'nc: {data['nc']}' must match."))
+    if "names" not in data:
+        data["names"] = [f"class_{i}" for i in range(data["nc"])]
+    else:
+        data["nc"] = len(data["names"])
+
+    data["names"] = check_class_names(data["names"])
+
+    # Resolve paths
+    path = Path(extract_dir or data.get("path") or Path(data.get("yaml_file", "")).parent)  # dataset root
+    if not path.is_absolute():
+        path = (DATASETS_DIR / path).resolve()
+
+    # Set paths
+    data["path"] = path  # download scripts
+    for k in "train", "val", "test", "minival":
+        if data.get(k):  # prepend path
+            if isinstance(data[k], str):
+                x = (path / data[k]).resolve()
+                if not x.exists() and data[k].startswith("../"):
+                    x = (path / data[k][3:]).resolve()
+                data[k] = str(x)
+            else:
+                data[k] = [str((path / x).resolve()) for x in data[k]]
+
+    # Parse YAML
+    val, s = (data.get(x) for x in ("val", "download"))
+    if val:
+        val = [Path(x).resolve() for x in (val if isinstance(val, list) else [val])]  # val path
+        if not all(x.exists() for x in val):
+            name = clean_url(dataset)  # dataset name with URL auth stripped
+            m = f"\nDataset '{name}' images not found ⚠️, missing path '{[x for x in val if not x.exists()][0]}'"
+            if s and autodownload:
+                LOGGER.warning(m)
+            else:
+                m += f"\nNote dataset download directory is '{DATASETS_DIR}'. You can update this in '{SETTINGS_FILE}'"
+                raise FileNotFoundError(m)
+            t = time.time()
+            r = None  # success
+            if s.startswith("http") and s.endswith(".zip"):  # URL
+                safe_download(url=s, dir=DATASETS_DIR, delete=True)
+            elif s.startswith("bash "):  # bash script
+                LOGGER.info(f"Running {s} ...")
+                r = os.system(s)
+            else:  # python script
+                exec(s, {"yaml": data})
+            dt = f"({round(time.time() - t, 1)}s)"
+            s = f"success ✅ {dt}, saved to {colorstr('bold', DATASETS_DIR)}" if r in {0, None} else f"failure {dt} ❌"
+            LOGGER.info(f"Dataset download {s}\n")
+    check_font("Arial.ttf" if is_ascii(data["names"]) else "Arial.Unicode.ttf")  # download fonts
+
+    return data  # dictionary
+
+
+def check_cls_dataset(dataset, split=""):
+    """
+    Checks a classification dataset such as Imagenet.
+
+    This function accepts a `dataset` name and attempts to retrieve the corresponding dataset information.
+    If the dataset is not found locally, it attempts to download the dataset from the internet and save it locally.
+
+    Args:
+        dataset (str | Path): The name of the dataset.
+        split (str, optional): The split of the dataset. Either 'val', 'test', or ''. Defaults to ''.
+
+    Returns:
+        (dict): A dictionary containing the following keys:
+            - 'train' (Path): The directory path containing the training set of the dataset.
+            - 'val' (Path): The directory path containing the validation set of the dataset.
+            - 'test' (Path): The directory path containing the test set of the dataset.
+            - 'nc' (int): The number of classes in the dataset.
+            - 'names' (dict): A dictionary of class names in the dataset.
+    """
+    # Download (optional if dataset=https://file.zip is passed directly)
+    if str(dataset).startswith(("http:/", "https:/")):
+        dataset = safe_download(dataset, dir=DATASETS_DIR, unzip=True, delete=False)
+    elif Path(dataset).suffix in {".zip", ".tar", ".gz"}:
+        file = check_file(dataset)
+        dataset = safe_download(file, dir=DATASETS_DIR, unzip=True, delete=False)
+
+    dataset = Path(dataset)
+    data_dir = (dataset if dataset.is_dir() else (DATASETS_DIR / dataset)).resolve()
+    if not data_dir.is_dir():
+        LOGGER.warning(f"\nDataset not found ⚠️, missing path {data_dir}, attempting download...")
+        t = time.time()
+        if str(dataset) == "imagenet":
+            subprocess.run(f"bash {ROOT / 'data/scripts/get_imagenet.sh'}", shell=True, check=True)
+        else:
+            url = f"https://github.com/ultralytics/assets/releases/download/v0.0.0/{dataset}.zip"
+            download(url, dir=data_dir.parent)
+        s = f"Dataset download success ✅ ({time.time() - t:.1f}s), saved to {colorstr('bold', data_dir)}\n"
+        LOGGER.info(s)
+    train_set = data_dir / "train"
+    val_set = (
+        data_dir / "val"
+        if (data_dir / "val").exists()
+        else data_dir / "validation"
+        if (data_dir / "validation").exists()
+        else None
+    )  # data/test or data/val
+    test_set = data_dir / "test" if (data_dir / "test").exists() else None  # data/val or data/test
+    if split == "val" and not val_set:
+        LOGGER.warning("WARNING ⚠️ Dataset 'split=val' not found, using 'split=test' instead.")
+    elif split == "test" and not test_set:
+        LOGGER.warning("WARNING ⚠️ Dataset 'split=test' not found, using 'split=val' instead.")
+
+    nc = len([x for x in (data_dir / "train").glob("*") if x.is_dir()])  # number of classes
+    names = [x.name for x in (data_dir / "train").iterdir() if x.is_dir()]  # class names list
+    names = dict(enumerate(sorted(names)))
+
+    # Print to console
+    for k, v in {"train": train_set, "val": val_set, "test": test_set}.items():
+        prefix = f"{colorstr(f'{k}:')} {v}..."
+        if v is None:
+            LOGGER.info(prefix)
+        else:
+            files = [path for path in v.rglob("*.*") if path.suffix[1:].lower() in IMG_FORMATS]
+            nf = len(files)  # number of files
+            nd = len({file.parent for file in files})  # number of directories
+            if nf == 0:
+                if k == "train":
+                    raise FileNotFoundError(emojis(f"{dataset} '{k}:' no training images found ❌ "))
+                else:
+                    LOGGER.warning(f"{prefix} found {nf} images in {nd} classes: WARNING ⚠️ no images found")
+            elif nd != nc:
+                LOGGER.warning(f"{prefix} found {nf} images in {nd} classes: ERROR ❌️ requires {nc} classes, not {nd}")
+            else:
+                LOGGER.info(f"{prefix} found {nf} images in {nd} classes ✅ ")
+
+    return {"train": train_set, "val": val_set, "test": test_set, "nc": nc, "names": names}
+
+
+class HUBDatasetStats:
+    """
+    A class for generating HUB dataset JSON and `-hub` dataset directory.
+
+    Args:
+        path (str): Path to data.yaml or data.zip (with data.yaml inside data.zip). Default is 'coco8.yaml'.
+        task (str): Dataset task. Options are 'detect', 'segment', 'pose', 'classify'. Default is 'detect'.
+        autodownload (bool): Attempt to download dataset if not found locally. Default is False.
+
+    Example:
+        Download *.zip files from https://github.com/ultralytics/hub/tree/main/example_datasets
+            i.e. https://github.com/ultralytics/hub/raw/main/example_datasets/coco8.zip for coco8.zip.
+        ```python
+        from ultralytics.data.utils import HUBDatasetStats
+
+        stats = HUBDatasetStats("path/to/coco8.zip", task="detect")  # detect dataset
+        stats = HUBDatasetStats("path/to/coco8-seg.zip", task="segment")  # segment dataset
+        stats = HUBDatasetStats("path/to/coco8-pose.zip", task="pose")  # pose dataset
+        stats = HUBDatasetStats("path/to/dota8.zip", task="obb")  # OBB dataset
+        stats = HUBDatasetStats("path/to/imagenet10.zip", task="classify")  # classification dataset
+
+        stats.get_json(save=True)
+        stats.process_images()
+        ```
+    """
+
+    def __init__(self, path="coco8.yaml", task="detect", autodownload=False):
+        """Initialize class."""
+        path = Path(path).resolve()
+        LOGGER.info(f"Starting HUB dataset checks for {path}....")
+
+        self.task = task  # detect, segment, pose, classify, obb
+        if self.task == "classify":
+            unzip_dir = unzip_file(path)
+            data = check_cls_dataset(unzip_dir)
+            data["path"] = unzip_dir
+        else:  # detect, segment, pose, obb
+            _, data_dir, yaml_path = self._unzip(Path(path))
+            try:
+                # Load YAML with checks
+                data = yaml_load(yaml_path)
+                data["path"] = ""  # strip path since YAML should be in dataset root for all HUB datasets
+                yaml_save(yaml_path, data)
+                data = check_det_dataset(yaml_path, autodownload)  # dict
+                data["path"] = data_dir  # YAML path should be set to '' (relative) or parent (absolute)
+            except Exception as e:
+                raise Exception("error/HUB/dataset_stats/init") from e
+
+        self.hub_dir = Path(f"{data['path']}-hub")
+        self.im_dir = self.hub_dir / "images"
+        self.stats = {"nc": len(data["names"]), "names": list(data["names"].values())}  # statistics dictionary
+        self.data = data
+
+    @staticmethod
+    def _unzip(path):
+        """Unzip data.zip."""
+        if not str(path).endswith(".zip"):  # path is data.yaml
+            return False, None, path
+        unzip_dir = unzip_file(path, path=path.parent)
+        assert unzip_dir.is_dir(), (
+            f"Error unzipping {path}, {unzip_dir} not found. path/to/abc.zip MUST unzip to path/to/abc/"
+        )
+        return True, str(unzip_dir), find_dataset_yaml(unzip_dir)  # zipped, data_dir, yaml_path
+
+    def _hub_ops(self, f):
+        """Saves a compressed image for HUB previews."""
+        compress_one_image(f, self.im_dir / Path(f).name)  # save to dataset-hub
+
+    def get_json(self, save=False, verbose=False):
+        """Return dataset JSON for Ultralytics HUB."""
+
+        def _round(labels):
+            """Update labels to integer class and 4 decimal place floats."""
+            if self.task == "detect":
+                coordinates = labels["bboxes"]
+            elif self.task in {"segment", "obb"}:  # Segment and OBB use segments. OBB segments are normalized xyxyxyxy
+                coordinates = [x.flatten() for x in labels["segments"]]
+            elif self.task == "pose":
+                n, nk, nd = labels["keypoints"].shape
+                coordinates = np.concatenate((labels["bboxes"], labels["keypoints"].reshape(n, nk * nd)), 1)
+            else:
+                raise ValueError(f"Undefined dataset task={self.task}.")
+            zipped = zip(labels["cls"], coordinates)
+            return [[int(c[0]), *(round(float(x), 4) for x in points)] for c, points in zipped]
+
+        for split in "train", "val", "test":
+            self.stats[split] = None  # predefine
+            path = self.data.get(split)
+
+            # Check split
+            if path is None:  # no split
+                continue
+            files = [f for f in Path(path).rglob("*.*") if f.suffix[1:].lower() in IMG_FORMATS]  # image files in split
+            if not files:  # no images
+                continue
+
+            # Get dataset statistics
+            if self.task == "classify":
+                from torchvision.datasets import ImageFolder  # scope for faster 'import ultralytics'
+
+                dataset = ImageFolder(self.data[split])
+
+                x = np.zeros(len(dataset.classes)).astype(int)
+                for im in dataset.imgs:
+                    x[im[1]] += 1
+
+                self.stats[split] = {
+                    "instance_stats": {"total": len(dataset), "per_class": x.tolist()},
+                    "image_stats": {"total": len(dataset), "unlabelled": 0, "per_class": x.tolist()},
+                    "labels": [{Path(k).name: v} for k, v in dataset.imgs],
+                }
+            else:
+                from ultralytics.data import YOLODataset
+
+                dataset = YOLODataset(img_path=self.data[split], data=self.data, task=self.task)
+                x = np.array(
+                    [
+                        np.bincount(label["cls"].astype(int).flatten(), minlength=self.data["nc"])
+                        for label in TQDM(dataset.labels, total=len(dataset), desc="Statistics")
+                    ]
+                )  # shape(128x80)
+                self.stats[split] = {
+                    "instance_stats": {"total": int(x.sum()), "per_class": x.sum(0).tolist()},
+                    "image_stats": {
+                        "total": len(dataset),
+                        "unlabelled": int(np.all(x == 0, 1).sum()),
+                        "per_class": (x > 0).sum(0).tolist(),
+                    },
+                    "labels": [{Path(k).name: _round(v)} for k, v in zip(dataset.im_files, dataset.labels)],
+                }
+
+        # Save, print and return
+        if save:
+            self.hub_dir.mkdir(parents=True, exist_ok=True)  # makes dataset-hub/
+            stats_path = self.hub_dir / "stats.json"
+            LOGGER.info(f"Saving {stats_path.resolve()}...")
+            with open(stats_path, "w") as f:
+                json.dump(self.stats, f)  # save stats.json
+        if verbose:
+            LOGGER.info(json.dumps(self.stats, indent=2, sort_keys=False))
+        return self.stats
+
+    def process_images(self):
+        """Compress images for Ultralytics HUB."""
+        from ultralytics.data import YOLODataset  # ClassificationDataset
+
+        self.im_dir.mkdir(parents=True, exist_ok=True)  # makes dataset-hub/images/
+        for split in "train", "val", "test":
+            if self.data.get(split) is None:
+                continue
+            dataset = YOLODataset(img_path=self.data[split], data=self.data)
+            with ThreadPool(NUM_THREADS) as pool:
+                for _ in TQDM(pool.imap(self._hub_ops, dataset.im_files), total=len(dataset), desc=f"{split} images"):
+                    pass
+        LOGGER.info(f"Done. All images saved to {self.im_dir}")
+        return self.im_dir
+
+
+def compress_one_image(f, f_new=None, max_dim=1920, quality=50):
+    """
+    Compresses a single image file to reduced size while preserving its aspect ratio and quality using either the Python
+    Imaging Library (PIL) or OpenCV library. If the input image is smaller than the maximum dimension, it will not be
+    resized.
+
+    Args:
+        f (str): The path to the input image file.
+        f_new (str, optional): The path to the output image file. If not specified, the input file will be overwritten.
+        max_dim (int, optional): The maximum dimension (width or height) of the output image. Default is 1920 pixels.
+        quality (int, optional): The image compression quality as a percentage. Default is 50%.
+
+    Example:
+        ```python
+        from pathlib import Path
+        from ultralytics.data.utils import compress_one_image
+
+        for f in Path("path/to/dataset").rglob("*.jpg"):
+            compress_one_image(f)
+        ```
+    """
+    try:  # use PIL
+        im = Image.open(f)
+        r = max_dim / max(im.height, im.width)  # ratio
+        if r < 1.0:  # image too large
+            im = im.resize((int(im.width * r), int(im.height * r)))
+        im.save(f_new or f, "JPEG", quality=quality, optimize=True)  # save
+    except Exception as e:  # use OpenCV
+        LOGGER.info(f"WARNING ⚠️ HUB ops PIL failure {f}: {e}")
+        im = cv2.imread(f)
+        im_height, im_width = im.shape[:2]
+        r = max_dim / max(im_height, im_width)  # ratio
+        if r < 1.0:  # image too large
+            im = cv2.resize(im, (int(im_width * r), int(im_height * r)), interpolation=cv2.INTER_AREA)
+        cv2.imwrite(str(f_new or f), im)
+
+
+def autosplit(path=DATASETS_DIR / "coco8/images", weights=(0.9, 0.1, 0.0), annotated_only=False):
+    """
+    Automatically split a dataset into train/val/test splits and save the resulting splits into autosplit_*.txt files.
+
+    Args:
+        path (Path, optional): Path to images directory. Defaults to DATASETS_DIR / 'coco8/images'.
+        weights (list | tuple, optional): Train, validation, and test split fractions. Defaults to (0.9, 0.1, 0.0).
+        annotated_only (bool, optional): If True, only images with an associated txt file are used. Defaults to False.
+
+    Example:
+        ```python
+        from ultralytics.data.utils import autosplit
+
+        autosplit()
+        ```
+    """
+    path = Path(path)  # images dir
+    files = sorted(x for x in path.rglob("*.*") if x.suffix[1:].lower() in IMG_FORMATS)  # image files only
+    n = len(files)  # number of files
+    random.seed(0)  # for reproducibility
+    indices = random.choices([0, 1, 2], weights=weights, k=n)  # assign each image to a split
+
+    txt = ["autosplit_train.txt", "autosplit_val.txt", "autosplit_test.txt"]  # 3 txt files
+    for x in txt:
+        if (path.parent / x).exists():
+            (path.parent / x).unlink()  # remove existing
+
+    LOGGER.info(f"Autosplitting images from {path}" + ", using *.txt labeled images only" * annotated_only)
+    for i, img in TQDM(zip(indices, files), total=n):
+        if not annotated_only or Path(img2label_paths([str(img)])[0]).exists():  # check label
+            with open(path.parent / txt[i], "a") as f:
+                f.write(f"./{img.relative_to(path.parent).as_posix()}" + "\n")  # add image to txt file
+
+
+def load_dataset_cache_file(path):
+    """Load an Ultralytics *.cache dictionary from path."""
+    import gc
+
+    gc.disable()  # reduce pickle load time https://github.com/ultralytics/ultralytics/pull/1585
+    cache = np.load(str(path), allow_pickle=True).item()  # load dict
+    gc.enable()
+    return cache
+
+
+def save_dataset_cache_file(prefix, path, x, version):
+    """Save an Ultralytics dataset *.cache dictionary x to path."""
+    x["version"] = version  # add cache version
+    if is_dir_writeable(path.parent):
+        if path.exists():
+            path.unlink()  # remove *.cache file if exists
+        np.save(str(path), x)  # save cache for next time
+        path.with_suffix(".cache.npy").rename(path)  # remove .npy suffix
+        LOGGER.info(f"{prefix}New cache created: {path}")
+    else:
+        LOGGER.warning(f"{prefix}WARNING ⚠️ Cache directory {path.parent} is not writeable, cache not saved.")