Upload 98 files

data/Argoverse.yaml

@@ -0,0 +1,67 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Argoverse-HD dataset (ring-front-center camera) http://www.cs.cmu.edu/~mengtial/proj/streaming/ by Argo AI
+# Example usage: python train.py --data Argoverse.yaml
+# parent
+# ├── yolov5
+# └── datasets
+#     └── Argoverse  ← downloads here (31.3 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/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
+nc: 8  # number of classes
+names: ['person',  'bicycle',  'car',  'motorcycle',  'bus',  'truck',  'traffic_light',  'stop_sign']  # class names
+
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import json
+
+  from tqdm import tqdm
+  from utils.general import download, 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
+  dir = Path('../datasets/Argoverse')  # dataset root dir
+  urls = ['https://argoverse-hd.s3.us-east-2.amazonaws.com/Argoverse-HD-Full.zip']
+  download(urls, dir=dir, delete=False)
+
+  # 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 VisDrone annotations to YOLO labels

data/GlobalWheat2020.yaml

@@ -0,0 +1,54 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Global Wheat 2020 dataset http://www.global-wheat.com/ by University of Saskatchewan
+# Example usage: python train.py --data GlobalWheat2020.yaml
+# parent
+# ├── yolov5
+# └── 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
+nc: 1  # number of classes
+names: ['wheat_head']  # class names
+
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  from utils.general import download, 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/yolov5/releases/download/v1.0/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 / p).rename(dir / 'images' / p)  # move to /images
+      f = (dir / p).with_suffix('.json')  # json file
+      if f.exists():
+          f.rename((dir / 'annotations' / p).with_suffix('.json'))  # move to /annotations

data/Objects365.yaml

@@ -0,0 +1,114 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Objects365 dataset https://www.objects365.org/ by Megvii
+# Example usage: python train.py --data Objects365.yaml
+# parent
+# ├── yolov5
+# └── 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
+nc: 365  # number of classes
+names: ['Person', 'Sneakers', 'Chair', 'Other Shoes', 'Hat', 'Car', 'Lamp', 'Glasses', 'Bottle', 'Desk', 'Cup',
+        'Street Lights', 'Cabinet/shelf', 'Handbag/Satchel', 'Bracelet', 'Plate', 'Picture/Frame', 'Helmet', 'Book',
+        'Gloves', 'Storage box', 'Boat', 'Leather Shoes', 'Flower', 'Bench', 'Potted Plant', 'Bowl/Basin', 'Flag',
+        'Pillow', 'Boots', 'Vase', 'Microphone', 'Necklace', 'Ring', 'SUV', 'Wine Glass', 'Belt', 'Monitor/TV',
+        'Backpack', 'Umbrella', 'Traffic Light', 'Speaker', 'Watch', 'Tie', 'Trash bin Can', 'Slippers', 'Bicycle',
+        'Stool', 'Barrel/bucket', 'Van', 'Couch', 'Sandals', 'Basket', 'Drum', 'Pen/Pencil', 'Bus', 'Wild Bird',
+        'High Heels', 'Motorcycle', 'Guitar', 'Carpet', 'Cell Phone', 'Bread', 'Camera', 'Canned', 'Truck',
+        'Traffic cone', 'Cymbal', 'Lifesaver', 'Towel', 'Stuffed Toy', 'Candle', 'Sailboat', 'Laptop', 'Awning',
+        'Bed', 'Faucet', 'Tent', 'Horse', 'Mirror', 'Power outlet', 'Sink', 'Apple', 'Air Conditioner', 'Knife',
+        'Hockey Stick', 'Paddle', 'Pickup Truck', 'Fork', 'Traffic Sign', 'Balloon', 'Tripod', 'Dog', 'Spoon', 'Clock',
+        'Pot', 'Cow', 'Cake', 'Dinning Table', 'Sheep', 'Hanger', 'Blackboard/Whiteboard', 'Napkin', 'Other Fish',
+        'Orange/Tangerine', 'Toiletry', 'Keyboard', 'Tomato', 'Lantern', 'Machinery Vehicle', 'Fan',
+        'Green Vegetables', 'Banana', 'Baseball Glove', 'Airplane', 'Mouse', 'Train', 'Pumpkin', 'Soccer', 'Skiboard',
+        'Luggage', 'Nightstand', 'Tea pot', 'Telephone', 'Trolley', 'Head Phone', 'Sports Car', 'Stop Sign',
+        'Dessert', 'Scooter', 'Stroller', 'Crane', 'Remote', 'Refrigerator', 'Oven', 'Lemon', 'Duck', 'Baseball Bat',
+        'Surveillance Camera', 'Cat', 'Jug', 'Broccoli', 'Piano', 'Pizza', 'Elephant', 'Skateboard', 'Surfboard',
+        'Gun', 'Skating and Skiing shoes', 'Gas stove', 'Donut', 'Bow Tie', 'Carrot', 'Toilet', 'Kite', 'Strawberry',
+        'Other Balls', 'Shovel', 'Pepper', 'Computer Box', 'Toilet Paper', 'Cleaning Products', 'Chopsticks',
+        'Microwave', 'Pigeon', 'Baseball', 'Cutting/chopping Board', 'Coffee Table', 'Side Table', 'Scissors',
+        'Marker', 'Pie', 'Ladder', 'Snowboard', 'Cookies', 'Radiator', 'Fire Hydrant', 'Basketball', 'Zebra', 'Grape',
+        'Giraffe', 'Potato', 'Sausage', 'Tricycle', 'Violin', 'Egg', 'Fire Extinguisher', 'Candy', 'Fire Truck',
+        'Billiards', 'Converter', 'Bathtub', 'Wheelchair', 'Golf Club', 'Briefcase', 'Cucumber', 'Cigar/Cigarette',
+        'Paint Brush', 'Pear', 'Heavy Truck', 'Hamburger', 'Extractor', 'Extension Cord', 'Tong', 'Tennis Racket',
+        'Folder', 'American Football', 'earphone', 'Mask', 'Kettle', 'Tennis', 'Ship', 'Swing', 'Coffee Machine',
+        'Slide', 'Carriage', 'Onion', 'Green beans', 'Projector', 'Frisbee', 'Washing Machine/Drying Machine',
+        'Chicken', 'Printer', 'Watermelon', 'Saxophone', 'Tissue', 'Toothbrush', 'Ice cream', 'Hot-air balloon',
+        'Cello', 'French Fries', 'Scale', 'Trophy', 'Cabbage', 'Hot dog', 'Blender', 'Peach', 'Rice', 'Wallet/Purse',
+        'Volleyball', 'Deer', 'Goose', 'Tape', 'Tablet', 'Cosmetics', 'Trumpet', 'Pineapple', 'Golf Ball',
+        'Ambulance', 'Parking meter', 'Mango', 'Key', 'Hurdle', 'Fishing Rod', 'Medal', 'Flute', 'Brush', 'Penguin',
+        'Megaphone', 'Corn', 'Lettuce', 'Garlic', 'Swan', 'Helicopter', 'Green Onion', 'Sandwich', 'Nuts',
+        'Speed Limit Sign', 'Induction Cooker', 'Broom', 'Trombone', 'Plum', 'Rickshaw', 'Goldfish', 'Kiwi fruit',
+        'Router/modem', 'Poker Card', 'Toaster', 'Shrimp', 'Sushi', 'Cheese', 'Notepaper', 'Cherry', 'Pliers', 'CD',
+        'Pasta', 'Hammer', 'Cue', 'Avocado', 'Hamimelon', 'Flask', 'Mushroom', 'Screwdriver', 'Soap', 'Recorder',
+        'Bear', 'Eggplant', 'Board Eraser', 'Coconut', 'Tape Measure/Ruler', 'Pig', 'Showerhead', 'Globe', 'Chips',
+        'Steak', 'Crosswalk Sign', 'Stapler', 'Camel', 'Formula 1', 'Pomegranate', 'Dishwasher', 'Crab',
+        'Hoverboard', 'Meat ball', 'Rice Cooker', 'Tuba', 'Calculator', 'Papaya', 'Antelope', 'Parrot', 'Seal',
+        'Butterfly', 'Dumbbell', 'Donkey', 'Lion', 'Urinal', 'Dolphin', 'Electric Drill', 'Hair Dryer', 'Egg tart',
+        'Jellyfish', 'Treadmill', 'Lighter', 'Grapefruit', 'Game board', 'Mop', 'Radish', 'Baozi', 'Target', 'French',
+        'Spring Rolls', 'Monkey', 'Rabbit', 'Pencil Case', 'Yak', 'Red Cabbage', 'Binoculars', 'Asparagus', 'Barbell',
+        'Scallop', 'Noddles', 'Comb', 'Dumpling', 'Oyster', 'Table Tennis paddle', 'Cosmetics Brush/Eyeliner Pencil',
+        'Chainsaw', 'Eraser', 'Lobster', 'Durian', 'Okra', 'Lipstick', 'Cosmetics Mirror', 'Curling', 'Table Tennis']
+
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  from tqdm import tqdm
+
+  from utils.general import Path, check_requirements, download, np, xyxy2xywhn
+
+  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, delete=False)  # annotations json
+          download([f'{url}patch{i}.tar.gz' for i in range(patches)], dir=images, curl=True, delete=False, threads=8)
+      elif split == 'val':
+          download([f'{url}zhiyuan_objv2_{split}.json'], dir=dir, delete=False)  # annotations json
+          download([f'{url}images/v1/patch{i}.tar.gz' for i in range(15 + 1)], dir=images, curl=True, delete=False, threads=8)
+          download([f'{url}images/v2/patch{i}.tar.gz' for i in range(16, patches)], dir=images, curl=True, delete=False, 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)

data/SKU-110K.yaml

@@ -0,0 +1,53 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# SKU-110K retail items dataset https://github.com/eg4000/SKU110K_CVPR19 by Trax Retail
+# Example usage: python train.py --data SKU-110K.yaml
+# parent
+# ├── yolov5
+# └── 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
+nc: 1  # number of classes
+names: ['object']  # class names
+
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import shutil
+  from tqdm import tqdm
+  from utils.general import np, pd, Path, download, 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, delete=False)
+
+  # 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

data/VOC.yaml

@@ -0,0 +1,81 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# PASCAL VOC dataset http://host.robots.ox.ac.uk/pascal/VOC by University of Oxford
+# Example usage: python train.py --data VOC.yaml
+# parent
+# ├── yolov5
+# └── 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
+nc: 20  # number of classes
+names: ['aeroplane', 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog',
+        'horse', 'motorbike', 'person', 'pottedplant', 'sheep', 'sofa', 'train', 'tvmonitor']  # class names
+
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  import xml.etree.ElementTree as ET
+
+  from tqdm import tqdm
+  from utils.general import download, 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)
+
+      for obj in root.iter('object'):
+          cls = obj.find('name').text
+          if cls in yaml['names'] and not 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 = yaml['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/yolov5/releases/download/v1.0/'
+  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', delete=False, curl=True, threads=3)
+
+  # 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

data/VisDrone.yaml

@@ -0,0 +1,61 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# VisDrone2019-DET dataset https://github.com/VisDrone/VisDrone-Dataset by Tianjin University
+# Example usage: python train.py --data VisDrone.yaml
+# parent
+# ├── yolov5
+# └── 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
+nc: 10  # number of classes
+names: ['pedestrian', 'people', 'bicycle', 'car', 'van', 'truck', 'tricycle', 'awning-tricycle', 'bus', 'motor']
+
+
+# Download script/URL (optional) ---------------------------------------------------------------------------------------
+download: |
+  from utils.general import download, os, Path
+
+  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(os.sep + 'annotations' + os.sep, 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/yolov5/releases/download/v1.0/VisDrone2019-DET-train.zip',
+          'https://github.com/ultralytics/yolov5/releases/download/v1.0/VisDrone2019-DET-val.zip',
+          'https://github.com/ultralytics/yolov5/releases/download/v1.0/VisDrone2019-DET-test-dev.zip',
+          'https://github.com/ultralytics/yolov5/releases/download/v1.0/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

data/coco.yaml

@@ -0,0 +1,45 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# COCO 2017 dataset http://cocodataset.org by Microsoft
+# Example usage: python train.py --data coco.yaml
+# parent
+# ├── yolov5
+# └── 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: ../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
+nc: 80  # number of classes
+names: ['person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train', 'truck', 'boat', 'traffic light',
+        'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
+        'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie', 'suitcase', 'frisbee',
+        'skis', 'snowboard', 'sports ball', 'kite', 'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
+        'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple',
+        'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch',
+        'potted plant', 'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote', 'keyboard', 'cell phone',
+        'microwave', 'oven', 'toaster', 'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
+        'hair drier', 'toothbrush']  # class names
+
+
+# Download script/URL (optional)
+download: |
+  from utils.general import download, Path
+
+
+  # Download labels
+  segments = False  # segment or box labels
+  dir = Path(yaml['path'])  # dataset root dir
+  url = 'https://github.com/ultralytics/yolov5/releases/download/v1.0/'
+  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)

data/coco128.yaml

@@ -0,0 +1,30 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# COCO128 dataset https://www.kaggle.com/ultralytics/coco128 (first 128 images from COCO train2017) by Ultralytics
+# Example usage: python train.py --data coco128.yaml
+# parent
+# ├── yolov5
+# └── 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
+nc: 80  # number of classes
+names: ['person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train', 'truck', 'boat', 'traffic light',
+        'fire hydrant', 'stop sign', 'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
+        'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag', 'tie', 'suitcase', 'frisbee',
+        'skis', 'snowboard', 'sports ball', 'kite', 'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
+        'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon', 'bowl', 'banana', 'apple',
+        'sandwich', 'orange', 'broccoli', 'carrot', 'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch',
+        'potted plant', 'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote', 'keyboard', 'cell phone',
+        'microwave', 'oven', 'toaster', 'sink', 'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
+        'hair drier', 'toothbrush']  # class names
+
+
+# Download script/URL (optional)
+download: https://ultralytics.com/assets/coco128.zip

data/hyps/hyp.Objects365.yaml

@@ -0,0 +1,34 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Hyperparameters for Objects365 training
+# python train.py --weights yolov5m.pt --data Objects365.yaml --evolve
+# See Hyperparameter Evolution tutorial for details https://github.com/ultralytics/yolov5#tutorials
+
+lr0: 0.00258
+lrf: 0.17
+momentum: 0.779
+weight_decay: 0.00058
+warmup_epochs: 1.33
+warmup_momentum: 0.86
+warmup_bias_lr: 0.0711
+box: 0.0539
+cls: 0.299
+cls_pw: 0.825
+obj: 0.632
+obj_pw: 1.0
+iou_t: 0.2
+anchor_t: 3.44
+anchors: 3.2
+fl_gamma: 0.0
+hsv_h: 0.0188
+hsv_s: 0.704
+hsv_v: 0.36
+degrees: 0.0
+translate: 0.0902
+scale: 0.491
+shear: 0.0
+perspective: 0.0
+flipud: 0.0
+fliplr: 0.5
+mosaic: 1.0
+mixup: 0.0
+copy_paste: 0.0

data/hyps/hyp.VOC.yaml

@@ -0,0 +1,40 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Hyperparameters for VOC training
+# python train.py --batch 128 --weights yolov5m6.pt --data VOC.yaml --epochs 50 --img 512 --hyp hyp.scratch-med.yaml --evolve
+# See Hyperparameter Evolution tutorial for details https://github.com/ultralytics/yolov5#tutorials
+
+# YOLOv5 Hyperparameter Evolution Results
+# Best generation: 467
+# Last generation: 996
+#    metrics/precision,       metrics/recall,      metrics/mAP_0.5, metrics/mAP_0.5:0.95,         val/box_loss,         val/obj_loss,         val/cls_loss
+#              0.87729,              0.85125,              0.91286,              0.72664,            0.0076739,            0.0042529,            0.0013865
+
+lr0: 0.00334
+lrf: 0.15135
+momentum: 0.74832
+weight_decay: 0.00025
+warmup_epochs: 3.3835
+warmup_momentum: 0.59462
+warmup_bias_lr: 0.18657
+box: 0.02
+cls: 0.21638
+cls_pw: 0.5
+obj: 0.51728
+obj_pw: 0.67198
+iou_t: 0.2
+anchor_t: 3.3744
+fl_gamma: 0.0
+hsv_h: 0.01041
+hsv_s: 0.54703
+hsv_v: 0.27739
+degrees: 0.0
+translate: 0.04591
+scale: 0.75544
+shear: 0.0
+perspective: 0.0
+flipud: 0.0
+fliplr: 0.5
+mosaic: 0.85834
+mixup: 0.04266
+copy_paste: 0.0
+anchors: 3.412

data/hyps/hyp.scratch-high.yaml

@@ -0,0 +1,34 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Hyperparameters for high-augmentation COCO training from scratch
+# python train.py --batch 32 --cfg yolov5m6.yaml --weights '' --data coco.yaml --img 1280 --epochs 300
+# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials
+
+lr0: 0.01  # initial learning rate (SGD=1E-2, Adam=1E-3)
+lrf: 0.1  # final OneCycleLR learning rate (lr0 * lrf)
+momentum: 0.937  # SGD momentum/Adam beta1
+weight_decay: 0.0005  # optimizer weight decay 5e-4
+warmup_epochs: 3.0  # warmup epochs (fractions ok)
+warmup_momentum: 0.8  # warmup initial momentum
+warmup_bias_lr: 0.1  # warmup initial bias lr
+box: 0.05  # box loss gain
+cls: 0.3  # cls loss gain
+cls_pw: 1.0  # cls BCELoss positive_weight
+obj: 0.7  # obj loss gain (scale with pixels)
+obj_pw: 1.0  # obj BCELoss positive_weight
+iou_t: 0.20  # IoU training threshold
+anchor_t: 4.0  # anchor-multiple threshold
+# anchors: 3  # anchors per output layer (0 to ignore)
+fl_gamma: 0.0  # focal loss gamma (efficientDet default gamma=1.5)
+hsv_h: 0.015  # image HSV-Hue augmentation (fraction)
+hsv_s: 0.7  # image HSV-Saturation augmentation (fraction)
+hsv_v: 0.4  # image HSV-Value augmentation (fraction)
+degrees: 0.0  # image rotation (+/- deg)
+translate: 0.1  # image translation (+/- fraction)
+scale: 0.9  # image scale (+/- gain)
+shear: 0.0  # image shear (+/- deg)
+perspective: 0.0  # image perspective (+/- fraction), range 0-0.001
+flipud: 0.0  # image flip up-down (probability)
+fliplr: 0.5  # image flip left-right (probability)
+mosaic: 1.0  # image mosaic (probability)
+mixup: 0.1  # image mixup (probability)
+copy_paste: 0.1  # segment copy-paste (probability)

data/hyps/hyp.scratch-low.yaml

@@ -0,0 +1,34 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Hyperparameters for low-augmentation COCO training from scratch
+# python train.py --batch 64 --cfg yolov5n6.yaml --weights '' --data coco.yaml --img 640 --epochs 300 --linear
+# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials
+
+lr0: 0.01  # initial learning rate (SGD=1E-2, Adam=1E-3)
+lrf: 0.01  # final OneCycleLR learning rate (lr0 * lrf)
+momentum: 0.937  # SGD momentum/Adam beta1
+weight_decay: 0.0005  # optimizer weight decay 5e-4
+warmup_epochs: 3.0  # warmup epochs (fractions ok)
+warmup_momentum: 0.8  # warmup initial momentum
+warmup_bias_lr: 0.1  # warmup initial bias lr
+box: 0.05  # box loss gain
+cls: 0.5  # cls loss gain
+cls_pw: 1.0  # cls BCELoss positive_weight
+obj: 1.0  # obj loss gain (scale with pixels)
+obj_pw: 1.0  # obj BCELoss positive_weight
+iou_t: 0.20  # IoU training threshold
+anchor_t: 4.0  # anchor-multiple threshold
+# anchors: 3  # anchors per output layer (0 to ignore)
+fl_gamma: 0.0  # focal loss gamma (efficientDet default gamma=1.5)
+hsv_h: 0.015  # image HSV-Hue augmentation (fraction)
+hsv_s: 0.7  # image HSV-Saturation augmentation (fraction)
+hsv_v: 0.4  # image HSV-Value augmentation (fraction)
+degrees: 0.0  # image rotation (+/- deg)
+translate: 0.1  # image translation (+/- fraction)
+scale: 0.5  # image scale (+/- gain)
+shear: 0.0  # image shear (+/- deg)
+perspective: 0.0  # image perspective (+/- fraction), range 0-0.001
+flipud: 0.0  # image flip up-down (probability)
+fliplr: 0.5  # image flip left-right (probability)
+mosaic: 1.0  # image mosaic (probability)
+mixup: 0.0  # image mixup (probability)
+copy_paste: 0.0  # segment copy-paste (probability)

data/hyps/hyp.scratch-med.yaml

@@ -0,0 +1,34 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Hyperparameters for medium-augmentation COCO training from scratch
+# python train.py --batch 32 --cfg yolov5m6.yaml --weights '' --data coco.yaml --img 1280 --epochs 300
+# See tutorials for hyperparameter evolution https://github.com/ultralytics/yolov5#tutorials
+
+lr0: 0.01  # initial learning rate (SGD=1E-2, Adam=1E-3)
+lrf: 0.1  # final OneCycleLR learning rate (lr0 * lrf)
+momentum: 0.937  # SGD momentum/Adam beta1
+weight_decay: 0.0005  # optimizer weight decay 5e-4
+warmup_epochs: 3.0  # warmup epochs (fractions ok)
+warmup_momentum: 0.8  # warmup initial momentum
+warmup_bias_lr: 0.1  # warmup initial bias lr
+box: 0.05  # box loss gain
+cls: 0.3  # cls loss gain
+cls_pw: 1.0  # cls BCELoss positive_weight
+obj: 0.7  # obj loss gain (scale with pixels)
+obj_pw: 1.0  # obj BCELoss positive_weight
+iou_t: 0.20  # IoU training threshold
+anchor_t: 4.0  # anchor-multiple threshold
+# anchors: 3  # anchors per output layer (0 to ignore)
+fl_gamma: 0.0  # focal loss gamma (efficientDet default gamma=1.5)
+hsv_h: 0.015  # image HSV-Hue augmentation (fraction)
+hsv_s: 0.7  # image HSV-Saturation augmentation (fraction)
+hsv_v: 0.4  # image HSV-Value augmentation (fraction)
+degrees: 0.0  # image rotation (+/- deg)
+translate: 0.1  # image translation (+/- fraction)
+scale: 0.9  # image scale (+/- gain)
+shear: 0.0  # image shear (+/- deg)
+perspective: 0.0  # image perspective (+/- fraction), range 0-0.001
+flipud: 0.0  # image flip up-down (probability)
+fliplr: 0.5  # image flip left-right (probability)
+mosaic: 1.0  # image mosaic (probability)
+mixup: 0.1  # image mixup (probability)
+copy_paste: 0.0  # segment copy-paste (probability)

data/scripts/download_weights.sh

@@ -0,0 +1,20 @@
+#!/bin/bash
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Download latest models from https://github.com/ultralytics/yolov5/releases
+# Example usage: bash path/to/download_weights.sh
+# parent
+# └── yolov5
+#     ├── yolov5s.pt  ← downloads here
+#     ├── yolov5m.pt
+#     └── ...
+
+python - <<EOF
+from utils.downloads import attempt_download
+
+models = ['n', 's', 'm', 'l', 'x']
+models.extend([x + '6' for x in models])  # add P6 models
+
+for x in models:
+    attempt_download(f'yolov5{x}.pt')
+
+EOF

data/scripts/get_coco.sh

@@ -0,0 +1,27 @@
+#!/bin/bash
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Download COCO 2017 dataset http://cocodataset.org
+# Example usage: bash data/scripts/get_coco.sh
+# parent
+# ├── yolov5
+# └── datasets
+#     └── coco  ← downloads here
+
+# Download/unzip labels
+d='../datasets' # unzip directory
+url=https://github.com/ultralytics/yolov5/releases/download/v1.0/
+f='coco2017labels.zip' # or 'coco2017labels-segments.zip', 68 MB
+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/
+f1='train2017.zip' # 19G, 118k images
+f2='val2017.zip'   # 1G, 5k images
+f3='test2017.zip'  # 7G, 41k images (optional)
+for f in $f1 $f2; do
+  echo 'Downloading' $url$f '...'
+  curl -L $url$f -o $f && unzip -q $f -d $d && rm $f &
+done
+wait # finish background tasks

data/scripts/get_coco128.sh

@@ -0,0 +1,17 @@
+#!/bin/bash
+# YOLOv5 🚀 by Ultralytics, GPL-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
+# ├── yolov5
+# └── datasets
+#     └── coco128  ← downloads here
+
+# Download/unzip images and labels
+d='../datasets' # unzip directory
+url=https://github.com/ultralytics/yolov5/releases/download/v1.0/
+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

data/xView.yaml

@@ -0,0 +1,102 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 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!  --------
+# Example usage: python train.py --data xView.yaml
+# parent
+# ├── yolov5
+# └── 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
+nc: 60  # number of classes
+names: ['Fixed-wing Aircraft', 'Small Aircraft', 'Cargo Plane', 'Helicopter', 'Passenger Vehicle', 'Small Car', 'Bus',
+        'Pickup Truck', 'Utility Truck', 'Truck', 'Cargo Truck', 'Truck w/Box', 'Truck Tractor', 'Trailer',
+        'Truck w/Flatbed', 'Truck w/Liquid', 'Crane Truck', 'Railway Vehicle', 'Passenger Car', 'Cargo Car',
+        'Flat Car', 'Tank car', 'Locomotive', 'Maritime Vessel', 'Motorboat', 'Sailboat', 'Tugboat', 'Barge',
+        'Fishing Vessel', 'Ferry', 'Yacht', 'Container Ship', 'Oil Tanker', 'Engineering Vehicle', 'Tower crane',
+        'Container Crane', 'Reach Stacker', 'Straddle Carrier', 'Mobile Crane', 'Dump Truck', 'Haul Truck',
+        'Scraper/Tractor', 'Front loader/Bulldozer', 'Excavator', 'Cement Mixer', 'Ground Grader', 'Hut/Tent', 'Shed',
+        'Building', 'Aircraft Hangar', 'Damaged Building', 'Facility', 'Construction Site', 'Vehicle Lot', 'Helipad',
+        'Storage Tank', 'Shipping container lot', 'Shipping Container', 'Pylon', 'Tower']  # class names
+
+
+# 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 utils.datasets import autosplit
+  from utils.general import download, 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, delete=False)
+
+  # 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')

models/common.py

@@ -0,0 +1,762 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+"""
+Common modules
+"""
+
+import json
+import math
+import platform
+import warnings
+from collections import OrderedDict, namedtuple
+from copy import copy
+from pathlib import Path
+
+import cv2
+import numpy as np
+import pandas as pd
+import requests
+import torch
+import torch.nn as nn
+import yaml
+from PIL import Image
+from torch.cuda import amp
+
+from utils.dataloaders import exif_transpose, letterbox
+from utils.general import (LOGGER, check_requirements, check_suffix, check_version, colorstr, increment_path,
+                           make_divisible, non_max_suppression, scale_coords, xywh2xyxy, xyxy2xywh)
+from utils.plots import Annotator, colors, save_one_box
+from utils.torch_utils import copy_attr, time_sync
+
+
+def autopad(k, p=None):  # kernel, padding
+    # Pad to 'same'
+    if p is None:
+        p = k // 2 if isinstance(k, int) else [x // 2 for x in k]  # auto-pad
+    return p
+
+
+class Conv(nn.Module):
+    # Standard convolution
+    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):  # ch_in, ch_out, kernel, stride, padding, groups
+        super().__init__()
+        self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False)
+        self.bn = nn.BatchNorm2d(c2)
+        self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
+
+    def forward(self, x):
+        return self.act(self.bn(self.conv(x)))
+
+    def forward_fuse(self, x):
+        return self.act(self.conv(x))
+
+
+class DWConv(Conv):
+    # Depth-wise convolution class
+    def __init__(self, c1, c2, k=1, s=1, act=True):  # ch_in, ch_out, kernel, stride, padding, groups
+        super().__init__(c1, c2, k, s, g=math.gcd(c1, c2), act=act)
+
+
+class DWConvTranspose2d(nn.ConvTranspose2d):
+    # Depth-wise transpose convolution class
+    def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0):  # ch_in, ch_out, kernel, stride, padding, padding_out
+        super().__init__(c1, c2, k, s, p1, p2, groups=math.gcd(c1, c2))
+
+
+class TransformerLayer(nn.Module):
+    # Transformer layer https://arxiv.org/abs/2010.11929 (LayerNorm layers removed for better performance)
+    def __init__(self, c, num_heads):
+        super().__init__()
+        self.q = nn.Linear(c, c, bias=False)
+        self.k = nn.Linear(c, c, bias=False)
+        self.v = nn.Linear(c, c, bias=False)
+        self.ma = nn.MultiheadAttention(embed_dim=c, num_heads=num_heads)
+        self.fc1 = nn.Linear(c, c, bias=False)
+        self.fc2 = nn.Linear(c, c, bias=False)
+
+    def forward(self, x):
+        x = self.ma(self.q(x), self.k(x), self.v(x))[0] + x
+        x = self.fc2(self.fc1(x)) + x
+        return x
+
+
+class TransformerBlock(nn.Module):
+    # Vision Transformer https://arxiv.org/abs/2010.11929
+    def __init__(self, c1, c2, num_heads, num_layers):
+        super().__init__()
+        self.conv = None
+        if c1 != c2:
+            self.conv = Conv(c1, c2)
+        self.linear = nn.Linear(c2, c2)  # learnable position embedding
+        self.tr = nn.Sequential(*(TransformerLayer(c2, num_heads) for _ in range(num_layers)))
+        self.c2 = c2
+
+    def forward(self, x):
+        if self.conv is not None:
+            x = self.conv(x)
+        b, _, w, h = x.shape
+        p = x.flatten(2).permute(2, 0, 1)
+        return self.tr(p + self.linear(p)).permute(1, 2, 0).reshape(b, self.c2, w, h)
+
+
+class Bottleneck(nn.Module):
+    # Standard bottleneck
+    def __init__(self, c1, c2, shortcut=True, g=1, e=0.5):  # ch_in, ch_out, shortcut, groups, expansion
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = Conv(c1, c_, 1, 1)
+        self.cv2 = Conv(c_, c2, 3, 1, g=g)
+        self.add = shortcut and c1 == c2
+
+    def forward(self, x):
+        return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
+
+
+class BottleneckCSP(nn.Module):
+    # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):  # ch_in, ch_out, number, shortcut, groups, expansion
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = Conv(c1, c_, 1, 1)
+        self.cv2 = nn.Conv2d(c1, c_, 1, 1, bias=False)
+        self.cv3 = nn.Conv2d(c_, c_, 1, 1, bias=False)
+        self.cv4 = Conv(2 * c_, c2, 1, 1)
+        self.bn = nn.BatchNorm2d(2 * c_)  # applied to cat(cv2, cv3)
+        self.act = nn.SiLU()
+        self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
+
+    def forward(self, x):
+        y1 = self.cv3(self.m(self.cv1(x)))
+        y2 = self.cv2(x)
+        return self.cv4(self.act(self.bn(torch.cat((y1, y2), 1))))
+
+
+class CrossConv(nn.Module):
+    # Cross Convolution Downsample
+    def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False):
+        # ch_in, ch_out, kernel, stride, groups, expansion, shortcut
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = Conv(c1, c_, (1, k), (1, s))
+        self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g)
+        self.add = shortcut and c1 == c2
+
+    def forward(self, x):
+        return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
+
+
+class C3(nn.Module):
+    # CSP Bottleneck with 3 convolutions
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):  # ch_in, ch_out, number, shortcut, groups, expansion
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = Conv(c1, c_, 1, 1)
+        self.cv2 = Conv(c1, c_, 1, 1)
+        self.cv3 = Conv(2 * c_, c2, 1)  # optional act=FReLU(c2)
+        self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
+
+    def forward(self, x):
+        return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), 1))
+
+
+class C3x(C3):
+    # C3 module with cross-convolutions
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
+        super().__init__(c1, c2, n, shortcut, g, e)
+        c_ = int(c2 * e)
+        self.m = nn.Sequential(*(CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)))
+
+
+class C3TR(C3):
+    # C3 module with TransformerBlock()
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
+        super().__init__(c1, c2, n, shortcut, g, e)
+        c_ = int(c2 * e)
+        self.m = TransformerBlock(c_, c_, 4, n)
+
+
+class C3SPP(C3):
+    # C3 module with SPP()
+    def __init__(self, c1, c2, k=(5, 9, 13), n=1, shortcut=True, g=1, e=0.5):
+        super().__init__(c1, c2, n, shortcut, g, e)
+        c_ = int(c2 * e)
+        self.m = SPP(c_, c_, k)
+
+
+class C3Ghost(C3):
+    # C3 module with GhostBottleneck()
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5):
+        super().__init__(c1, c2, n, shortcut, g, e)
+        c_ = int(c2 * e)  # hidden channels
+        self.m = nn.Sequential(*(GhostBottleneck(c_, c_) for _ in range(n)))
+
+
+class SPP(nn.Module):
+    # Spatial Pyramid Pooling (SPP) layer https://arxiv.org/abs/1406.4729
+    def __init__(self, c1, c2, k=(5, 9, 13)):
+        super().__init__()
+        c_ = c1 // 2  # hidden channels
+        self.cv1 = Conv(c1, c_, 1, 1)
+        self.cv2 = Conv(c_ * (len(k) + 1), c2, 1, 1)
+        self.m = nn.ModuleList([nn.MaxPool2d(kernel_size=x, stride=1, padding=x // 2) for x in k])
+
+    def forward(self, x):
+        x = self.cv1(x)
+        with warnings.catch_warnings():
+            warnings.simplefilter('ignore')  # suppress torch 1.9.0 max_pool2d() warning
+            return self.cv2(torch.cat([x] + [m(x) for m in self.m], 1))
+
+
+class SPPF(nn.Module):
+    # Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher
+    def __init__(self, c1, c2, k=5):  # equivalent to SPP(k=(5, 9, 13))
+        super().__init__()
+        c_ = c1 // 2  # hidden channels
+        self.cv1 = Conv(c1, c_, 1, 1)
+        self.cv2 = Conv(c_ * 4, c2, 1, 1)
+        self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2)
+
+    def forward(self, x):
+        x = self.cv1(x)
+        with warnings.catch_warnings():
+            warnings.simplefilter('ignore')  # suppress torch 1.9.0 max_pool2d() warning
+            y1 = self.m(x)
+            y2 = self.m(y1)
+            return self.cv2(torch.cat((x, y1, y2, self.m(y2)), 1))
+
+
+class Focus(nn.Module):
+    # Focus wh information into c-space
+    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True):  # ch_in, ch_out, kernel, stride, padding, groups
+        super().__init__()
+        self.conv = Conv(c1 * 4, c2, k, s, p, g, act)
+        # self.contract = Contract(gain=2)
+
+    def forward(self, x):  # x(b,c,w,h) -> y(b,4c,w/2,h/2)
+        return self.conv(torch.cat((x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]), 1))
+        # return self.conv(self.contract(x))
+
+
+class GhostConv(nn.Module):
+    # Ghost Convolution https://github.com/huawei-noah/ghostnet
+    def __init__(self, c1, c2, k=1, s=1, g=1, act=True):  # ch_in, ch_out, kernel, stride, groups
+        super().__init__()
+        c_ = c2 // 2  # hidden channels
+        self.cv1 = Conv(c1, c_, k, s, None, g, act)
+        self.cv2 = Conv(c_, c_, 5, 1, None, c_, act)
+
+    def forward(self, x):
+        y = self.cv1(x)
+        return torch.cat((y, self.cv2(y)), 1)
+
+
+class GhostBottleneck(nn.Module):
+    # Ghost Bottleneck https://github.com/huawei-noah/ghostnet
+    def __init__(self, c1, c2, k=3, s=1):  # ch_in, ch_out, kernel, stride
+        super().__init__()
+        c_ = c2 // 2
+        self.conv = nn.Sequential(
+            GhostConv(c1, c_, 1, 1),  # pw
+            DWConv(c_, c_, k, s, act=False) if s == 2 else nn.Identity(),  # dw
+            GhostConv(c_, c2, 1, 1, act=False))  # pw-linear
+        self.shortcut = nn.Sequential(DWConv(c1, c1, k, s, act=False), Conv(c1, c2, 1, 1,
+                                                                            act=False)) if s == 2 else nn.Identity()
+
+    def forward(self, x):
+        return self.conv(x) + self.shortcut(x)
+
+
+class Contract(nn.Module):
+    # Contract width-height into channels, i.e. x(1,64,80,80) to x(1,256,40,40)
+    def __init__(self, gain=2):
+        super().__init__()
+        self.gain = gain
+
+    def forward(self, x):
+        b, c, h, w = x.size()  # assert (h / s == 0) and (W / s == 0), 'Indivisible gain'
+        s = self.gain
+        x = x.view(b, c, h // s, s, w // s, s)  # x(1,64,40,2,40,2)
+        x = x.permute(0, 3, 5, 1, 2, 4).contiguous()  # x(1,2,2,64,40,40)
+        return x.view(b, c * s * s, h // s, w // s)  # x(1,256,40,40)
+
+
+class Expand(nn.Module):
+    # Expand channels into width-height, i.e. x(1,64,80,80) to x(1,16,160,160)
+    def __init__(self, gain=2):
+        super().__init__()
+        self.gain = gain
+
+    def forward(self, x):
+        b, c, h, w = x.size()  # assert C / s ** 2 == 0, 'Indivisible gain'
+        s = self.gain
+        x = x.view(b, s, s, c // s ** 2, h, w)  # x(1,2,2,16,80,80)
+        x = x.permute(0, 3, 4, 1, 5, 2).contiguous()  # x(1,16,80,2,80,2)
+        return x.view(b, c // s ** 2, h * s, w * s)  # x(1,16,160,160)
+
+
+class Concat(nn.Module):
+    # Concatenate a list of tensors along dimension
+    def __init__(self, dimension=1):
+        super().__init__()
+        self.d = dimension
+
+    def forward(self, x):
+        return torch.cat(x, self.d)
+
+
+class DetectMultiBackend(nn.Module):
+    # YOLOv5 MultiBackend class for python inference on various backends
+    def __init__(self, weights='yolov5s.pt', device=torch.device('cpu'), dnn=False, data=None, fp16=False, fuse=True):
+        # Usage:
+        #   PyTorch:              weights = *.pt
+        #   TorchScript:                    *.torchscript
+        #   ONNX Runtime:                   *.onnx
+        #   ONNX OpenCV DNN:                *.onnx with --dnn
+        #   OpenVINO:                       *.xml
+        #   CoreML:                         *.mlmodel
+        #   TensorRT:                       *.engine
+        #   TensorFlow SavedModel:          *_saved_model
+        #   TensorFlow GraphDef:            *.pb
+        #   TensorFlow Lite:                *.tflite
+        #   TensorFlow Edge TPU:            *_edgetpu.tflite
+        from models.experimental import attempt_download, attempt_load  # scoped to avoid circular import
+
+        super().__init__()
+        w = str(weights[0] if isinstance(weights, list) else weights)
+        pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs = self.model_type(w)  # get backend
+        w = attempt_download(w)  # download if not local
+        fp16 &= (pt or jit or onnx or engine) and device.type != 'cpu'  # FP16
+        stride, names = 32, [f'class{i}' for i in range(1000)]  # assign defaults
+        if data:  # assign class names (optional)
+            with open(data, errors='ignore') as f:
+                names = yaml.safe_load(f)['names']
+
+        if pt:  # PyTorch
+            model = attempt_load(weights if isinstance(weights, list) else w, device=device, inplace=True, fuse=fuse)
+            stride = max(int(model.stride.max()), 32)  # model stride
+            names = model.module.names if hasattr(model, 'module') else model.names  # get class names
+            model.half() if fp16 else model.float()
+            self.model = model  # explicitly assign for to(), cpu(), cuda(), half()
+        elif jit:  # TorchScript
+            LOGGER.info(f'Loading {w} for TorchScript inference...')
+            extra_files = {'config.txt': ''}  # model metadata
+            model = torch.jit.load(w, _extra_files=extra_files)
+            model.half() if fp16 else model.float()
+            if extra_files['config.txt']:
+                d = json.loads(extra_files['config.txt'])  # extra_files dict
+                stride, names = int(d['stride']), d['names']
+        elif dnn:  # ONNX OpenCV DNN
+            LOGGER.info(f'Loading {w} for ONNX OpenCV DNN inference...')
+            check_requirements(('opencv-python>=4.5.4',))
+            net = cv2.dnn.readNetFromONNX(w)
+        elif onnx:  # ONNX Runtime
+            LOGGER.info(f'Loading {w} for ONNX Runtime inference...')
+            cuda = torch.cuda.is_available()
+            check_requirements(('onnx', 'onnxruntime-gpu' if cuda else 'onnxruntime'))
+            import onnxruntime
+            providers = ['CUDAExecutionProvider', 'CPUExecutionProvider'] if cuda else ['CPUExecutionProvider']
+            session = onnxruntime.InferenceSession(w, providers=providers)
+            meta = session.get_modelmeta().custom_metadata_map  # metadata
+            if 'stride' in meta:
+                stride, names = int(meta['stride']), eval(meta['names'])
+        elif xml:  # OpenVINO
+            LOGGER.info(f'Loading {w} for OpenVINO inference...')
+            check_requirements(('openvino',))  # requires openvino-dev: https://pypi.org/project/openvino-dev/
+            from openvino.runtime import Core, Layout, get_batch
+            ie = Core()
+            if not Path(w).is_file():  # if not *.xml
+                w = next(Path(w).glob('*.xml'))  # get *.xml file from *_openvino_model dir
+            network = ie.read_model(model=w, weights=Path(w).with_suffix('.bin'))
+            if network.get_parameters()[0].get_layout().empty:
+                network.get_parameters()[0].set_layout(Layout("NCHW"))
+            batch_dim = get_batch(network)
+            if batch_dim.is_static:
+                batch_size = batch_dim.get_length()
+            executable_network = ie.compile_model(network, device_name="CPU")  # device_name="MYRIAD" for Intel NCS2
+            output_layer = next(iter(executable_network.outputs))
+            meta = Path(w).with_suffix('.yaml')
+            if meta.exists():
+                stride, names = self._load_metadata(meta)  # load metadata
+        elif engine:  # TensorRT
+            LOGGER.info(f'Loading {w} for TensorRT inference...')
+            import tensorrt as trt  # https://developer.nvidia.com/nvidia-tensorrt-download
+            check_version(trt.__version__, '7.0.0', hard=True)  # require tensorrt>=7.0.0
+            Binding = namedtuple('Binding', ('name', 'dtype', 'shape', 'data', 'ptr'))
+            logger = trt.Logger(trt.Logger.INFO)
+            with open(w, 'rb') as f, trt.Runtime(logger) as runtime:
+                model = runtime.deserialize_cuda_engine(f.read())
+            context = model.create_execution_context()
+            bindings = OrderedDict()
+            fp16 = False  # default updated below
+            dynamic = False
+            for index in range(model.num_bindings):
+                name = model.get_binding_name(index)
+                dtype = trt.nptype(model.get_binding_dtype(index))
+                if model.binding_is_input(index):
+                    if -1 in tuple(model.get_binding_shape(index)):  # dynamic
+                        dynamic = True
+                        context.set_binding_shape(index, tuple(model.get_profile_shape(0, index)[2]))
+                    if dtype == np.float16:
+                        fp16 = True
+                shape = tuple(context.get_binding_shape(index))
+                data = torch.from_numpy(np.empty(shape, dtype=np.dtype(dtype))).to(device)
+                bindings[name] = Binding(name, dtype, shape, data, int(data.data_ptr()))
+            binding_addrs = OrderedDict((n, d.ptr) for n, d in bindings.items())
+            batch_size = bindings['images'].shape[0]  # if dynamic, this is instead max batch size
+        elif coreml:  # CoreML
+            LOGGER.info(f'Loading {w} for CoreML inference...')
+            import coremltools as ct
+            model = ct.models.MLModel(w)
+        else:  # TensorFlow (SavedModel, GraphDef, Lite, Edge TPU)
+            if saved_model:  # SavedModel
+                LOGGER.info(f'Loading {w} for TensorFlow SavedModel inference...')
+                import tensorflow as tf
+                keras = False  # assume TF1 saved_model
+                model = tf.keras.models.load_model(w) if keras else tf.saved_model.load(w)
+            elif pb:  # GraphDef https://www.tensorflow.org/guide/migrate#a_graphpb_or_graphpbtxt
+                LOGGER.info(f'Loading {w} for TensorFlow GraphDef inference...')
+                import tensorflow as tf
+
+                def wrap_frozen_graph(gd, inputs, outputs):
+                    x = tf.compat.v1.wrap_function(lambda: tf.compat.v1.import_graph_def(gd, name=""), [])  # wrapped
+                    ge = x.graph.as_graph_element
+                    return x.prune(tf.nest.map_structure(ge, inputs), tf.nest.map_structure(ge, outputs))
+
+                gd = tf.Graph().as_graph_def()  # graph_def
+                with open(w, 'rb') as f:
+                    gd.ParseFromString(f.read())
+                frozen_func = wrap_frozen_graph(gd, inputs="x:0", outputs="Identity:0")
+            elif tflite or edgetpu:  # https://www.tensorflow.org/lite/guide/python#install_tensorflow_lite_for_python
+                try:  # https://coral.ai/docs/edgetpu/tflite-python/#update-existing-tf-lite-code-for-the-edge-tpu
+                    from tflite_runtime.interpreter import Interpreter, load_delegate
+                except ImportError:
+                    import tensorflow as tf
+                    Interpreter, load_delegate = tf.lite.Interpreter, tf.lite.experimental.load_delegate,
+                if edgetpu:  # Edge TPU https://coral.ai/software/#edgetpu-runtime
+                    LOGGER.info(f'Loading {w} for TensorFlow Lite Edge TPU inference...')
+                    delegate = {
+                        'Linux': 'libedgetpu.so.1',
+                        'Darwin': 'libedgetpu.1.dylib',
+                        'Windows': 'edgetpu.dll'}[platform.system()]
+                    interpreter = Interpreter(model_path=w, experimental_delegates=[load_delegate(delegate)])
+                else:  # Lite
+                    LOGGER.info(f'Loading {w} for TensorFlow Lite inference...')
+                    interpreter = Interpreter(model_path=w)  # load TFLite model
+                interpreter.allocate_tensors()  # allocate
+                input_details = interpreter.get_input_details()  # inputs
+                output_details = interpreter.get_output_details()  # outputs
+            elif tfjs:
+                raise Exception('ERROR: YOLOv5 TF.js inference is not supported')
+            else:
+                raise Exception(f'ERROR: {w} is not a supported format')
+        self.__dict__.update(locals())  # assign all variables to self
+
+    def forward(self, im, augment=False, visualize=False, val=False):
+        # YOLOv5 MultiBackend inference
+        b, ch, h, w = im.shape  # batch, channel, height, width
+        if self.fp16 and im.dtype != torch.float16:
+            im = im.half()  # to FP16
+
+        if self.pt:  # PyTorch
+            y = self.model(im, augment=augment, visualize=visualize)[0]
+        elif self.jit:  # TorchScript
+            y = self.model(im)[0]
+        elif self.dnn:  # ONNX OpenCV DNN
+            im = im.cpu().numpy()  # torch to numpy
+            self.net.setInput(im)
+            y = self.net.forward()
+        elif self.onnx:  # ONNX Runtime
+            im = im.cpu().numpy()  # torch to numpy
+            y = self.session.run([self.session.get_outputs()[0].name], {self.session.get_inputs()[0].name: im})[0]
+        elif self.xml:  # OpenVINO
+            im = im.cpu().numpy()  # FP32
+            y = self.executable_network([im])[self.output_layer]
+        elif self.engine:  # TensorRT
+            if self.dynamic and im.shape != self.bindings['images'].shape:
+                i_in, i_out = (self.model.get_binding_index(x) for x in ('images', 'output'))
+                self.context.set_binding_shape(i_in, im.shape)  # reshape if dynamic
+                self.bindings['images'] = self.bindings['images']._replace(shape=im.shape)
+                self.bindings['output'].data.resize_(tuple(self.context.get_binding_shape(i_out)))
+            s = self.bindings['images'].shape
+            assert im.shape == s, f"input size {im.shape} {'>' if self.dynamic else 'not equal to'} max model size {s}"
+            self.binding_addrs['images'] = int(im.data_ptr())
+            self.context.execute_v2(list(self.binding_addrs.values()))
+            y = self.bindings['output'].data
+        elif self.coreml:  # CoreML
+            im = im.permute(0, 2, 3, 1).cpu().numpy()  # torch BCHW to numpy BHWC shape(1,320,192,3)
+            im = Image.fromarray((im[0] * 255).astype('uint8'))
+            # im = im.resize((192, 320), Image.ANTIALIAS)
+            y = self.model.predict({'image': im})  # coordinates are xywh normalized
+            if 'confidence' in y:
+                box = xywh2xyxy(y['coordinates'] * [[w, h, w, h]])  # xyxy pixels
+                conf, cls = y['confidence'].max(1), y['confidence'].argmax(1).astype(np.float)
+                y = np.concatenate((box, conf.reshape(-1, 1), cls.reshape(-1, 1)), 1)
+            else:
+                k = 'var_' + str(sorted(int(k.replace('var_', '')) for k in y)[-1])  # output key
+                y = y[k]  # output
+        else:  # TensorFlow (SavedModel, GraphDef, Lite, Edge TPU)
+            im = im.permute(0, 2, 3, 1).cpu().numpy()  # torch BCHW to numpy BHWC shape(1,320,192,3)
+            if self.saved_model:  # SavedModel
+                y = (self.model(im, training=False) if self.keras else self.model(im)).numpy()
+            elif self.pb:  # GraphDef
+                y = self.frozen_func(x=self.tf.constant(im)).numpy()
+            else:  # Lite or Edge TPU
+                input, output = self.input_details[0], self.output_details[0]
+                int8 = input['dtype'] == np.uint8  # is TFLite quantized uint8 model
+                if int8:
+                    scale, zero_point = input['quantization']
+                    im = (im / scale + zero_point).astype(np.uint8)  # de-scale
+                self.interpreter.set_tensor(input['index'], im)
+                self.interpreter.invoke()
+                y = self.interpreter.get_tensor(output['index'])
+                if int8:
+                    scale, zero_point = output['quantization']
+                    y = (y.astype(np.float32) - zero_point) * scale  # re-scale
+            y[..., :4] *= [w, h, w, h]  # xywh normalized to pixels
+
+        if isinstance(y, np.ndarray):
+            y = torch.tensor(y, device=self.device)
+        return (y, []) if val else y
+
+    def warmup(self, imgsz=(1, 3, 640, 640)):
+        # Warmup model by running inference once
+        warmup_types = self.pt, self.jit, self.onnx, self.engine, self.saved_model, self.pb
+        if any(warmup_types) and self.device.type != 'cpu':
+            im = torch.zeros(*imgsz, dtype=torch.half if self.fp16 else torch.float, device=self.device)  # input
+            for _ in range(2 if self.jit else 1):  #
+                self.forward(im)  # warmup
+
+    @staticmethod
+    def model_type(p='path/to/model.pt'):
+        # Return model type from model path, i.e. path='path/to/model.onnx' -> type=onnx
+        from export import export_formats
+        suffixes = list(export_formats().Suffix) + ['.xml']  # export suffixes
+        check_suffix(p, suffixes)  # checks
+        p = Path(p).name  # eliminate trailing separators
+        pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, xml2 = (s in p for s in suffixes)
+        xml |= xml2  # *_openvino_model or *.xml
+        tflite &= not edgetpu  # *.tflite
+        return pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs
+
+    @staticmethod
+    def _load_metadata(f='path/to/meta.yaml'):
+        # Load metadata from meta.yaml if it exists
+        with open(f, errors='ignore') as f:
+            d = yaml.safe_load(f)
+        return d['stride'], d['names']  # assign stride, names
+
+
+class AutoShape(nn.Module):
+    # YOLOv5 input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS
+    conf = 0.25  # NMS confidence threshold
+    iou = 0.45  # NMS IoU threshold
+    agnostic = False  # NMS class-agnostic
+    multi_label = False  # NMS multiple labels per box
+    classes = None  # (optional list) filter by class, i.e. = [0, 15, 16] for COCO persons, cats and dogs
+    max_det = 1000  # maximum number of detections per image
+    amp = False  # Automatic Mixed Precision (AMP) inference
+
+    def __init__(self, model, verbose=True):
+        super().__init__()
+        if verbose:
+            LOGGER.info('Adding AutoShape... ')
+        copy_attr(self, model, include=('yaml', 'nc', 'hyp', 'names', 'stride', 'abc'), exclude=())  # copy attributes
+        self.dmb = isinstance(model, DetectMultiBackend)  # DetectMultiBackend() instance
+        self.pt = not self.dmb or model.pt  # PyTorch model
+        self.model = model.eval()
+        if self.pt:
+            m = self.model.model.model[-1] if self.dmb else self.model.model[-1]  # Detect()
+            m.inplace = False  # Detect.inplace=False for safe multithread inference
+
+    def _apply(self, fn):
+        # Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers
+        self = super()._apply(fn)
+        if self.pt:
+            m = self.model.model.model[-1] if self.dmb else self.model.model[-1]  # Detect()
+            m.stride = fn(m.stride)
+            m.grid = list(map(fn, m.grid))
+            if isinstance(m.anchor_grid, list):
+                m.anchor_grid = list(map(fn, m.anchor_grid))
+        return self
+
+    @torch.no_grad()
+    def forward(self, imgs, size=640, augment=False, profile=False):
+        # Inference from various sources. For height=640, width=1280, RGB images example inputs are:
+        #   file:       imgs = 'data/images/zidane.jpg'  # str or PosixPath
+        #   URI:             = 'https://ultralytics.com/images/zidane.jpg'
+        #   OpenCV:          = cv2.imread('image.jpg')[:,:,::-1]  # HWC BGR to RGB x(640,1280,3)
+        #   PIL:             = Image.open('image.jpg') or ImageGrab.grab()  # HWC x(640,1280,3)
+        #   numpy:           = np.zeros((640,1280,3))  # HWC
+        #   torch:           = torch.zeros(16,3,320,640)  # BCHW (scaled to size=640, 0-1 values)
+        #   multiple:        = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...]  # list of images
+
+        t = [time_sync()]
+        p = next(self.model.parameters()) if self.pt else torch.zeros(1, device=self.model.device)  # for device, type
+        autocast = self.amp and (p.device.type != 'cpu')  # Automatic Mixed Precision (AMP) inference
+        if isinstance(imgs, torch.Tensor):  # torch
+            with amp.autocast(autocast):
+                return self.model(imgs.to(p.device).type_as(p), augment, profile)  # inference
+
+        # Pre-process
+        n, imgs = (len(imgs), list(imgs)) if isinstance(imgs, (list, tuple)) else (1, [imgs])  # number, list of images
+        shape0, shape1, files = [], [], []  # image and inference shapes, filenames
+        for i, im in enumerate(imgs):
+            f = f'image{i}'  # filename
+            if isinstance(im, (str, Path)):  # filename or uri
+                im, f = Image.open(requests.get(im, stream=True).raw if str(im).startswith('http') else im), im
+                im = np.asarray(exif_transpose(im))
+            elif isinstance(im, Image.Image):  # PIL Image
+                im, f = np.asarray(exif_transpose(im)), getattr(im, 'filename', f) or f
+            files.append(Path(f).with_suffix('.jpg').name)
+            if im.shape[0] < 5:  # image in CHW
+                im = im.transpose((1, 2, 0))  # reverse dataloader .transpose(2, 0, 1)
+            im = im[..., :3] if im.ndim == 3 else np.tile(im[..., None], 3)  # enforce 3ch input
+            s = im.shape[:2]  # HWC
+            shape0.append(s)  # image shape
+            g = (size / max(s))  # gain
+            shape1.append([y * g for y in s])
+            imgs[i] = im if im.data.contiguous else np.ascontiguousarray(im)  # update
+        shape1 = [make_divisible(x, self.stride) if self.pt else size for x in np.array(shape1).max(0)]  # inf shape
+        x = [letterbox(im, shape1, auto=False)[0] for im in imgs]  # pad
+        x = np.ascontiguousarray(np.array(x).transpose((0, 3, 1, 2)))  # stack and BHWC to BCHW
+        x = torch.from_numpy(x).to(p.device).type_as(p) / 255  # uint8 to fp16/32
+        t.append(time_sync())
+
+        with amp.autocast(autocast):
+            # Inference
+            y = self.model(x, augment, profile)  # forward
+            t.append(time_sync())
+
+            # Post-process
+            y = non_max_suppression(y if self.dmb else y[0],
+                                    self.conf,
+                                    self.iou,
+                                    self.classes,
+                                    self.agnostic,
+                                    self.multi_label,
+                                    max_det=self.max_det)  # NMS
+            for i in range(n):
+                scale_coords(shape1, y[i][:, :4], shape0[i])
+
+            t.append(time_sync())
+            return Detections(imgs, y, files, t, self.names, x.shape)
+
+
+class Detections:
+    # YOLOv5 detections class for inference results
+    def __init__(self, imgs, pred, files, times=(0, 0, 0, 0), names=None, shape=None):
+        super().__init__()
+        d = pred[0].device  # device
+        gn = [torch.tensor([*(im.shape[i] for i in [1, 0, 1, 0]), 1, 1], device=d) for im in imgs]  # normalizations
+        self.imgs = imgs  # list of images as numpy arrays
+        self.pred = pred  # list of tensors pred[0] = (xyxy, conf, cls)
+        self.names = names  # class names
+        self.files = files  # image filenames
+        self.times = times  # profiling times
+        self.xyxy = pred  # xyxy pixels
+        self.xywh = [xyxy2xywh(x) for x in pred]  # xywh pixels
+        self.xyxyn = [x / g for x, g in zip(self.xyxy, gn)]  # xyxy normalized
+        self.xywhn = [x / g for x, g in zip(self.xywh, gn)]  # xywh normalized
+        self.n = len(self.pred)  # number of images (batch size)
+        self.t = tuple((times[i + 1] - times[i]) * 1000 / self.n for i in range(3))  # timestamps (ms)
+        self.s = shape  # inference BCHW shape
+
+    def display(self, pprint=False, show=False, save=False, crop=False, render=False, labels=True, save_dir=Path('')):
+        crops = []
+        for i, (im, pred) in enumerate(zip(self.imgs, self.pred)):
+            s = f'image {i + 1}/{len(self.pred)}: {im.shape[0]}x{im.shape[1]} '  # string
+            if pred.shape[0]:
+                for c in pred[:, -1].unique():
+                    n = (pred[:, -1] == c).sum()  # detections per class
+                    s += f"{n} {self.names[int(c)]}{'s' * (n > 1)}, "  # add to string
+                if show or save or render or crop:
+                    annotator = Annotator(im, example=str(self.names))
+                    for *box, conf, cls in reversed(pred):  # xyxy, confidence, class
+                        label = f'{self.names[int(cls)]} {conf:.2f}'
+                        if crop:
+                            file = save_dir / 'crops' / self.names[int(cls)] / self.files[i] if save else None
+                            crops.append({
+                                'box': box,
+                                'conf': conf,
+                                'cls': cls,
+                                'label': label,
+                                'im': save_one_box(box, im, file=file, save=save)})
+                        else:  # all others
+                            annotator.box_label(box, label if labels else '', color=colors(cls))
+                    im = annotator.im
+            else:
+                s += '(no detections)'
+
+            im = Image.fromarray(im.astype(np.uint8)) if isinstance(im, np.ndarray) else im  # from np
+            if pprint:
+                print(s.rstrip(', '))
+            if show:
+                im.show(self.files[i])  # show
+            if save:
+                f = self.files[i]
+                im.save(save_dir / f)  # save
+                if i == self.n - 1:
+                    LOGGER.info(f"Saved {self.n} image{'s' * (self.n > 1)} to {colorstr('bold', save_dir)}")
+            if render:
+                self.imgs[i] = np.asarray(im)
+        if crop:
+            if save:
+                LOGGER.info(f'Saved results to {save_dir}\n')
+            return crops
+
+    def print(self):
+        self.display(pprint=True)  # print results
+        print(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {tuple(self.s)}' % self.t)
+
+    def show(self, labels=True):
+        self.display(show=True, labels=labels)  # show results
+
+    def save(self, labels=True, save_dir='runs/detect/exp'):
+        save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True)  # increment save_dir
+        self.display(save=True, labels=labels, save_dir=save_dir)  # save results
+
+    def crop(self, save=True, save_dir='runs/detect/exp'):
+        save_dir = increment_path(save_dir, exist_ok=save_dir != 'runs/detect/exp', mkdir=True) if save else None
+        return self.display(crop=True, save=save, save_dir=save_dir)  # crop results
+
+    def render(self, labels=True):
+        self.display(render=True, labels=labels)  # render results
+        return self.imgs
+
+    def pandas(self):
+        # return detections as pandas DataFrames, i.e. print(results.pandas().xyxy[0])
+        new = copy(self)  # return copy
+        ca = 'xmin', 'ymin', 'xmax', 'ymax', 'confidence', 'class', 'name'  # xyxy columns
+        cb = 'xcenter', 'ycenter', 'width', 'height', 'confidence', 'class', 'name'  # xywh columns
+        for k, c in zip(['xyxy', 'xyxyn', 'xywh', 'xywhn'], [ca, ca, cb, cb]):
+            a = [[x[:5] + [int(x[5]), self.names[int(x[5])]] for x in x.tolist()] for x in getattr(self, k)]  # update
+            setattr(new, k, [pd.DataFrame(x, columns=c) for x in a])
+        return new
+
+    def tolist(self):
+        # return a list of Detections objects, i.e. 'for result in results.tolist():'
+        r = range(self.n)  # iterable
+        x = [Detections([self.imgs[i]], [self.pred[i]], [self.files[i]], self.times, self.names, self.s) for i in r]
+        # for d in x:
+        #    for k in ['imgs', 'pred', 'xyxy', 'xyxyn', 'xywh', 'xywhn']:
+        #        setattr(d, k, getattr(d, k)[0])  # pop out of list
+        return x
+
+    def __len__(self):
+        return self.n  # override len(results)
+
+    def __str__(self):
+        self.print()  # override print(results)
+        return ''
+
+
+class Classify(nn.Module):
+    # Classification head, i.e. x(b,c1,20,20) to x(b,c2)
+    def __init__(self, c1, c2, k=1, s=1, p=None, g=1):  # ch_in, ch_out, kernel, stride, padding, groups
+        super().__init__()
+        self.aap = nn.AdaptiveAvgPool2d(1)  # to x(b,c1,1,1)
+        self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g)  # to x(b,c2,1,1)
+        self.flat = nn.Flatten()
+
+    def forward(self, x):
+        z = torch.cat([self.aap(y) for y in (x if isinstance(x, list) else [x])], 1)  # cat if list
+        return self.flat(self.conv(z))  # flatten to x(b,c2)

models/experimental.py

@@ -0,0 +1,102 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+"""
+Experimental modules
+"""
+import math
+
+import numpy as np
+import torch
+import torch.nn as nn
+
+from models.common import Conv
+from utils.downloads import attempt_download
+
+
+class Sum(nn.Module):
+    # Weighted sum of 2 or more layers https://arxiv.org/abs/1911.09070
+    def __init__(self, n, weight=False):  # n: number of inputs
+        super().__init__()
+        self.weight = weight  # apply weights boolean
+        self.iter = range(n - 1)  # iter object
+        if weight:
+            self.w = nn.Parameter(-torch.arange(1.0, n) / 2, requires_grad=True)  # layer weights
+
+    def forward(self, x):
+        y = x[0]  # no weight
+        if self.weight:
+            w = torch.sigmoid(self.w) * 2
+            for i in self.iter:
+                y = y + x[i + 1] * w[i]
+        else:
+            for i in self.iter:
+                y = y + x[i + 1]
+        return y
+
+
+class MixConv2d(nn.Module):
+    # Mixed Depth-wise Conv https://arxiv.org/abs/1907.09595
+    def __init__(self, c1, c2, k=(1, 3), s=1, equal_ch=True):  # ch_in, ch_out, kernel, stride, ch_strategy
+        super().__init__()
+        n = len(k)  # number of convolutions
+        if equal_ch:  # equal c_ per group
+            i = torch.linspace(0, n - 1E-6, c2).floor()  # c2 indices
+            c_ = [(i == g).sum() for g in range(n)]  # intermediate channels
+        else:  # equal weight.numel() per group
+            b = [c2] + [0] * n
+            a = np.eye(n + 1, n, k=-1)
+            a -= np.roll(a, 1, axis=1)
+            a *= np.array(k) ** 2
+            a[0] = 1
+            c_ = np.linalg.lstsq(a, b, rcond=None)[0].round()  # solve for equal weight indices, ax = b
+
+        self.m = nn.ModuleList([
+            nn.Conv2d(c1, int(c_), k, s, k // 2, groups=math.gcd(c1, int(c_)), bias=False) for k, c_ in zip(k, c_)])
+        self.bn = nn.BatchNorm2d(c2)
+        self.act = nn.SiLU()
+
+    def forward(self, x):
+        return self.act(self.bn(torch.cat([m(x) for m in self.m], 1)))
+
+
+class Ensemble(nn.ModuleList):
+    # Ensemble of models
+    def __init__(self):
+        super().__init__()
+
+    def forward(self, x, augment=False, profile=False, visualize=False):
+        y = [module(x, augment, profile, visualize)[0] for module in self]
+        # y = torch.stack(y).max(0)[0]  # max ensemble
+        # y = torch.stack(y).mean(0)  # mean ensemble
+        y = torch.cat(y, 1)  # nms ensemble
+        return y, None  # inference, train output
+
+
+def attempt_load(weights, device=None, inplace=True, fuse=True):
+    # Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a
+    from models.yolo import Detect, Model
+
+    model = Ensemble()
+    for w in weights if isinstance(weights, list) else [weights]:
+        ckpt = torch.load(attempt_download(w), map_location='cpu')  # load
+        ckpt = (ckpt.get('ema') or ckpt['model']).to(device).float()  # FP32 model
+        model.append(ckpt.fuse().eval() if fuse else ckpt.eval())  # fused or un-fused model in eval mode
+
+    # Compatibility updates
+    for m in model.modules():
+        t = type(m)
+        if t in (nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU, Detect, Model):
+            m.inplace = inplace  # torch 1.7.0 compatibility
+            if t is Detect and not isinstance(m.anchor_grid, list):
+                delattr(m, 'anchor_grid')
+                setattr(m, 'anchor_grid', [torch.zeros(1)] * m.nl)
+        elif t is nn.Upsample and not hasattr(m, 'recompute_scale_factor'):
+            m.recompute_scale_factor = None  # torch 1.11.0 compatibility
+
+    if len(model) == 1:
+        return model[-1]  # return model
+    print(f'Ensemble created with {weights}\n')
+    for k in 'names', 'nc', 'yaml':
+        setattr(model, k, getattr(model[0], k))
+    model.stride = model[torch.argmax(torch.tensor([m.stride.max() for m in model])).int()].stride  # max stride
+    assert all(model[0].nc == m.nc for m in model), f'Models have different class counts: {[m.nc for m in model]}'
+    return model  # return ensemble

models/hub/anchors.yaml

@@ -0,0 +1,59 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# Default anchors for COCO data
+
+
+# P5 -------------------------------------------------------------------------------------------------------------------
+# P5-640:
+anchors_p5_640:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+
+# P6 -------------------------------------------------------------------------------------------------------------------
+# P6-640:  thr=0.25: 0.9964 BPR, 5.54 anchors past thr, n=12, img_size=640, metric_all=0.281/0.716-mean/best, past_thr=0.469-mean: 9,11,  21,19,  17,41,  43,32,  39,70,  86,64,  65,131,  134,130,  120,265,  282,180,  247,354,  512,387
+anchors_p6_640:
+  - [9,11,  21,19,  17,41]  # P3/8
+  - [43,32,  39,70,  86,64]  # P4/16
+  - [65,131,  134,130,  120,265]  # P5/32
+  - [282,180,  247,354,  512,387]  # P6/64
+
+# P6-1280:  thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1280, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 19,27,  44,40,  38,94,  96,68,  86,152,  180,137,  140,301,  303,264,  238,542,  436,615,  739,380,  925,792
+anchors_p6_1280:
+  - [19,27,  44,40,  38,94]  # P3/8
+  - [96,68,  86,152,  180,137]  # P4/16
+  - [140,301,  303,264,  238,542]  # P5/32
+  - [436,615,  739,380,  925,792]  # P6/64
+
+# P6-1920:  thr=0.25: 0.9950 BPR, 5.55 anchors past thr, n=12, img_size=1920, metric_all=0.281/0.714-mean/best, past_thr=0.468-mean: 28,41,  67,59,  57,141,  144,103,  129,227,  270,205,  209,452,  455,396,  358,812,  653,922,  1109,570,  1387,1187
+anchors_p6_1920:
+  - [28,41,  67,59,  57,141]  # P3/8
+  - [144,103,  129,227,  270,205]  # P4/16
+  - [209,452,  455,396,  358,812]  # P5/32
+  - [653,922,  1109,570,  1387,1187]  # P6/64
+
+
+# P7 -------------------------------------------------------------------------------------------------------------------
+# P7-640:  thr=0.25: 0.9962 BPR, 6.76 anchors past thr, n=15, img_size=640, metric_all=0.275/0.733-mean/best, past_thr=0.466-mean: 11,11,  13,30,  29,20,  30,46,  61,38,  39,92,  78,80,  146,66,  79,163,  149,150,  321,143,  157,303,  257,402,  359,290,  524,372
+anchors_p7_640:
+  - [11,11,  13,30,  29,20]  # P3/8
+  - [30,46,  61,38,  39,92]  # P4/16
+  - [78,80,  146,66,  79,163]  # P5/32
+  - [149,150,  321,143,  157,303]  # P6/64
+  - [257,402,  359,290,  524,372]  # P7/128
+
+# P7-1280:  thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1280, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 19,22,  54,36,  32,77,  70,83,  138,71,  75,173,  165,159,  148,334,  375,151,  334,317,  251,626,  499,474,  750,326,  534,814,  1079,818
+anchors_p7_1280:
+  - [19,22,  54,36,  32,77]  # P3/8
+  - [70,83,  138,71,  75,173]  # P4/16
+  - [165,159,  148,334,  375,151]  # P5/32
+  - [334,317,  251,626,  499,474]  # P6/64
+  - [750,326,  534,814,  1079,818]  # P7/128
+
+# P7-1920:  thr=0.25: 0.9968 BPR, 6.71 anchors past thr, n=15, img_size=1920, metric_all=0.273/0.732-mean/best, past_thr=0.463-mean: 29,34,  81,55,  47,115,  105,124,  207,107,  113,259,  247,238,  222,500,  563,227,  501,476,  376,939,  749,711,  1126,489,  801,1222,  1618,1227
+anchors_p7_1920:
+  - [29,34,  81,55,  47,115]  # P3/8
+  - [105,124,  207,107,  113,259]  # P4/16
+  - [247,238,  222,500,  563,227]  # P5/32
+  - [501,476,  376,939,  749,711]  # P6/64
+  - [1126,489,  801,1222,  1618,1227]  # P7/128

models/hub/yolov3-spp.yaml

@@ -0,0 +1,51 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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, anchors]],   # Detect(P3, P4, P5)
+  ]

models/hub/yolov3-tiny.yaml

@@ -0,0 +1,41 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [10,14, 23,27, 37,58]  # P4/16
+  - [81,82, 135,169, 344,319]  # P5/32
+
+# 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, anchors]],  # Detect(P4, P5)
+  ]

models/hub/yolov3.yaml

@@ -0,0 +1,51 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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, anchors]],   # Detect(P3, P4, P5)
+  ]

models/hub/yolov5-bifpn.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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 BiFPN 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, 6], 1, Concat, [1]],  # cat P4 <--- BiFPN change
+   [-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, anchors]],  # Detect(P3, P4, P5)
+  ]

models/hub/yolov5-fpn.yaml

@@ -0,0 +1,42 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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 FPN head
+head:
+  [[-1, 3, C3, [1024, False]],  # 10 (P5/32-large)
+
+   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+   [[-1, 6], 1, Concat, [1]],  # cat backbone P4
+   [-1, 1, Conv, [512, 1, 1]],
+   [-1, 3, C3, [512, False]],  # 14 (P4/16-medium)
+
+   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+   [[-1, 4], 1, Concat, [1]],  # cat backbone P3
+   [-1, 1, Conv, [256, 1, 1]],
+   [-1, 3, C3, [256, False]],  # 18 (P3/8-small)
+
+   [[18, 14, 10], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5)
+  ]

models/hub/yolov5-p2.yaml

@@ -0,0 +1,54 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors: 3  # AutoAnchor evolves 3 anchors per P output layer
+
+# 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 with (P2, P3, P4, P5) outputs
+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, [128, 1, 1]],
+   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+   [[-1, 2], 1, Concat, [1]],  # cat backbone P2
+   [-1, 1, C3, [128, False]],  # 21 (P2/4-xsmall)
+
+   [-1, 1, Conv, [128, 3, 2]],
+   [[-1, 18], 1, Concat, [1]],  # cat head P3
+   [-1, 3, C3, [256, False]],  # 24 (P3/8-small)
+
+   [-1, 1, Conv, [256, 3, 2]],
+   [[-1, 14], 1, Concat, [1]],  # cat head P4
+   [-1, 3, C3, [512, False]],  # 27 (P4/16-medium)
+
+   [-1, 1, Conv, [512, 3, 2]],
+   [[-1, 10], 1, Concat, [1]],  # cat head P5
+   [-1, 3, C3, [1024, False]],  # 30 (P5/32-large)
+
+   [[21, 24, 27, 30], 1, Detect, [nc, anchors]],  # Detect(P2, P3, P4, P5)
+  ]

models/hub/yolov5-p34.yaml

@@ -0,0 +1,41 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.33  # model depth multiple
+width_multiple: 0.50  # layer channel multiple
+anchors: 3  # AutoAnchor evolves 3 anchors per P output layer
+
+# 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 with (P3, P4) outputs
+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)
+
+    [ [ 17, 20 ], 1, Detect, [ nc, anchors ] ],  # Detect(P3, P4)
+  ]

models/hub/yolov5-p6.yaml

@@ -0,0 +1,56 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors: 3  # AutoAnchor evolves 3 anchors per P output layer
+
+# 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 with (P3, P4, P5, P6) outputs
+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, anchors]],  # Detect(P3, P4, P5, P6)
+  ]

models/hub/yolov5-p7.yaml

@@ -0,0 +1,67 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors: 3  # AutoAnchor evolves 3 anchors per P output layer
+
+# 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, Conv, [1280, 3, 2]],  # 11-P7/128
+   [-1, 3, C3, [1280]],
+   [-1, 1, SPPF, [1280, 5]],  # 13
+  ]
+
+# YOLOv5 v6.0 head with (P3, P4, P5, P6, P7) outputs
+head:
+  [[-1, 1, Conv, [1024, 1, 1]],
+   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+   [[-1, 10], 1, Concat, [1]],  # cat backbone P6
+   [-1, 3, C3, [1024, False]],  # 17
+
+   [-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]],  # 21
+
+   [-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]],  # 25
+
+   [-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]],  # 29 (P3/8-small)
+
+   [-1, 1, Conv, [256, 3, 2]],
+   [[-1, 26], 1, Concat, [1]],  # cat head P4
+   [-1, 3, C3, [512, False]],  # 32 (P4/16-medium)
+
+   [-1, 1, Conv, [512, 3, 2]],
+   [[-1, 22], 1, Concat, [1]],  # cat head P5
+   [-1, 3, C3, [768, False]],  # 35 (P5/32-large)
+
+   [-1, 1, Conv, [768, 3, 2]],
+   [[-1, 18], 1, Concat, [1]],  # cat head P6
+   [-1, 3, C3, [1024, False]],  # 38 (P6/64-xlarge)
+
+   [-1, 1, Conv, [1024, 3, 2]],
+   [[-1, 14], 1, Concat, [1]],  # cat head P7
+   [-1, 3, C3, [1280, False]],  # 41 (P7/128-xxlarge)
+
+   [[29, 32, 35, 38, 41], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5, P6, P7)
+  ]

models/hub/yolov5-panet.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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 PANet 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, anchors]],  # Detect(P3, P4, P5)
+  ]

models/hub/yolov5l6.yaml

@@ -0,0 +1,60 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [19,27,  44,40,  38,94]  # P3/8
+  - [96,68,  86,152,  180,137]  # P4/16
+  - [140,301,  303,264,  238,542]  # P5/32
+  - [436,615,  739,380,  925,792]  # P6/64
+
+# 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, anchors]],  # Detect(P3, P4, P5, P6)
+  ]

models/hub/yolov5m6.yaml

@@ -0,0 +1,60 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.67  # model depth multiple
+width_multiple: 0.75  # layer channel multiple
+anchors:
+  - [19,27,  44,40,  38,94]  # P3/8
+  - [96,68,  86,152,  180,137]  # P4/16
+  - [140,301,  303,264,  238,542]  # P5/32
+  - [436,615,  739,380,  925,792]  # P6/64
+
+# 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, anchors]],  # Detect(P3, P4, P5, P6)
+  ]

models/hub/yolov5n6.yaml

@@ -0,0 +1,60 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.33  # model depth multiple
+width_multiple: 0.25  # layer channel multiple
+anchors:
+  - [19,27,  44,40,  38,94]  # P3/8
+  - [96,68,  86,152,  180,137]  # P4/16
+  - [140,301,  303,264,  238,542]  # P5/32
+  - [436,615,  739,380,  925,792]  # P6/64
+
+# 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, anchors]],  # Detect(P3, P4, P5, P6)
+  ]

models/hub/yolov5s-ghost.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.33  # model depth multiple
+width_multiple: 0.50  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# YOLOv5 v6.0 backbone
+backbone:
+  # [from, number, module, args]
+  [[-1, 1, Conv, [64, 6, 2, 2]],  # 0-P1/2
+   [-1, 1, GhostConv, [128, 3, 2]],  # 1-P2/4
+   [-1, 3, C3Ghost, [128]],
+   [-1, 1, GhostConv, [256, 3, 2]],  # 3-P3/8
+   [-1, 6, C3Ghost, [256]],
+   [-1, 1, GhostConv, [512, 3, 2]],  # 5-P4/16
+   [-1, 9, C3Ghost, [512]],
+   [-1, 1, GhostConv, [1024, 3, 2]],  # 7-P5/32
+   [-1, 3, C3Ghost, [1024]],
+   [-1, 1, SPPF, [1024, 5]],  # 9
+  ]
+
+# YOLOv5 v6.0 head
+head:
+  [[-1, 1, GhostConv, [512, 1, 1]],
+   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+   [[-1, 6], 1, Concat, [1]],  # cat backbone P4
+   [-1, 3, C3Ghost, [512, False]],  # 13
+
+   [-1, 1, GhostConv, [256, 1, 1]],
+   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+   [[-1, 4], 1, Concat, [1]],  # cat backbone P3
+   [-1, 3, C3Ghost, [256, False]],  # 17 (P3/8-small)
+
+   [-1, 1, GhostConv, [256, 3, 2]],
+   [[-1, 14], 1, Concat, [1]],  # cat head P4
+   [-1, 3, C3Ghost, [512, False]],  # 20 (P4/16-medium)
+
+   [-1, 1, GhostConv, [512, 3, 2]],
+   [[-1, 10], 1, Concat, [1]],  # cat head P5
+   [-1, 3, C3Ghost, [1024, False]],  # 23 (P5/32-large)
+
+   [[17, 20, 23], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5)
+  ]

models/hub/yolov5s-transformer.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.33  # model depth multiple
+width_multiple: 0.50  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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, C3TR, [1024]],  # 9 <--- C3TR() Transformer module
+   [-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, anchors]],  # Detect(P3, P4, P5)
+  ]

models/hub/yolov5s6.yaml

@@ -0,0 +1,60 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.33  # model depth multiple
+width_multiple: 0.50  # layer channel multiple
+anchors:
+  - [19,27,  44,40,  38,94]  # P3/8
+  - [96,68,  86,152,  180,137]  # P4/16
+  - [140,301,  303,264,  238,542]  # P5/32
+  - [436,615,  739,380,  925,792]  # P6/64
+
+# 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, anchors]],  # Detect(P3, P4, P5, P6)
+  ]

models/hub/yolov5x6.yaml

@@ -0,0 +1,60 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.33  # model depth multiple
+width_multiple: 1.25  # layer channel multiple
+anchors:
+  - [19,27,  44,40,  38,94]  # P3/8
+  - [96,68,  86,152,  180,137]  # P4/16
+  - [140,301,  303,264,  238,542]  # P5/32
+  - [436,615,  739,380,  925,792]  # P6/64
+
+# 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, anchors]],  # Detect(P3, P4, P5, P6)
+  ]

models/tf.py

@@ -0,0 +1,574 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+"""
+TensorFlow, Keras and TFLite versions of YOLOv5
+Authored by https://github.com/zldrobit in PR https://github.com/ultralytics/yolov5/pull/1127
+
+Usage:
+    $ python models/tf.py --weights yolov5s.pt
+
+Export:
+    $ python path/to/export.py --weights yolov5s.pt --include saved_model pb tflite tfjs
+"""
+
+import argparse
+import sys
+from copy import deepcopy
+from pathlib import Path
+
+FILE = Path(__file__).resolve()
+ROOT = FILE.parents[1]  # YOLOv5 root directory
+if str(ROOT) not in sys.path:
+    sys.path.append(str(ROOT))  # add ROOT to PATH
+# ROOT = ROOT.relative_to(Path.cwd())  # relative
+
+import numpy as np
+import tensorflow as tf
+import torch
+import torch.nn as nn
+from tensorflow import keras
+
+from models.common import (C3, SPP, SPPF, Bottleneck, BottleneckCSP, C3x, Concat, Conv, CrossConv, DWConv,
+                           DWConvTranspose2d, Focus, autopad)
+from models.experimental import MixConv2d, attempt_load
+from models.yolo import Detect
+from utils.activations import SiLU
+from utils.general import LOGGER, make_divisible, print_args
+
+
+class TFBN(keras.layers.Layer):
+    # TensorFlow BatchNormalization wrapper
+    def __init__(self, w=None):
+        super().__init__()
+        self.bn = keras.layers.BatchNormalization(
+            beta_initializer=keras.initializers.Constant(w.bias.numpy()),
+            gamma_initializer=keras.initializers.Constant(w.weight.numpy()),
+            moving_mean_initializer=keras.initializers.Constant(w.running_mean.numpy()),
+            moving_variance_initializer=keras.initializers.Constant(w.running_var.numpy()),
+            epsilon=w.eps)
+
+    def call(self, inputs):
+        return self.bn(inputs)
+
+
+class TFPad(keras.layers.Layer):
+    # Pad inputs in spatial dimensions 1 and 2
+    def __init__(self, pad):
+        super().__init__()
+        if isinstance(pad, int):
+            self.pad = tf.constant([[0, 0], [pad, pad], [pad, pad], [0, 0]])
+        else:  # tuple/list
+            self.pad = tf.constant([[0, 0], [pad[0], pad[0]], [pad[1], pad[1]], [0, 0]])
+
+    def call(self, inputs):
+        return tf.pad(inputs, self.pad, mode='constant', constant_values=0)
+
+
+class TFConv(keras.layers.Layer):
+    # Standard convolution
+    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None):
+        # ch_in, ch_out, weights, kernel, stride, padding, groups
+        super().__init__()
+        assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument"
+        # TensorFlow convolution padding is inconsistent with PyTorch (e.g. k=3 s=2 'SAME' padding)
+        # see https://stackoverflow.com/questions/52975843/comparing-conv2d-with-padding-between-tensorflow-and-pytorch
+        conv = keras.layers.Conv2D(
+            filters=c2,
+            kernel_size=k,
+            strides=s,
+            padding='SAME' if s == 1 else 'VALID',
+            use_bias=not hasattr(w, 'bn'),
+            kernel_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()),
+            bias_initializer='zeros' if hasattr(w, 'bn') else keras.initializers.Constant(w.conv.bias.numpy()))
+        self.conv = conv if s == 1 else keras.Sequential([TFPad(autopad(k, p)), conv])
+        self.bn = TFBN(w.bn) if hasattr(w, 'bn') else tf.identity
+        self.act = activations(w.act) if act else tf.identity
+
+    def call(self, inputs):
+        return self.act(self.bn(self.conv(inputs)))
+
+
+class TFDWConv(keras.layers.Layer):
+    # Depthwise convolution
+    def __init__(self, c1, c2, k=1, s=1, p=None, act=True, w=None):
+        # ch_in, ch_out, weights, kernel, stride, padding, groups
+        super().__init__()
+        assert c2 % c1 == 0, f'TFDWConv() output={c2} must be a multiple of input={c1} channels'
+        conv = keras.layers.DepthwiseConv2D(
+            kernel_size=k,
+            depth_multiplier=c2 // c1,
+            strides=s,
+            padding='SAME' if s == 1 else 'VALID',
+            use_bias=not hasattr(w, 'bn'),
+            depthwise_initializer=keras.initializers.Constant(w.conv.weight.permute(2, 3, 1, 0).numpy()),
+            bias_initializer='zeros' if hasattr(w, 'bn') else keras.initializers.Constant(w.conv.bias.numpy()))
+        self.conv = conv if s == 1 else keras.Sequential([TFPad(autopad(k, p)), conv])
+        self.bn = TFBN(w.bn) if hasattr(w, 'bn') else tf.identity
+        self.act = activations(w.act) if act else tf.identity
+
+    def call(self, inputs):
+        return self.act(self.bn(self.conv(inputs)))
+
+
+class TFDWConvTranspose2d(keras.layers.Layer):
+    # Depthwise ConvTranspose2d
+    def __init__(self, c1, c2, k=1, s=1, p1=0, p2=0, w=None):
+        # ch_in, ch_out, weights, kernel, stride, padding, groups
+        super().__init__()
+        assert c1 == c2, f'TFDWConv() output={c2} must be equal to input={c1} channels'
+        assert k == 4 and p1 == 1, 'TFDWConv() only valid for k=4 and p1=1'
+        weight, bias = w.weight.permute(2, 3, 1, 0).numpy(), w.bias.numpy()
+        self.c1 = c1
+        self.conv = [
+            keras.layers.Conv2DTranspose(filters=1,
+                                         kernel_size=k,
+                                         strides=s,
+                                         padding='VALID',
+                                         output_padding=p2,
+                                         use_bias=True,
+                                         kernel_initializer=keras.initializers.Constant(weight[..., i:i + 1]),
+                                         bias_initializer=keras.initializers.Constant(bias[i])) for i in range(c1)]
+
+    def call(self, inputs):
+        return tf.concat([m(x) for m, x in zip(self.conv, tf.split(inputs, self.c1, 3))], 3)[:, 1:-1, 1:-1]
+
+
+class TFFocus(keras.layers.Layer):
+    # Focus wh information into c-space
+    def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True, w=None):
+        # ch_in, ch_out, kernel, stride, padding, groups
+        super().__init__()
+        self.conv = TFConv(c1 * 4, c2, k, s, p, g, act, w.conv)
+
+    def call(self, inputs):  # x(b,w,h,c) -> y(b,w/2,h/2,4c)
+        # inputs = inputs / 255  # normalize 0-255 to 0-1
+        inputs = [inputs[:, ::2, ::2, :], inputs[:, 1::2, ::2, :], inputs[:, ::2, 1::2, :], inputs[:, 1::2, 1::2, :]]
+        return self.conv(tf.concat(inputs, 3))
+
+
+class TFBottleneck(keras.layers.Layer):
+    # Standard bottleneck
+    def __init__(self, c1, c2, shortcut=True, g=1, e=0.5, w=None):  # ch_in, ch_out, shortcut, groups, expansion
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
+        self.cv2 = TFConv(c_, c2, 3, 1, g=g, w=w.cv2)
+        self.add = shortcut and c1 == c2
+
+    def call(self, inputs):
+        return inputs + self.cv2(self.cv1(inputs)) if self.add else self.cv2(self.cv1(inputs))
+
+
+class TFCrossConv(keras.layers.Layer):
+    # Cross Convolution
+    def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False, w=None):
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = TFConv(c1, c_, (1, k), (1, s), w=w.cv1)
+        self.cv2 = TFConv(c_, c2, (k, 1), (s, 1), g=g, w=w.cv2)
+        self.add = shortcut and c1 == c2
+
+    def call(self, inputs):
+        return inputs + self.cv2(self.cv1(inputs)) if self.add else self.cv2(self.cv1(inputs))
+
+
+class TFConv2d(keras.layers.Layer):
+    # Substitution for PyTorch nn.Conv2D
+    def __init__(self, c1, c2, k, s=1, g=1, bias=True, w=None):
+        super().__init__()
+        assert g == 1, "TF v2.2 Conv2D does not support 'groups' argument"
+        self.conv = keras.layers.Conv2D(filters=c2,
+                                        kernel_size=k,
+                                        strides=s,
+                                        padding='VALID',
+                                        use_bias=bias,
+                                        kernel_initializer=keras.initializers.Constant(
+                                            w.weight.permute(2, 3, 1, 0).numpy()),
+                                        bias_initializer=keras.initializers.Constant(w.bias.numpy()) if bias else None)
+
+    def call(self, inputs):
+        return self.conv(inputs)
+
+
+class TFBottleneckCSP(keras.layers.Layer):
+    # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None):
+        # ch_in, ch_out, number, shortcut, groups, expansion
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
+        self.cv2 = TFConv2d(c1, c_, 1, 1, bias=False, w=w.cv2)
+        self.cv3 = TFConv2d(c_, c_, 1, 1, bias=False, w=w.cv3)
+        self.cv4 = TFConv(2 * c_, c2, 1, 1, w=w.cv4)
+        self.bn = TFBN(w.bn)
+        self.act = lambda x: keras.activations.swish(x)
+        self.m = keras.Sequential([TFBottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)])
+
+    def call(self, inputs):
+        y1 = self.cv3(self.m(self.cv1(inputs)))
+        y2 = self.cv2(inputs)
+        return self.cv4(self.act(self.bn(tf.concat((y1, y2), axis=3))))
+
+
+class TFC3(keras.layers.Layer):
+    # CSP Bottleneck with 3 convolutions
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None):
+        # ch_in, ch_out, number, shortcut, groups, expansion
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
+        self.cv2 = TFConv(c1, c_, 1, 1, w=w.cv2)
+        self.cv3 = TFConv(2 * c_, c2, 1, 1, w=w.cv3)
+        self.m = keras.Sequential([TFBottleneck(c_, c_, shortcut, g, e=1.0, w=w.m[j]) for j in range(n)])
+
+    def call(self, inputs):
+        return self.cv3(tf.concat((self.m(self.cv1(inputs)), self.cv2(inputs)), axis=3))
+
+
+class TFC3x(keras.layers.Layer):
+    # 3 module with cross-convolutions
+    def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5, w=None):
+        # ch_in, ch_out, number, shortcut, groups, expansion
+        super().__init__()
+        c_ = int(c2 * e)  # hidden channels
+        self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
+        self.cv2 = TFConv(c1, c_, 1, 1, w=w.cv2)
+        self.cv3 = TFConv(2 * c_, c2, 1, 1, w=w.cv3)
+        self.m = keras.Sequential([
+            TFCrossConv(c_, c_, k=3, s=1, g=g, e=1.0, shortcut=shortcut, w=w.m[j]) for j in range(n)])
+
+    def call(self, inputs):
+        return self.cv3(tf.concat((self.m(self.cv1(inputs)), self.cv2(inputs)), axis=3))
+
+
+class TFSPP(keras.layers.Layer):
+    # Spatial pyramid pooling layer used in YOLOv3-SPP
+    def __init__(self, c1, c2, k=(5, 9, 13), w=None):
+        super().__init__()
+        c_ = c1 // 2  # hidden channels
+        self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
+        self.cv2 = TFConv(c_ * (len(k) + 1), c2, 1, 1, w=w.cv2)
+        self.m = [keras.layers.MaxPool2D(pool_size=x, strides=1, padding='SAME') for x in k]
+
+    def call(self, inputs):
+        x = self.cv1(inputs)
+        return self.cv2(tf.concat([x] + [m(x) for m in self.m], 3))
+
+
+class TFSPPF(keras.layers.Layer):
+    # Spatial pyramid pooling-Fast layer
+    def __init__(self, c1, c2, k=5, w=None):
+        super().__init__()
+        c_ = c1 // 2  # hidden channels
+        self.cv1 = TFConv(c1, c_, 1, 1, w=w.cv1)
+        self.cv2 = TFConv(c_ * 4, c2, 1, 1, w=w.cv2)
+        self.m = keras.layers.MaxPool2D(pool_size=k, strides=1, padding='SAME')
+
+    def call(self, inputs):
+        x = self.cv1(inputs)
+        y1 = self.m(x)
+        y2 = self.m(y1)
+        return self.cv2(tf.concat([x, y1, y2, self.m(y2)], 3))
+
+
+class TFDetect(keras.layers.Layer):
+    # TF YOLOv5 Detect layer
+    def __init__(self, nc=80, anchors=(), ch=(), imgsz=(640, 640), w=None):  # detection layer
+        super().__init__()
+        self.stride = tf.convert_to_tensor(w.stride.numpy(), dtype=tf.float32)
+        self.nc = nc  # number of classes
+        self.no = nc + 5  # number of outputs per anchor
+        self.nl = len(anchors)  # number of detection layers
+        self.na = len(anchors[0]) // 2  # number of anchors
+        self.grid = [tf.zeros(1)] * self.nl  # init grid
+        self.anchors = tf.convert_to_tensor(w.anchors.numpy(), dtype=tf.float32)
+        self.anchor_grid = tf.reshape(self.anchors * tf.reshape(self.stride, [self.nl, 1, 1]), [self.nl, 1, -1, 1, 2])
+        self.m = [TFConv2d(x, self.no * self.na, 1, w=w.m[i]) for i, x in enumerate(ch)]
+        self.training = False  # set to False after building model
+        self.imgsz = imgsz
+        for i in range(self.nl):
+            ny, nx = self.imgsz[0] // self.stride[i], self.imgsz[1] // self.stride[i]
+            self.grid[i] = self._make_grid(nx, ny)
+
+    def call(self, inputs):
+        z = []  # inference output
+        x = []
+        for i in range(self.nl):
+            x.append(self.m[i](inputs[i]))
+            # x(bs,20,20,255) to x(bs,3,20,20,85)
+            ny, nx = self.imgsz[0] // self.stride[i], self.imgsz[1] // self.stride[i]
+            x[i] = tf.reshape(x[i], [-1, ny * nx, self.na, self.no])
+
+            if not self.training:  # inference
+                y = tf.sigmoid(x[i])
+                grid = tf.transpose(self.grid[i], [0, 2, 1, 3]) - 0.5
+                anchor_grid = tf.transpose(self.anchor_grid[i], [0, 2, 1, 3]) * 4
+                xy = (y[..., 0:2] * 2 + grid) * self.stride[i]  # xy
+                wh = y[..., 2:4] ** 2 * anchor_grid
+                # Normalize xywh to 0-1 to reduce calibration error
+                xy /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32)
+                wh /= tf.constant([[self.imgsz[1], self.imgsz[0]]], dtype=tf.float32)
+                y = tf.concat([xy, wh, y[..., 4:]], -1)
+                z.append(tf.reshape(y, [-1, self.na * ny * nx, self.no]))
+
+        return tf.transpose(x, [0, 2, 1, 3]) if self.training else (tf.concat(z, 1), x)
+
+    @staticmethod
+    def _make_grid(nx=20, ny=20):
+        # yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)])
+        # return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float()
+        xv, yv = tf.meshgrid(tf.range(nx), tf.range(ny))
+        return tf.cast(tf.reshape(tf.stack([xv, yv], 2), [1, 1, ny * nx, 2]), dtype=tf.float32)
+
+
+class TFUpsample(keras.layers.Layer):
+    # TF version of torch.nn.Upsample()
+    def __init__(self, size, scale_factor, mode, w=None):  # warning: all arguments needed including 'w'
+        super().__init__()
+        assert scale_factor == 2, "scale_factor must be 2"
+        self.upsample = lambda x: tf.image.resize(x, (x.shape[1] * 2, x.shape[2] * 2), method=mode)
+        # self.upsample = keras.layers.UpSampling2D(size=scale_factor, interpolation=mode)
+        # with default arguments: align_corners=False, half_pixel_centers=False
+        # self.upsample = lambda x: tf.raw_ops.ResizeNearestNeighbor(images=x,
+        #                                                            size=(x.shape[1] * 2, x.shape[2] * 2))
+
+    def call(self, inputs):
+        return self.upsample(inputs)
+
+
+class TFConcat(keras.layers.Layer):
+    # TF version of torch.concat()
+    def __init__(self, dimension=1, w=None):
+        super().__init__()
+        assert dimension == 1, "convert only NCHW to NHWC concat"
+        self.d = 3
+
+    def call(self, inputs):
+        return tf.concat(inputs, self.d)
+
+
+def parse_model(d, ch, model, imgsz):  # model_dict, input_channels(3)
+    LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10}  {'module':<40}{'arguments':<30}")
+    anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple']
+    na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors  # number of anchors
+    no = na * (nc + 5)  # number of outputs = anchors * (classes + 5)
+
+    layers, save, c2 = [], [], ch[-1]  # layers, savelist, ch out
+    for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']):  # from, number, module, args
+        m_str = m
+        m = eval(m) if isinstance(m, str) else m  # eval strings
+        for j, a in enumerate(args):
+            try:
+                args[j] = eval(a) if isinstance(a, str) else a  # eval strings
+            except NameError:
+                pass
+
+        n = max(round(n * gd), 1) if n > 1 else n  # depth gain
+        if m in [
+                nn.Conv2d, Conv, DWConv, DWConvTranspose2d, Bottleneck, SPP, SPPF, MixConv2d, Focus, CrossConv,
+                BottleneckCSP, C3, C3x]:
+            c1, c2 = ch[f], args[0]
+            c2 = make_divisible(c2 * gw, 8) if c2 != no else c2
+
+            args = [c1, c2, *args[1:]]
+            if m in [BottleneckCSP, C3, C3x]:
+                args.insert(2, n)
+                n = 1
+        elif m is nn.BatchNorm2d:
+            args = [ch[f]]
+        elif m is Concat:
+            c2 = sum(ch[-1 if x == -1 else x + 1] for x in f)
+        elif m is Detect:
+            args.append([ch[x + 1] for x in f])
+            if isinstance(args[1], int):  # number of anchors
+                args[1] = [list(range(args[1] * 2))] * len(f)
+            args.append(imgsz)
+        else:
+            c2 = ch[f]
+
+        tf_m = eval('TF' + m_str.replace('nn.', ''))
+        m_ = keras.Sequential([tf_m(*args, w=model.model[i][j]) for j in range(n)]) if n > 1 \
+            else tf_m(*args, w=model.model[i])  # module
+
+        torch_m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args)  # module
+        t = str(m)[8:-2].replace('__main__.', '')  # module type
+        np = sum(x.numel() for x in torch_m_.parameters())  # number params
+        m_.i, m_.f, m_.type, m_.np = i, f, t, np  # attach index, 'from' index, type, number params
+        LOGGER.info(f'{i:>3}{str(f):>18}{str(n):>3}{np:>10}  {t:<40}{str(args):<30}')  # print
+        save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1)  # append to savelist
+        layers.append(m_)
+        ch.append(c2)
+    return keras.Sequential(layers), sorted(save)
+
+
+class TFModel:
+    # TF YOLOv5 model
+    def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, model=None, imgsz=(640, 640)):  # model, channels, classes
+        super().__init__()
+        if isinstance(cfg, dict):
+            self.yaml = cfg  # model dict
+        else:  # is *.yaml
+            import yaml  # for torch hub
+            self.yaml_file = Path(cfg).name
+            with open(cfg) as f:
+                self.yaml = yaml.load(f, Loader=yaml.FullLoader)  # model dict
+
+        # Define model
+        if nc and nc != self.yaml['nc']:
+            LOGGER.info(f"Overriding {cfg} nc={self.yaml['nc']} with nc={nc}")
+            self.yaml['nc'] = nc  # override yaml value
+        self.model, self.savelist = parse_model(deepcopy(self.yaml), ch=[ch], model=model, imgsz=imgsz)
+
+    def predict(self,
+                inputs,
+                tf_nms=False,
+                agnostic_nms=False,
+                topk_per_class=100,
+                topk_all=100,
+                iou_thres=0.45,
+                conf_thres=0.25):
+        y = []  # outputs
+        x = inputs
+        for m in self.model.layers:
+            if m.f != -1:  # if not from previous layer
+                x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f]  # from earlier layers
+
+            x = m(x)  # run
+            y.append(x if m.i in self.savelist else None)  # save output
+
+        # Add TensorFlow NMS
+        if tf_nms:
+            boxes = self._xywh2xyxy(x[0][..., :4])
+            probs = x[0][:, :, 4:5]
+            classes = x[0][:, :, 5:]
+            scores = probs * classes
+            if agnostic_nms:
+                nms = AgnosticNMS()((boxes, classes, scores), topk_all, iou_thres, conf_thres)
+            else:
+                boxes = tf.expand_dims(boxes, 2)
+                nms = tf.image.combined_non_max_suppression(boxes,
+                                                            scores,
+                                                            topk_per_class,
+                                                            topk_all,
+                                                            iou_thres,
+                                                            conf_thres,
+                                                            clip_boxes=False)
+            return nms, x[1]
+        return x[0]  # output only first tensor [1,6300,85] = [xywh, conf, class0, class1, ...]
+        # x = x[0][0]  # [x(1,6300,85), ...] to x(6300,85)
+        # xywh = x[..., :4]  # x(6300,4) boxes
+        # conf = x[..., 4:5]  # x(6300,1) confidences
+        # cls = tf.reshape(tf.cast(tf.argmax(x[..., 5:], axis=1), tf.float32), (-1, 1))  # x(6300,1)  classes
+        # return tf.concat([conf, cls, xywh], 1)
+
+    @staticmethod
+    def _xywh2xyxy(xywh):
+        # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right
+        x, y, w, h = tf.split(xywh, num_or_size_splits=4, axis=-1)
+        return tf.concat([x - w / 2, y - h / 2, x + w / 2, y + h / 2], axis=-1)
+
+
+class AgnosticNMS(keras.layers.Layer):
+    # TF Agnostic NMS
+    def call(self, input, topk_all, iou_thres, conf_thres):
+        # wrap map_fn to avoid TypeSpec related error https://stackoverflow.com/a/65809989/3036450
+        return tf.map_fn(lambda x: self._nms(x, topk_all, iou_thres, conf_thres),
+                         input,
+                         fn_output_signature=(tf.float32, tf.float32, tf.float32, tf.int32),
+                         name='agnostic_nms')
+
+    @staticmethod
+    def _nms(x, topk_all=100, iou_thres=0.45, conf_thres=0.25):  # agnostic NMS
+        boxes, classes, scores = x
+        class_inds = tf.cast(tf.argmax(classes, axis=-1), tf.float32)
+        scores_inp = tf.reduce_max(scores, -1)
+        selected_inds = tf.image.non_max_suppression(boxes,
+                                                     scores_inp,
+                                                     max_output_size=topk_all,
+                                                     iou_threshold=iou_thres,
+                                                     score_threshold=conf_thres)
+        selected_boxes = tf.gather(boxes, selected_inds)
+        padded_boxes = tf.pad(selected_boxes,
+                              paddings=[[0, topk_all - tf.shape(selected_boxes)[0]], [0, 0]],
+                              mode="CONSTANT",
+                              constant_values=0.0)
+        selected_scores = tf.gather(scores_inp, selected_inds)
+        padded_scores = tf.pad(selected_scores,
+                               paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]],
+                               mode="CONSTANT",
+                               constant_values=-1.0)
+        selected_classes = tf.gather(class_inds, selected_inds)
+        padded_classes = tf.pad(selected_classes,
+                                paddings=[[0, topk_all - tf.shape(selected_boxes)[0]]],
+                                mode="CONSTANT",
+                                constant_values=-1.0)
+        valid_detections = tf.shape(selected_inds)[0]
+        return padded_boxes, padded_scores, padded_classes, valid_detections
+
+
+def activations(act=nn.SiLU):
+    # Returns TF activation from input PyTorch activation
+    if isinstance(act, nn.LeakyReLU):
+        return lambda x: keras.activations.relu(x, alpha=0.1)
+    elif isinstance(act, nn.Hardswish):
+        return lambda x: x * tf.nn.relu6(x + 3) * 0.166666667
+    elif isinstance(act, (nn.SiLU, SiLU)):
+        return lambda x: keras.activations.swish(x)
+    else:
+        raise Exception(f'no matching TensorFlow activation found for PyTorch activation {act}')
+
+
+def representative_dataset_gen(dataset, ncalib=100):
+    # Representative dataset generator for use with converter.representative_dataset, returns a generator of np arrays
+    for n, (path, img, im0s, vid_cap, string) in enumerate(dataset):
+        im = np.transpose(img, [1, 2, 0])
+        im = np.expand_dims(im, axis=0).astype(np.float32)
+        im /= 255
+        yield [im]
+        if n >= ncalib:
+            break
+
+
+def run(
+        weights=ROOT / 'yolov5s.pt',  # weights path
+        imgsz=(640, 640),  # inference size h,w
+        batch_size=1,  # batch size
+        dynamic=False,  # dynamic batch size
+):
+    # PyTorch model
+    im = torch.zeros((batch_size, 3, *imgsz))  # BCHW image
+    model = attempt_load(weights, device=torch.device('cpu'), inplace=True, fuse=False)
+    _ = model(im)  # inference
+    model.info()
+
+    # TensorFlow model
+    im = tf.zeros((batch_size, *imgsz, 3))  # BHWC image
+    tf_model = TFModel(cfg=model.yaml, model=model, nc=model.nc, imgsz=imgsz)
+    _ = tf_model.predict(im)  # inference
+
+    # Keras model
+    im = keras.Input(shape=(*imgsz, 3), batch_size=None if dynamic else batch_size)
+    keras_model = keras.Model(inputs=im, outputs=tf_model.predict(im))
+    keras_model.summary()
+
+    LOGGER.info('PyTorch, TensorFlow and Keras models successfully verified.\nUse export.py for TF model export.')
+
+
+def parse_opt():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--weights', type=str, default=ROOT / 'yolov5s.pt', help='weights path')
+    parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], help='inference size h,w')
+    parser.add_argument('--batch-size', type=int, default=1, help='batch size')
+    parser.add_argument('--dynamic', action='store_true', help='dynamic batch size')
+    opt = parser.parse_args()
+    opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1  # expand
+    print_args(vars(opt))
+    return opt
+
+
+def main(opt):
+    run(**vars(opt))
+
+
+if __name__ == "__main__":
+    opt = parse_opt()
+    main(opt)

models/yolo.py

@@ -0,0 +1,337 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+"""
+YOLO-specific modules
+
+Usage:
+    $ python path/to/models/yolo.py --cfg yolov5s.yaml
+"""
+
+import argparse
+import contextlib
+import os
+import platform
+import sys
+from copy import deepcopy
+from pathlib import Path
+
+FILE = Path(__file__).resolve()
+ROOT = FILE.parents[1]  # YOLOv5 root directory
+if str(ROOT) not in sys.path:
+    sys.path.append(str(ROOT))  # add ROOT to PATH
+if platform.system() != 'Windows':
+    ROOT = Path(os.path.relpath(ROOT, Path.cwd()))  # relative
+
+from models.common import *
+from models.experimental import *
+from utils.autoanchor import check_anchor_order
+from utils.general import LOGGER, check_version, check_yaml, make_divisible, print_args
+from utils.plots import feature_visualization
+from utils.torch_utils import (fuse_conv_and_bn, initialize_weights, model_info, profile, scale_img, select_device,
+                               time_sync)
+
+try:
+    import thop  # for FLOPs computation
+except ImportError:
+    thop = None
+
+
+class Detect(nn.Module):
+    stride = None  # strides computed during build
+    onnx_dynamic = False  # ONNX export parameter
+    export = False  # export mode
+
+    def __init__(self, nc=80, anchors=(), ch=(), inplace=True):  # detection layer
+        super().__init__()
+        self.nc = nc  # number of classes
+        self.no = nc + 5  # number of outputs per anchor
+        self.nl = len(anchors)  # number of detection layers
+        self.na = len(anchors[0]) // 2  # number of anchors
+        self.grid = [torch.zeros(1)] * self.nl  # init grid
+        self.anchor_grid = [torch.zeros(1)] * self.nl  # init anchor grid
+        self.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2))  # shape(nl,na,2)
+        self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch)  # output conv
+        self.inplace = inplace  # use inplace ops (e.g. slice assignment)
+
+    def forward(self, x):
+        z = []  # inference output
+        for i in range(self.nl):
+            x[i] = self.m[i](x[i])  # conv
+            bs, _, ny, nx = x[i].shape  # x(bs,255,20,20) to x(bs,3,20,20,85)
+            x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
+
+            if not self.training:  # inference
+                if self.onnx_dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]:
+                    self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i)
+
+                y = x[i].sigmoid()
+                if self.inplace:
+                    y[..., 0:2] = (y[..., 0:2] * 2 + self.grid[i]) * self.stride[i]  # xy
+                    y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i]  # wh
+                else:  # for YOLOv5 on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953
+                    xy, wh, conf = y.split((2, 2, self.nc + 1), 4)  # y.tensor_split((2, 4, 5), 4)  # torch 1.8.0
+                    xy = (xy * 2 + self.grid[i]) * self.stride[i]  # xy
+                    wh = (wh * 2) ** 2 * self.anchor_grid[i]  # wh
+                    y = torch.cat((xy, wh, conf), 4)
+                z.append(y.view(bs, -1, self.no))
+
+        return x if self.training else (torch.cat(z, 1),) if self.export else (torch.cat(z, 1), x)
+
+    def _make_grid(self, nx=20, ny=20, i=0):
+        d = self.anchors[i].device
+        t = self.anchors[i].dtype
+        shape = 1, self.na, ny, nx, 2  # grid shape
+        y, x = torch.arange(ny, device=d, dtype=t), torch.arange(nx, device=d, dtype=t)
+        if check_version(torch.__version__, '1.10.0'):  # torch>=1.10.0 meshgrid workaround for torch>=0.7 compatibility
+            yv, xv = torch.meshgrid(y, x, indexing='ij')
+        else:
+            yv, xv = torch.meshgrid(y, x)
+        grid = torch.stack((xv, yv), 2).expand(shape) - 0.5  # add grid offset, i.e. y = 2.0 * x - 0.5
+        anchor_grid = (self.anchors[i] * self.stride[i]).view((1, self.na, 1, 1, 2)).expand(shape)
+        return grid, anchor_grid
+
+
+class Model(nn.Module):
+    # YOLOv5 model
+    def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None):  # model, input channels, number of classes
+        super().__init__()
+        if isinstance(cfg, dict):
+            self.yaml = cfg  # model dict
+        else:  # is *.yaml
+            import yaml  # for torch hub
+            self.yaml_file = Path(cfg).name
+            with open(cfg, encoding='ascii', errors='ignore') as f:
+                self.yaml = yaml.safe_load(f)  # model dict
+
+        # Define model
+        ch = self.yaml['ch'] = self.yaml.get('ch', ch)  # input channels
+        if nc and nc != self.yaml['nc']:
+            LOGGER.info(f"Overriding model.yaml nc={self.yaml['nc']} with nc={nc}")
+            self.yaml['nc'] = nc  # override yaml value
+        if anchors:
+            LOGGER.info(f'Overriding model.yaml anchors with anchors={anchors}')
+            self.yaml['anchors'] = round(anchors)  # override yaml value
+        self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch])  # model, savelist
+        self.names = [str(i) for i in range(self.yaml['nc'])]  # default names
+        self.inplace = self.yaml.get('inplace', True)
+
+        # Build strides, anchors
+        m = self.model[-1]  # Detect()
+        if isinstance(m, Detect):
+            s = 256  # 2x min stride
+            m.inplace = self.inplace
+            m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))])  # forward
+            check_anchor_order(m)  # must be in pixel-space (not grid-space)
+            m.anchors /= m.stride.view(-1, 1, 1)
+            self.stride = m.stride
+            self._initialize_biases()  # only run once
+
+        # Init weights, biases
+        initialize_weights(self)
+        self.info()
+        LOGGER.info('')
+
+    def forward(self, x, augment=False, profile=False, visualize=False):
+        if augment:
+            return self._forward_augment(x)  # augmented inference, None
+        return self._forward_once(x, profile, visualize)  # single-scale inference, train
+
+    def _forward_augment(self, x):
+        img_size = x.shape[-2:]  # height, width
+        s = [1, 0.83, 0.67]  # scales
+        f = [None, 3, None]  # flips (2-ud, 3-lr)
+        y = []  # outputs
+        for si, fi in zip(s, f):
+            xi = scale_img(x.flip(fi) if fi else x, si, gs=int(self.stride.max()))
+            yi = self._forward_once(xi)[0]  # forward
+            # cv2.imwrite(f'img_{si}.jpg', 255 * xi[0].cpu().numpy().transpose((1, 2, 0))[:, :, ::-1])  # save
+            yi = self._descale_pred(yi, fi, si, img_size)
+            y.append(yi)
+        y = self._clip_augmented(y)  # clip augmented tails
+        return torch.cat(y, 1), None  # augmented inference, train
+
+    def _forward_once(self, x, profile=False, visualize=False):
+        y, dt = [], []  # outputs
+        for m in self.model:
+            if m.f != -1:  # if not from previous layer
+                x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f]  # from earlier layers
+            if profile:
+                self._profile_one_layer(m, x, dt)
+            x = m(x)  # run
+            y.append(x if m.i in self.save else None)  # save output
+            if visualize:
+                feature_visualization(x, m.type, m.i, save_dir=visualize)
+        return x
+
+    def _descale_pred(self, p, flips, scale, img_size):
+        # de-scale predictions following augmented inference (inverse operation)
+        if self.inplace:
+            p[..., :4] /= scale  # de-scale
+            if flips == 2:
+                p[..., 1] = img_size[0] - p[..., 1]  # de-flip ud
+            elif flips == 3:
+                p[..., 0] = img_size[1] - p[..., 0]  # de-flip lr
+        else:
+            x, y, wh = p[..., 0:1] / scale, p[..., 1:2] / scale, p[..., 2:4] / scale  # de-scale
+            if flips == 2:
+                y = img_size[0] - y  # de-flip ud
+            elif flips == 3:
+                x = img_size[1] - x  # de-flip lr
+            p = torch.cat((x, y, wh, p[..., 4:]), -1)
+        return p
+
+    def _clip_augmented(self, y):
+        # Clip YOLOv5 augmented inference tails
+        nl = self.model[-1].nl  # number of detection layers (P3-P5)
+        g = sum(4 ** x for x in range(nl))  # grid points
+        e = 1  # exclude layer count
+        i = (y[0].shape[1] // g) * sum(4 ** x for x in range(e))  # indices
+        y[0] = y[0][:, :-i]  # large
+        i = (y[-1].shape[1] // g) * sum(4 ** (nl - 1 - x) for x in range(e))  # indices
+        y[-1] = y[-1][:, i:]  # small
+        return y
+
+    def _profile_one_layer(self, m, x, dt):
+        c = isinstance(m, Detect)  # is final layer, copy input as inplace fix
+        o = thop.profile(m, inputs=(x.copy() if c else x,), verbose=False)[0] / 1E9 * 2 if thop else 0  # FLOPs
+        t = time_sync()
+        for _ in range(10):
+            m(x.copy() if c else x)
+        dt.append((time_sync() - t) * 100)
+        if m == self.model[0]:
+            LOGGER.info(f"{'time (ms)':>10s} {'GFLOPs':>10s} {'params':>10s}  module")
+        LOGGER.info(f'{dt[-1]:10.2f} {o:10.2f} {m.np:10.0f}  {m.type}')
+        if c:
+            LOGGER.info(f"{sum(dt):10.2f} {'-':>10s} {'-':>10s}  Total")
+
+    def _initialize_biases(self, cf=None):  # initialize biases into Detect(), cf is class frequency
+        # https://arxiv.org/abs/1708.02002 section 3.3
+        # cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1.
+        m = self.model[-1]  # Detect() module
+        for mi, s in zip(m.m, m.stride):  # from
+            b = mi.bias.view(m.na, -1).detach()  # conv.bias(255) to (3,85)
+            b[:, 4] += math.log(8 / (640 / s) ** 2)  # obj (8 objects per 640 image)
+            b[:, 5:] += math.log(0.6 / (m.nc - 0.999999)) if cf is None else torch.log(cf / cf.sum())  # cls
+            mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)
+
+    def _print_biases(self):
+        m = self.model[-1]  # Detect() module
+        for mi in m.m:  # from
+            b = mi.bias.detach().view(m.na, -1).T  # conv.bias(255) to (3,85)
+            LOGGER.info(
+                ('%6g Conv2d.bias:' + '%10.3g' * 6) % (mi.weight.shape[1], *b[:5].mean(1).tolist(), b[5:].mean()))
+
+    # def _print_weights(self):
+    #     for m in self.model.modules():
+    #         if type(m) is Bottleneck:
+    #             LOGGER.info('%10.3g' % (m.w.detach().sigmoid() * 2))  # shortcut weights
+
+    def fuse(self):  # fuse model Conv2d() + BatchNorm2d() layers
+        LOGGER.info('Fusing layers... ')
+        for m in self.model.modules():
+            if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'):
+                m.conv = fuse_conv_and_bn(m.conv, m.bn)  # update conv
+                delattr(m, 'bn')  # remove batchnorm
+                m.forward = m.forward_fuse  # update forward
+        self.info()
+        return self
+
+    def info(self, verbose=False, img_size=640):  # print model information
+        model_info(self, verbose, img_size)
+
+    def _apply(self, fn):
+        # Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffers
+        self = super()._apply(fn)
+        m = self.model[-1]  # Detect()
+        if isinstance(m, Detect):
+            m.stride = fn(m.stride)
+            m.grid = list(map(fn, m.grid))
+            if isinstance(m.anchor_grid, list):
+                m.anchor_grid = list(map(fn, m.anchor_grid))
+        return self
+
+
+def parse_model(d, ch):  # model_dict, input_channels(3)
+    LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10}  {'module':<40}{'arguments':<30}")
+    anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple']
+    na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors  # number of anchors
+    no = na * (nc + 5)  # number of outputs = anchors * (classes + 5)
+
+    layers, save, c2 = [], [], ch[-1]  # layers, savelist, ch out
+    for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']):  # from, number, module, args
+        m = eval(m) if isinstance(m, str) else m  # eval strings
+        for j, a in enumerate(args):
+            with contextlib.suppress(NameError):
+                args[j] = eval(a) if isinstance(a, str) else a  # eval strings
+
+        n = n_ = max(round(n * gd), 1) if n > 1 else n  # depth gain
+        if m in (Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv,
+                 BottleneckCSP, C3, C3TR, C3SPP, C3Ghost, nn.ConvTranspose2d, DWConvTranspose2d, C3x):
+            c1, c2 = ch[f], args[0]
+            if c2 != no:  # if not output
+                c2 = make_divisible(c2 * gw, 8)
+
+            args = [c1, c2, *args[1:]]
+            if m in [BottleneckCSP, C3, C3TR, C3Ghost, C3x]:
+                args.insert(2, n)  # number of repeats
+                n = 1
+        elif m is nn.BatchNorm2d:
+            args = [ch[f]]
+        elif m is Concat:
+            c2 = sum(ch[x] for x in f)
+        elif m is Detect:
+            args.append([ch[x] for x in f])
+            if isinstance(args[1], int):  # number of anchors
+                args[1] = [list(range(args[1] * 2))] * len(f)
+        elif m is Contract:
+            c2 = ch[f] * args[0] ** 2
+        elif m is Expand:
+            c2 = ch[f] // args[0] ** 2
+        else:
+            c2 = ch[f]
+
+        m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args)  # module
+        t = str(m)[8:-2].replace('__main__.', '')  # module type
+        np = sum(x.numel() for x in m_.parameters())  # number params
+        m_.i, m_.f, m_.type, m_.np = i, f, t, np  # attach index, 'from' index, type, number params
+        LOGGER.info(f'{i:>3}{str(f):>18}{n_:>3}{np:10.0f}  {t:<40}{str(args):<30}')  # print
+        save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1)  # append to savelist
+        layers.append(m_)
+        if i == 0:
+            ch = []
+        ch.append(c2)
+    return nn.Sequential(*layers), sorted(save)
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml')
+    parser.add_argument('--batch-size', type=int, default=1, help='total batch size for all GPUs')
+    parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
+    parser.add_argument('--profile', action='store_true', help='profile model speed')
+    parser.add_argument('--line-profile', action='store_true', help='profile model speed layer by layer')
+    parser.add_argument('--test', action='store_true', help='test all yolo*.yaml')
+    opt = parser.parse_args()
+    opt.cfg = check_yaml(opt.cfg)  # check YAML
+    print_args(vars(opt))
+    device = select_device(opt.device)
+
+    # Create model
+    im = torch.rand(opt.batch_size, 3, 640, 640).to(device)
+    model = Model(opt.cfg).to(device)
+
+    # Options
+    if opt.line_profile:  # profile layer by layer
+        _ = model(im, profile=True)
+
+    elif opt.profile:  # profile forward-backward
+        results = profile(input=im, ops=[model], n=3)
+
+    elif opt.test:  # test all models
+        for cfg in Path(ROOT / 'models').rglob('yolo*.yaml'):
+            try:
+                _ = Model(cfg)
+            except Exception as e:
+                print(f'Error in {cfg}: {e}')
+
+    else:  # report fused model summary
+        model.fuse()

models/yolov5l.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 1.0  # model depth multiple
+width_multiple: 1.0  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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, anchors]],  # Detect(P3, P4, P5)
+  ]

models/yolov5m.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.67  # model depth multiple
+width_multiple: 0.75  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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, anchors]],  # Detect(P3, P4, P5)
+  ]

models/yolov5n.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.33  # model depth multiple
+width_multiple: 0.25  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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, anchors]],  # Detect(P3, P4, P5)
+  ]

models/yolov5s.yaml

@@ -0,0 +1,48 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+
+# Parameters
+nc: 80  # number of classes
+depth_multiple: 0.33  # model depth multiple
+width_multiple: 0.50  # layer channel multiple
+anchors:
+  - [10,13, 16,30, 33,23]  # P3/8
+  - [30,61, 62,45, 59,119]  # P4/16
+  - [116,90, 156,198, 373,326]  # P5/32
+
+# 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, anchors]],  # Detect(P3, P4, P5)
+  ]