macOS - 快速上手使用 YOLO

在這里插入圖片描述

文章目錄

- 一、關于 yolo
- 二、安裝
- 三、命令行使用
- - 官方示例
  - yolo cfg
  - yolo predict
- 四、Python 調用
- - results 數據

一、關于 yolo

YOLO(YOLO（You Only Look Once）是一種流行的物體檢測和圖像分割模型，由華盛頓大學的約瑟夫-雷德蒙（Joseph Redmon）和阿里-法哈迪（Ali Farhadi）開發。YOLO 于 2015 年推出，因其高速度和高精確度而廣受歡迎。

Docs: https://docs.ultralytics.com
Solutions: https://docs.ultralytics.com/solutions/
Community: https://community.ultralytics.com
GitHub: https://github.com/ultralytics/ultralytics

PS : Flash-Attention 官方明確要求 CUDA 環境23，而 macOS 不支持 NVIDIA CUDA，所以無法安裝。
本教程以 Mac Mini M4 為例。

二、安裝

安裝 torch

https://pytorch.org/get-started/locally/

安裝 ultralytics

pip install ultralytics

驗證 yolo 安裝 – 查看版本

yolo --version

打印信息：

$ yolo --version 
Creating new Ultralytics Settings v0.0.6 file ? 
View Ultralytics Settings with 'yolo settings' or at '/Users/es/Library/Application Support/Ultralytics/settings.json'
Update Settings with 'yolo settings key=value', i.e. 'yolo settings runs_dir=path/to/dir'. For help see https://docs.ultralytics.com/quickstart/#ultralytics-settings.
WARNING ?? argument '--version' does not require leading dashes '--', updating to 'version'.
8.3.158

三、命令行使用

常用命令

yolo help
yolo checks
yolo version
yolo settings
yolo copy-cfg
yolo cfg
yolo solutions help

語法格式：

yolo TASK MODE ARGS

TASK (optional) is one of [‘classify’, ‘segment’, ‘pose’, ‘detect’, ‘obb’]
MODE (required) is one of [‘benchmark’, ‘track’, ‘export’, ‘val’, ‘train’, ‘predict’]
ARGS (optional) are any number of custom ‘arg=value’ pairs like ‘imgsz=320’ that override defaults.

更多參數詳見：https://docs.ultralytics.com/usage/cfg

官方示例

Train a detection model for 10 epochs with an initial learning_rate of 0.01

yolo train data=coco8.yaml model=yolo11n.pt epochs=10 lr0=0.01

Predict a YouTube video using a pretrained segmentation model at image size 320:

yolo predict model=yolo11n-seg.pt source='https://youtu.be/LNwODJXcvt4' imgsz=320

Val a pretrained detection model at batch-size 1 and image size 640:

yolo val model=yolo11n.pt data=coco8.yaml batch=1 imgsz=640

Export a YOLO11n classification model to ONNX format at image size 224 by 128 (no TASK required)

yolo export model=yolo11n-cls.pt format=onnx imgsz=224,128

Ultralytics solutions usage

yolo solutions count or in ['crop', 'blur', 'workout', 'heatmap', 'isegment', 'visioneye', 'speed', 'queue', 'analytics', 'inference', 'trackzone'] source="path/to/video.mp4"

yolo cfg

$ yolo cfg
Printing '/opt/miniconda3/lib/python3.13/site-packages/ultralytics/cfg/default.yaml'task: detect 
mode: train
model: null
data: null
epochs: 100
time: null
patience: 100
batch: 16
imgsz: 640
save: true
save_period: -1
cache: false
device: null
workers: 8
project: null
name: null
exist_ok: false
pretrained: true
optimizer: auto
verbose: true
seed: 0
deterministic: true
single_cls: false
rect: false
cos_lr: false
close_mosaic: 10
resume: false
amp: true
fraction: 1.0
profile: false
freeze: null
multi_scale: false
overlap_mask: true
mask_ratio: 4
dropout: 0.0
val: true
split: val
save_json: false
conf: null
iou: 0.7
max_det: 300
half: false
dnn: false
plots: true
source: null
vid_stride: 1
stream_buffer: false
visualize: false
augment: false
agnostic_nms: false
classes: null
retina_masks: false
embed: null
show: false
save_frames: false
save_txt: false
save_conf: false
save_crop: false
show_labels: true
show_conf: true
show_boxes: true
line_width: null
format: torchscript
keras: false
optimize: false
int8: false
dynamic: false
simplify: true
opset: null
workspace: null
nms: false
lr0: 0.01
lrf: 0.01
momentum: 0.937
weight_decay: 0.0005
warmup_epochs: 3.0
warmup_momentum: 0.8
warmup_bias_lr: 0.1
box: 7.5
cls: 0.5
dfl: 1.5
pose: 12.0
kobj: 1.0
nbs: 64
hsv_h: 0.015
hsv_s: 0.7
hsv_v: 0.4
degrees: 0.0
translate: 0.1
scale: 0.5
shear: 0.0
perspective: 0.0
flipud: 0.0
fliplr: 0.5
bgr: 0.0
mosaic: 1.0
mixup: 0.0
cutmix: 0.0
copy_paste: 0.0
copy_paste_mode: flip
auto_augment: randaugment
erasing: 0.4
cfg: null
tracker: botsort.yaml

yolo predict

$ yolo predict model=yolo11n.pt imgsz=640 conf=0.25
Downloading https://github.com/ultralytics/assets/releases/download/v8.3.0/yolo11n.pt to 'yolo11n.pt'...
100%|████████████████████████████████████████████| 5.35M/5.35M [00:08<00:00, 655kB/s]
WARNING ?? 'source' argument is missing. Using default 'source=/opt/miniconda3/lib/python3.13/site-packages/ultralytics/assets'.
Ultralytics 8.3.158 🚀 Python-3.13.2 torch-2.7.1 CPU (Apple M4)
YOLO11n summary (fused): 100 layers, 2,616,248 parameters, 0 gradients, 6.5 GFLOPsimage 1/2 /opt/miniconda3/lib/python3.13/site-packages/ultralytics/assets/bus.jpg: 640x480 4 persons, 1 bus, 28.8ms
image 2/2 /opt/miniconda3/lib/python3.13/site-packages/ultralytics/assets/zidane.jpg: 384x640 2 persons, 1 tie, 20.5ms  
Speed: 1.9ms preprocess, 24.6ms inference, 2.2ms postprocess per image at shape (1, 3, 384, 640)  
Results saved to runs/detect/predict  
💡 Learn more at https://docs.ultralytics.com/modes/predict

執行日志說明

從 releases 下載模型：https://github.com/ultralytics/assets/releases，下載完后，保存在了當前執行腳本的目錄下。
沒有設置圖片，默認使用 ultralytics 庫下的圖片： /opt/miniconda3/lib/python3.13/site-packages/ultralytics/assets
識別結果在 runs/detect/predict 文件夾下
更多 predict 的使用可見：https://docs.ultralytics.com/modes/predict

四、Python 調用

from ultralytics import YOLO# Load a model
model = YOLO("yolo11n-seg.pt")  # load an official modelmodel_path = 'yolo11n-seg.pt'
# model_path = ''
model = YOLO(model_path)  # load a custom model# Predict with the model
results = model("https://ultralytics.com/images/bus.jpg")  # predict on an image# Access the results
for result in results:xy = result.masks.xy  # mask in polygon formatxyn = result.masks.xyn  # normalizedmasks = result.masks.data  # mask in matrix format (num_objects x H x W)print('\n-- results : ', results)

results 數據

此處 results 列表只有一個元素

[ultralytics.engine.results.Results object with attributes:boxes: ultralytics.engine.results.Boxes object
keypoints: None
masks: ultralytics.engine.results.Masks object
names: {0: 'person', 1: 'bicycle', 2: 'car', 3: 'motorcycle', 4: 'airplane', 5: 'bus', 6: 'train', 7: 'truck', 8: 'boat', 9: 'traffic light', 10: 'fire hydrant', 11: 'stop sign', 12: 'parking meter', 13: 'bench', 14: 'bird', 15: 'cat', 16: 'dog', 17: 'horse', 18: 'sheep', 19: 'cow', 20: 'elephant', 21: 'bear', 22: 'zebra', 23: 'giraffe', 24: 'backpack', 25: 'umbrella', 26: 'handbag', 27: 'tie', 28: 'suitcase', 29: 'frisbee', 30: 'skis', 31: 'snowboard', 32: 'sports ball', 33: 'kite', 34: 'baseball bat', 35: 'baseball glove', 36: 'skateboard', 37: 'surfboard', 38: 'tennis racket', 39: 'bottle', 40: 'wine glass', 41: 'cup', 42: 'fork', 43: 'knife', 44: 'spoon', 45: 'bowl', 46: 'banana', 47: 'apple', 48: 'sandwich', 49: 'orange', 50: 'broccoli', 51: 'carrot', 52: 'hot dog', 53: 'pizza', 54: 'donut', 55: 'cake', 56: 'chair', 57: 'couch', 58: 'potted plant', 59: 'bed', 60: 'dining table', 61: 'toilet', 62: 'tv', 63: 'laptop', 64: 'mouse', 65: 'remote', 66: 'keyboard', 67: 'cell phone', 68: 'microwave', 69: 'oven', 70: 'toaster', 71: 'sink', 72: 'refrigerator', 73: 'book', 74: 'clock', 75: 'vase', 76: 'scissors', 77: 'teddy bear', 78: 'hair drier', 79: 'toothbrush'}
obb: None
orig_img: array([[[119, 146, 172],[121, 148, 174],[122, 152, 177],...,[161, 171, 188],[160, 170, 187],[160, 170, 187]],[[120, 147, 173],[122, 149, 175],[123, 153, 178],...,[161, 171, 188],[160, 170, 187],[160, 170, 187]],[[123, 150, 176],[124, 151, 177],[125, 155, 180],...,[161, 171, 188],[160, 170, 187],[160, 170, 187]],...,[[183, 182, 186],[179, 178, 182],[180, 179, 183],...,[121, 111, 117],[113, 103, 109],[115, 105, 111]],[[165, 164, 168],[173, 172, 176],[187, 186, 190],...,[102,  92,  98],[101,  91,  97],[103,  93,  99]],[[123, 122, 126],[145, 144, 148],[176, 175, 179],...,[ 95,  85,  91],[ 96,  86,  92],[ 98,  88,  94]]], dtype=uint8) 
orig_shape: (1080, 810)
path: '/Users/es/Documents/code/code24/bus.jpg'
probs: None
save_dir: 'runs/segment/predict'
speed: {'preprocess': 2.8683749842457473, 'inference': 56.7722920095548, 'postprocess': 5.611875036265701}]