Link to this sectionTFLite、ONNX、CoreML、TensorRT 导出#

📚 本指南介绍了如何将训练好的 YOLOv5 🚀 模型从 PyTorch 导出为多种部署格式，包括 ONNX、TensorRT、CoreML 等。

Link to this section开始之前#

Clone repo and install requirements.txt in a Python>=3.8.0 environment, including PyTorch>=1.8. Models and datasets download automatically from the latest YOLOv5 release.

git clone https://github.com/ultralytics/yolov5 # clone
cd yolov5
pip install -r requirements.txt # install

有关 TensorRT 导出示例（需要 GPU），请参阅我们的 Colab 笔记本附录部分。

Link to this section支持的导出格式#

YOLOv5 推理官方支持 12 种格式：

性能提示

导出为 ONNX 或 OpenVINO 可获得高达 3 倍的 CPU 加速。请参阅 CPU 基准测试。
导出为 TensorRT 可获得高达 5 倍的 GPU 加速。请参阅 GPU 基准测试。

格式	`export.py --include`	模型
PyTorch	-	`yolov5s.pt`
TorchScript	`torchscript`	`yolov5s.torchscript`
ONNX	`onnx`	`yolov5s.onnx`
OpenVINO	`openvino`	`yolov5s_openvino_model/`
TensorRT	`engine`	`yolov5s.engine`
CoreML	`coreml`	`yolov5s.mlmodel`
TensorFlow SavedModel	`saved_model`	`yolov5s_saved_model/`
TensorFlow GraphDef	`pb`	`yolov5s.pb`
TensorFlow Lite	`tflite`	`yolov5s.tflite`
TensorFlow Edge TPU	`edgetpu`	`yolov5s_edgetpu.tflite`
TensorFlow.js	`tfjs`	`yolov5s_web_model/`
PaddlePaddle	`paddle`	`yolov5s_paddle_model/`

Link to this section基准测试#

以下基准测试在 Colab Pro 上运行，使用 YOLOv5 教程笔记本。若要重现：

python benchmarks.py --weights yolov5s.pt --imgsz 640 --device 0

Link to this sectionColab Pro V100 GPU#

benchmarks: weights=/content/yolov5/yolov5s.pt, imgsz=640, batch_size=1, data=/content/yolov5/data/coco128.yaml, device=0, half=False, test=False
Checking setup...
YOLOv5 🚀 v6.1-135-g7926afc torch 1.10.0+cu111 CUDA:0 (Tesla V100-SXM2-16GB, 16160MiB)
Setup complete ✅ (8 CPUs, 51.0 GB RAM, 46.7/166.8 GB disk)

Benchmarks complete (458.07s)
                   Format  mAP@0.5:0.95  Inference time (ms)
0                 PyTorch        0.4623                10.19
1             TorchScript        0.4623                 6.85
2                    ONNX        0.4623                14.63
3                OpenVINO           NaN                  NaN
4                TensorRT        0.4617                 1.89
5                  CoreML           NaN                  NaN
6   TensorFlow SavedModel        0.4623                21.28
7     TensorFlow GraphDef        0.4623                21.22
8         TensorFlow Lite           NaN                  NaN
9     TensorFlow Edge TPU           NaN                  NaN
10          TensorFlow.js           NaN                  NaN

Link to this sectionColab Pro CPU#

benchmarks: weights=/content/yolov5/yolov5s.pt, imgsz=640, batch_size=1, data=/content/yolov5/data/coco128.yaml, device=cpu, half=False, test=False
Checking setup...
YOLOv5 🚀 v6.1-135-g7926afc torch 1.10.0+cu111 CPU
Setup complete ✅ (8 CPUs, 51.0 GB RAM, 41.5/166.8 GB disk)

Benchmarks complete (241.20s)
                   Format  mAP@0.5:0.95  Inference time (ms)
0                 PyTorch        0.4623               127.61
1             TorchScript        0.4623               131.23
2                    ONNX        0.4623                69.34
3                OpenVINO        0.4623                66.52
4                TensorRT           NaN                  NaN
5                  CoreML           NaN                  NaN
6   TensorFlow SavedModel        0.4623               123.79
7     TensorFlow GraphDef        0.4623               121.57
8         TensorFlow Lite        0.4623               316.61
9     TensorFlow Edge TPU           NaN                  NaN
10          TensorFlow.js           NaN                  NaN

Link to this section导出训练好的 YOLOv5 模型#

此命令将预训练的 YOLOv5s 模型导出为 TorchScript 和 ONNX 格式。yolov5s.pt 是“小型”模型，也是可用的第二小模型。其他选项包括 yolov5n.pt、yolov5m.pt、yolov5l.pt 和 yolov5x.pt，以及它们的 P6 对应版本（如 yolov5s6.pt），或者你自己的自定义训练检查点（如 runs/exp/weights/best.pt）。有关所有可用模型的详细信息，请参阅我们的 README 表格。

python export.py --weights yolov5s.pt --include torchscript onnx

提示

添加 --half 以 FP16 半精度导出模型，从而减小文件大小

输出：

export: data=data/coco128.yaml, weights=['yolov5s.pt'], imgsz=[640, 640], batch_size=1, device=cpu, half=False, inplace=False, train=False, keras=False, optimize=False, int8=False, dynamic=False, simplify=False, opset=12, verbose=False, workspace=4, nms=False, agnostic_nms=False, topk_per_class=100, topk_all=100, iou_thres=0.45, conf_thres=0.25, include=['torchscript', 'onnx']
YOLOv5 🚀 v6.2-104-ge3e5122 Python-3.8.0 torch-1.12.1+cu113 CPU

Downloading https://github.com/ultralytics/yolov5/releases/download/v6.2/yolov5s.pt to yolov5s.pt...
100% 14.1M/14.1M [00:00<00:00, 274MB/s]

Fusing layers...
YOLOv5s summary: 213 layers, 7225885 parameters, 0 gradients

PyTorch: starting from yolov5s.pt with output shape (1, 25200, 85) (14.1 MB)

TorchScript: starting export with torch 1.12.1+cu113...
TorchScript: export success ✅ 1.7s, saved as yolov5s.torchscript (28.1 MB)

ONNX: starting export with onnx 1.12.0...
ONNX: export success ✅ 2.3s, saved as yolov5s.onnx (28.0 MB)

Export complete (5.5s)
Results saved to /content/yolov5
Detect:          python detect.py --weights yolov5s.onnx
Validate:        python val.py --weights yolov5s.onnx
PyTorch Hub:     model = torch.hub.load('ultralytics/yolov5', 'custom', 'yolov5s.onnx')
Visualize:       https://netron.app/

这 3 个导出模型将与原始 PyTorch 模型一起保存：

YOLO export locations

推荐使用 Netron Viewer 来可视化导出的模型：

YOLO model visualization

Link to this section导出模型的使用示例#

detect.py 在导出模型上运行推理：

python detect.py --weights yolov5s.pt             # PyTorch
python detect.py --weights yolov5s.torchscript    # TorchScript
python detect.py --weights yolov5s.onnx           # ONNX Runtime or OpenCV DNN with dnn=True
python detect.py --weights yolov5s_openvino_model # OpenVINO
python detect.py --weights yolov5s.engine         # TensorRT
python detect.py --weights yolov5s.mlmodel        # CoreML (macOS only)
python detect.py --weights yolov5s_saved_model    # TensorFlow SavedModel
python detect.py --weights yolov5s.pb             # TensorFlow GraphDef
python detect.py --weights yolov5s.tflite         # TensorFlow Lite
python detect.py --weights yolov5s_edgetpu.tflite # TensorFlow Edge TPU
python detect.py --weights yolov5s_paddle_model   # PaddlePaddle

val.py 在导出模型上运行验证：

python val.py --weights yolov5s.pt             # PyTorch
python val.py --weights yolov5s.torchscript    # TorchScript
python val.py --weights yolov5s.onnx           # ONNX Runtime or OpenCV DNN with dnn=True
python val.py --weights yolov5s_openvino_model # OpenVINO
python val.py --weights yolov5s.engine         # TensorRT
python val.py --weights yolov5s.mlmodel        # CoreML (macOS Only)
python val.py --weights yolov5s_saved_model    # TensorFlow SavedModel
python val.py --weights yolov5s.pb             # TensorFlow GraphDef
python val.py --weights yolov5s.tflite         # TensorFlow Lite
python val.py --weights yolov5s_edgetpu.tflite # TensorFlow Edge TPU
python val.py --weights yolov5s_paddle_model   # PaddlePaddle

将 PyTorch Hub 与导出的 YOLOv5 模型结合使用：

import torch

# Model
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s.pt")
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s.torchscript")  # TorchScript
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s.onnx")  # ONNX Runtime
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s_openvino_model")  # OpenVINO
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s.engine")  # TensorRT
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s.mlmodel")  # CoreML (macOS Only)
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s_saved_model")  # TensorFlow SavedModel
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s.pb")  # TensorFlow GraphDef
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s.tflite")  # TensorFlow Lite
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s_edgetpu.tflite")  # TensorFlow Edge TPU
model = torch.hub.load("ultralytics/yolov5", "custom", "yolov5s_paddle_model")  # PaddlePaddle

# Images
img = "https://ultralytics.com/images/zidane.jpg"  # or file, Path, PIL, OpenCV, numpy, list

# Inference
results = model(img)

# Results
results.print()  # or .show(), .save(), .crop(), .pandas(), etc.

Link to this sectionOpenCV DNN 推理#

使用 ONNX 模型进行 OpenCV 推理：

python export.py --weights yolov5s.pt --include onnx

python detect.py --weights yolov5s.onnx --dnn # detect
python val.py --weights yolov5s.onnx --dnn    # validate

Link to this sectionC++ 推理#

YOLOv5 OpenCV DNN C++ 在导出 ONNX 模型上的推理示例：

YOLOv5 OpenVINO C++ 推理示例：

Link to this sectionTensorFlow.js Web 浏览器推理#

https://aukerul-shuvo.github.io/YOLOv5_TensorFlow-JS/

Link to this section支持的环境#

Ultralytics 提供了一系列即用型环境，预装了 CUDA、CUDNN、Python 和 PyTorch 等关键依赖，助你快速启动项目。

免费 GPU 笔记本：
Google Cloud： GCP 快速入门指南
Amazon： AWS 快速入门指南
Azure： AzureML 快速入门指南
Docker： Docker 快速入门指南

Link to this section项目状态#

此徽章表示所有 YOLOv5 GitHub Actions 持续集成 (CI) 测试均已成功通过。这些 CI 测试严格检查 YOLOv5 在训练、验证、推理、导出和基准测试等各个关键方面的功能和性能。它们确保了在 macOS、Windows 和 Ubuntu 上的稳定可靠运行，测试每 24 小时进行一次，并会在每次新提交代码时自动触发。

贡献者

GLglenn-jocher¹⁶ RIRizwanMunawar³ LALaughing-q¹

创建于 2023年11月12日更新于前天