@ -23,7 +23,7 @@ This comprehensive guide provides a detailed walkthrough for deploying Ultralyti
!!! Note
This guide has been tested with both [Seeed Studio reComputer J4012](https://www.seeedstudio.com/reComputer-J4012-p-5586.html) which is based on NVIDIA Jetson Orin NX 16GB running the latest stable JetPack release of [JP5.1.3](https://developer.nvidia.com/embedded/jetpack-sdk-513) and [Seeed Studio reComputer J1020 v2](https://www.seeedstudio.com/reComputer-J1020-v2-p-5498.html) which is based on NVIDIA Jetson Nano 4GB running JetPack release of [JP4.6.1](https://developer.nvidia.com/embedded/jetpack-sdk-461). It is expected to work across all the NVIDIA Jetson hardware lineup including latest and legacy.
This guide has been tested with both [Seeed Studio reComputer J4012](https://www.seeedstudio.com/reComputer-J4012-p-5586.html) which is based on NVIDIA Jetson Orin NX 16GB running the latest stable JetPack release of [JP6.0](https://developer.nvidia.com/embedded/jetpack-sdk-60), JetPack release of [JP5.1.3](https://developer.nvidia.com/embedded/jetpack-sdk-513) and [Seeed Studio reComputer J1020 v2](https://www.seeedstudio.com/reComputer-J1020-v2-p-5498.html) which is based on NVIDIA Jetson Nano 4GB running JetPack release of [JP4.6.1](https://developer.nvidia.com/embedded/jetpack-sdk-461). It is expected to work across all the NVIDIA Jetson hardware lineup including latest and legacy.
## What is NVIDIA Jetson?
@ -61,32 +61,111 @@ The first step after getting your hands on an NVIDIA Jetson device is to flash N
For methods 3 and 4 above, after flashing the system and booting the device, please enter "sudo apt update && sudo apt install nvidia-jetpack -y" on the device terminal to install all the remaining JetPack components needed.
## Run on JetPack 5.x
## JetPack Support Based on Jetson Device
If you own a Jetson Xavier NX, AGX Xavier, AGX Orin, Orin Nano or Orin NX which supports JetPack 5.x, you can continue to follow this guide. However, if you have a legacy device such as Jetson Nano, please skip to [Run on JetPack 4.x](#run-on-jetpack-4x).
The below table highlights NVIDIA JetPack versions supported by different NVIDIA Jetson devices.
There are two ways of setting up Ultralytics package on NVIDIA Jetson to build your next Computer Vision project. You can use either of them.
## Quick Start with Docker
- [Start with Docker](#start-with-docker)
- [Start without Docker](#start-without-docker)
The fastest way to get started with Ultralytics YOLOv8 on NVIDIA Jetson is to run with pre-built docker images for Jetson. Refer to the table above and choose the JetPack version according to the Jetson device you own.
#### Start with Docker
=== "JetPack 4"
The fastest way to get started with Ultralytics YOLOv8 on NVIDIA Jetson is to run with pre-built docker image for Jetson.
After this is done, skip to [Use TensorRT on NVIDIA Jetson section](#use-tensorrt-on-nvidia-jetson).
## Start with Native Installation
For a native installation without Docker, please refer to the steps below.
### Run on JetPack 6.x
#### Install Ultralytics Package
Here we will install Ultralytics package on the Jetson with optional dependencies so that we can export the PyTorch models to other different formats. We will mainly focus on [NVIDIA TensorRT exports](../integrations/tensorrt.md) because TensorRT will make sure we can get the maximum performance out of the Jetson devices.
1. Update packages list, install pip and upgrade to latest
```bash
sudo apt update
sudo apt install python3-pip -y
pip install -U pip
```
2. Install `ultralytics` pip package with optional dependencies
```bash
pip install ultralytics[export]
```
3. Reboot the device
```bash
sudo reboot
```
#### Install PyTorch and Torchvision
Execute the below command to pull the Docker container and run on Jetson. This is based on [l4t-pytorch](https://catalog.ngc.nvidia.com/orgs/nvidia/containers/l4t-pytorch) docker image which contains PyTorch and Torchvision in a Python3 environment.
The above ultralytics installation will install Torch and Torchvision. However, these 2 packages installed via pip are not compatible to run on Jetson platform which is based on ARM64 architecture. Therefore, we need to manually install pre-built PyTorch pip wheel and compile/ install Torchvision from source.
Install `torch 2.3.0` and `torchvision 0.18` according to JP6.0
After this is done, skip to [Use TensorRT on NVIDIA Jetson section](#use-tensorrt-on-nvidia-jetson).
Visit the [PyTorch for Jetson page](https://forums.developer.nvidia.com/t/pytorch-for-jetson/72048) to access all different versions of PyTorch for different JetPack versions. For a more detailed list on the PyTorch, Torchvision compatibility, visit the [PyTorch and Torchvision compatibility page](https://github.com/pytorch/vision).
#### Start without Docker
#### Install `onnxruntime-gpu`
##### Install Ultralytics Package
The [onnxruntime-gpu](https://pypi.org/project/onnxruntime-gpu/) package hosted in PyPI does not have `aarch64` binaries for the Jetson. So we need to manually install this package. This package is needed for some of the exports.
All different `onnxruntime-gpu` packages corresponding to different JetPack and Python versions are listed [here](https://elinux.org/Jetson_Zoo#ONNX_Runtime). However, here we will download and install `onnxruntime-gpu 1.18.0` with `Python3.10` support.
`onnxruntime-gpu` will automatically revert back the numpy version to latest. So we need to reinstall numpy to `1.23.5` to fix an issue by executing:
`pip install numpy==1.23.5`
### Run on JetPack 5.x
#### Install Ultralytics Package
Here we will install Ultralytics package on the Jetson with optional dependencies so that we can export the PyTorch models to other different formats. We will mainly focus on [NVIDIA TensorRT exports](../integrations/tensorrt.md) because TensorRT will make sure we can get the maximum performance out of the Jetson devices.
@ -110,7 +189,7 @@ Here we will install Ultralytics package on the Jetson with optional dependencie
sudo reboot
```
##### Install PyTorch and Torchvision
#### Install PyTorch and Torchvision
The above ultralytics installation will install Torch and Torchvision. However, these 2 packages installed via pip are not compatible to run on Jetson platform which is based on ARM64 architecture. Therefore, we need to manually install pre-built PyTorch pip wheel and compile/ install Torchvision from source.
@ -140,11 +219,11 @@ The above ultralytics installation will install Torch and Torchvision. However,
Visit the [PyTorch for Jetson page](https://forums.developer.nvidia.com/t/pytorch-for-jetson/72048) to access all different versions of PyTorch for different JetPack versions. For a more detailed list on the PyTorch, Torchvision compatibility, visit the [PyTorch and Torchvision compatibility page](https://github.com/pytorch/vision).
##### Install `onnxruntime-gpu`
#### Install `onnxruntime-gpu`
The [onnxruntime-gpu](https://pypi.org/project/onnxruntime-gpu/) package hosted in PyPI does not have `aarch64` binaries for the Jetson. So we need to manually install this package. This package is needed for some of the exports.
All different `onnxruntime-gpu` packages corresponding to different JetPack and Python versions are listed [here](https://elinux.org/Jetson_Zoo#ONNX_Runtime). However, here we will download and install `onnxruntime-gpu 1.17.0` with `Python3.8` support for the JetPack we are using for this guide.
All different `onnxruntime-gpu` packages corresponding to different JetPack and Python versions are listed [here](https://elinux.org/Jetson_Zoo#ONNX_Runtime). However, here we will download and install `onnxruntime-gpu 1.17.0` with `Python3.8` support.
Here we support to run Ultralytics on legacy hardware such as the Jetson Nano. Currently we use Docker to achieve this.
Execute the below command to pull the Docker container and run on Jetson. This is based on [l4t-cuda](https://catalog.ngc.nvidia.com/orgs/nvidia/containers/l4t-cuda) docker image which contains CUDA in a L4T environment.
Out of all the model export formats supported by Ultralytics, TensorRT delivers the best inference performance when working with NVIDIA Jetson devices and our recommendation is to use TensorRT with Jetson. We also have a detailed document on TensorRT [here](../integrations/tensorrt.md).
@ -372,7 +441,7 @@ Congratulations on successfully setting up YOLOv8 on your NVIDIA Jetson! For fur
### How do I deploy Ultralytics YOLOv8 on NVIDIA Jetson devices?
Deploying Ultralytics YOLOv8 on NVIDIA Jetson devices is a straightforward process. First, flash your Jetson device with the NVIDIA JetPack SDK. Then, either use a pre-built Docker image for quick setup or manually install the required packages. Detailed steps for each approach can be found in sections [Start with Docker](#start-with-docker) and [Start without Docker](#start-without-docker).
Deploying Ultralytics YOLOv8 on NVIDIA Jetson devices is a straightforward process. First, flash your Jetson device with the NVIDIA JetPack SDK. Then, either use a pre-built Docker image for quick setup or manually install the required packages. Detailed steps for each approach can be found in sections [Quick Start with Docker](#quick-start-with-docker) and [Start with Native Installation](#start-with-native-installation).
### What performance benchmarks can I expect from YOLOv8 models on NVIDIA Jetson devices?
@ -82,7 +82,7 @@ This table details the configurations and options available for exporting YOLO m
| `optimize` | `bool` | `False` | Applies optimization for mobile devices when exporting to TorchScript, potentially reducing model size and improving performance. |
| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. |
| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal accuracy loss, primarily for edge devices. |
| `dynamic` | `bool` | `False` | Allows dynamic input sizes for ONNX and TensorRT exports, enhancing flexibility in handling varying image dimensions. |
| `dynamic` | `bool` | `False` | Allows dynamic input sizes for ONNX, TensorRT and OpenVINO exports, enhancing flexibility in handling varying image dimensions. |
| `simplify` | `bool` | `False` | Simplifies the model graph for ONNX exports with `onnxslim`, potentially improving performance and compatibility. |
| `opset` | `int` | `None` | Specifies the ONNX opset version for compatibility with different ONNX parsers and runtimes. If not set, uses the latest supported version. |
| `workspace` | `float` | `4.0` | Sets the maximum workspace size in GiB for TensorRT optimizations, balancing memory usage and performance. |
@ -100,7 +100,7 @@ Available YOLOv8 export formats are in the table below. You can export to any fo
@ -223,7 +223,7 @@ Export settings for YOLO models encompass configurations and options related to
| `optimize` | `bool` | `False` | Applies optimization for mobile devices when exporting to TorchScript, potentially reducing model size and improving performance. |
| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. |
| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal accuracy loss, primarily for edge devices. |
| `dynamic` | `bool` | `False` | Allows dynamic input sizes for ONNX and TensorRT exports, enhancing flexibility in handling varying image dimensions. |
| `dynamic` | `bool` | `False` | Allows dynamic input sizes for ONNX, TensorRT and OpenVINO exports, enhancing flexibility in handling varying image dimensions. |
| `simplify` | `bool` | `False` | Simplifies the model graph for ONNX exports, potentially improving performance and compatibility. |
| `opset` | `int` | `None` | Specifies the ONNX opset version for compatibility with different ONNX parsers and runtimes. If not set, uses the latest supported version. |
| `workspace` | `float` | `4.0` | Sets the maximum workspace size in GB for TensorRT optimizations, balancing memory usage and performance. |
Ultralytics Assistant
4 months ago
committed by
GitHub