## PyTorch Container for Jetson and JetPack

The `l4t-pytorch` docker image contains PyTorch and torchvision pre-installed in a Python 3 environment to get up & running quickly with PyTorch on Jetson.  These containers support the following releases of [JetPack](https://developer.nvidia.com/embedded/jetpack) for Jetson Nano, TX1/TX2, Xavier NX, AGX Xavier, AGX Orin, and Orin NX:

* JetPack 5.1 (L4T R35.2.1)
* JetPack 5.0.2 (L4T R35.1.0)
* JetPack 5.0.1 Developer Preview (L4T R34.1.1)
* JetPack 5.0.0 Developer Preview (L4T R34.1.0)
* JetPack 4.6.1 (L4T R32.7.1)
* JetPack 4.6 (L4T R32.6.1)
* JetPack 4.5 (L4T R32.5.0)
* JetPack 4.4.1 (L4T R32.4.4)
* JetPack 4.4 (L4T R32.4.3)
* JetPack 4.4 Developer Preview (L4T R32.4.2)

For additional machine learning containers for Jetson, see the [`l4t-ml`](https://ngc.nvidia.com/catalog/containers/nvidia:l4t-ml) and [`l4t-tensorflow`](https://ngc.nvidia.com/catalog/containers/nvidia:l4t-tensorflow) images. Note that the PyTorch pip wheel installers for aarch64 used by these containers are available to download independently from the [Jetson Zoo](https://elinux.org/Jetson_Zoo#PyTorch_.28Caffe2.29).

### Package Versions

Depending on your version of JetPack-L4T, different tags of the `l4t-pytorch` container are available, each with support for Python 3.  Be sure to clone a tag that matches the version of JetPack-L4T that you have installed on your Jetson.

* JetPack 5.1 (L4T R35.2.1)

  * `l4t-pytorch:r35.2.1-pth2.0-py3`
    * PyTorch v2.0.0
    * torchvision v0.14.1
    * torchaudio v0.13.1
* JetPack 5.0.2 (L4T R35.1.0)

  * `l4t-pytorch:r35.1.0-pth1.11-py3`
    * PyTorch v1.11.0
    * torchvision v0.12.0
    * torchaudio v0.11.0
  * `l4t-pytorch:r35.1.0-pth1.12-py3`
    * PyTorch v1.12.0
    * torchvision v0.13.0
    * torchaudio v0.12.0
  * `l4t-pytorch:r35.1.0-pth1.13-py3`
    * PyTorch v1.13.0
    * torchvision v0.13.0
    * torchaudio v0.12.0
* JetPack 5.0.1 Developer Preview (L4T R34.1.1)

  * `l4t-pytorch:r34.1.1-pth1.11-py3`
    * PyTorch v1.11.0
    * torchvision v0.12.0
    * torchaudio v0.11.0
  * `l4t-pytorch:r34.1.1-pth1.12-py3`
    * PyTorch v1.12.0
    * torchvision v0.12.0
    * torchaudio v0.11.0
* JetPack 5.0.0 Developer Preview (L4T R34.1.0)

  * `l4t-pytorch:r34.1.0-pth1.12-py3`
    * PyTorch v1.12.0
    * torchvision v0.12.0
    * torchaudio v0.11.0
* JetPack 4.6.1 (L4T R32.7.1)

  * `l4t-pytorch:r32.7.1-pth1.9-py3`
    * PyTorch v1.9.0
    * torchvision v0.10.0
    * torchaudio v0.9.0
  * `l4t-pytorch:r32.7.1-pth1.10-py3`
    * PyTorch v1.10.0
    * torchvision v0.11.0
    * torchaudio v0.10.0
* JetPack 4.6 (L4T R32.6.1)

  * `l4t-pytorch:r32.6.1-pth1.8-py3`
    * PyTorch v1.8.0
    * torchvision v0.9.0
    * torchaudio v0.8.0
  * `l4t-pytorch:r32.6.1-pth1.9-py3`
    * PyTorch v1.9.0
    * torchvision v0.10.0
    * torchaudio v0.9.0
* JetPack 4.5 (L4T R32.5.0)

  * `l4t-pytorch:r32.5.0-pth1.6-py3`
    * PyTorch v1.6.0
    * torchvision v0.7.0
    * torchaudio v0.6.0
  * `l4t-pytorch:r32.5.0-pth1.7-py3`
    * PyTorch v1.7.0
    * torchvision v0.8.0
    * torchaudio v0.7.0
* JetPack 4.4.1 (L4T R32.4.4)

  * `l4t-pytorch:r32.4.4-pth1.6-py3`
    * PyTorch v1.6.0
    * torchvision v0.7.0
    * torchaudio v0.6.0
* JetPack 4.4 (L4T R32.4.3)

  * `l4t-pytorch:r32.4.3-pth1.6-py3`
    * PyTorch v1.6.0
    * torchvision v0.7.0
    * torchaudio v0.6.0
* JetPack 4.4 Developer Preview (L4T R32.4.2)

  * `l4t-pytorch:r32.4.2-pth1.5-py3`
    * PyTorch v1.5.0
    * torchvision v0.6.0
  * `l4t-pytorch:r32.4.2-pth1.4-py3`
    * PyTorch v1.4.0
    * torchvision v0.5.0
  * `l4t-pytorch:r32.4.2-pth1.3-py3`
    * PyTorch v1.3.0
    * torchvision v0.4.2
  * `l4t-pytorch:r32.4.2-pth1.2-py3`
    * PyTorch v1.2.0
    * torchvision v0.4.0

**note:** the `l4t-pytorch` containers require [JetPack 4.4](https://developer.nvidia.com/embedded/jetpack) or newer

### Running the Container

First pull one of the `l4t-pytorch` container tags from above, corresponding to the version of JetPack-L4T that you have installed on your Jetson.  For example, if you are running the latest JetPack 5.1 (L4T R35.2.1) release:

```
sudo docker pull nvcr.io/nvidia/l4t-pytorch:r35.2.1-pth2.0-py3
```

Then to start an interactive session in the container, run the following command:

```
sudo docker run -it --rm --runtime nvidia --network host nvcr.io/nvidia/l4t-pytorch:r35.2.1-pth2.0-py3
```

You should then be able to start a Python3 interpreter and `import torch` and `import torchvision`.

### Mounting Directories from the Host Device

To mount scripts, data, ect. from your Jetson's filesystem to run inside the container, use Docker's [`-v` flag](https://docs.docker.com/storage/bind-mounts/) when starting your Docker instance:

```
sudo docker run -it --rm --runtime nvidia --network host -v /home/user/project:/location/in/container nvcr.io/nvidia/l4t-pytorch:r35.2.1-pth2.0-py3
```

### Dockerfiles

To access or modify the Dockerfiles and scripts used to build this container, see this [GitHub repo](https://github.com/dusty-nv/jetson-containers).

### License

The `l4t-pytorch` container includes various software packages with their respective licenses included within the container.

### Getting Help & Support

If you have any questions or need help, please visit the [Jetson Developer Forums](https://forums.developer.nvidia.com/c/agx-autonomous-machines/jetson-embedded-systems/70) and the [PyTorch for Jetson](https://forums.developer.nvidia.com/t/pytorch-for-jetson-nano-version-1-5-0-now-available/72048) topic.

l4t-pytorch

PyTorch is a GPU accelerated tensor computational framework with a Python front end. This container contains PyTorch and torchvision pre-installed in a Python 3 environment to get up & running quickly with PyTorch on Jetson.

NVIDIA L4T PyTorch

Description

Publisher

Latest Tag

Modified

Compressed Size

Multinode Support

Multi-Arch Support

r35.2.1-pth2.0-py3 (Latest) Scan Results