Get Started Manually
In this tutorial, you will manually set up the build environment, create your first PyTorch Live project, and run it in an emulator or on your phone.
If you ended up here, your operating system is either Windows or Linux, or the PyTorch Live command line tool torchlive-cli
failed for you. No worries, we have prepared manual steps for you to get started with PyTorch Live!
Installation
PyTorch Live is powered by PyTorch Mobile and React Native. Both are active and flourishing open source projects and require the following dependencies:
PyTorch Mobile
React Native
- Java Development Kit (any of the following)
- Yarn
- Watchman
- Android Command Line Tools
- Android SDK
- Android SDK Manager
- Android Virtual Device Manager
- Android Emulator
- CocoaPods
The PyTorch Live CLI currently does not support the Apple M1 chipset. However, there are instructions by the community for how to install React Native on an Apple M1 without Rosetta.
As a first step, let's follow their installation instructions to set up the development environment for Android and iOS.
Set up the PyTorch
PyTorch Live uses python3
to export PyTorch models for the PyTorch Mobile runtime. The export is part of each PyTorch Live app project initialization (see later steps). Run the following command to check if python3 is installed:
❯ python3 --version
Python 3.9.4
The version installed on your machine might be slightly different, but should at least be version 3.7
or above.
If python3
is not installed, you can download Python from the python.org website, or install it via Chocolatey (Windows), apt (Linux), or Homebrew (macOS).
- Chocolatey
- Apt
- Homebrew
choco install python3
apt install python3
brew install python3
Set up the React Native development environment
The React Native website has a very detailed step by step guide to install React Native on macOS, Windows, and Linux and for each of the two target OS Android and iOS.
Follow their installation instructions on: https://reactnative.dev/docs/environment-setup
Create a New Project
Now that we have set up the build environment, let's create a new PyTorch Live project and run it in an emulator or on your device.
Initialize Your Project
For PyTorch Live beta, we provide a React Native app template, packed with AI demos and practical examples. Use the npx react-native init
command to initialize your first PyTorch Live project.
npx react-native init MyFirstProject --template react-native-template-pytorch-live
Behind the scenes, the tool created a new React Native project and configured the build tools for mobile development. The project contains on-device AI models and interactive examples that use them.
This is what you should see after the project is initialized:
❯ npx react-native init MyFirstProject --template react-native-template-pytorch-live
npx: installed 637 in 16.171s
###### ######
### #### #### ###
## ### ### ##
## #### ##
## #### ##
## ## ## ##
## ### ### ##
## ######################## ##
###### ### ### ######
### ## ## ## ## ###
### ## ### #### ### ## ###
## #### ######## #### ##
## ### ########## ### ##
## #### ######## #### ##
### ## ### #### ### ## ###
### ## ## ## ## ###
###### ### ### ######
## ######################## ##
## ### ### ##
## ## ## ##
## #### ##
## #### ##
## ### ### ##
### #### #### ###
###### ######
Welcome to React Native!
Learn once, write anywhere
✔ Downloading template
✔ Copying template
✔ Processing template
✔ Executing post init script
✔ Installing CocoaPods dependencies (this may take a few minutes)
Run instructions for Android:
• Have an Android emulator running (quickest way to get started), or a device connected.
• cd "/Users/raedle/Projects/MyFirstProject" && npx react-native run-android
Run instructions for iOS:
• cd "/Users/raedle/Projects/MyFirstProject" && npx react-native run-ios
- or -
• Open MyFirstProject/ios/MyFirstProject.xcworkspace in Xcode or run "xed -b ios"
• Hit the Run button
Run instructions for macOS:
• See https://aka.ms/ReactNativeGuideMacOS for the latest up-to-date instructions.
Run Your Project
We are almost there. Now let's navigate to the newly created PyTorch Live project folder:
cd MyFirstProject
And run the PyTorch Live project on Android:
npx react-native run-android
By default, the PyTorch Live app will run in an emulator. You can also run it on your own phone which may improve performance. To run it on your phone, connect your phone to your computer and enable Android's developer mode.
On Android 4.1 and lower, the Developer options screen is available by default. On Android 4.2 and higher, you must enable this screen. To enable developer options, go to Settings > Developer options and tap the Build Number option 7 times.
More details are on the Android configure on-device developer options.
Welcome to PyTorch Live
Start by trying out the various Examples in the app. When you are ready to experiment, take a look at the Toolbox demos and follow along with a tutorial: recognizing objects in photographs or answering questions about some text.
PyTorch Live is in beta. Your feedback will help us improve tremendously. Please file issues and share ideas on GitHub.
Show us what you've made with PyTorch Live by tagging #playtorch in your social media posts.