PyTorch Mobile

on YouTube ↗09m

" summary

PyTorch Mobile provides an end-to-end path from Python to iOS and Android by running TorchScript interpreter models without a Python runtime, supporting traced or scripted LibTorch models. Releases include pre-built binaries in JCenter and CocoaPods and Java bindings for Android; the demo apps are a Reddit LSTM classifier and a quantized MobileNetV2 ImageNet classifier. Build size is about 5 MB per architecture, Android demo is 100% Java and iOS demo ~85% Swift.

AI-written summary · about

" chapters (5)

1. 0:00Introduction
2. 1:35Using PyTorch
3. 2:45Java Code
4. 5:15Code
5. 6:30Upcoming

" related talks

• 09mPyTorch Mobile: Past, present and 2.x futureRAZIEL ALVAREZ GUEVARA · PyTorch Conference 2022
• 08mPyTorch MobileDavid Reiss · PyTorch Developer Day 2020
• 06mPyTorch Mobile and Android Neural Networks APIOli Gaymond · PyTorch Developer Day 2020
• 27mMachine Learning and MobileBrett Koonce · Scale By The Bay 2019
• 35mClient Side Deep Learning Optimization with PyTorchShane Caldwell, Tyler Kirby · Strange Loop 2021
• 10mExecuTorch on Microcontrollers: Deploying PyTorch To...RJ Ascani, Matthias Cremon · PyTorch Conference Europe 2026
• 43mTensorFlow for Mobile DevelopersEnrique López Mañas · Mobile Era 2017
• 17mDeploy a custom Machine Learning model to mobileArun Venkatesan, Yu-Cheng Ling, Adam Koch · Google I/O 2022