Increasing the Impact of AI Through Better Software
summary
PyTorch explains how researchers move from prototype to production by compiling models to an intermediate representation and running them in high‑performance runtimes, claiming teams at Facebook use the TorchScript workflow to share one codebase between research and product and serve billions of predictions via C++/mobile runtimes. It demonstrates easy tracing with torch.jit.trace and scripting for branching, describes DistributedDataParallel (multi‑process) plus ModelParallel for oversized models, and highlights torchtext/torchaudio and spectrogram modules implemented as nn.Modules.
AI-written summary · about
chapters (4)
- 0:00Introduction to PyTorch
- 7:04PyTorch at Facebook
- 14:26New PyTorch features
- 20:42Getting involved with PyTorch
related talks
19mIncreasing the Impact of AI Through Better Software (Pytorch Updates)
10mResearch to Production: PyTorch JIT/TorchScript Updates
19mGet to know PyTorch - a cutting edge open source AI framework for Machine Learning
47mGet to Know PyTorch - A Cutting Edge Open Source AI Framework for Deep Learning
43mExploring Deep Learning Framework PyTorch
24mAdvancements in High-Performance AI/ML through PyTorch's Python Compiler
09mTogether Goes Brrr: Threading Research & Production with Torch Compile
12mKeynote: How PyTorch Became the Foundation of the AI Revolution