✳flâneur — a map of the web's best reading

PyTorch's design origins

soumith.ch · 1,113 words · saved by 1 readers

0.1 Origins 0.2 Torch7 (~2010-2017) 0.3 Refactoring LuaTorch to be language agonstic (late 2015 to mid 2016) 0.4 Writing a new Python based Torch (mid 2016) 0.5 The connection to JAX: inspiration of HIPS/autograd 0.6 The inspiration from Chainer → PyTorch and the inspiration for PyTorch → Chainer v2 0.7 Post-launch evolution (2017 to present) 0.8 Attributing ideas is healthy, awesome and should be done more often 0.9 References: The groundwork for PyTorch originally started in early 2016, online, among a band of Torch7's contributors. These days, we also commonly refer to Torch7 as LuaTorch, as it was used via Lua. Torch7 was written by Ronan Collobert, Clement Farabet and Koray Kavuckuoglu in ~2010. I was deeply involved in Torch7 since 2012, with official “maintainer” status, joining these three original authors in April 2014. LuaTorch's C backend with all the CPU and CUDA code for Linear Algebra and Neural Networks was deeply intertwined with Lua. So, a bunc

**PyTorch's design origins** *its connection to Lua, its intertwined deep connection to JAX, its symbiotic connection to Chainer* **December 17th, 2023** [https://soumith.ch/blog.html](https://soumith.ch/blog.html) ## Origins The groundwork for PyTorch originally started in early 2016, online, among a band of Torch7's contributors. ## Torch7 (~2010-2017) These days, we also commonly refer to [Torch7](http://torch.ch/) as LuaTorch, as it was used via Lua. Torch7 was written by Ronan Collobert, [Clement Farabet](https://twitter.com/clmnt) and [Koray Kavuckuoglu](https://twitter.com/koraykv) in ~

Explore this link on the map →

related reading