4 ms·
TL;DR - New JIT feature that lets you run your model without python. It now seems trivial to load a pytorch model in C++ - New distributed computation package
by l3robot 8y ago
TL;DR
- New JIT feature that lets you run your model without python. It now seems trivial to load a pytorch model in C++
- New distributed computation package. Major redesign.
- C++ frontend
- New torch hub feature to load models from github easily
- make3 8y agoI love PyTorch, but my experience with jits embedded in Python (eg. Numba) has been everything but simple, nevermind trivial. I'll really have to try it to believe it.
- mlthoughts2018 8y agoI’ve had the opposite experience with numba in production. It works almost flawlessly, very easy to reason about the generated code and inspect annotations, easy to debug.
- alexbw 8y agoCheck out github.com/google/jax, it’s NumPy on the GPU with automatic differentiation, JIT and autobatching.
- mlthoughts2018 8y agoThat’s very cool. Numba and Cython work extremely well with virtually no overhead or extra effort on my part, so jax doesn’t seem like it would buy me much for most of my work. But I can imagine a lot of projects where jax woukd be useful, and I plan to keep current on best practices for it.
- make3 8y agoDo you interact with numpy or other compiled numerical packages? this is where it usually breaks for me. The thing is that I use numpy, scipy, keras, tensorflow, etc. in literally every project, making numba not too useful