Our great sponsors
-
diffrax
Numerical differential equation solvers in JAX. Autodifferentiable and GPU-capable. https://docs.kidger.site/diffrax/
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
-
tinygrad
Discontinued You like pytorch? You like micrograd? You love tinygrad! ❤️ [Moved to: https://github.com/tinygrad/tinygrad] (by geohot)
On the other hand, there is just no MD implemented with PyTorch.
At least prior to this announcement: JAX was much faster than PyTorch for differentiable physics. (Better JIT compiler; reduced Python-level overhead.)
E.g for numerical ODE simulation, I've found that Diffrax (https://github.com/patrick-kidger/diffrax) is ~100 times faster than torchdiffeq on the forward pass. The backward pass is much closer, and for this Diffrax is about 1.5 times faster.
It remains to be seen how PyTorch 2.0 will compare, or course!
Right now my job is actually building out the scientific computing ecosystem in JAX, so feel free to ping me with any other questions.
Related posts
- Show HN: Optimistix: Nonlinear Optimisation in Jax+Equinox
- [D] Adjoint Sensitivity Method vs Reverse Mode Autodiff
- Solving system of coupled differential equations using Runge-Kutta in python
- [D] What useful personal projects have you made?
- Why is Python used by lots of scientists to simulate and calculate things, although it is pretty slow in comparison to other languages?