torchopt
Adan-pytorch
| torchopt | Adan-pytorch | |
|---|---|---|
| 1 | 1 | |
| 501 | 240 | |
| 4.4% | - | |
| 7.1 | 3.2 | |
| 3 days ago | over 1 year ago | |
| Python | Python | |
| Apache License 2.0 | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
torchopt
-
What is the most efficient approach to ensemble a pytorch actor-critic model?
I would suggest checking https://pytorch.org/functorch/ and https://github.com/metaopt/torchopt for efficient inference and training with ensembles (e.g., t be independent actors in a multi-agent setting or multiple critics).
Adan-pytorch
-
I tried the new adan optimizer..... Promising results(?)
Sometimes, google tries to be useful by recommending stuff like this https://github.com/lucidrains/Adan-pytorch
What are some alternatives?
jaxopt - Hardware accelerated, batchable and differentiable optimizers in JAX.
Sophia - Effortless plugin and play Optimizer to cut model training costs by 50%. New optimizer that is 2x faster than Adam on LLMs.
symbolicai - Compositional Differentiable Programming Library
Adan - Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models
theseus - A library for differentiable nonlinear optimization
adan-experiments - Checking out the new adan optimizer
functorch - functorch is JAX-like composable function transforms for PyTorch.
AdamP - AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights (ICLR 2021)
pennylane - PennyLane is a cross-platform Python library for differentiable programming of quantum computers. Train a quantum computer the same way as a neural network.