long-range-arena
elegy
Our great sponsors
long-range-arena | elegy | |
---|---|---|
6 | 5 | |
682 | 463 | |
2.9% | 1.5% | |
0.0 | 0.0 | |
4 months 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.
long-range-arena
-
The Secret Sauce behind 100K context window in LLMs: all tricks in one place
https://github.com/google-research/long-range-arena
-
[R] The Annotated S4: Efficiently Modeling Long Sequences with Structured State Spaces
The Structured State Space for Sequence Modeling (S4) architecture is a new approach to very long-range sequence modeling tasks for vision, language, and audio, showing a capacity to capture dependencies over tens of thousands of steps. Especially impressive are the model’s results on the challenging Long Range Arena benchmark, showing an ability to reason over sequences of up to 16,000+ elements with high accuracy.
-
[D] Is there a repo on which many light-weight self-attention mechanism are introduced?
1.1 Long Range Arena: A Benchmark for Efficient Transformers. From authors of above, they proposed a benchmark for modeling long range interactions. It also inlcudes a repository
- [R] Google’s H-Transformer-1D: Fast One-Dimensional Hierarchical Attention With Linear Complexity for Long Sequence Processing
- [2107.11906] H-Transformer-1D: Fast One-Dimensional Hierarchical Attention for Sequences
-
[R][D] Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. Zhou et al. AAAI21 Best Paper. ProbSparse self-attention reduces complexity to O(nlogn), generative style decoder to obtainsequence output in one step, and self-attention distilling for further reducing memory
I think the paper is written in a clear style and I like that the authors included many experiments, including hyperparameter effects, ablations and extensive baseline comparisons. One thing I would have liked is them comparing their Informer to more efficient transformers (they compared only against logtrans and reformer) using the LRA (https://github.com/google-research/long-range-arena) benchmark.
elegy
-
is Elegy framework for JAX abandoned?
I wonder if https://github.com/poets-ai/elegy is still an active project or dead because it hasn't had a commit in almost a year. Would be too bad if abandoned because I like it.
-
[D] Any less-boilerplate framework for Jax/Flax/Haiku?
Elegy might be worth a look.
-
PyTorch vs. TensorFlow in Academic Papers
JAX is really cool, but still somewhat immature. I would love to see it taking more ground and improving wrt e.g. integration with tensorboard and getting all the goodies we have in tensorflow. If you are looking for a higher level framework, I would recommend elegy [0] which is very close to the keras API.
[0] https://github.com/poets-ai/elegy
-
[D] Should We Be Using JAX in 2022?
What's your favorite Deep Learning API for JAX - Flax, Haiku, Elegy, something else?
-
Best sources to learn JAX?
For a Module library checkout Flax or Haiku, they are well maintained. For a Trainer interface like Keras / Pytorch Lightning checkout Elegy: https://github.com/poets-ai/elegy
What are some alternatives?
performer-pytorch - An implementation of Performer, a linear attention-based transformer, in Pytorch
dm-haiku - JAX-based neural network library
attention-is-all-you-need-pytorch - A PyTorch implementation of the Transformer model in "Attention is All You Need".
jax-resnet - Implementations and checkpoints for ResNet, Wide ResNet, ResNeXt, ResNet-D, and ResNeSt in JAX (Flax).
HJxB - Continuous-Time/State/Action Fitted Value Iteration via Hamilton-Jacobi-Bellman (HJB)
equinox - Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/
flax - Flax is a neural network library for JAX that is designed for flexibility.
tldr-transformers - The "tl;dr" on a few notable transformer papers (pre-2022).
scenic - Scenic: A Jax Library for Computer Vision Research and Beyond
LFattNet - Attention-based View Selection Networks for Light-field Disparity Estimation
runtime - A performant and modular runtime for TensorFlow