jax-models
GradCache
| jax-models | GradCache | |
|---|---|---|
| 6 | 1 | |
| 138 | 310 | |
| - | - | |
| 0.0 | 4.5 | |
| almost 2 years ago | about 2 months ago | |
| Python | Python | |
| Apache License 2.0 | Apache License 2.0 |
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.
jax-models
-
[D] How to contribute to open source ML and DL without having access to high quality setup?
I was in the same position as you are and the best thing you can do is to start reproducing papers (that's what I did with jax-models). This will
-
[D] Should We Be Using JAX in 2022?
I've been using JAX, especially Flax for quite some time now for my reproducibility initiative (jax_models) and this is what I really appreciate about the framework
- Weekly updated open sourced model implementations in Flax
- Weekly updated open sourced deep learning model implementations in Flax
- [P] Weekly updated open sourced model implementations in Flax
GradCache
-
[D] How to handle absurd batch sizes in SimCLR / OpenAI's CLIP?
Maybe you should try this thing: https://github.com/luyug/GradCache
What are some alternatives?
datasets - TFDS is a collection of datasets ready to use with TensorFlow, Jax, ...
h-former - H-Former is a VAE for generating in-between fonts (or combining fonts). Its encoder uses a Point net and transformer to compute a code vector of glyph. Its decoder is composed of multiple independent decoders which act on a code vector to reconstruct a point cloud representing a glpyh.
flax - Flax is a neural network library for JAX that is designed for flexibility.
transformers - 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
flaxmodels - Pretrained deep learning models for Jax/Flax: StyleGAN2, GPT2, VGG, ResNet, etc.
EasyLM - Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Flax.
equinox - Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/
long-range-arena - Long Range Arena for Benchmarking Efficient Transformers
elegy - A High Level API for Deep Learning in JAX
fortuna - A Library for Uncertainty Quantification.
jax - Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more