🚀 Feature request

Fast Tokenizer for DeBERTA-V3 and mDeBERTa-V3

Motivation

DeBERTa V3 is an improved version of DeBERTa. With the V3 version, the authors also released a multilingual model "mDeBERTa-base" that outperforms XLM-R-base. However, DeBERTa V3 currently lacks a FastTokenizer implementation which makes it impossible to use with some of the example scripts (They require a Fa

It would be great if you could add the JAX equivalent of scipy.stats.mode, which is currently unavailable.

A use case may be a ML classification task with ensembles, where multiple models give different predictions, and we are interested in finding the most common one.

As an example, consider predictions to be a two-dimensional DeviceArray of predictions, with
shape = (number of model

Is there a way to train a bidirectional RNN (like LSTM or GRU) on trax nowadays?

Love the book! Ran into this display issue when viewing it on Chrome's PDF viewer:

Dear Numpyro developers,

Please develop Euler Maruyama features in numpyro similar to features found in PyMC.

Thanks alot.

Dear Brax team,

Since Brax is fully differentiable, I thought it'd be possible to use it like DiffTaichi or GradSim for system identification (e.g. determining the mass of an object from a trajectory and known force) but I couldn't find any example for this.
Do you happen to have any demo or tips for this?

From the top of my head I would do something like this:
Let's say the task is to es

when running the deep learning examples (say) deep_learning/flax_image_classif.py , the GPU utilization is never above 5%, while for the equivalent flax example the GPU utilization is around 90%, and the example runs more than 20x faster.

My guess is that there's a crucial @jax.jit directive missing somewhere.