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

Per a question asked by a user in the forum, it would be nice to have a tutorial/example for this type of regression. I searched for some examples available and found some nice ones below.

• https://github.com/AsaCooperStickland/Spike_And_Slab with Gibbs sampling
• http://www.batisengul.co.uk/post/spike-and-slab-bayesian-linear-regr

Description

This feature request was a suggestion of @lukasheinrich's:

I wonder whether we should have something like

$ pyhf contrib download http://pyhf.scikit-hep.org/examples/shapesysexample

or similar, as a way to get starter workspaces.

This should be easy to implement and I think mostly will just require making

https://github.com/scikit-hep/pyhf/bl

Because / at end of question is missing

Description
Update README and add experiments.

Solution
Create experiments or examples package to work