How to concatenate distribution?

I want to create a 2-D array x, and x[0] ~ N(0, 1), x[1] ~ StudentT(3, 0, 1), can I use a single line to do it? For example,

x = numpyro.sample('x', concatenate([N(0, 1), StudentT(3, 0, 1)], axis=0))

The goal is to use a single site name to include the coefficients with different types of priors. An alternative is to use numpyro.deterministic('x', jnp.concatenate(...)), but that requires extra site names for elements in array.

I guess tfd.Blockwise is what you need (I haven’t tried it yet). Hopefully it will work with numpyro.

Thank you for the answer!