Ordered transform doesn't work without testval

[impact27]

If you have questions about a specific use case, or you are not sure whether this is a bug or not, please post it to our discourse channel: https://discourse.pymc.io

Description of your problem

Please provide a minimal, self-contained, and reproducible example.

import pymc3 as pm
import theano.tensor as tt
import pymc3.distributions.transforms as tr
from theano.graph.op import Op
import theano
import numpy as np

tr_order = tr.Ordered()
tr_bounds = tr.interval(0, 0.5)

with pm.Model() as model:
    DRi = pm.Uniform("DRi", lower=0, upper=0.5, shape=2,
                      transform=tr.Chain([tr_bounds, tr_order]),
                       # testval=[0.1, 0.4]
                     )
    pm.Normal("A",
              mu=tt.sum(DRi),
              sigma=1,
              observed=np.random.uniform(0, 0.5, size=10))
    trace = pm.sample(return_inferencedata=True, cores=1)

Please provide the full traceback.

Traceback (most recent call last):

  File "C:\Users\quentin.peter\Desktop\aSyn_Kd_sizing\untitled0.py", line 19, in <module>
    DRi = pm.Uniform("DRi", lower=0, upper=0.5, shape=2,

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\distributions\distribution.py", line 122, in __new__
    return model.Var(name, dist, data, total_size, dims=dims)

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\model.py", line 1142, in Var
    var = TransformedRV(

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\model.py", line 2013, in __init__
    transformed_name, transform.apply(distribution), total_size=total_size

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\distributions\transforms.py", line 126, in apply
    return TransformedDistribution.dist(dist, self)

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\distributions\distribution.py", line 130, in dist
    dist.__init__(*args, **kwargs)

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\distributions\transforms.py", line 150, in __init__
    testval = forward(dist.default())

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\distributions\transforms.py", line 536, in forward
    y = transf.forward(y)

  File "C:\ProgramData\Anaconda3\lib\site-packages\pymc3\distributions\transforms.py", line 385, in forward
    y = tt.inc_subtensor(y[..., 0], x[..., 0])

  File "C:\ProgramData\Anaconda3\lib\site-packages\theano\tensor\var.py", line 493, in __getitem__
    raise IndexError("too many indices for array")

IndexError: too many indices for array

Please provide any additional information below.
This works if I give testval=[0.1, 0.4]

Versions and main components

PyMC3 Version: 3.11.2
Aesara Version: NA
Python Version: 3.8.5
Operating system: windows
How did you install PyMC3: (conda/pip) pip

[OriolAbril]

We need the versions of the packages you are using to be able to help

[ricardoV94]

This is a known limitation of the ordered transform: #3680

You have to manually specify a testval. In the future we will probably implement a pm.Sort that takes care of this sort of low level details for the user, see #4530