Remove some dev envs for older python versions.

[twiecki]

…ove pandas from envs.

Thank your for opening a PR!

Before you proceed, please make sure that the pre-commit linting/style checks pass.

Depending on what your PR does, here are a few things you might want to address in the description:

• what are the (breaking) changes that this PR makes?
• important background, or details about the implementation
• are the changes—especially new features—covered by tests and docstrings?
• consider adding/updating relevant example notebooks
• right before it's ready to merge, mention the PR in the RELEASE-NOTES.md

[codecov]

Codecov Report

Merging #5842 (d11aad2) into main (3bf95ce) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             main    #5842   +/-   ##
=======================================
  Coverage   89.50%   89.50%           
=======================================
  Files          73       73           
  Lines       13267    13267           
=======================================
  Hits        11874    11874           
  Misses       1393     1393           

[twiecki]

Weird windows error:

C:\Miniconda3\envs\pymc-test-py38\lib\site-packages\aesara\graph\basic.py:1169: TypeError
---------------------------- Captured stdout call -----------------------------
\u2588\r\u2588\r\u2588\r\u2588\r
---------------------------- Captured stderr call -----------------------------
Auto-assigning NUTS sampler...
Initializing NUTS using jitter+adapt_diag...
Sequential sampling (2 chains in 1 job)
NUTS: [a]
Sampling 2 chains for 1_000 tune and 1_000 draw iterations (2_000 + 2_000 draws total) took 3 seconds.
Auto-assigning NUTS sampler...
Initializing NUTS using jitter+adapt_diag...
Sequential sampling (2 chains in 1 job)
NUTS: [a]
Sampling 2 chains for 1_000 tune and 1_000 draw iterations (2_000 + 2_000 draws total) took 3 seconds.
------------------------------ Captured log call ------------------------------
INFO     pymc:sampling.py:532 Auto-assigning NUTS sampler...
INFO     pymc:sampling.py:2480 Initializing NUTS using jitter+adapt_diag...
INFO     pymc:sampling.py:647 Sequential sampling (2 chains in 1 job)
INFO     pymc:sampling.py:244 NUTS: [a]
INFO     pymc:sampling.py:676 Sampling 2 chains for 1_000 tune and 1_000 draw iterations (2_000 + 2_000 draws total) took 3 seconds.
INFO     pymc:sampling.py:532 Auto-assigning NUTS sampler...
INFO     pymc:sampling.py:2480 Initializing NUTS using jitter+adapt_diag...
INFO     pymc:sampling.py:647 Sequential sampling (2 chains in 1 job)
INFO     pymc:sampling.py:244 NUTS: [a]
INFO     pymc:sampling.py:676 Sampling 2 chains for 1_000 tune and 1_000 draw iterations (2_000 + 2_000 draws total) took 3 seconds.
_________________________ test_no_init_nuts_compound __________________________

caplog = <_pytest.logging.LogCaptureFixture object at 0x0000014D8A9ED1F0>

    def test_no_init_nuts_compound(caplog):
        with pm.Model() as model:
            a = pm.Normal("a")
            b = pm.Poisson("b", 1)
>           pm.sample(10, tune=10)

pymc\tests\test_sampling.py:1488: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
pymc\sampling.py:524: in sample
    step = assign_step_methods(model, step, methods=pm.STEP_METHODS, step_kwargs=kwargs)
pymc\sampling.py:229: in assign_step_methods
    return instantiate_steppers(model, steps, selected_steps, step_kwargs)
pymc\sampling.py:147: in instantiate_steppers
    step = step_class(vars=vars, model=model, **args)
pymc\step_methods\metropolis.py:229: in __init__
    self.delta_logp = delta_logp(initial_values, model.logpt(), vars, shared)
pymc\step_methods\metropolis.py:1049: in delta_logp
    [logp0], inarray0 = pm.join_nonshared_inputs(point, [logp], vars, shared)
pymc\aesaraf.py:668: in join_nonshared_inputs
    xs_special = [aesara.clone_replace(x, replace, strict=False) for x in xs]
pymc\aesaraf.py:668: in <listcomp>
    xs_special = [aesara.clone_replace(x, replace, strict=False) for x in xs]
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

output = __logp, replace = {b: Subtensor{int64}.0, a: a_shared}
rebuild_kwds = {'strict': False}
rebuild_collect_shared = <function rebuild_collect_shared at 0x0000014DFAF24820>
items = [(b, Subtensor{int64}.0), (a, a_shared)]
tmp_replace = [(b, <TensorType(int64, ())>), (a, <TensorType(float32, ())>)]
new_replace = [(<TensorType(int64, ())>, Subtensor{int64}.0), (<TensorType(float32, ())>, a_shared)]

    def clone_replace(
        output: Collection[Variable],
        replace: Optional[
            Union[Iterable[Tuple[Variable, Variable]], Dict[Variable, Variable]]
        ] = None,
        **rebuild_kwds,
    ) -> List[Variable]:
        """Clone a graph and replace subgraphs within it.
    
        It returns a copy of the initial subgraph with the corresponding
        substitutions.
    
        Parameters
        ----------
        output
            Aesara expression that represents the computational graph.
        replace
            Dictionary describing which subgraphs should be replaced by what.
        rebuild_kwds
            Keywords to `rebuild_collect_shared`.
    
        """
        from aesara.compile.function.pfunc import rebuild_collect_shared
    
        items: Union[List[Tuple[Variable, Variable]], Tuple[Tuple[Variable, Variable], ...]]
        if isinstance(replace, dict):
            items = list(replace.items())
        elif isinstance(replace, (list, tuple)):
            items = replace
        elif replace is None:
            items = []
        else:
            raise ValueError(
                (
                    "replace is neither a dictionary, list, "
                    f"tuple or None ! The value provided is {replace},"
                    f"of type {type(replace)}"
                )
            )
        tmp_replace = [(x, x.type()) for x, y in items]
        new_replace = [(x, y) for ((_, x), (_, y)) in zip(tmp_replace, items)]
>       _, _outs, _ = rebuild_collect_shared(output, [], tmp_replace, [], **rebuild_kwds)
E       TypeError: rebuild_collect_shared() got an unexpected keyword argument 'strict'

[twiecki]

@brandonwillard Could this be an aesara bug that got introduced in 2.7.0?

[brandonwillard]

@brandonwillard Could this be an aesara bug that got introduced in 2.7.0?

No, that's a change in the clone_replace API.

[brandonwillard]

#5849 should fix that issue.

[twiecki]

That looks like a legit error:

__________________________ test_empirical_from_trace __________________________
another_simple_model = <pymc.model.Model object at 0x00000210ABB23CA0>
    def test_empirical_from_trace(another_simple_model):
        with another_simple_model:
>           step = pm.Metropolis()
pymc\tests\test_variational_inference.py:785: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
pymc\step_methods\metropolis.py:229: in __init__
    self.delta_logp = delta_logp(initial_values, model.logpt(), vars, shared)
pymc\step_methods\metropolis.py:1049: in delta_logp
    [logp0], inarray0 = pm.join_nonshared_inputs(point, [logp], vars, shared)
pymc\aesaraf.py:668: in join_nonshared_inputs
    xs_special = [aesara.clone_replace(x, replace, strict=False) for x in xs]
pymc\aesaraf.py:668: in <listcomp>
    xs_special = [aesara.clone_replace(x, replace, strict=False) for x in xs]
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
output = __logp, replace = {x: Reshape{1}.0}, rebuild_kwds = {'strict': False}
rebuild_collect_shared = <function rebuild_collect_shared at 0x000002109D7A4820>
items = [(x, Reshape{1}.0)], tmp_replace = [(x, <TensorType(float64, (None,))>)]
new_replace = [(<TensorType(float64, (None,))>, Reshape{1}.0)]
    def clone_replace(
        output: Collection[Variable],
        replace: Optional[
            Union[Iterable[Tuple[Variable, Variable]], Dict[Variable, Variable]]
        ] = None,
        **rebuild_kwds,
    ) -> List[Variable]:
        """Clone a graph and replace subgraphs within it.
        It returns a copy of the initial subgraph with the corresponding
        substitutions.
        Parameters
        ----------
        output
            Aesara expression that represents the computational graph.
        replace
            Dictionary describing which subgraphs should be replaced by what.
        rebuild_kwds
            Keywords to `rebuild_collect_shared`.
        """
        from aesara.compile.function.pfunc import rebuild_collect_shared
        items: Union[List[Tuple[Variable, Variable]], Tuple[Tuple[Variable, Variable], ...]]
        if isinstance(replace, dict):
            items = list(replace.items())
        elif isinstance(replace, (list, tuple)):
            items = replace
        elif replace is None:
            items = []
        else:
            raise ValueError(
                (
                    "replace is neither a dictionary, list, "
                    f"tuple or None ! The value provided is {replace},"
                    f"of type {type(replace)}"
                )
            )
        tmp_replace = [(x, x.type()) for x, y in items]
        new_replace = [(x, y) for ((_, x), (_, y)) in zip(tmp_replace, items)]
>       _, _outs, _ = rebuild_collect_shared(output, [], tmp_replace, [], **rebuild_kwds)
E       TypeError: rebuild_collect_shared() got an unexpected keyword argument 'strict'

[ricardoV94]

The kwarg changed, #5849

[ricardoV94]

Don't we have tests that cover pandas functionality? We would still want to test even if it's a user optional dependency

[ricardoV94]

Closing this in favor of #5911