Set theano config in model context

pymc3/model.py

@@ -8,6 +8,7 @@
 from theano import theano, tensor as tt
 from theano.tensor.var import TensorVariable
 
+from pymc3.theanof import set_theano_conf
 import pymc3 as pm
 from pymc3.math import flatten_list
 from .memoize import memoize
@@ -108,10 +109,16 @@ class Context(object):
 
     def __enter__(self):
         type(self).get_contexts().append(self)
+        # self._theano_config is set in Model.__new__
+        if hasattr(self, '_theano_config'):
+            self._old_theano_config = set_theano_conf(self._theano_config)
         return self
 
     def __exit__(self, typ, value, traceback):
         type(self).get_contexts().pop()
+        # self._theano_config is set in Model.__new__
+        if hasattr(self, '_old_theano_config'):
+            set_theano_conf(self._old_theano_config)
 
     @classmethod
     def get_contexts(cls):
@@ -301,6 +308,11 @@ class Model(six.with_metaclass(InitContextMeta, Context, Factor)):
         will be passed to the parent instance. So that 'nested' model
         contributes to the variables and likelihood factors of
         parent model.
+    theano_config : dict, default=None
+        A dictionary of theano config values that should be set
+        temporarily in the model context. See the documentation
+        of theano for a complete list. Set `compute_test_value` to
+        `raise` if it is None.
 
     Examples
     --------
@@ -367,9 +379,13 @@ def __new__(cls, *args, **kwargs):
             instance._parent = cls.get_contexts()[-1]
         else:
             instance._parent = None
+        theano_config = kwargs.get('theano_config', None)
+        if theano_config is None or 'compute_test_value' not in theano_config:
+            theano_config = {'compute_test_value': 'raise'}
+        instance._theano_config = theano_config
         return instance
 
-    def __init__(self, name='', model=None):
+    def __init__(self, name='', model=None, theano_config=None):
         self.name = name
         if self.parent is not None:
             self.named_vars = treedict(parent=self.parent.named_vars)
@@ -1032,7 +1048,3 @@ def all_continuous(vars):
         return False
     else:
         return True
-
-# theano stuff
-theano.config.warn.sum_div_dimshuffle_bug = False
-theano.config.compute_test_value = 'raise'

pymc3/tests/conftest.py

@@ -0,0 +1,9 @@
+import theano
+import pytest
+
+
+@pytest.fixture(scope="session", autouse=True)
+def theano_config():
+    config = theano.configparser.change_flags(compute_test_value='raise')
+    with config:
+        yield

pymc3/tests/test_model.py

@@ -22,6 +22,7 @@ def gen2():
         yield np.ones((20, 100)) * i
         i += 1
 
+
 class NewModel(pm.Model):
     def __init__(self, name='', model=None):
         super(NewModel, self).__init__(name, model)
@@ -133,13 +134,15 @@ def test_model_root(self):
             with pm.Model() as sub:
                 assert model is sub.root
 
+
 class TestObserved(object):
     def test_observed_rv_fail(self):
         with pytest.raises(TypeError):
-            with pm.Model() as model:
+            with pm.Model():
                 x = Normal('x')
                 Normal('n', observed=x)
 
+
 class TestScaling(object):
     def test_density_scaling(self):
         with pm.Model() as model1:
@@ -171,7 +174,8 @@ def true_dens():
 
         for i in range(10):
             _1, _2, _t = p1(), p2(), next(t)
-            np.testing.assert_almost_equal(_1, _t, decimal=select_by_precision(float64=7, float32=2))  # Value O(-50,000)
+            decimals = select_by_precision(float64=7, float32=2)
+            np.testing.assert_almost_equal(_1, _t, decimal=decimals)  # Value O(-50,000)
             np.testing.assert_almost_equal(_1, _2)
         # Done
 
@@ -192,3 +196,20 @@ def test_gradient_with_scaling(self):
             g1 = grad1(1)
             g2 = grad2(1)
             np.testing.assert_almost_equal(g1, g2)
+
+
+class TestTheanoConfig(object):
+    def test_set_testval_raise(self):
+        with theano.configparser.change_flags(compute_test_value='off'):
+            with pm.Model():
+                assert theano.config.compute_test_value == 'raise'
+            assert theano.config.compute_test_value == 'off'
+
+    def test_nested(self):
+        with theano.configparser.change_flags(compute_test_value='off'):
+            with pm.Model(theano_config={'compute_test_value': 'ignore'}):
+                assert theano.config.compute_test_value == 'ignore'
+                with pm.Model(theano_config={'compute_test_value': 'warn'}):
+                    assert theano.config.compute_test_value == 'warn'
+                assert theano.config.compute_test_value == 'ignore'
+            assert theano.config.compute_test_value == 'off'

pymc3/tests/test_special_functions.py

@@ -21,9 +21,9 @@ def test_functions():
     psi = function([x], ps.psi(x))
     function([x, p], ps.multigammaln(x, p))
     for x in xvals:
-        yield check_vals, gammaln, ss.gammaln, x
+        check_vals(gammaln, ss.gammaln, x)
     for x in xvals[1:]:
-        yield check_vals, psi, ss.psi, x
+        check_vals(psi, ss.psi, x)
 
 """
 scipy.special.multigammaln gives bad values if you pass a non scalar to a
@@ -52,7 +52,7 @@ def ssmultigammaln(a, b):
 
     for p in [0, 1, 2, 3, 4, 100]:
         for x in xvals:
-            yield check_vals, multigammaln, ssmultigammaln, x, p
+            check_vals(multigammaln, ssmultigammaln, x, p)
 
 
 def check_vals(fn1, fn2, *args):

pymc3/tests/test_step.py

@@ -136,7 +136,7 @@ def teardown_class(self):
 
     def test_sample_exact(self):
         for step_method in self.master_samples:
-            yield self.check_trace, step_method
+            self.check_trace(step_method)
 
     def check_trace(self, step_method):
         """Tests whether the trace for step methods is exactly the same as on master.
@@ -196,7 +196,7 @@ def test_step_continuous(self):
             )
         for step in steps:
             trace = sample(8000, step=step, start=start, model=model, random_seed=1)
-            yield self.check_stat, check, trace, step.__class__.__name__
+            self.check_stat(check, trace, step.__class__.__name__)
 
     def test_step_discrete(self):
         if theano.config.floatX == "float32":
@@ -211,7 +211,7 @@ def test_step_discrete(self):
             )
         for step in steps:
             trace = sample(20000, step=step, start=start, model=model, random_seed=1)
-            yield self.check_stat, check, trace, step.__class__.__name__
+            self.check_stat(check, trace, step.__class__.__name__)
 
     def test_step_categorical(self):
         start, model, (mu, C) = simple_categorical()
@@ -225,7 +225,7 @@ def test_step_categorical(self):
             )
         for step in steps:
             trace = sample(8000, step=step, start=start, model=model, random_seed=1)
-            yield self.check_stat, check, trace, step.__class__.__name__
+            self.check_stat(check, trace, step.__class__.__name__)
 
     def test_step_elliptical_slice(self):
         start, model, (K, L, mu, std, noise) = mv_prior_simple()
@@ -239,7 +239,7 @@ def test_step_elliptical_slice(self):
             )
         for step in steps:
             trace = sample(5000, step=step, start=start, model=model, random_seed=1)
-            yield self.check_stat, check, trace, step.__class__.__name__
+            self.check_stat(check, trace, step.__class__.__name__)
 
 
 class TestMetropolisProposal(object):

pymc3/tests/test_theanof.py

@@ -1,8 +1,9 @@
 import pickle
 import itertools
+import collections
 import numpy as np
 from theano import theano
-from pymc3.theanof import GeneratorOp, generator, tt_rng, floatX
+from pymc3.theanof import GeneratorOp, generator, tt_rng, floatX, set_theano_conf
 from pymc3.data import DataSampler, GeneratorAdapter
 import pytest
 
@@ -99,3 +100,25 @@ def test_gen_cloning_with_shape_change(self):
         shared = theano.shared(data)
         res2 = theano.clone(res, {gen: shared**2})
         assert res2.eval().shape == (1000,)
+
+
+class TestSetTheanoConfig(object):
+    def test_invalid_key(self):
+        with pytest.raises(ValueError) as e:
+            set_theano_conf({'bad_key': True})
+        e.match('Unknown')
+
+    def test_restore_when_bad_key(self):
+        with theano.configparser.change_flags(compute_test_value='off'):
+            with pytest.raises(ValueError):
+                conf = collections.OrderedDict(
+                    [('compute_test_value', 'raise'), ('bad_key', True)])
+                set_theano_conf(conf)
+            assert theano.config.compute_test_value == 'off'
+
+    def test_restore(self):
+        with theano.configparser.change_flags(compute_test_value='off'):
+            conf = set_theano_conf({'compute_test_value': 'raise'})
+            assert conf == {'compute_test_value': 'off'}
+            conf = set_theano_conf(conf)
+            assert conf == {'compute_test_value': 'raise'}

pymc3/theanof.py

@@ -427,3 +427,31 @@ def grad(self, args, g_outs):
 
 def floatX_array(x):
     return floatX(np.array(x))
+
+
+def set_theano_conf(values):
+    """Change the theano configuration and return old values.
+
+    This is similar to `theano.configparser.change_flags`, but it
+    returns the original values in a pickleable form.
+    """
+    variables = {}
+    unknown = set(values.keys())
+    for variable in theano.configparser._config_var_list:
+        if variable.fullname in values:
+            variables[variable.fullname] = variable
+            unknown.remove(variable.fullname)
+    if len(unknown) > 0:
+        raise ValueError("Unknown theano config settings: %s" % unknown)
+
+    old = {}
+    for name, variable in variables.items():
+        old_value = variable.__get__(True, None)
+        try:
+            variable.__set__(None, values[name])
+        except Exception:
+            for key, old_value in old.items():
+                variables[key].__set__(None, old_value)
+            raise
+        old[name] = old_value
+    return old