Merge pull request #1495 from pymc-devs/constant

Renamed ConstantDist to Constant

Author: fonnesbeck

pymc3/distributions/__init__.py

@@ -31,6 +31,7 @@
 from .discrete import Poisson
 from .discrete import NegativeBinomial
 from .discrete import ConstantDist
+from .discrete import Constant
 from .discrete import ZeroInflatedPoisson
 from .discrete import ZeroInflatedNegativeBinomial
 from .discrete import DiscreteUniform
@@ -94,6 +95,7 @@
            'Poisson',
            'NegativeBinomial',
            'ConstantDist',
+           'Constant',
            'ZeroInflatedPoisson',
            'ZeroInflatedNegativeBinomial',
            'DiscreteUniform',

pymc3/distributions/discrete.py

@@ -9,7 +9,7 @@
 from .distribution import Discrete, draw_values, generate_samples
 
 __all__ = ['Binomial',  'BetaBinomial',  'Bernoulli',  'Poisson',
-           'NegativeBinomial', 'ConstantDist', 'ZeroInflatedPoisson',
+           'NegativeBinomial', 'ConstantDist', 'Constant', 'ZeroInflatedPoisson',
            'ZeroInflatedNegativeBinomial', 'DiscreteUniform', 'Geometric',
            'Categorical']
 
@@ -413,7 +413,7 @@ def logp(self, value):
                      sumto1)
 
 
-class ConstantDist(Discrete):
+class Constant(Discrete):
     """
     Constant log-likelihood.
 
@@ -424,7 +424,7 @@ class ConstantDist(Discrete):
     """
 
     def __init__(self, c, *args, **kwargs):
-        super(ConstantDist, self).__init__(*args, **kwargs)
+        super(Constant, self).__init__(*args, **kwargs)
         self.mean = self.median = self.mode = self.c = c
 
     def random(self, point=None, size=None, repeat=None):
@@ -441,6 +441,11 @@ def logp(self, value):
         c = self.c
         return bound(0, tt.eq(value, c))
 
+def ConstantDist(*args, **kwargs):
+    warnings.warn("ConstantDist has been deprecated. In future, use Constant instead.",
+                DeprecationWarning)
+    return Constant(*args, **kwargs)
+
 
 class ZeroInflatedPoisson(Discrete):
     R"""

pymc3/step_methods/__init__.py

@@ -16,6 +16,4 @@
 
 from .slicer import Slice
 
-from .nuts import NUTS
-
-from .arraystep import Constant
+from .nuts import NUTS

pymc3/step_methods/arraystep.py

@@ -8,7 +8,7 @@
 from enum import IntEnum, unique
 
 __all__ = ['ArrayStep', 'ArrayStepShared', 'metrop_select', 'SamplerHist',
-           'Competence', 'Constant']
+           'Competence']
 
 
 @unique
@@ -163,30 +163,3 @@ def __init__(self):
     def acceptr(self):
         return np.minimum(np.exp(self.metrops), 1)
 
-
-class Constant(ArrayStep):
-    """
-    Dummy sampler that returns the current value at every iteration. Useful for
-    fixing parameters at a particular value.
-
-    Parameters
-    ----------
-    vars : list
-        List of variables for sampler.
-    model : PyMC Model
-        Optional model for sampling step. Defaults to None (taken from context).
-    """
-
-    def __init__(self, vars, model=None, **kwargs):
-
-        model = modelcontext(model)
-
-        self.model = model
-
-        vars = inputvars(vars)
-
-        super(Constant, self).__init__(vars, [model.fastlogp], **kwargs)
-
-    def astep(self, q0, logp):
-
-        return q0

pymc3/tests/test_distributions.py

@@ -7,7 +7,7 @@
 from ..blocking import DictToVarBijection, DictToArrayBijection, ArrayOrdering
 from ..distributions import (DensityDist, Categorical, Multinomial, VonMises, Dirichlet,
                              MvStudentT, MvNormal, ZeroInflatedPoisson,
-                             ZeroInflatedNegativeBinomial, ConstantDist, Poisson, Bernoulli, Beta,
+                             ZeroInflatedNegativeBinomial, ConstantDist, Constant, Poisson, Bernoulli, Beta,
                              BetaBinomial, HalfStudentT, StudentT, Weibull, Pareto, InverseGamma,
                              Gamma, Cauchy, HalfCauchy, Lognormal, Laplace, NegativeBinomial,
                              Geometric, Exponential, ExGaussian, Normal, Flat, LKJCorr, Wald,
@@ -479,7 +479,7 @@ def test_poisson(self):
                                  lambda value, mu: sp.poisson.logpmf(value, mu))
 
     def test_constantdist(self):
-        self.pymc3_matches_scipy(ConstantDist, I, {'c': I},
+        self.pymc3_matches_scipy(Constant, I, {'c': I},
                                  lambda value, c: np.log(c == value))
 
     def test_zeroinflatedpoisson(self):

pymc3/tests/test_distributions_random.py

@@ -281,8 +281,8 @@ class TestNegativeBinomial(BaseTestCases.BaseTestCase):
     params = {'mu': 1., 'alpha': 1.}
 
 
-class TestConstantDist(BaseTestCases.BaseTestCase):
-    distribution = pm.ConstantDist
+class TestConstant(BaseTestCases.BaseTestCase):
+    distribution = pm.Constant
     params = {'c': 3}
 
 
@@ -471,7 +471,7 @@ def ref_rand(size, p):
     def test_constant_dist(self):
         def ref_rand(size, c):
             return c * np.ones(size, dtype=int)
-        pymc3_random_discrete(pm.ConstantDist, {'c': I}, ref_rand=ref_rand)
+        pymc3_random_discrete(pm.Constant, {'c': I}, ref_rand=ref_rand)
 
     def test_mv_normal(self):
         def ref_rand(size, mu, cov):

pymc3/tests/test_step.py

@@ -5,7 +5,7 @@
 from pymc3.sampling import assign_step_methods, sample
 from pymc3.model import Model
 from pymc3.step_methods import (NUTS, BinaryGibbsMetropolis, CategoricalGibbsMetropolis,
-                                Metropolis, Constant, Slice, CompoundStep,
+                                Metropolis, Slice, CompoundStep,
                                 MultivariateNormalProposal, HamiltonianMC)
 from pymc3.distributions import Binomial, Normal, Bernoulli, Categorical
 from numpy.testing import assert_almost_equal
@@ -68,14 +68,6 @@ def test_step_categorical(self):
             self.check_stat(check, trace)
 
 
-    def test_constant_step(self):
-        with Model():
-            x = Normal('x', 0, 1)
-            start = {'x': -1}
-            tr = sample(10, step=Constant([x]), start=start)
-        assert_almost_equal(tr['x'], start['x'], decimal=10)
-
-
 class TestCompoundStep(unittest.TestCase):
     samplers = (Metropolis, Slice, HamiltonianMC, NUTS)