Move all custom Exceptions to exceptions.py

pymc/backends/base.py

@@ -45,10 +45,6 @@
 logger = logging.getLogger(__name__)
 
 
-class BackendError(Exception):
-    pass
-
-
 class IBaseTrace(ABC, Sized):
     """Minimal interface needed to record and access draws and stats for one MCMC chain."""
 

pymc/data.py

@@ -36,6 +36,7 @@
 
 import pymc as pm
 
+from pymc.exceptions import ShapeError
 from pymc.pytensorf import convert_observed_data
 
 __all__ = [
@@ -237,7 +238,7 @@ def determine_coords(
 
     if isinstance(value, np.ndarray) and dims is not None:
         if len(dims) != value.ndim:
-            raise pm.exceptions.ShapeError(
+            raise ShapeError(
                 "Invalid data shape. The rank of the dataset must match the " "length of `dims`.",
                 actual=value.shape,
                 expected=value.ndim,
@@ -445,7 +446,7 @@ def Data(
     if isinstance(dims, str):
         dims = (dims,)
     if not (dims is None or len(dims) == x.ndim):
-        raise pm.exceptions.ShapeError(
+        raise ShapeError(
             "Length of `dims` must match the dimensions of the dataset.",
             actual=len(dims),
             expected=x.ndim,

pymc/exceptions.py

@@ -12,13 +12,6 @@
 #   See the License for the specific language governing permissions and
 #   limitations under the License.
 
-__all__ = [
-    "SamplingError",
-    "ImputationWarning",
-    "ShapeWarning",
-    "ShapeError",
-]
-
 
 class SamplingError(RuntimeError):
     pass
@@ -74,3 +67,55 @@ class NotConstantValueError(ValueError):
 
 class BlockModelAccessError(RuntimeError):
     pass
+
+
+class ParallelSamplingError(Exception):
+    def __init__(self, message, chain):
+        super().__init__(message)
+        self._chain = chain
+
+
+class RemoteTraceback(Exception):
+    def __init__(self, tb):
+        self.tb = tb
+
+    def __str__(self):
+        return self.tb
+
+
+class VariationalInferenceError(Exception):
+    """Exception for VI specific cases"""
+
+
+class NotImplementedInference(VariationalInferenceError, NotImplementedError):
+    """Marking non functional parts of code"""
+
+
+class ExplicitInferenceError(VariationalInferenceError, TypeError):
+    """Exception for bad explicit inference"""
+
+
+class ParametrizationError(VariationalInferenceError, ValueError):
+    """Error raised in case of bad parametrization"""
+
+
+class GroupError(VariationalInferenceError, TypeError):
+    """Error related to VI groups"""
+
+
+class IntegrationError(RuntimeError):
+    pass
+
+
+class PositiveDefiniteError(ValueError):
+    def __init__(self, msg, idx):
+        super().__init__(msg)
+        self.idx = idx
+        self.msg = msg
+
+    def __str__(self):
+        return f"Scaling is not positive definite: {self.msg}. Check indexes {self.idx}."
+
+
+class ParameterValueError(ValueError):
+    """Exception for invalid parameters values in logprob graphs"""

pymc/logprob/utils.py

@@ -63,6 +63,7 @@
 from pytensor.tensor.random.op import RandomVariable
 from pytensor.tensor.variable import TensorVariable
 
+from pymc.exceptions import ParameterValueError
 from pymc.logprob.abstract import MeasurableVariable, _logprob
 from pymc.util import makeiter
 
@@ -231,10 +232,6 @@ def check_potential_measurability(
     return False
 
 
-class ParameterValueError(ValueError):
-    """Exception for invalid parameters values in logprob graphs"""
-
-
 class CheckParameterValue(CheckAndRaise):
     """Implements a parameter value check in a logprob graph.
 

pymc/model/core.py

@@ -59,13 +59,13 @@
 from pymc.exceptions import (
     BlockModelAccessError,
     ImputationWarning,
+    ParameterValueError,
     SamplingError,
     ShapeError,
     ShapeWarning,
 )
 from pymc.initial_point import make_initial_point_fn
 from pymc.logprob.basic import transformed_conditional_logp
-from pymc.logprob.utils import ParameterValueError
 from pymc.model_graph import model_to_graphviz
 from pymc.pytensorf import (
     PointFunc,

pymc/sampling/mcmc.py

@@ -56,7 +56,7 @@
 )
 from pymc.backends.base import IBaseTrace, MultiTrace, _choose_chains
 from pymc.blocking import DictToArrayBijection
-from pymc.exceptions import SamplingError
+from pymc.exceptions import ParallelSamplingError, SamplingError
 from pymc.initial_point import PointType, StartDict, make_initial_point_fns_per_chain
 from pymc.model import Model, modelcontext
 from pymc.sampling.parallel import Draw, _cpu_count
@@ -1199,7 +1199,7 @@ def _mp_sample(
                     if callback is not None:
                         callback(trace=strace, draw=draw)
 
-        except ps.ParallelSamplingError as error:
+        except ParallelSamplingError as error:
             strace = traces[error._chain]
             for strace in traces:
                 strace.close()

pymc/sampling/parallel.py

@@ -29,25 +29,13 @@
 from fastprogress.fastprogress import progress_bar
 
 from pymc.blocking import DictToArrayBijection
-from pymc.exceptions import SamplingError
+from pymc.exceptions import ParallelSamplingError, RemoteTraceback, SamplingError
 from pymc.util import RandomSeed
 
 logger = logging.getLogger(__name__)
 
 
-class ParallelSamplingError(Exception):
-    def __init__(self, message, chain):
-        super().__init__(message)
-        self._chain = chain
-
-
 # Taken from https://hg.python.org/cpython/rev/c4f92b597074
-class RemoteTraceback(Exception):
-    def __init__(self, tb):
-        self.tb = tb
-
-    def __str__(self):
-        return self.tb
 
 
 class ExceptionWithTraceback:

pymc/step_methods/hmc/base_hmc.py

@@ -23,14 +23,14 @@
 import numpy as np
 
 from pymc.blocking import DictToArrayBijection, RaveledVars, StatsType
-from pymc.exceptions import SamplingError
+from pymc.exceptions import IntegrationError, SamplingError
 from pymc.model import Point, modelcontext
 from pymc.pytensorf import floatX
 from pymc.stats.convergence import SamplerWarning, WarningType
 from pymc.step_methods import step_sizes
 from pymc.step_methods.arraystep import GradientSharedStep
 from pymc.step_methods.hmc import integration
-from pymc.step_methods.hmc.integration import IntegrationError, State
+from pymc.step_methods.hmc.integration import State
 from pymc.step_methods.hmc.quadpotential import QuadPotentialDiagAdapt, quad_potential
 from pymc.tuning import guess_scaling
 from pymc.util import get_value_vars_from_user_vars

pymc/step_methods/hmc/hmc.py

@@ -18,10 +18,11 @@
 
 import numpy as np
 
+from pymc.exceptions import IntegrationError
 from pymc.stats.convergence import SamplerWarning
 from pymc.step_methods.compound import Competence
 from pymc.step_methods.hmc.base_hmc import BaseHMC, DivergenceInfo, HMCStepData
-from pymc.step_methods.hmc.integration import IntegrationError, State
+from pymc.step_methods.hmc.integration import State
 from pymc.vartypes import discrete_types
 
 __all__ = ["HamiltonianMC"]

pymc/step_methods/hmc/integration.py

@@ -19,6 +19,7 @@
 from scipy import linalg
 
 from pymc.blocking import RaveledVars
+from pymc.exceptions import IntegrationError
 from pymc.step_methods.hmc.quadpotential import QuadPotential
 
 
@@ -32,10 +33,6 @@ class State(NamedTuple):
     index_in_trajectory: int
 
 
-class IntegrationError(RuntimeError):
-    pass
-
-
 class CpuLeapfrogIntegrator:
     def __init__(self, potential: QuadPotential, logp_dlogp_func):
         """Leapfrog integrator using CPU."""

pymc/step_methods/hmc/nuts.py

@@ -18,13 +18,14 @@
 
 import numpy as np
 
+from pymc.exceptions import IntegrationError
 from pymc.math import logbern
 from pymc.pytensorf import floatX
 from pymc.stats.convergence import SamplerWarning
 from pymc.step_methods.compound import Competence
 from pymc.step_methods.hmc import integration
 from pymc.step_methods.hmc.base_hmc import BaseHMC, DivergenceInfo, HMCStepData
-from pymc.step_methods.hmc.integration import IntegrationError, State
+from pymc.step_methods.hmc.integration import State
 from pymc.vartypes import continuous_types
 
 __all__ = ["NUTS"]

pymc/step_methods/hmc/quadpotential.py

@@ -25,6 +25,7 @@
 from numpy.random import normal
 from scipy.sparse import issparse
 
+from pymc.exceptions import PositiveDefiniteError
 from pymc.pytensorf import floatX
 
 __all__ = [
@@ -87,16 +88,6 @@ def partial_check_positive_definite(C):
         raise PositiveDefiniteError("Simple check failed. Diagonal contains negatives", i)
 
 
-class PositiveDefiniteError(ValueError):
-    def __init__(self, msg, idx):
-        super().__init__(msg)
-        self.idx = idx
-        self.msg = msg
-
-    def __str__(self):
-        return f"Scaling is not positive definite: {self.msg}. Check indexes {self.idx}."
-
-
 class QuadPotential:
     dtype: np.dtype
 

pymc/testing.py

@@ -34,9 +34,10 @@
 
 from pymc.distributions.distribution import Distribution
 from pymc.distributions.shape_utils import change_dist_size
+from pymc.exceptions import ParameterValueError
 from pymc.initial_point import make_initial_point_fn
 from pymc.logprob.basic import icdf, logcdf, logp, transformed_conditional_logp
-from pymc.logprob.utils import ParameterValueError, find_rvs_in_graph
+from pymc.logprob.utils import find_rvs_in_graph
 from pymc.pytensorf import (
     compile_pymc,
     floatX,

pymc/variational/approximations.py

@@ -26,12 +26,11 @@
 
 from pymc.blocking import DictToArrayBijection
 from pymc.distributions.dist_math import rho2sigma
+from pymc.exceptions import NotImplementedInference
 from pymc.util import makeiter
-from pymc.variational import opvi
 from pymc.variational.opvi import (
     Approximation,
     Group,
-    NotImplementedInference,
     _known_scan_ignored_inputs,
     node_property,
 )
@@ -212,7 +211,7 @@ def __init_group__(self, group):
     def create_shared_params(self, trace=None, size=None, jitter=1, start=None):
         if trace is None:
             if size is None:
-                raise opvi.ParametrizationError("Need `trace` or `size` to initialize")
+                raise pymc.exceptions.ParametrizationError("Need `trace` or `size` to initialize")
             else:
                 start = self._prepare_start(start)
                 # Initialize particles

pymc/variational/operators.py

@@ -19,9 +19,9 @@
 
 import pymc as pm
 
+from pymc.exceptions import NotImplementedInference, ParametrizationError
 from pymc.variational import opvi
 from pymc.variational.opvi import (
-    NotImplementedInference,
     ObjectiveFunction,
     Operator,
     _known_scan_ignored_inputs,
@@ -81,7 +81,7 @@ class KSDObjective(ObjectiveFunction):
 
     def __init__(self, op: KSD, tf: opvi.TestFunction):
         if not isinstance(op, KSD):
-            raise opvi.ParametrizationError("Op should be KSD")
+            raise ParametrizationError("Op should be KSD")
         super().__init__(op, tf)
 
     @pytensor.config.change_flags(compute_test_value="off")

pymc/variational/opvi.py

@@ -68,6 +68,12 @@
 from pymc.backends.base import MultiTrace
 from pymc.backends.ndarray import NDArray
 from pymc.blocking import DictToArrayBijection
+from pymc.exceptions import (
+    ExplicitInferenceError,
+    GroupError,
+    ParametrizationError,
+    VariationalInferenceError,
+)
 from pymc.initial_point import make_initial_point_fn
 from pymc.model import modelcontext
 from pymc.pytensorf import (
@@ -91,30 +97,6 @@
 __all__ = ["ObjectiveFunction", "Operator", "TestFunction", "Group", "Approximation"]
 
 
-class VariationalInferenceError(Exception):
-    """Exception for VI specific cases"""
-
-
-class NotImplementedInference(VariationalInferenceError, NotImplementedError):
-    """Marking non functional parts of code"""
-
-
-class ExplicitInferenceError(VariationalInferenceError, TypeError):
-    """Exception for bad explicit inference"""
-
-
-class AEVBInferenceError(VariationalInferenceError, TypeError):
-    """Exception for bad aevb inference"""
-
-
-class ParametrizationError(VariationalInferenceError, ValueError):
-    """Error raised in case of bad parametrization"""
-
-
-class GroupError(VariationalInferenceError, TypeError):
-    """Error related to VI groups"""
-
-
 def _known_scan_ignored_inputs(terms):
     # TODO: remove when scan issue with grads is fixed
     from pymc.data import MinibatchIndexRV

tests/distributions/test_continuous.py

@@ -29,8 +29,8 @@
 
 from pymc.distributions.continuous import Normal, Uniform, get_tau_sigma, interpolated
 from pymc.distributions.dist_math import clipped_beta_rvs
+from pymc.exceptions import ParameterValueError
 from pymc.logprob.basic import icdf, logcdf, logp
-from pymc.logprob.utils import ParameterValueError
 from pymc.pytensorf import floatX
 from pymc.testing import (
     BaseTestDistributionRandom,

tests/distributions/test_discrete.py

@@ -29,8 +29,8 @@
 import pymc as pm
 
 from pymc.distributions.discrete import Geometric, _OrderedLogistic, _OrderedProbit
+from pymc.exceptions import ParameterValueError
 from pymc.logprob.basic import icdf, logcdf, logp
-from pymc.logprob.utils import ParameterValueError
 from pymc.pytensorf import floatX
 from pymc.testing import (
     BaseTestDistributionRandom,