Merge pull request #7 from msamsami/improve-dist-mixins

Improve distribution mixins

README.md

@@ -1,6 +1,6 @@
 # WNB: General and weighted naive Bayes classifiers
 
-![](https://img.shields.io/badge/version-v0.1.6-green)
+![](https://img.shields.io/badge/version-v0.1.7-green)
 ![](https://img.shields.io/badge/python-3.7%20%7C%203.8%20%7C%203.9-blue)
 
 <p>

setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name='wnb',
-    version='0.1.6',
+    version='0.1.7',
     description='Python library for the implementations of general and weighted naive Bayes (WNB) classifiers.',
     keywords=['python', 'bayes', 'naivebayes', 'classifier', 'probabilistic'],
     author='Mehdi Samsami',

wnb/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "0.1.6"
+__version__ = "0.1.7"
 __author__ = "Mehdi Samsami"
 
 __all__ = [

wnb/_base.py

@@ -1,5 +1,7 @@
 from abc import ABCMeta
+import inspect
 from functools import wraps
+from numbers import Number
 
 import numpy as np
 
@@ -26,25 +28,82 @@ def wrapper(*args):
     return decorator
 
 
-class ContinuousDistMixin(metaclass=ABCMeta):
+class DistMixin(metaclass=ABCMeta):
     """
-    Mixin class for all continuous probability distributions in wnb.
+    Mixin class for probability distributions in wnb.
     """
 
     name = None
 
-    def __init__(self, **kwargs):
-        """Initializes an instance of the probability distribution with given parameters.
+    @classmethod
+    def from_data(cls, data):
+        """Creates an instance of the class from given data. Distribution parameters will be estimated from the data.
 
+        Returns:
+            self: An instance of the class.
         """
         pass
 
     @classmethod
-    def from_data(cls, data):
-        """Creates an instance of the class from given data. Distribution parameters will be estimated from the data.
+    def _get_param_names(cls):
+        """Gets parameter names for the distribution instance.
+
+        """
+        init = getattr(cls.__init__, "deprecated_original", cls.__init__)
+        if init is object.__init__:
+            return []
+
+        init_signature = inspect.signature(init)
+        parameters = [
+            p
+            for p in init_signature.parameters.values()
+            if p.name != "self" and p.kind != p.VAR_KEYWORD
+        ]
+
+        for p in parameters:
+            if p.kind == p.VAR_POSITIONAL:
+                raise RuntimeError(
+                    "wnb estimators should always "
+                    "specify their parameters in the signature "
+                    "of their __init__ (no varargs). "
+                    "%s with constructor %s doesn't "
+                    "follow this convention." % (cls, init_signature)
+                )
+
+        return sorted([p.name for p in parameters])
+
+    def get_params(self) -> dict:
+        """Gets parameters for this distribution instance.
 
         Returns:
-            self: An instance of the class.
+            dict: Parameter names mapped to their values.
+        """
+        out = dict()
+        for key in self._get_param_names():
+            value = getattr(self, key)
+            out[key] = value
+        return out
+
+    def __repr__(self) -> str:
+        return "".join([
+            "<",
+            self.__class__.__name__,
+            "(",
+            ", ".join(
+                [f"{k}={v:.4f}" if isinstance(v, Number) else f"{k}={v}" for k, v in self.get_params().items()]
+            ),
+            ")>"
+        ])
+
+
+class ContinuousDistMixin(DistMixin, metaclass=ABCMeta):
+    """
+    Mixin class for all continuous probability distributions in wnb.
+    """
+
+    def __init__(self, **kwargs):
+        """Initializes an instance of the continuous probability distribution with given parameters.
+
         """
         pass
 
@@ -63,32 +122,18 @@ def pdf(self, x: float) -> float:
     def __call__(self, x: float) -> float:
         return self.pdf(x)
 
-    def __repr__(self) -> str:
-        pass
-
 
-class DiscreteDistMixin(metaclass=ABCMeta):
+class DiscreteDistMixin(DistMixin, metaclass=ABCMeta):
     """
     Mixin class for all discrete probability distributions in wnb.
     """
 
-    name = None
-
     def __init__(self, **kwargs):
-        """Initializes an instance of the probability distribution with given parameters.
+        """Initializes an instance of the discrete probability distribution with given parameters.
 
         """
         pass
 
-    @classmethod
-    def from_data(cls, data):
-        """Creates an instance of the class from given data. Distribution parameters will be estimated from the data.
-
-        Returns:
-            self: An instance of the class.
-        """
-        pass
-
     def pmf(self, x: float) -> float:
         """Returns the value of probability mass function (PMF) at x.
 
@@ -103,6 +148,3 @@ def pmf(self, x: float) -> float:
     @vectorize(signature="(),()->()")
     def __call__(self, x: float) -> float:
         return self.pmf(x)
-
-    def __repr__(self) -> str:
-        pass

wnb/dist.py

@@ -38,9 +38,6 @@ def from_data(cls, data: np.ndarray):
     def pdf(self, x: float) -> float:
         return (1.0 / np.sqrt(2 * np.pi * self.sigma**2)) * np.exp(-0.5 * (((x - self.mu) / self.sigma)**2))
 
-    def __repr__(self) -> str:
-        return f"<NormalDist(mu={self.mu:.4f}, sigma={self.sigma:.4f})>"
-
 
 class LognormalDist(ContinuousDistMixin):
     name = D.LOGNORMAL
@@ -59,9 +56,6 @@ def from_data(cls, data: np.ndarray):
     def pdf(self, x: float) -> float:
         return (1.0 / (x * np.sigma * np.sqrt(2 * np.pi))) * np.exp(-0.5 * ((np.log(x - self.mu) / self.sigma)**2))
 
-    def __repr__(self) -> str:
-        return f"<LognormalDist(mu={self.mu:.4f}, sigma={self.sigma:.4f})>"
-
 
 class ExponentialDist(ContinuousDistMixin):
     name = D.EXPONENTIAL
@@ -77,9 +71,6 @@ def from_data(cls, data: np.ndarray):
     def pdf(self, x: float) -> float:
         return self.rate * np.exp(-self.rate * x) if x >= 0 else 0.0
 
-    def __repr__(self) -> str:
-        return f"<ExponentialDist(rate={self.rate:.4f})>"
-
 
 class UniformDist(ContinuousDistMixin):
     name = D.UNIFORM
@@ -96,9 +87,6 @@ def from_data(cls, data):
     def pdf(self, x: float) -> float:
         return 1 / (self.b - self.a) if self.a <= x <= self.b else 0.0
 
-    def __repr__(self) -> str:
-        return f"<UniformDist(a={self.a:.4f}, b={self.b:.4f})>"
-
 
 class ParetoDist(ContinuousDistMixin):
     name = D.PARETO
@@ -116,9 +104,6 @@ def from_data(cls, data):
     def pdf(self, x: float) -> float:
         return (self.alpha * self.x_m**self.alpha) / x**(self.alpha + 1) if x >= self.x_m else 0.0
 
-    def __repr__(self) -> str:
-        return f"<ParetoDist(x_m={self.x_m:.4f}, alpha={self.alpha:.4f})>"
-
 
 class GammaDist(ContinuousDistMixin):
     name = D.GAMMA
@@ -139,9 +124,6 @@ def from_data(cls, data):
     def pdf(self, x: float) -> float:
         return (x ** (self.k-1) * np.exp(-x / self.theta)) / (gamma(self.k) * self.theta ** self.k)
 
-    def __repr__(self) -> str:
-        return f"<GammaDist(k={self.k:.4f}, theta={self.theta:.4f})>"
-
 
 class BernoulliDist(DiscreteDistMixin):
     name = D.BERNOULLI
@@ -160,9 +142,6 @@ def from_data(cls, data):
     def pmf(self, x: int) -> float:
         return 0.0 if x not in [0, 1] else self.p if x == 1 else 1 - self.p
 
-    def __repr__(self) -> str:
-        return f"<BernoulliDist(p={self.p:.4f})>"
-
 
 class CategoricalDist(DiscreteDistMixin):
     name = D.CATEGORICAL
@@ -179,9 +158,6 @@ def from_data(cls, data):
     def pmf(self, x: Any) -> float:
         return self.prob.get(x)
 
-    def __repr__(self) -> str:
-        return f"<CategoricalDist(prob={self.prob})>"
-
 
 class MultinomialDist(DiscreteDistMixin):
     name = D.MULTINOMIAL
@@ -206,9 +182,6 @@ def pmf(self, x: Sequence[int]) -> float:
             return np.math.factorial(self.n) * np.product([p**v for v, p in self.prob.items()]) / \
                 np.product([np.math.factorial(v) for v in self.prob.keys()])
 
-    def __repr__(self) -> str:
-        return f"<MultinomialDist(n={self.n}, prob={self.prob})>"
-
 
 class GeometricDist(DiscreteDistMixin):
     name = D.GEOMETRIC
@@ -227,9 +200,6 @@ def from_data(cls, data):
     def pmf(self, x: int) -> float:
         return self.p * (1 - self.p)**(x-1) if x >= 1 else 0.0
 
-    def __repr__(self) -> str:
-        return f"<GeometricDist(p={self.p:.4f})>"
-
 
 class PoissonDist(DiscreteDistMixin):
     name = D.POISSON
@@ -245,9 +215,6 @@ def from_data(cls, data):
     def pmf(self, x: int) -> float:
         return (np.exp(-self.rate) * self.rate**x) / np.math.factorial(x)
 
-    def __repr__(self) -> str:
-        return f"<PoissonDist(rate={self.rate:.4f})>"
-
 
 AllDistributions = {
     eval(cls).name: eval(cls)