Merge pull request #227 from galenseilis/patch-6

Mixtures and __repr__ enhancements.

ciw/dists/distributions.py

@@ -1,7 +1,7 @@
-from ciw.auxiliary import *
-from itertools import cycle
-import numpy as np
+'''Distributions available in Ciw.'''
+
 import copy
+from itertools import cycle
 from operator import add, mul, sub, truediv
 from random import (
     expovariate,
@@ -11,7 +11,12 @@
     lognormvariate,
     weibullvariate,
 )
+from typing import List, NoReturn
+
+import numpy as np
 
+from ciw.auxiliary import *
+from ciw.individual import Individual
 
 class Distribution(object):
     """
@@ -99,7 +104,7 @@ def __init__(self, lower, upper):
         self.upper = upper
 
     def __repr__(self):
-        return "Uniform: {0}, {1}".format(self.lower, self.upper)
+        return f"Uniform({self.lower}, {self.upper})"
 
     def sample(self, t=None, ind=None):
         return uniform(self.lower, self.upper)
@@ -121,7 +126,7 @@ def __init__(self, value):
         self.value = value
 
     def __repr__(self):
-        return "Deterministic: {0}".format(self.value)
+        return f"Deterministic({self.value})"
 
     def sample(self, t=None, ind=None):
         return self.value
@@ -151,7 +156,7 @@ def __init__(self, lower, mode, upper):
         self.upper = upper
 
     def __repr__(self):
-        return "Triangular: {0}, {1}, {2}".format(self.lower, self.mode, self.upper)
+        return f"Triangular({self.lower}, {self.mode}, {self.upper})"
 
     def sample(self, t=None, ind=None):
         return triangular(self.lower, self.upper, self.mode)
@@ -173,7 +178,7 @@ def __init__(self, rate):
         self.rate = rate
 
     def __repr__(self):
-        return "Exponential: {0}".format(self.rate)
+        return f"Exponential({self.rate})"
 
     def sample(self, t=None, ind=None):
         return expovariate(self.rate)
@@ -193,7 +198,7 @@ def __init__(self, shape, scale):
         self.scale = scale
 
     def __repr__(self):
-        return "Gamma: {0}, {1}".format(self.shape, self.scale)
+        return f"Gamma({self.shape}, {self.scale})"
 
     def sample(self, t=None, ind=None):
         return gammavariate(self.shape, self.scale)
@@ -213,7 +218,7 @@ def __init__(self, mean, sd):
         self.sd = sd
 
     def __repr__(self):
-        return "Normal: {0}, {1}".format(self.mean, self.sd)
+        return f"Normal({self.mean}, {self.sd})"
 
     def sample(self, t=None, ind=None):
         return truncated_normal(self.mean, self.sd)
@@ -233,7 +238,7 @@ def __init__(self, mean, sd):
         self.sd = sd
 
     def __repr__(self):
-        return "Lognormal: {0}, {1}".format(self.mean, self.sd)
+        return f"Lognormal({self.mean}, {self.sd})"
 
     def sample(self, t=None, ind=None):
         return lognormvariate(self.mean, self.sd)
@@ -253,7 +258,7 @@ def __init__(self, scale, shape):
         self.shape = shape
 
     def __repr__(self):
-        return "Weibull: {0}, {1}".format(self.scale, self.shape)
+        return f"Weibull({self.scale}, {self.shape})"
 
     def sample(self, t=None, ind=None):
         return weibullvariate(self.scale, self.shape)
@@ -298,7 +303,10 @@ def __init__(self, sequence):
         self.generator = cycle(self.sequence)
 
     def __repr__(self):
-        return "Sequential"
+        if len(self.sequence) <= 3:
+            return f"Sequential({self.sequence})"
+        else:
+            return f"Sequential({self.sequence[0]}, ..., {self.sequence[-1]})"
 
     def sample(self, t=None, ind=None):
         return next(self.generator)
@@ -324,7 +332,7 @@ def __init__(self, values, probs):
         self.probs = probs
 
     def __repr__(self):
-        return "Pmf"
+        return f"Pmf({self.values}, {self.probs})"
 
     def sample(self, t=None, ind=None):
         return random_choice(self.values, self.probs)
@@ -420,7 +428,7 @@ def __init__(self, rate, num_phases):
         super().__init__(initial_state, absorbing_matrix)
 
     def __repr__(self):
-        return f"Erlang: {self.rate}, {self.num_phases}"
+        return f"Erlang({self.rate}, {self.num_phases})"
 
 
 class HyperExponential(PhaseType):
@@ -611,7 +619,7 @@ def sample(self, t=None, ind=None):
         return ciw.rng.poisson(lam=self.rate)
 
     def __repr__(self):
-        return f"Poisson: {self.rate}"
+        return f"Poisson({self.rate})"
 
 
 class Geometric(Distribution):
@@ -634,7 +642,7 @@ def sample(self, t=None, ind=None):
         return ciw.rng.geometric(p=self.prob)
 
     def __repr__(self):
-        return f"Geometric: {self.prob}"
+        return f"Geometric({self.prob})"
 
 
 class Binomial(Distribution):
@@ -663,4 +671,83 @@ def sample(self, t=None, ind=None):
         return ciw.rng.binomial(n=self.n, p=self.prob)
 
     def __repr__(self):
-        return f"Binomial: {self.n}, {self.prob}"
+        return f"Binomial({self.n}, {self.prob})"
+
+
+class MixtureDistribution(Distribution):
+    """
+    A mixture distribution combining multiple probability distributions.
+
+    Parameters
+    ----------
+    dists : List[Distribution]
+        A list of probability distributions to be combined in the mixture.
+    rhos : List[float]
+        A list of weights corresponding to the importance of each distribution in the mixture.
+        The weights must sum to 1.
+
+    Attributes
+    ----------
+    rhos : List[float]
+        List of weights assigned to each distribution in the mixture.
+    dists : List[Distribution]
+        List of probability distributions in the mixture.
+
+    Methods
+    -------
+    sample(t: float, inds: List[Individual] = None) -> float:
+        Generate a random sample from the mixture distribution.
+
+    Notes
+    -----
+    The weights in `rhos` should sum to 1, indicating the relative importance of each distribution
+    in the mixture. The distributions in `dists` should be instances of `ciw.dists.Distribution`.
+    """
+
+    def __init__(self, dists: List[Distribution], rhos: List[float]) -> NoReturn:
+        """
+        Initialize the MixtureDistribution.
+
+        Parameters
+        ----------
+        dists : List[Distribution]
+            A list of probability distributions to be combined in the mixture.
+        rhos : List[float]
+            A list of weights corresponding to the importance of each distribution in the mixture.
+            The weights must sum to 1.
+        """
+        self.rhos = rhos
+        self.dists = dists
+
+    def sample(self, t: float, inds: List[Individual] = None) -> float:
+        """
+        Generate a random sample from the mixture distribution.
+
+        Parameters
+        ----------
+        t : float
+            The time parameter for the sample generation.
+        inds : List[Individual], optional
+            List of individuals associated with the sample, if applicable.
+
+        Returns
+        -------
+        float
+            A random sample from the mixture distribution.
+        """
+        chosen_dist = random.choices(
+            population=self.dists,
+            weights=self.rhos,
+            k=1)[0]
+
+        return chosen_dist.sample(t, inds)
+
+    def __repr__(self):
+
+        dist_strs = [f'{rho} * {dist}' for rho,dist in zip(self.rhos, self.dists)]
+
+        if len(dist_strs) <= 3:
+            inside = ', '.join(dist_strs)
+            return f"Mixture({inside})"
+        else:
+            return f"Mixture({dist_strs[0]}, ..., {dist_strs[-1]})"