v2: Additional probability distributions

Add additional probability distributions as required for PEtab-dev/PEtab#595.
See #374.

Author: dweindl

petab/v1/distributions.py

@@ -5,7 +5,7 @@
 import abc
 
 import numpy as np
-from scipy.stats import laplace, norm, uniform
+from scipy.stats import cauchy, chi2, laplace, norm, uniform
 
 __all__ = [
     "Distribution",
@@ -366,7 +366,12 @@ def __init__(
         super().__init__(log=log)
 
     def __repr__(self):
-        log = f", log={self._logbase}" if self._logbase else ""
+        if self._logbase is False:
+            log = ""
+        if self._logbase == np.exp(1):
+            log = ", log=True"
+        else:
+            log = f", log={self._logbase}"
         return f"Uniform(low={self._low}, high={self._high}{log})"
 
     def _sample(self, shape=None) -> np.ndarray | float:
@@ -411,9 +416,16 @@ def __init__(
         super().__init__(log=log, trunc=trunc)
 
     def __repr__(self):
+        if self._logbase is False:
+            log = ""
+        if self._logbase == np.exp(1):
+            log = ", log=True"
+        else:
+            log = f", log={self._logbase}"
+
         trunc = f", trunc={self._trunc}" if self._trunc else ""
-        log = f", log={self._logbase}" if self._logbase else ""
-        return f"Laplace(loc={self._loc}, scale={self._scale}{trunc}{log})"
+
+        return f"Laplace(loc={self._loc}, scale={self._scale}{log}{trunc})"
 
     def _sample(self, shape=None) -> np.ndarray | float:
         return np.random.laplace(loc=self._loc, scale=self._scale, size=shape)
@@ -436,3 +448,97 @@ def loc(self) -> float:
     def scale(self) -> float:
         """The scale parameter of the underlying distribution."""
         return self._scale
+
+
+class Cauchy(Distribution):
+    """A Cauchy distribution."""
+
+    def __init__(
+        self,
+        loc: float,
+        scale: float,
+        trunc: tuple[float, float] | None = None,
+        log: bool | float = False,
+    ):
+        self._loc = loc
+        self._scale = scale
+        super().__init__(log=log, trunc=trunc)
+
+    def __repr__(self):
+        if self._logbase is False:
+            log = ""
+        if self._logbase == np.exp(1):
+            log = ", log=True"
+        else:
+            log = f", log={self._logbase}"
+
+        trunc = f", trunc={self._trunc}" if self._trunc else ""
+
+        cls = self.__class__.__name__
+        return f"{cls}(loc={self._loc}, scale={self._scale}{log}{trunc})"
+
+    def _pdf_untransformed_untruncated(self, x) -> np.ndarray | float:
+        return cauchy.pdf(x, loc=self._loc, scale=self._scale)
+
+    def _cdf_untransformed_untruncated(self, x) -> np.ndarray | float:
+        return cauchy.cdf(x, loc=self._loc, scale=self._scale)
+
+    def _ppf_untransformed_untruncated(self, q) -> np.ndarray | float:
+        return cauchy.ppf(q, loc=self._loc, scale=self._scale)
+
+    @property
+    def loc(self) -> float:
+        """The location parameter of the underlying distribution."""
+        return self._loc
+
+    @property
+    def scale(self) -> float:
+        """The scale parameter of the underlying distribution."""
+        return self._scale
+
+
+class ChiSquare(Distribution):
+    """A chi-squared distribution.
+
+    :param k: Degrees of freedom.
+    """
+
+    def __init__(
+        self,
+        k: int,
+        trunc: tuple[float, float] | None = None,
+    ):
+        if not k.is_integer() or k < 1:
+            raise ValueError("k must be a positive integer.")
+
+        self._k = k
+        super().__init__(log=False, trunc=trunc)
+
+    def __repr__(self):
+        trunc = f", trunc={self._trunc}" if self._trunc else ""
+
+        cls = self.__class__.__name__
+        return f"{cls}(k={self._k}, {trunc})"
+
+    def _pdf_untransformed_untruncated(self, x) -> np.ndarray | float:
+        return chi2.pdf(x, df=self._k)
+
+    def _cdf_untransformed_untruncated(self, x) -> np.ndarray | float:
+        return chi2.cdf(x, df=self._k)
+
+    def _ppf_untransformed_untruncated(self, q) -> np.ndarray | float:
+        return chi2.ppf(q, df=self._k)
+
+    @property
+    def k(self) -> int:
+        """The degrees of freedom parameter."""
+        return self._k
+
+
+class Exponential(Distribution): ...
+
+
+class Gamma(Distribution): ...
+
+
+class Rayleigh(Distribution): ...