[ENH] Added Inverse Gamma distribution

skpro/distributions/__init__.py

@@ -6,6 +6,7 @@
 __all__ = [
     "Alpha",
     "Beta",
+    "Binomial",
     "ChiSquared",
     "Delta",
     "Empirical",
@@ -14,6 +15,7 @@
     "Gamma",
     "HalfNormal",
     "IID",
+    "InverseGamma",
     "Laplace",
     "Logistic",
     "LogNormal",
@@ -32,6 +34,7 @@
 
 from skpro.distributions.alpha import Alpha
 from skpro.distributions.beta import Beta
+from skpro.distributions.binomial import Binomial
 from skpro.distributions.chi_squared import ChiSquared
 from skpro.distributions.compose import IID
 from skpro.distributions.delta import Delta
@@ -40,6 +43,7 @@
 from skpro.distributions.fisk import Fisk
 from skpro.distributions.gamma import Gamma
 from skpro.distributions.halfnormal import HalfNormal
+from skpro.distributions.inversegamma import InverseGamma
 from skpro.distributions.laplace import Laplace
 from skpro.distributions.logistic import Logistic
 from skpro.distributions.lognormal import LogNormal

skpro/distributions/binomial.py

@@ -0,0 +1,73 @@
+# copyright: skpro developers, BSD-3-Clause License (see LICENSE file)
+"""Binomial probability distribution."""
+
+__author__ = ["meraldoantonio"]
+
+import pandas as pd
+from scipy.stats import binom, rv_discrete
+
+from skpro.distributions.adapters.scipy import _ScipyAdapter
+
+
+class Binomial(_ScipyAdapter):
+    r"""Binomial distribution.
+
+    Most methods wrap ``scipy.stats.binom``.
+    The Binomial distribution is parameterized by the number of trials :math:`n`
+    and the probability of success :math:`p`,
+    such that the probability mass function (PMF) is given by:
+
+    .. math:: P(X = k) = \binom{n}{k} p^k (1-p)^{n-k}
+
+    Parameters
+    ----------
+    n : int or array of int (1D or 2D), must be non-negative
+    p : float or array of float (1D or 2D), must be in [0, 1]
+    index : pd.Index, optional, default = RangeIndex
+    columns : pd.Index, optional, default = RangeIndex
+
+    Example
+    -------
+    >>> from skpro.distributions.binomial import Binomial
+
+    >>> d = Binomial(n=[[10, 10], [20, 30], [40, 50]], p=0.5)
+    """
+
+    _tags = {
+        "capabilities:approx": ["pmf"],
+        "capabilities:exact": ["mean", "var", "pmf", "log_pmf", "cdf", "ppf"],
+        "distr:measuretype": "discrete",
+        "distr:paramtype": "parametric",
+        "broadcast_init": "on",
+    }
+
+    def __init__(self, n, p, index=None, columns=None):
+        self.n = n
+        self.p = p
+
+        super().__init__(index=index, columns=columns)
+
+    def _get_scipy_object(self) -> rv_discrete:
+        return binom
+
+    def _get_scipy_param(self):
+        n = self._bc_params["n"]
+        p = self._bc_params["p"]
+
+        return [], {"n": n, "p": p}
+
+    @classmethod
+    def get_test_params(cls, parameter_set="default"):
+        """Return testing parameter settings for the estimator."""
+        # array case examples
+        params1 = {"n": [[10, 10], [20, 30], [40, 50]], "p": 0.5}
+        params2 = {
+            "n": 10,
+            "p": 0.5,
+            "index": pd.Index([1, 2, 5]),
+            "columns": pd.Index(["a", "b"]),
+        }
+        # scalar case examples
+        params3 = {"n": 15, "p": 0.7}
+
+        return [params1, params2, params3]

skpro/distributions/inversegamma.py

@@ -0,0 +1,79 @@
+# copyright: skpro developers, BSD-3-Clause License (see LICENSE file)
+"""Inverse Gamma probability distribution."""
+
+__author__ = ["meraldoantonio"]
+
+import pandas as pd
+from scipy.stats import invgamma, rv_continuous
+
+from skpro.distributions.adapters.scipy import _ScipyAdapter
+
+
+class InverseGamma(_ScipyAdapter):
+    r"""Inverse Gamma Distribution.
+
+    Most methods wrap ``scipy.stats.invgamma``.
+
+    The Inverse Gamma Distribution is parameterized by shape :math:`\alpha` and
+    scale :math:`\beta`, such that the pdf is
+
+    .. math:: f(x) = \frac{\beta^{\alpha} x^{-\alpha-1} \exp\left(-\frac{\beta}{x}\right)}{\tau(\alpha)}
+
+    where :math:`\tau(\alpha)` is the Gamma function.
+    For all positive integers, :math:`\tau(\alpha) = (\alpha-1)!`.
+
+    Parameters
+    ----------
+    alpha : float or array of float (1D or 2D)
+        It represents the shape parameter.
+    beta : float or array of float (1D or 2D)
+        It represents the scale parameter.
+    index : pd.Index, optional, default = RangeIndex
+    columns : pd.Index, optional, default = RangeIndex
+
+    Example
+    -------
+    >>> from skpro.distributions.invgamma import InverseGamma
+
+    >>> d = InverseGamma(beta=[[1, 1], [2, 3], [4, 5]], alpha=2)
+    """  # noqa: E501
+
+    _tags = {
+        "capabilities:approx": ["energy", "pdfnorm"],
+        "capabilities:exact": ["mean", "var", "pdf", "log_pdf", "cdf", "ppf"],
+        "distr:measuretype": "continuous",
+        "distr:paramtype": "parametric",
+        "broadcast_init": "on",
+    }
+
+    def __init__(self, alpha, beta, index=None, columns=None):
+        self.alpha = alpha
+        self.beta = beta
+
+        super().__init__(index=index, columns=columns)
+
+    def _get_scipy_object(self) -> rv_continuous:
+        return invgamma
+
+    def _get_scipy_param(self):
+        alpha = self._bc_params["alpha"]
+        beta = self._bc_params["beta"]
+        scale = beta
+
+        return [], {"a": alpha, "scale": scale}
+
+    @classmethod
+    def get_test_params(cls, parameter_set="default"):
+        """Return testing parameter settings for the estimator."""
+        # array case examples
+        params1 = {"alpha": [6, 2.5], "beta": [[1, 1], [2, 3], [4, 5]]}
+        params2 = {
+            "alpha": 2,
+            "beta": 3,
+            "index": pd.Index([1, 2, 5]),
+            "columns": pd.Index(["a", "b"]),
+        }
+        # scalar case examples
+        params3 = {"alpha": 1.5, "beta": 2.1}
+
+        return [params1, params2, params3]