[ENH] Implement Multivariatenormal

skpro/distributions/__init__.py

@@ -26,6 +26,7 @@
     "LogNormal",
     "MeanScale",
     "Mixture",
+    "MultiVariate_Normal",
     "Normal",
     "Pareto",
     "Poisson",

skpro/distributions/multivariatenormal.py

@@ -0,0 +1,95 @@
+# copyright: skpro developers, BSD-3-Clause License (see LICENSE file)
+"""Multivariate Normal probability distribution."""
+
+__author__ = ["HarshvirSandhu"]
+
+import pandas as pd
+from scipy.stats import multivariate_normal
+
+from skpro.distributions.base import BaseDistribution
+
+
+class MultiVariate_Normal(BaseDistribution):
+    r"""Multivariate Normal distribution, aka log-logistic distribution.
+
+    Most methods wrap ``scipy.stats.multivariate_normal``.
+
+    The mean keyword specifies the mean and cov specifies the covariance matrix.
+    The probability distribution function (PDF) is given by:
+
+    .. math::
+
+        f(x) = \frac{1}{\sqrt{(2 \pi)^k \det \Sigma}}
+               \exp\left( -\frac{1}{2} (x - \mu)^T \Sigma^{-1} (x - \mu) \right),
+
+    where :math:`\mu` is the mean, :math:`\Sigma` the covariance matrix,
+    :math:`k` the rank of :math:`\Sigma`.
+
+    Parameters
+    ----------
+    mean : array_like, default = ``[0]``
+        Mean of the distribution.
+    cov : array_like or `Covariance`, default = ``[1]``
+        Symmetric positive (semi)definite covariance matrix of the distribution.
+    index : pd.Index, optional, default = RangeIndex
+    columns : pd.Index, optional, default = RangeIndex
+
+    Examples
+    --------
+    >>> from skpro.distributions.multivariatenormal import MultiVariate_Normal
+
+    >>> d = MultiVariate_Normal(mean=[1, 2, 3, 4, 5], cov=3)
+    """
+
+    _tags = {
+        "capabilities:approx": ["energy", "pdfnorm"],
+        "capabilities:exact": [
+            "mean",
+            "var",
+            "pdf",
+            "log_pdf",
+            "cdf",
+        ],
+        "distr:measuretype": "continuous",
+        "distr:paramtype": "parametric",
+        "broadcast_init": "on",
+    }
+
+    def __init__(self, mean=None, cov=None, index=None, columns=None):
+        if mean is None:
+            mean = [0]
+        if cov is None:
+            cov = [1]
+        self.mean = mean
+        self.cov = cov
+        self.scipy_obj = multivariate_normal(mean=self.mean, cov=self.cov)
+
+        super().__init__(index=index, columns=columns)
+
+    def pdf(self, x):
+        """Probability density function."""
+        return self.scipy_obj.pdf(x)
+
+    def logpdf(self, x):
+        """Logarithmic probability density function."""
+        return self.scipy_obj.logpdf(x)
+
+    def cdf(self, x):
+        """Cumulative distribution function."""
+        return self.scipy_obj.cdf(x)
+
+    @classmethod
+    def get_test_params(cls, parameter_set="default"):
+        """Return testing parameter settings for the estimator."""
+        # array case examples
+        params1 = {"mean": [1, 2, 3, 4, 5], "cov": 3}
+        params2 = {
+            "mean": 2,
+            "cov": 3,
+            "index": pd.Index([1, 2, 5]),
+            "columns": pd.Index(["a", "b"]),
+        }
+        # scalar case examples
+        params3 = {"mean": 1.5, "cov": 2.1}
+
+        return [params1, params2, params3]