Apéndice B — Código

El objetivo de este apéndice es describir algunas de las clases y métodos de Python utilizadas en el curso.

Paquetes usados

from scipy.stats import multivariate_normal
from scipy.special import logsumexp
import numpy as np

B.1 Clasificador Bayesiano Gausiano

class GaussianBayes(object):
    def __init__(self, naive=False) -> None:
        self._naive = naive

    @property
    def naive(self):
        return self._naive
    
    @property
    def labels(self):
        return self._labels

    @labels.setter
    def labels(self, labels):
        self._labels = labels

B.1.1 Estimación de Parámetros

    def fit(self, X, y):
        self.prior = y
        self.likelihood = (X, y)
        return self

    @property
    def prior(self):
        return self._prior

    @prior.setter
    def prior(self, y):
        labels, counts = np.unique(y, return_counts=True)
        prior = counts / counts.sum()        
        self.labels = labels
        self._prior = np.log(prior)

    @property
    def likelihood(self):
        return self._likelihood

    @likelihood.setter
    def likelihood(self, D):
        X, y = D
        likelihood = []
        for k in self.labels:
            mask = y == k
            mu = np.mean(X[mask], axis=0)
            if self.naive:
                cov = np.var(X[mask], axis=0, ddof=1)
            else:
                cov = np.cov(X[mask], rowvar=False)
            _ = multivariate_normal(mean=mu,
                                    cov=cov,
                                    allow_singular=True)
            likelihood.append(_)
        self._likelihood = likelihood

B.1.2 Predicción

    def predict(self, X):
        hy = self.predict_log_proba(X)
        _ = np.argmax(hy, axis=1)
        return self.labels[_]

    def predict_proba(self, X):
        _ = self.predict_log_proba(X)
        return np.exp(_)

    def predict_log_proba(self, X):
        log_ll = np.vstack([m.logpdf(X) 
                            for m in self.likelihood]).T
        prior = self.prior
        posterior = log_ll + prior
        evidence = np.atleast_2d(logsumexp(posterior,
                                           axis=1)).T
        return posterior - evidence

B.1.3 Uso

bayes = GaussianBayes().fit(T, y_t)
hy = bayes.predict(G)

# Código {#sec-clasificador-bayesiano-gausiano} El **objetivo** de este apéndice es describir algunas de las clases y métodos de Python utilizadas en el curso. ## Paquetes usados {.unnumbered} ```{python} #| echo: true from scipy.stats import multivariate_normal from scipy.special import logsumexp import numpy as np ``` ## Clasificador Bayesiano Gausiano ```python class GaussianBayes(object): def __init__(self, naive=False) -> None: self._naive = naive @property def naive(self): return self._naive @property def labels(self): return self._labels @labels.setter def labels(self, labels): self._labels = labels ``` ### Estimación de Parámetros ```python def fit(self, X, y): self.prior = y self.likelihood = (X, y) return self ``` ```python @property def prior(self): return self._prior @prior.setter def prior(self, y): labels, counts = np.unique(y, return_counts=True) prior = counts / counts.sum() self.labels = labels self._prior = np.log(prior) ``` ```python @property def likelihood(self): return self._likelihood @likelihood.setter def likelihood(self, D): X, y = D likelihood = [] for k in self.labels: mask = y == k mu = np.mean(X[mask], axis=0) if self.naive: cov = np.var(X[mask], axis=0, ddof=1) else: cov = np.cov(X[mask], rowvar=False) _ = multivariate_normal(mean=mu, cov=cov, allow_singular=True) likelihood.append(_) self._likelihood = likelihood ``` ### Predicción ```python def predict(self, X): hy = self.predict_log_proba(X) _ = np.argmax(hy, axis=1) return self.labels[_] ``` ```python def predict_proba(self, X): _ = self.predict_log_proba(X) return np.exp(_) ``` ```python def predict_log_proba(self, X): log_ll = np.vstack([m.logpdf(X) for m in self.likelihood]).T prior = self.prior posterior = log_ll + prior evidence = np.atleast_2d(logsumexp(posterior, axis=1)).T return posterior - evidence ``` ### Uso ```python bayes = GaussianBayes().fit(T, y_t) hy = bayes.predict(G) ```