Байесовский классификатор и ирисы Фишера

from collections import defaultdict
import numpy as np
from sklearn.datasets import load_iris
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split
 
class NaiveBayesClassifier(object):
    def init(self):  
        self.__class_freq = defaultdict(lambda:0)
        self.__feat_freq = defaultdict(lambda:0)
 
 
    def fit(self, X, y):
        # calculate classes and features frequencies
        for feature, label in zip(X, y):
            self.__class_freq[label] += 1
            for value in feature:
                self.__feat_freq[(value, label)] += 1
 
        # normalizate values
        num_samples = len(X)
        for k in self.__class_freq:
            self.__class_freq[k] /= num_samples
 
        for value, label in self.__feat_freq:
            self.__feat_freq[(value, label)] /= self.__class_freq[label]
 
        return self
 
 
def predict(self, X):
    # return argmin of classes 
               return min(self.__class_freq.keys(), 
               key=lambda c : self.__calculate_class_freq(X, c)) 
 
def __calculate_class_freq(self, X, clss):
    # calculate frequence for current class
    freq = - np.log(self.__class_freq[clss])
   
    for feat in X: 
        freq += - np.log(self.__feat_freq.get((feat, clss), 10 ** (-7)))
    return freq
 
data = load_iris()
 
X, y = data.data, data.target
 
X_train, X_test, y_train, y_test = train_test_split(X, y,
                                                    test_size=0.2,
                                                    random_state=42)
 
model = NaiveBayesClassifier().fit(X_train, y_train)
 
predictions = [model.predict(x) for x in X_test]
print(accuracy_score(predictions, y_test))