DataScience Task3 - Python

import numpy as np

class Normalizer:
    def fit(self, X):
        X = np.array(X, float)
        self.stats = np.arange(len(X), dtype=float)
        i = 0
        while i < len(X):
            tab = X[:, i]
            j, count, sum = (0, 0, 0)
            while j < len(tab):
                if not np.isnan(tab[j]):
                    sum = sum + tab[j]
                    count = count + 1
                j = j + 1
            self.stats[i] = sum / count
            i = i + 1

    def transform(self, X):
        X = np.array(X, float)
        i = 0
        newX = X.copy()
        while i < len(self.stats):
            tab = newX[:, i]
            j = 0
            while j < len(tab):
                if np.isnan(tab[j]):
                    tab[j] = self.stats[i]
                j = j + 1
            i = i + 1
        return newX

    def fit_transform(self, X):
        self.fit(X)
        return self.transform(X)

normalizer = Normalizer()
normalizer.fit([[7, 2, 3], [4, np.nan, 6], [10, 5, 9]])
print(normalizer.transform([[np.nan, 2, 3], [4, np.nan, 6], [10, np.nan, 9]]))