count_ordinal_encoder.py

import numpy as np
import pandas as pd
from sklearn.preprocessing import OrdinalEncoder

class CountOrdinalEncoder(OrdinalEncoder):
    """Encode categorical features as an integer array
    usint count information.
    """
    def __init__(self, categories='auto', dtype=np.float64):
        self.categories = categories
        self.dtype = dtype

    def fit(self, X, y=None):
        """Fit the OrdinalEncoder to X.

        Parameters
        ----------
        X : array-like, shape [n_samples, n_features]
            The data to determine the categories of each feature.

        Returns
        -------
        self
        """
        super().fit(X)
        X_list, _, _ = self._check_X(X)
        # now we'll reorder by counts
        for k, cat in enumerate(self.categories_):
            counts = []
            for c in cat:
                counts.append(np.sum(X_list[k] == c))
            order = np.argsort(counts)
            self.categories_[k] = cat[order]
        return self

coe = CountOrdinalEncoder()
coe.fit_transform(pd.DataFrame(['fr', 'fr', 'fr', 'en', 'en', 'es', 'es']))