{"cells": [{"cell_type": "markdown", "metadata": {}, "source": ["# LogisticRegression and Clustering\n", "\n", "A logistic regression implements a convex partition of the features spaces. A clustering algorithm applied before the trainer modifies the feature space in way the partition is not necessarily convex in the initial features. Let's see how. "]}, {"cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [{"data": {"text/html": ["
\n", ""], "text/plain": ["\n", " | nb_clusters | \n", "score | \n", "
---|---|---|
0 | \n", "1 | \n", "0.4475 | \n", "
1 | \n", "2 | \n", "0.6600 | \n", "
2 | \n", "3 | \n", "0.7475 | \n", "
3 | \n", "4 | \n", "0.8400 | \n", "
4 | \n", "5 | \n", "0.9200 | \n", "