comparison.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from main_test.will_gen_data import *\n",
    "import main_test.run as run\n",
    "from my_mdnd import mdnd\n",
    "import networkx as nx\n",
    "import pygraphviz\n",
    "import time"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[15  1  5  3 17 16  4  6 14  2 18  0 19 33 21 36 37 35  9 39 34 40 25 20\n",
      " 43 12 45 29 13 27 22 41 26 24 30 11  7 48  8 49 44 38 47 31 52 50 42 32\n",
      " 28 10 46 23 51]\n",
      "[11, 1, 9, 3, 6, 2, 7, 36, 38, 18, 49, 35, 25, 28, 8, 0, 5, 4, 10, 12, 23, 14, 30, 51, 33, 22, 32, 29, 48, 27, 34, 43, 47, 13, 20, 17, 15, 16, 41, 19, 21, 31, 46, 24, 40, 26, 50, 42, 37, 39, 45, 52, 44]\n"
     ]
    }
   ],
   "source": [
    "params = {\n",
    "    'num_clusters': 5,\n",
    "    'cluster_size': [10,10,10,10,10],\n",
    "    'num_per_cluster': [100,100,100,100,100],\n",
    "    'overlap': 3,\n",
    "    'bg_prob': 0.2\n",
    "}\n",
    "will_gen_data(params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_color(cluster):\n",
    "    \"\"\"\n",
    "    return a mapping of 9 colors\n",
    "    \"\"\"\n",
    "    c = ['red', 'green', 'blue','yellow','purple','orange','brown','pink','black']\n",
    "    count = Counter(cluster).most_common()\n",
    "    count = [c[0] for c in count]\n",
    "    color = {count[x]: c[x] if x < 8 else c[8] for x in range(len(count))}\n",
    "    res = [color[c] for c in cluster]\n",
    "    return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "500\n"
     ]
    }
   ],
   "source": [
    "links_train,links_test,clusters_train,clusters_test,nodes,node_clusters = run.get_data_will('toy_test',ratio=0.9)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dirichlet Process gibbs sampler\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/drproduck/dnm/my_mdnd.py:173: RuntimeWarning: divide by zero encountered in log\n",
      "  return np.log(Nk) + np.log(wu) + np.log(wv)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "burn in | iter 0, number of clusters 6\n",
      "[0 3 2 0 0 3 0 1 1 0 1 1 1 2 0 0 3 3 2 1 3 2 1 1 1 1 0 2 3 2 2 2 3 1 1 1 0\n",
      " 2 0 2 0 3 2 2 2 0 3 0 2 3 2 1 2 2 0 1 2 0 1 3 2 0 1 1 2 0 1 1 0 2 3 0 0 1\n",
      " 2 0 3 3 1 0 3 2 1 2 2 2 2 1 0 2 1 1 2 2 1 0 2 2 2 2 2 1 0 5 0 3 0 0 1 1 0\n",
      " 2 0 2 0 1 1 0 0 1 2 1 0 0 2 2 2 0 1 3 1 0 1 2 2 3 1 2 3 2 1 1 1 0 3 3 1 2\n",
      " 1 1 2 1 3 0 2 0 1 1 2 0 1 1 1 2 3 0 2 0 2 1 2 0 2 0 1 1 1 3 1 1 0 0 3 1 2\n",
      " 0 1 2 3 2 2 1 0 5 2 1 1 1 2 3 0 1 1 2 2 1 2 0 0 1 0 2 2 3 2 0 0 1 2 2 3 2\n",
      " 3 3 0 1 0 2 1 2 1 0 1 1 2 1 2 1 2 2 2 1 0 0 0 1 2 1 0 0 2 0 0 2 1 0 1 2 3\n",
      " 1 2 0 0 1 2 2 2 1 2 2 1 3 2 2 1 3 1 3 3 2 2 2 3 1 2 0 1 1 0 1 2 3 2 0 2 1\n",
      " 0 2 2 2 0 1 0 2 1 2 0 1 3 1 2 1 2 1 0 0 3 0 1 2 2 0 2 1 2 2 3 1 1 3 1 1 1\n",
      " 0 3 2 2 1 0 2 2 0 2 1 1 1 1 2 2 0 1 3 3 3 2 2 0 0 1 0 2 5 1 2 3 0 1 2 3 3\n",
      " 2 1 1 1 0 0 2 0 0 1 0 0 1 0 2 1 3 2 0 0 0 1 1 0 1 0 1 3 2 2 3 1 0 0 2 0 0\n",
      " 1 1 1 2 0 2 0 1 3 0 1 2 3 0 3 2 2 3 0 2 1 1 0 0 1 0 0 2 0 2 1 2 3 0 1 0 0\n",
      " 1 2 0 1 1 4]\n",
      "burn in | iter 1, number of clusters 10\n",
      "[3 3 2 0 0 0 1 1 1 4 1 3 0 2 0 0 0 2 2 1 3 1 1 1 1 1 2 2 0 2 1 2 4 1 1 1 0\n",
      " 2 0 2 3 3 0 2 1 0 2 1 2 3 2 1 2 2 0 1 2 0 3 2 2 0 1 2 3 0 1 1 2 0 0 2 0 1\n",
      " 2 0 0 1 1 2 3 2 1 0 2 2 1 1 0 0 1 2 2 2 1 0 2 0 0 2 1 1 2 2 0 3 2 0 1 1 0\n",
      " 2 0 1 0 1 1 1 0 0 2 1 0 3 2 1 2 0 1 0 1 0 3 2 2 6 1 2 3 2 3 1 1 3 5 2 1 2\n",
      " 1 1 1 1 3 0 0 0 1 2 7 0 1 1 2 2 0 0 2 1 2 0 2 1 2 0 2 1 2 3 1 1 0 2 3 3 2\n",
      " 1 1 2 3 2 1 1 0 2 0 1 1 1 2 3 1 1 1 1 2 1 2 0 0 1 3 2 0 0 2 2 0 1 1 2 3 2\n",
      " 2 3 2 1 0 2 2 2 0 0 1 1 2 1 2 1 2 2 1 3 0 0 2 0 0 1 2 0 2 0 0 1 1 0 1 2 1\n",
      " 1 2 2 0 1 1 2 2 0 2 2 0 3 6 2 0 3 1 3 3 2 1 2 3 1 0 0 1 1 0 2 2 1 2 0 1 0\n",
      " 0 1 1 2 0 1 0 2 1 0 0 1 3 3 2 1 0 0 4 0 2 2 2 2 2 2 2 0 2 2 3 1 2 3 1 1 1\n",
      " 2 2 2 2 1 0 1 2 0 2 1 1 1 1 2 0 0 1 6 0 3 2 2 0 1 2 1 2 6 1 0 3 0 1 2 1 1\n",
      " 2 1 1 1 1 8 1 0 2 2 0 2 1 0 2 8 3 2 1 2 0 1 1 6 1 0 1 3 9 2 3 1 0 0 2 1 0\n",
      " 1 1 2 2 0 2 0 1 3 0 2 0 3 2 3 2 2 1 3 2 1 1 3 0 1 1 0 0 0 2 1 2 3 0 1 1 0\n",
      " 1 2 1 1 1 3]\n",
      "burn in | iter 2, number of clusters 17\n",
      "[ 0  0  2  1  0  0  1  1  1  4  1  1  2  2  1  0  0  2  2  2  0  1  1  1\n",
      "  1  1  2  2  0  2  1  2  4  1  1  1  0  2 10  2  3  3  0  7  1  1  2  0\n",
      "  2  1  2  1  2  2  0  1  2  0  3 11  2  0  2  2 13  0  1  1  2  0  0  0\n",
      "  0  1  2  0  0  1  1  2  0  2  1  0  2  7  1  1  0  0  1  0  2  2  1  0\n",
      "  2 16  0  2  1  1  2  2  0  3  0  0  1  2  0  2  1  1  0  1  1  1  0  0\n",
      "  1  4  0  0  2  1  2  0  1  2  1  0  1  2  2  3  0  1 11  1  3  1  1  3\n",
      "  2  2  1  2  1  2  1  1  1  2  0  0  1  2  2  8  1  1  2  2 12  0  2  0\n",
      "  1  1  1  1  2  0  1  1  1  3  1  1  1  2  1  3  2  0  1  2  3  9  2  1\n",
      "  0  2  2  1  1  2  2  1  1  1  1  1  2  1  1  0  0  1  3  2  0  5  2  2\n",
      "  0  1  1  2  3  2  3  3  0  1  5  2  2  2  0  0  1  1  2  1  0  1  2 15\n",
      "  1  3  0  0  2  1 12  1  2  0  2  0  0  1  1  0  2  2  1  2  2  2  1  1\n",
      "  1  2  2  0  2  2  0  3  2  2  0  3  1  3  3 13  1  2  3  1  0  0  2  1\n",
      "  1  2  2  1  2  0  2  1  0  1  1  2  0  1  0  1  1  0  0  1  1  3  2  1\n",
      "  0  1  4 12  3  2  2  2  2  0  2  0  2  2  3  1  2  3  1  1  1  2  2  2\n",
      "  0  1  0  2  2  0  2  1  1  2  1  2  0  5  0 10  8  3  2  2  0  2  2  1\n",
      "  2  6  1  0  3  0  0  2  1  1  2  1  2  1  1  0  1  0  0  1  1  5  1  1\n",
      "  2 14  2  2  1  0  2  1  1 10  1  0 10  2  2  2  3  2  0  0  0  1  0  1\n",
      "  1  2  2  0  2  1  1  2  0  2  1  1  1  3  2  2  1  3  0  1  1  3  0  1\n",
      "  1  0  0  0  2  1  2  3  0  1  1  0  0 12  1  2  1  4]\n",
      "burn in | iter 3, number of clusters 17\n",
      "[ 0  2  2  1  0  0  1  1  1  8  1  6  2  2  1  0  0  2  2  2 10  1  1  1\n",
      "  1  1  2  2  0  2  2  2  4  1  1  1  0  2  3  2  0  0  2 10  1  1  2  2\n",
      "  2  1  2  1  2  3  0  1  2  0  3  2  2  0  1  2 15  0  1  1  2  0  0  0\n",
      "  0  1  2  0  0  1  1  2  0  2  1  0  2  0  1  1  0  0  4  2  2  2  1  0\n",
      "  2  0  0  2  1  1  0  3  0  1  0 13  1  2  0  3  3  1  0  1  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  2  1  0  1  2  2  3  0  1  3  2  3  1  1  0\n",
      "  2  2  1  2  1  1  1  1  1  0  0  0  1  2  2  0  1  1  2  2  6  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  1  2  1  3  2  0  1  2  3 11  2  1\n",
      "  0  2  0  1  1  0  2 12  1  1  1  1  2  1  2  0  0  1  3  2  0  5  2  2\n",
      "  0  1  1  2  3  2 12  3  2  1  5  2  2  2  0  0  1  1  2  1  2  1  1  2\n",
      "  1  3  0  0  2  1  2  1  2  0  2  0  0  1  1  0  2  2  1  9  2  2  1  2\n",
      "  1  2  2  0  2  2  0 13  5  2  0  3  1  3  3 15  1  2 14  1  0  0  2  1\n",
      "  1  2  2  1  2  0  2  1  0  1  1  2  0  1  1  1  1  0  0  1  1  3  2  1\n",
      "  0  0  8  6  3  2  2  2  2  1  2  0  2  2  1  2  1  1  1  1  1  2  2  2\n",
      "  0  1  0  2  2  0  2  1  1 16  2  2  0  1  1 10  4  5  2  2  0  2  2  1\n",
      "  2  2  1  0 12  0  0  2  1  1  1  1  2  1  1  1  2  0  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  0  1  1 10  1  0  3  0  7  2  3  1  0  0  0  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0  2 12  3  1  3  2  2  1  0  0  1  1  3 12  1\n",
      "  1  0  2  0  2  1  2  9  0  1  1  0  0  4  1  2  1  8]\n",
      "burn in | iter 4, number of clusters 21\n",
      "[ 1  2  2  1  0  0  1  2  1  8  1 12  2  2  1  0  0  2  2  2  3  1  1  1\n",
      "  1  1  2  2  0  2  2  2  4  1  1  1  0  2 11  2  0  3  2 18  1  1  2  2\n",
      "  2  1  2  1  2  0  0  1  2  0  3  2  2  0  1  2 19  0  1  1  2  0  0  0\n",
      "  0  1  2  0  0  1  1  2  0  2  1  0  2  0  1  1  0  0  1  0  4  2  1  0\n",
      "  2  0  0  2  1 14  0  3  0  1  0 15  1  2  0  2  0  1  0 17  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  0  1  0  1  2  2  3  0  1  9  2  1 14  1  0\n",
      "  2  2  1  2  1  2  2  1  1  0  0  0  1  2  2  0  1  1  5  2  6  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  0  2  1  3  2  0  1  2  3 13  1  1\n",
      "  0  2  0  1  1  2  2 12  1  1  1  1  2  1  2  0  0  1  3  2  2  5  2  2\n",
      "  0  1  1  2  1  2  3  1  2  1  5  2  2  2  0  0  1  1  2  1  1  1  1  2\n",
      "  1 11  0  0  2  1 12  1  2  0  2  0  0  1  1  0  2  2  1  1  2  2  1  2\n",
      "  2  2  2  0  2  2  0 15  5  2  0  3  1  3  3  2  1  2  3  1  0  0  2  1\n",
      "  1  2  1  1  2  0  2  1  0  1  2  2  0  1  1  1  1  1  0  1  1  3  2  1\n",
      "  0  0  8 20  3  2  2  2  2  1  2  2  2  2  3  2  1  1  1  1  1  2  2  2\n",
      "  4  1  1  2  2  0  2  1  1  2  1  2  0  1  9 10  7  1  2  2  0  2  2  1\n",
      "  2  2  1  0  4  0  0  2  1  1  1  1  2  1  1  1  1  0  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  0  1  1  0 13  0 16  2 12  2  3  1  0  3  0  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0  2 12  1  2  3  2  2  1  0  0  1  1  3  0  1\n",
      "  1  0  2  0  2  1  2  3  0  1  1  0  0 12  1  2  1 14]\n",
      "train | iter 0, number of clusters 21, cluster indices [0, 1, 2, 3, 4, 10, 5, 12, 11, 15, 17, 14, 20, 8, 6, 7, 9, 16, 18, 19, 21]\n",
      "[ 1  2  2  1  0  0  1  1  1 14  1 12  2  2  1  0  0  2  2  2  2  1  1  1\n",
      "  1  1  2  2  0  2  1  2 14  1  1  1  0  2  0  2  0  3  2 10  1  1  2  1\n",
      "  1  1  2  1  2  3  0  1  2  1  3  2  2  0  1  2 15  0  1  1  2  0  0  0\n",
      "  0  1  2  0  0  1  1  2  0  2  1  0  2 16  1  1  0  0  1  2  4  2  1  0\n",
      "  2  0  0  2 12  9  0  3  0  1  0 18  1  2  0  2  2  1  0  1  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  2  1  0  1  2  2  3  0  1  3  2 17 14  1 15\n",
      "  2  2  1  2  1  2  2 12  1  0  0  0  1  2  3  4  1  1  2  2 12  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  0  2  1  3  2  0  1  2  3 11  2  1\n",
      "  0  2  0  1  1  0  5  1  1  1  1  1  2  1  2  0  0  1  3  2  0  0  2  2\n",
      "  0  1  1  2  1  2 12  3  2  0  5  2  2  2  0  0  1  1  2  1  5  1  1  2\n",
      "  1  3  0  0  2  1  8  1  2  0  2  0  0  1  1  0  2  2  1 19  2  2  1  2\n",
      "  2  2  2  0  2  2  0 12  5  2  0  3  1  1  3  0  1  5  6  1  0  0  2  1\n",
      "  1  2  1  1  2  0  2  1  0  8  2  2  0  1  1  1  1  0  0  1  1  3  2  1\n",
      "  0  0 14  2  3  2  2 14  2  1  2  0  2  2  1  2  1 17  1  1  1  2  3  2\n",
      "  7  1  1  2  2  0  2  1  1 21  1  2  0  1  2  3  2  1  2  2  0  2  2  1\n",
      "  2  2  1  0  3  0  0  2  1  1  1  1  2  1  1  1  1  0  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  0  1  1 10  1  0  3  2  1  2  3  1 15  6  7  1  2  1\n",
      "  1  2  1  0  2  1  1  1  0  2  1  1  1  3  2  2  1  3  0  1  1  3  0  1\n",
      "  1  0  5  0  2  1  2  3  0  1  1  0  1 20  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 1, number of clusters 21, cluster indices [0, 1, 2, 3, 5, 12, 15, 17, 14, 20, 8, 7, 18, 19, 13, 9, 4, 10, 11, 16, 21]\n",
      "[ 1  2  2  1  0  0  1  1  1 14  1  2  2  2  1  0  0  2  2  2  3  1  1  1\n",
      "  1  1  2  2  0  2  1  2 16  1  1  1  0  2  3  5  0  3  2  3  1  1  2  2\n",
      "  2  1  2  1  2  3 18  1  2  0  3  2  2  0  1  2  9  0  1  1  2  0  0  0\n",
      "  0  1  2  2  0  1 11  2  2  2  1  0  2 12  1  1  0  0  1  0  7  2  1  0\n",
      "  2  0  0  2  8 14  0  3  0  1  7 15  1  2  0  2  2  1  0  1  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  7  1  0  1  2  2  3  0  1  0  2 17 14  1  0\n",
      "  2  2  1  2  1  2  2  8  1  0  0  0  1  9  3  0  1  1  5  2  2  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  0  2  1  3  2  0  1  2  0  1  2  1\n",
      "  0  2  0  1  1  2  5  1  1  1  1  1  2  1  2  0  0  1  3  2  0  0 14  2\n",
      "  0  1  1  2  1  2  2  3  2  8  0  2  2  2  0  0  1  1  2  1  5  1  1  1\n",
      "  1  3  0  0  2  1 20  1  2  2  2  0  0  1  1  0  2  2  1  2  2  2  1  2\n",
      "  2  2  2  0  2  2  0  0  5  2  0  3  1  1  3  4  1  5 15  1  0  0  2  1\n",
      "  1  2  2  1  2  0  2  1  0 21  2  2  0  1  1  1  1  0  0  1  1  3  2  1\n",
      "  0  0 14  2  3  2  2 14  2  0  2  0  2  2  1  2  1 17  1  1  1  2  2  2\n",
      "  0  1  0  2  2  0  2  1  1  0  3  2  0  1  2  3  0  1  2  2  0  2  2  1\n",
      "  2  2 13  0  1  0  0  2  1  1  1  1  1  1  1  1  2  0  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  0  1  1 19  1  0  3  2 10  2  3  2  0  3  3  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0  2  1 14  2  3  2  2  1  3  0  1  1  3  0  1\n",
      "  1  0  2  0  2  1  2  2  0  1  1  0  0  0  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 2, number of clusters 22, cluster indices [0, 1, 2, 3, 5, 12, 15, 17, 14, 8, 7, 19, 13, 9, 4, 10, 6, 11, 18, 21, 22, 23]\n",
      "[ 1  2  2  1  0  0  1  5  1 14  1  2  2  2  7  0  0  2  2  2  3  1  1  1\n",
      "  1  1  2  2  0  2  1  2 14  1  1  1  0  2 19 23  0  3  2 12  1  1  2  2\n",
      "  2  1  2  1  2  3  9  1  2  1  3  2  2  0  1  2  4  1  1  1  2  0  0  2\n",
      "  0  1  2  2  0  1  2  2  0  2  1  0  2 18  1  1  0  0  1  0  7  2  1  0\n",
      "  2  0  0  2  6  1  0  2  0  1  7  4  1  1  0  2  2  1  0 17  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  2  1  0  1  2  2  3  0  1  2  2 17 14  1  0\n",
      "  2  2  1  2  1  2  2  6  1  0  0  0  1  2 18  0  1  1  2  2  2  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  0  2  1  3  2  2  1  2  2  2  2  1\n",
      "  0  2  2  1  1  2  5  1  1  1  1  1  2  1  2  0  0  1  3  2  0  0 21  2\n",
      "  0  1  1  2  1  2  2  3  2  0  5  2  2  2  0  0  1  1  2  1  5  1  1  2\n",
      "  1 10  0  0  2  1  1  1  2  0  2  0  0  1 17  0  2  2  1 11  2  2  0  2\n",
      "  2  2  2  0  2  2  0  9  2  2  0  3  1  1  3  0  1  5  2  1  1  0  2  1\n",
      "  1  2  1  1  2  0  2  1  0  8  2  2  0  1  1  1  1  0  0  1  1  3  2  1\n",
      "  0  0 14  2  3  2  2 14  2  0  2  0  2  2 17  2  1 17  1  1  1  1  2  2\n",
      "  0  1  0  2  2  0  2  1  1  2  1  2  0  1  0 23  4  1  2  2  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  1  1  1  1  2  1  1  1  2  0  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  2  1  1 12  1  0 19  3  8  0  3 22  0 15  7  1  2  1\n",
      " 13  2  2  0  2  1  1  7  0  2  1  1  2  3  2  2  1  0  0  1  1  3  0  1\n",
      "  1  0  2  0  2  1  2  2  0  1  1  0  0  1  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 3, number of clusters 19, cluster indices [0, 1, 2, 3, 5, 12, 17, 14, 8, 7, 19, 6, 22, 23, 16, 11, 4, 15, 10]\n",
      "[ 1  2  2  1  0  0  1  1  1 14  1  2  2  2  1  0  0  2  2  2 11  1  1  1\n",
      "  1  1  2  2  0  2  2  2  0  1  1  1  0  2  3  5  0  3  2 11  1  1  2  2\n",
      "  2  1  2  1  2 11  0  1  2  0  3  2  2  0  1  2  0  0  1  1  2  0  0  2\n",
      "  0  1  2  0  0  1 12  2  0  2  1  0  2 12  1  1  0  0  1  0 22  2  1  0\n",
      "  2  0  0  2  1 14  0  3  0  1  0  7  1  2  0  2  2  1  0 17  1  1  0  0\n",
      "  1  1  0  0  2  1  5  0  1  2  1  0  1  2  2  3  0  1  2  2 17 14  1 16\n",
      "  2  2  1  2  1  2  2  1  1  0  0  0  1  2 12  0  1  1  2  2  2  0  2  0\n",
      "  2  1  1  1  2  0  2  1  2  3  1  1  0  2  1  3  2  0  1  2  2  2  2  1\n",
      "  0  2  2  1  1  2  5  1  1  1  1  1  2  1  2  0  0  1  3  2  0  0 14  2\n",
      "  0  1  1  0  1  2  2  3  2  8  0  2  2  2  0  0  1  1  2  1  5  1  1  1\n",
      "  1 19  0  0  2  1  4  1  2  0  2  0  0  1  0  0  2  2  1  1  2  2  0  2\n",
      "  2  2  2  0  2  2  0 15  2  2  0  3  1  1  3  0  1  5 16  1  1  0  2  1\n",
      "  1  2  1  1  2  0  2  1  0 23  2  2  0  1  1  1  1  0  0  1  1 16  2  1\n",
      "  0  0 14  2  3  2  2 14  2  0  2  0  2  2  1  2 22 17  1  1  1  2  2  2\n",
      "  0  1  0  2  2  0  2  1  1  0  1  2  0  1 10 11  1  1  1  2  0 14  2  1\n",
      "  2  2  1  0  1  0  0  2  1  1  1  1  1  1  1  1  2  0  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  2  1  1  1  2  0  3  0  6  2  3 22 16 16  3  1  2  1\n",
      "  1  2  5  0  2  1  1  1  0  2  1  1  1  3  2  2  1  0  0  1  1  0  2  1\n",
      "  1  0  2  0  2  1  2  2  0  1  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAL0AAAC7CAYAAAAwjp8tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAADndJREFUeJztnU+sHdV9xz/fGAioaUQcDLIwrkFFEVApRL6iSN5QUipKopBFUgERygLJm0QiSqRAuvGz1AXZJNlkY4WoXuACIomKWDRFBFRVqhDPIWrjuNQOshLLFoYKFNpFkZNfF3ee31y/Gc+5987cuXPP9yNd3flz5pwz9/3eb75z5nd+o4jAmJz4UN8dMGbR2OhNdtjoTXbY6E122OhNdtjoTXbY6E12zGX0ku6V9Iakk5Ieb6tTxnSJZn04JWkb8F/APcBp4DXgwYj4VXvdM6Z95vH0dwAnI+LNiPgAeBq4v51uGdMdl81x7PXAb0vrp4E/v9QB11xzTezZs2e8cvRodaG9e7dsKhet2D1bGbYekFKHWV5OnTrFO++8o6Zy8xh9VeVbtJKk/cB+gN27d7O+vr6xo7rWjf0TdVxy92xl2HpASh1meRmNRknl5jH608ANpfVdwJmLC0XEIeBQ0ako7biwqIMlaysvr8XFReFIaf9DpeqeKv8T1dynrJXKHNhaxrF3eTCPpn8NuFnSjZKuAB4Anm+nW8Z0x8yePiLOS/oq8FNgG/DDiDjWWs+M6YiZhyxnYTQaxQVNT4Ic2ShZKlrX3cky6XW3xTR9tIzqhtFoxPr6euONrJ/Imuyw0ZvsmGf0Zj5Us1Jx7a+VAyVNMSnTllM/VJ5Hii6agparW0ns6U122OhNdvQnb9q49ibUUffgt+1L/8z1tdwRS5pm7OlNdvTn6esouWZREYZQU3ZaF+cbvnyxpzfZYaM32bEU8mYiynJtczEONBxXLlsTfTkRWVmL9U1O2NOb7LDRm+xYrLw5erRy4HxCXDQNpdSGHpSLNJeZ6EeDjGqdAQ8dbfy2Q852bU9vssNGb7JjKUZvppE0adUt+aV32ft3CZb+t03Ant5kh43eZMdi5c3evdV5bQ5Wy5fYSNNRvqQeqZE6dQ+n6pjiMl3u34U+DYxVOIe2aPT0kn4o6ZykX5a2bZf0oqQTxffHuu2mMe2RIm/+Hrj3om2PAy9FxM3AS8W6MYMgKQWIpD3ACxHxZ8X6G8BdEXFW0k7glYj4RFM9Iykas+WVM59NkTJj2jQhNU1OVd+0LDwFyIAfgs1C1ylArouIswDF97Uz1mPMwul89EbSfknrktbf7roxYxKYdfTmLUk7S/LmXF3BcgLXm666Ko7cdNOWMg8dq84GOM0VOS3zWfX2Jo7cdtuF5XJfy3U8dWt1mV7JQNLMwqye/nngy8Xyl4F/bKc7xnRPypDlPwD/BnxC0mlJjwBPAPdIOsH49TtPdNtNY9qjUd5ExIM1uz49bWPbb7uNhyoTuG4yzUOU+rLVM7F0cHO5adRqcneNXCnNynqo1H5dv9pRG4tPTrtqOAzBZIeN3mRHj6HF1ZfmaeJC6stWS50+Yk7aj3lZAUnT80Mze3qTHcsxiWRBtOF1J8f3Szeva6XN5WWzlZ6fH9jTm+yw0ZvsWHl5U5YxE9KkIu3HPPdXk+X7u3x3fY/oFCDGDBAbvcmOlZc3dclhJ0ISDjTkwS8xEal5sDokoG6UaGN7l88LulYd08iaZZ2Xa09vssNGb7Jj5eVNmbYvsct0yV5GlvX3sac32WGjN9mx1PKmKWXGwdIQzIHSu3pS8tNPlRO2/AqfteZJIVUjNhPHLjof/kVklhlkC/b0Jjts9CY7kjKctcVoNIr1igSui6JJ0tSlCElJHTJtChLTPq1lOJN0g6SXJR2XdEzSo8V2J3E1gyRF3pwHvhERtwB3Al+RdCtO4moGSkoKkLPARt7K9yUdB64H7gfuKoodBl4BHuukl3Mwa9LW1ljbGp8z2U7CmxBrqIptsZxqZqob2SJ78aeAV0lM4jqRy/JtZ7M0/ZNs9JI+AvwI+FpE/C71uIg4FBGjiBjt2LFjlj4a0ypJD6ckXc7Y4J+KiB8Xm5OTuC4LTaMz04YWl6mTFU3xJ2VJ0/gaIpY3XHdIpIzeCHgSOB4R3yntchJXM0hSPP0+4GHgPyT9otj2t4yTtj5bJHT9DfDFbrpoTLukjN78K3XZVmdI4rpoamc6rZW2K33mVJmUmVMTfWmYoZUiV+rKbLa5GCk0ZJnlMASTHVmFITTRffqMS9edciPbVPfEccNywHPT9YvWjBksNnqTHUs9iWTRdJ8+Y/ydMqZfljpNKUXG9WwNQxjyzWaX2NOb7LDRm+ywvOmBpHCHxvH4lLotaaqwpzfZYaM32WF50zFVcmQyZ35CupIGSdMFqzzyY09vssNGb7LD8qYD6iWLLvpOmxc7KS+qIzs3t7cjRdqRNHV97Rd7epMdNnqTHUshb1ZtpKBu3mvTuSXF2NTWMevLoGdPQdLMcv4t7elNdtjoTXY0yhtJVwL/Any4KP9cRByQdCPwNLAd+DnwcER8kNxyp5fVfplKriX8DrNKvpR+rNpvn0KKp/8/4O6I+CRwO3CvpDuBbwPfLXJZvgs80l03jWmPRqOPMf9TrF5efAK4G3iu2H4Y+HwnPTSmZVIznG0DjgJ/Cnwf+DXwXkScL4qcZpzUNZ0FXVb7GBmaNSNZL6NYUybcb+UF0G3MwJ+jjqQb2Yj4fUTcDuwC7gBuqSpW3TcncDXLxVTj9BHxnqRXGOepv1rSZYW33wWcqTnmEHAIxilAqsp06eHaqG/a/k1VvuSl6kqmRFnOfBVJec1KVTvTvpJl2jId1pGSy3KHpKuL5auAvwSOAy8DXyiKOZelGQwpnn4ncLjQ9R8Cno2IFyT9Cnha0t8BrzNO8mrM0pOSy/LfGb+I4eLtbzLW93Oz7KEH8/RvUTeyVTeYSXWkvH6l6aZxYGP9fiJrssNGb7JjKaIsV416WXHpiR61oQJTSR1VLrY1qrIRlRl1oz5dSp15RoxK2NOb7LDRm+wYnLwZ9oST7vpbJTtUeodtbcsJaUcmpEyxvXZ/XR1t/K1akk729CY7bPQmO4Yhb1RzmT7QdjPdTmxpJUKxhk4ng5RHTdaK9tZK+9eqR2+WVX7a05vssNGb7Ojv7YI9TyTIjil/K9VJyqZ3CPX4N/HbBY2pwUZvsqO/0ZueZ88sO20/2NHa5nJKbY2yt+6B1JTt9IE9vckOG73Jjv5Gb0y7FKMmKomKydT4pZcxJ5SZVTr2OaDm0RtjarDRm+xINnpJ2yS9LumFYv1GSa9KOiHpGUlXdNdNk0rE5kfShc/EDrT5UekzUaZEuUwVpf3B5mdZmcbTP8o4380GTuBqBkmS0UvaBXwG+EGxLpzAdemJiAufi/aUPh1Sd4WQqj/ztnP0aFLxVE//PeCbwB+K9Y8zbwJXY3oiJa3fZ4FzEVH+N6r6t3QCVzMIUsIQ9gGfk3QfcCXwUcaev7UErrkxz1h2UyLWqetOibgsj+tX9aM8L7em/YM6eGH5wBRzasvHra2tbdk/CykvZfhWROyKiD3AA8DPIuJLOIGrGSjzjNM/Bnxd0knGGt8JXM0gWGwYghQXghBWOEKyFy6EIWxSOfkjoY5LlVfFKEutDaVorRbjFhyGYEwNNnqTHYudRLJ3LzjKshsKaTCXQEiQF5VSpk6ilJYnR2/Sc7fMetylsKc32WGjN9nhSSRmNlJiZVoaMUptcwSsR3j0xpiLsdGb7BhGAlczN33niq8dhSnH7TT0cSK9SNU5jEZJfbGnN9lhozfZYXmTCW3nip9WLrXxYKnczjxyzZ7eZIeN3mTH8smb3HLOD/R8kyRF6dwObry3h3qpM1HnxrE1ue/neQ2TPb3JDhu9yY7lkzcDusQn0SRfhnq+dXEwNaHFEwokRdJ1+FvZ05vsWD5PP1Bq30G77J581hvplDmvdeWnib7sYG5tktFLOgW8D/weOB8RI0nbgWeAPcAp4G8i4t2pWjemB6aRN38REbdHxEZUz+PAS0UC15eKdWOWnnnkzf3AXcXyYeAVxrlwzJBoe6JH2/WlMGXy11RPH8A/SzoqaX+x7bqIOAtQfF87VcvG9ESqp98XEWckXQu8KOk/Uxso/kn2A+zevXuGLhrTLkmePiLOFN/ngJ8AdwBvSdoJUHyfqzn2UESMImK0Y8eOdnq9hNTngl8B6t5QskFKHvq6+prK1NXX1KdLkJKq+48k/fHGMvBXwC+B5xknbgUncDUDIkXeXAf8pBiHvgw4EhH/JOk14FlJjwC/Ab7YXTeNaY9Go4+IN4FPVmz/b+DTXXTKDIyaea6lwMr6zGszzpGlKj++58gaU42N3mTHQjOcSXob+F/gnYU12h/XsPrnuWzn+CcR0ThEuFCjB5C0XgplWFlyOM+hnqPljckOG73Jjj6M/lAPbfZBDuc5yHNcuKY3pm8sb0x2LNToJd0r6Q1JJyWtxKQTSTdIelnScUnHJD1abN8u6UVJJ4rvj/Xd13mRtE3S65JeKNZvlPRqcY7PSLqi7z6msDCjl7QN+D7w18CtwIOSbl1U+x1yHvhGRNwC3Al8pTivVZxZ9ihwvLT+beC7xTm+CzzSS6+mZJGe/g7gZES8GREfAE8znn01aCLibET8vFh+n7FRXM/43A4XxQ4Dn++nh+0gaRfwGeAHxbqAu4HniiKDOcdFGv31wG9L66eLbSuDpD3Ap4BXWb2ZZd8Dvgn8oVj/OPBeRJwv1gfz91yk0VdNZFyZoSNJHwF+BHwtIn7Xd3/aRNJngXMRcbS8uaLoIP6ei8x7cxq4obS+CzizwPY7Q9LljA3+qYj4cbH5LUk7I+LspWaWDYR9wOck3QdcCXyUsee/WtJlhbcfzN9zkZ7+NeDm4o7/CuABxrOvBk2hbZ8EjkfEd0q7VmZmWUR8KyJ2RcQexn+3n0XEl4CXgS8UxQZzjgsz+sIbfBX4KeObvWcj4tii2u+QfcDDwN2SflF87gOeAO6RdAK4p1hfNR4Dvi7pJGON/2TP/UnCT2RNdviJrMkOG73JDhu9yQ4bvckOG73JDhu9yQ4bvckOG73Jjv8H+4oHSfC52xQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 4, number of clusters 20, cluster indices [0, 1, 2, 3, 5, 12, 17, 14, 22, 23, 16, 11, 15, 9, 13, 18, 6, 7, 8, 19]\n",
      "[ 1  2  2  1  0  0  1 18  1 14  1  2  2  2  1  0  0  2  2  2  0  1  1  1\n",
      "  1  1  2  2  0  2  2  2  1  1  1  1  0  2 16  6 15  3  2 11  1  1  2  2\n",
      "  2  1  2  1  2 22  0  1  0  1  3  2  2  0  1  2  0  1  1  1  2  0  0  0\n",
      "  0 13  2  0  0  1  2  2  2  2  1  0  2 11  1  1  0  0  1  2  0  2  1  0\n",
      "  2  2  0  2  1  1  0  2  0  1  0 15  1  2  0  2  2  1  0 17  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  2  2  1  0  1  2  2  3  0  1  2  2 17 14  1  0\n",
      "  2  2  1  2  1  2 12  2  1  0  0  0  1  2 22  0  1  1  2  2  2  0  2  0\n",
      "  2  1  1  1  2  0  2  1  2  3  1  1  0  2  1  3  2  0  1  2  2  2  2  1\n",
      "  0  2  0  1  1  2  5  1  8  1  2  1  2  1  2  0  0  1  3  1  2  0  7  2\n",
      "  0  1  1  0 14  2  2  3  2 23  5  2  8  2  0  0  1  1  2  1  5  1  1  1\n",
      "  1  2  0  0  2  1  2  1  2  0  2  0  0  1 17  0  2  2  1  9  2  2  1  2\n",
      "  2  2  2  0  2  2  0 19  2  2  0  3  1  1  3  0  1  5  2  1  1  0  2  1\n",
      "  1  2  2  1  2  0  8  1  0  1  2  2  0  1  1  1  1  0  0  1  1  0  2  1\n",
      "  0  0 14  2  3  2  2  1  2  0  2  0  2  2  1  2 22 17  1  1  1  1  2  2\n",
      "  0  1  1  2  2  0  2  1  1 18  1  2  0  1  1 11  1  1 22  2  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  1  1  1  1  2  2  1  1  2  0  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  2  1  1 16  1  0  2  2  2  0  3 22  0  1 22  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0  2  1 14  1  3  2  2  1  0  0  1  1  3  2  1\n",
      "  8  0  2  0  2  1  2  2  0  1  1  0  0  1  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 5, number of clusters 18, cluster indices [0, 1, 2, 3, 5, 17, 14, 22, 11, 15, 13, 18, 8, 4, 6, 10, 7, 9]\n",
      "[ 1  2  2  8  0  0  1  1  1 14  1  2  2  2  1 22  0  2  2  2  3  1  1  1\n",
      "  1  1  2  2  0  2  1  2  1  1  1  1  0  2  3  5  0  3  2 22 22  1  2  2\n",
      "  2  1  2  1  2 22  0  8  0  1  3  2  2  0  2  2  0  1  1  1  2  0  0  0\n",
      "  0  1  2  2  0  1  1  2  0  2  1  0  2  2  1  1  0  0  1  2 22  2  1  0\n",
      "  2  2  0  2  4  1  2  2  0  1  0  9  1  2  0  2  2  1  0  0  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  2 17  1  0  1  2  2  3  0  1  2  2 17  1  1  0\n",
      "  2  2  1  2  1  2  2  2  1 22  0  0  1  2 22  2  1  1  2  8  2  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1 18  0  2  1  3 13  0  1  2  2  8  2 14\n",
      "  0  2  0  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  3  1  0  0  2  2\n",
      "  0  1  1  0 14  2  2  3  2  1  2  2  8  2  0  0  1  1  2  1  5  3  1  2\n",
      "  1  8  0  0  2  1  2  1  2  0  2  1  0  1 10  0  2  2  1  7 11  2  0  2\n",
      "  0  2  2  0  2  2  0 15  2  2  0  3  1  1  2  0  1  2  0  1  1  0  2  1\n",
      "  1  2  2  1  2  0  2  1  0  2 13  2  0  1  1  1  1  0  0  1  1  6  2  1\n",
      "  0  0 14  2  3  2  2  1 11  0  2  0  2  2  1  2 22 17  1  1  1  2  2  2\n",
      "  0  1  1  2  2  0  2  1  1  0  1  2  0  1  2  3  1  1  2  2  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  1  1  1  1  2  2  1  1  2 15  2  1  1  2  1  1\n",
      "  2  1  2  2  1  0  0  1  1  0  1  0  2  0  2  0  2 22  0  6 22  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0  2  1 14  2  3  2  2  1  0  0  1  1  3  2  1\n",
      "  8  0  2  0  2  1  2  8  0  1  1  0  0  1  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAL0AAAC7CAYAAAAwjp8tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAADkxJREFUeJztnU/sHVUVxz/HAmJEhUohlVJLIiHgQkxfkKQbrGIQCbgQAiWEBbEuIIFAAuimr4kmuBE2xNiIsQsqENCIhqgEIcTEEN4PiFIqUkkDtQ2FCAFdSCrHxZvHb6a/md/ceW/mzcy730/SdP7cuXPnzfmd+c65954xd0eImPhI2w0QYt7I6EV0yOhFdMjoRXTI6EV0yOhFdMjoRXTMZPRmdomZvWxm+83szroaJUST2LSdU2a2Bvg7cDFwEHgWuMbdX6qveULUzyye/gJgv7u/6u7vAw8AV9TTLCGa47gZjj0DeD21fhD40moHnHrqqb5p06ZkbWl5x79ShdZuXnHc0qHlsps/s3I/wFKqus35RTJl0kzKh9QhusuBAwd46623rKzcLEafV/kKrWRm24HtABs3bmQ0Gq08fE/qgG0jjsV2Lpcd7Vi5f3ye5eVRfpFMmTST8iF1iO4yGAyCys1i9AeBM1PrG4BDxxZy913ArqRRqT+K1OK1lr+8XMfyStoyM+8j6ePy31OcomMLN4kFZBZN/yxwtpmdZWYnAFcDj9bTLCGaY2pP7+5Hzewm4PfAGuBn7r63tpYJ0RCzyBvc/THgsemOXl1qZEpmihaVDZA6DeqXQtWVLTUp0Vg7RDnqkRXRIaMX0TGTvJkF27m87GentMG28KhKOpSZlT1tyIfy6FHe9sw17Ji93TtTP+wO3zFzfYuIPL2IDhm9iI6pB5xNw2Aw8NGcuzqLemHVEbV4DAYDRqNR6TAEeXoRHa29yBaRfrFjOHbHQS+yFV8Cw+LqYhGRpxfRIaMX0dENeZPSGhmlUSI7fJiqYpgfs/fcEdAragooIxYFeXoRHTJ6ER1zlTdLh5ay0ZkJw+XFsihM0dCD7FHloZlMPauesX76HDmaDHPo8xAHeXoRHTJ6ER2diN5UkTSBNU7fmDnQN0mTps+yZoI8vYgOGb2Ijm6MstyTTgGS2p7XtqLQR5MhkT6HWyYswjWUUNsoSzP7mZkdMbMXU9vWmtnjZvZK8v8pszZYiHkRIm9+DlxyzLY7gSfc/WzgiWRdiF5QGr1x96fNbNMxm68ALkqWdwNPAXeU1VXYOZU934fLkydydvprQWcTRR1VqTLpUw9XH5YclnakGrnX0yDFc4jjZtoX2dPd/TBA8v9p9TVJiGZpPE6fTuDKp5o+mxDlTGv0b5jZenc/bGbrgSNFBdMJXM3M0+NsUmUKjg1vUOHsqsJgT2qlpL8lJGAU0pZ5U0dKkUVkWnnzKHB9snw98Ot6miNE84SELH8B/Bk4x8wOmtkNwF3AxWb2CuPP79zVbDOFqI+Q6M01Bbu+UvVkmzdvzv8oQ5oqnSgBueozNVSIyIRIlMJ89wXtqkX2RNDJ1DQahiCiQ0YvoqPFocUFj+Y6QjZFuerbkAN1y5EFkDRm7XaaydOL6OjEJJImSQ9PqMPpZ2PzBXVPV3U0tD0kQp5eRIeMXkTHwsubkGEDefurPoGz5dt7fDcexp/3UNEGkKcX0SGjF9Gx8PKmMDls5rtoq+fBT5Md7VDwcbWyYZkNSoPGVUeVE3R0yIQ8vYgOGb2IjsWXN2nqfsQW1Xd/vafpLR2SNGnk6UV0yOhFdHRa3pQGOwImkRSmDKmQEzY9WSSTaqSoXdvyIzmTY9t+6Hc0qDI35OlFdMjoRXR0Wt6UPnqDJpEsUyZpCsfpZHbklylMQRIih+ZMV9rRFiHZEM40syfNbJ+Z7TWzm5PtSuIqekmIvDkK3Obu5wIXAjea2XkoiavoKSEpQA4Dk7yV75nZPuAMpkziOm+qZCGrEtGZhSqJZFevZ+Vc09gjMyFUepFNshd/EXiGwCSuZrbdzEZmNnrzzTdna60QNRBs9GZ2EvAIcIu7vxt6nLvvcveBuw/WrVs3TRuFqJWg6I2ZHc/Y4O93918mm4OTuLZJURayPBlQeWhxiiJZUVZnRtKEaLGW02csAiHRGwPuA/a5+49Su5TEVfSSEE+/BbgO+KuZvZBs+x7jpK0PJQldXwOubKaJQtRLSPTmTxRmW62exHXuFEiaNNNOaCqM9pTIlOI+tTr0Vfp6G5RCPQ4TaRiCiI5OD0Oom6kzf1eur8Qbh7wNh51oVRp90e2Zd08jTy+iQ0YvoiMqeVPGvNJnFMqoomGeIV96S7YPbWdq/7DgRHEjTy+iQ0YvokPypgWClEZAhCcvv9oOT38Ut+QDuZEiTy+iQ0YvokPypmnyOpwCvjlbWgcNpxLp8TCDMuTpRXTI6EV0SN40wZ6yzGszDPIpzepWkxSpQ9J0VCLJ04vokNGL6OiGvOnoY3BqChK4Tj22OaiO6X432zl9CpJSOnov5elFdMjoRXSUyhszOxF4GvhoUv5hd99hZmcBDwBrgeeA69z9/dATZx6rHX0MTk1h9GYlQb/DlL9PiHSpXdL0gBBP/19gq7t/ATgfuMTMLgR+CNyd5LJ8G7ihuWYKUR+lRu9j/p2sHp/8c2Ar8HCyfTfwzUZaKETNhGY4WwMsAZ8D7gX+Abzj7keTIgcZJ3UNZm6P1TYiQ3VHb6bEh6mVglHGaQmUObbo/tTwAehaIkYz/FZBL7Lu/j93Px/YAFwAnJtXLL9tSuAqukWlOL27v2NmTzHOU3+ymR2XePsNwKGCY3YBuwAGg0H+H0bXY8VVvUqF8pnrLZrzEZJDfMqnSPr8RV4/9zzpySzD1O6A+1fLPZ7hvobkslxnZicnyx8DvgrsA54EvpUUUy5L0RtCPP16YHei6z8CPOTuvzWzl4AHzOz7wPOMk7wK0XlCcln+hfGHGI7d/ipjfT8znY8VzyKR5vQiO5Emmd8yoI4QSVMmPzt//45BPbIiOmT0Ijq6Mcpy0SiUFSUTPWYYhjApsTOV4Ww4HC7vrymqMql/h+dHfZqUOpX7FAqQpxfRIaMX0WHzHOE4GAx8NBrNVsmiTTipi7zhAbP8ViETWsrqnvO9GgwGjEaj0nCUPL2IDhm9iI5eRG8y0YFGT9Tw47iGEYqFNCgf8qImXrC/audYG8jTi+iQ0YvoaE3eVOnQKBx+W7ccafpxPO/HfV2RnEyVK4cWZ+5PD6Jr8vQiOmT0Ijpakzdtz57pPG3LhPTsqpL9fUOeXkSHjF5ERy86p6JkyplTDFOyJB1UyRTOP00dQ4SNElnUAeTpRXTI6EV0BMubJBvCCPinu182awJX0QzZ/qj85LDpwFCIpCkrkxmbM8y0prS9bVDF09/MON/NBCVwFb0kNJflBuAbwA+AW23sQrYC25Iiuxn/jf+4gTaKKSmeIBSQyawGip4Qdc11XVFfbo69lYR6+nuA24EPkvVPM2MCVyHaIiSt32XAEXdfSm/OKaoErqIXhMibLcDlZnYpcCLwScaev7YErrEx0wiDkkSsYXWnj6t4/px2FL0kV+4DyDl4z+c//+Gma696qbyOAEI+yvBdd9/g7puAq4E/uvu1KIGr6CmzxOnvYPxSu5+xxlcCV9EL5poCxD5jznfGy31L+tl1JvIhndWsaoazEAmSzqA2YUeBRqqjD6AKSgEiRAEyehEd/ctwJjpHOsKybe/e3DJFMqbs2JC6J0jeCFGAjF5Eh+SNmIqQz/bUFTEKPudPwA+55I0QxyKjF9GhObKxUHNKkcodSQEfby5row9Zdf/gN4OgpsjTi+iQ0YvokLyJhbqjdFXlUt0JdmeQa/L0Ijpk9CI6Oidv5vUh3q7Q2+sNkBSVJ4DnfJC5cEL5MHVYaUuyyNOL6JDRi+jQ2JuG6a18KWGW3DVN/SYaWixEAZ17ke0rRXkju+7dp/W6IXNei8pXeRo0Mbc2NK3fAeA94H/AUXcfmNla4EFgE3AAuMrd3650diFaoIq8+bK7n+/uk1E9dwJPJAlcn0jWheg8s8ibK4CLkuXdwFOMc+GIHlH3i2cbL7JNJXB14A9mtmRm25Ntp7v7YYDk/9OqNFSItgj19Fvc/ZCZnQY8bmZ/Cz1B8keyHWDjxo1TNFGIeqkcpzezIfBv4NvARe5+2MzWA0+5+zmrHRtjnD4GQvLQ1xGFKT1PXXNkzezjZvaJyTLwNeBF4FHGiVtBCVxFjwiRN6cDv0ri0McBe9z9d2b2LPCQmd0AvAZc2VwzhagPDUMQ9WIF6iLEzsomhpTUrWEIQhQgoxfRMd/89GZvAv8B3prbSdvjVBb/Ort2jZ9193VlheZq9ABmNkoNZVhYYrjOvl6j5I2IDhm9iI42jH5XC+dsgxius5fXOHdNL0TbSN6I6Jir0ZvZJWb2spntN7OFmHRiZmea2ZNmts/M9prZzcn2tWb2uJm9kvx/StttnRUzW2Nmz5vZb5P1s8zsmeQaHzSzE9puYwhzM3ozWwPcC3wdOA+4xszOm9f5G+QocJu7nwtcCNyYXNciziy7GdiXWv8hcHdyjW8DN7TSqorM09NfAOx391fd/X3gAcazr3qNux929+eS5fcYG8UZjK9td1JsN/DNdlpYD2a2AfgG8NNk3YCtwMNJkd5c4zyN/gzg9dT6wWTbwmBmm4AvAs+weDPL7gFuBz5I1j8NvOPuR5P13tzPeRp93ui3hQkdmdlJwCPALe7+btvtqRMzuww44u5L6c05RXtxP+eZ9+YgcGZqfQPBU3m7jZkdz9jg73f3Xyab3zCz9amZZUfaa+HMbAEuN7NLgROBTzL2/Ceb2XGJt+/N/Zynp38WODt54z8BuJrx7Ktek2jb+4B97v6j1K6FmVnm7t919w3uvonxffuju18LPAl8KynWm2ucm9En3uAm4PeMX/YecvfVv3veD7YA1wFbzeyF5N+lwF3AxWb2CnBxsr5o3AHcamb7GWv8+1puTxDqkRXRoR5ZER0yehEdMnoRHTJ6ER0yehEdMnoRHTJ6ER0yehEd/wdTFMfpS5KScAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 6, number of clusters 21, cluster indices [0, 1, 2, 3, 5, 17, 14, 22, 11, 18, 8, 6, 10, 9, 12, 16, 4, 7, 15, 13, 19]\n",
      "[ 1 12  2  8  0  0  1 18  1 14  1  2  2  2  1  8  0  2  2  2 12  1  1  1\n",
      "  1  1  2  2  0  2  2  2  1  0  1  1  0  2  2  7  0  3  0  3  0  1  2  1\n",
      "  2  1  2  1  2 22  0  8  0  0  3  2  2  0  2  0  0  1 13  1  2 16  0  0\n",
      "  0  1  2  1  0  1 15  2  0  2  1  0  2 22  1  1  0  0  1  0 22  2  1  0\n",
      "  2  2  0  2 16  1  0  2  0  1  0  1  1  2  0  2  2  1  0 10  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  2  2  1  0  1  2  2  3  0  1  2  2 17  1  1  0\n",
      "  2  2 15  2  1  2  2  2  1  0  0  0  1  2 22  1  1  1  2  8  2  0  2  0\n",
      "  2  0 11  1  2  0  2  1  2  3  1  1  0  2  1  3  2  0  1  0  2  1  2  1\n",
      "  0  2  0  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  3  2  0  0  2  2\n",
      "  0  1  1  2 14  2  2  3  2  0  5  2  8 11  0  0  1  1  2  1  5  0 11  2\n",
      "  1  2  0  0  2  1  2  1  2  0  2  0  0  1  4  0  2  2  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  9  2  2  0  3  1  1  2  0  4  7  6  1  1  0  2  1\n",
      "  1  2 11  1  1  0  2  1  0 22  8  2  2  1  1  1  1  0  0  1  1  0  2  1\n",
      "  0  0 14  2  3  2  2  2 11  0  2  0  2  2  1  2  1 17  1  1  1  1  2  2\n",
      "  0  1  0  2  2  0  2  4  1  0  1  2  0  1 11  3  1  0  1  2  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  1  1 11  1  1  2 18  1  2  9  2  1  1  2  1  1\n",
      "  2  1 12  2  1  0  0  3  1  0  1  0  3  0  2  2  2  1  0 19 22  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0  2  1 14  1  3  2  2  1  0  0  4  1  0  2  1\n",
      "  8  0  5  2  2  1  2  2  0  1  1  0  0  1  1 18  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 7, number of clusters 20, cluster indices [0, 1, 2, 3, 5, 17, 14, 22, 11, 18, 8, 6, 10, 12, 4, 7, 13, 19, 16, 15]\n",
      "[ 1 12  2  1  0  0  1 18  1 14  1  2  2  2  1 19  0  2  2  2  0  1  1  1\n",
      "  1  1  2  2  0  2  1  2  1  0  1  1  0  2  2  5  0  3 22  6  0  1  2  1\n",
      "  2  1  0  1  2 22  0  8  0  0  3  2  2  0  2  0  0  1  0  1  2 19  0  0\n",
      "  0  1  2  1  0  1 16  2  0  2  1  0  2  6  1  1  0  0  1  0 22  2  1  0\n",
      "  2  2  0  2  2  1  2  2  0  1  0  1  1  1  2  2  2  1  0 17  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  2  1  0  1 16  2  3  0  1  2  2  1  1  1  0\n",
      "  2  2  1  2  1  2  2  2  1  0  0  0  1  2 22  1  1  1  8  8  2  0  2  0\n",
      "  2  0 11  1  2  2  2  1  2  3  1 18 10  2  1  2  2  0  1  0  2 14  2  1\n",
      "  2  2  2  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  3  2  0  0  2  2\n",
      "  0  1  1  0 14  2  2  0  2 15  0  2  8  7  0  0  1  1  2  1  5  0 11  1\n",
      "  1  2  0  0  2  1  2  1  2  0  2  0  0  1 17  0  2  2  1  4 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1  2  2  0  3  1  1  2  0  1  5 12  1  1  0  2 19\n",
      "  1  2  1  1  1  0  2  1  0  2 15  2  0  8  1  1  1  0  0  1  1  0  8  1\n",
      "  0  0 14  2  3  2  2  2  2  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  0  1  0  2  2  0  2  1  1  0  1  2  0  1  0  3  1  1  2  2  0 12  2  1\n",
      "  2  2  1  0  1  0  0  2  0  1  1  1  1  2 18  1  2  1  2  1  1  2  1  1\n",
      "  2  1 12  2  1  0  2  3  1  0  1  0  2  0  2  0  2  1  0  6 22  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0 12  1 14  1  3  2 22  1  0  0  1  1  3  2 13\n",
      "  8  0  2  1  2  1  2  2  0  1  1  0  0  1  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 8, number of clusters 20, cluster indices [0, 1, 2, 3, 5, 17, 14, 22, 11, 8, 6, 12, 19, 9, 7, 10, 20, 13, 4, 15]\n",
      "[ 1 12  2  1  0  0  1  2  1 14  1  2  2  2  1 19  0  2  2  2  3  1  1  1\n",
      "  1  1  2  2  0  2  2  2  1  0  1  3  0  2  2 19  0  3  2  1  0  1  2  1\n",
      "  2  1  2  1  2  3  0  8  0  0  3  2  2  0  1  0  0  1  0  1  2 20  0  0\n",
      "  0  1  2  1  0  1 10  2  0  2  1  0  2  0  1  1  0  0  1  2 22  2  1  0\n",
      "  2  2  0  2  9 14  0  2  0  1  0  1  1  2  2  2  2  1  0 15  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  2  1  0  1  6  2  3  0  1  2  2  1 14  1  0\n",
      "  2  2  1  2  1  2  2  2  1  0  0  0  1  2 22  1  1  1  2  8  2  0  2  0\n",
      "  2  0 11  1  2  0  2  1  2  3  1  1 17  2  1  3  2  0  1  0  2  1  2  1\n",
      "  0  2  0  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  2  0  2  0  2  2\n",
      "  0  1  1  0 14  0  2  0  2  1  0  2  8  2  0  0  1  1  2  1  5  0 11  1\n",
      "  1  2  0  0  2  1  2  1  2  0  2  0  0  1  7  0  2  2  1 20 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1  2  2  0  3  1  1  2  0  1  8  8  1  1  0  2  1\n",
      "  1  2  1  1  1  0  2  1  0 13  2  2  0  1  1  1  1  5  0  1  1  0  8  1\n",
      "  5  0 14  2  3  2  2 14 11  8  2  0  2  2  1  2 22  1  1  1  1  1  2  2\n",
      "  0  1  0  2  2  0  2  1  1  0  1  2  0  1 11  3  1  0  1  2  0  0  2  1\n",
      "  2  2  1  0  1  0  0  2  0  1 11  1  1  1  1  0  2  1  2  1  1  2  1  1\n",
      "  2  0 12  2  1  0  0  3  1  0  1  0  2  0  2  0  2  1  0  8 22  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0  4  1 14  1  2  2  1  1  0  0  1  1  3  2  1\n",
      "  8  0  0  1  2  1  2  2  0  1  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 9, number of clusters 18, cluster indices [0, 1, 2, 3, 5, 14, 11, 8, 12, 7, 20, 13, 4, 16, 18, 6, 9, 10]\n",
      "[ 1 12  2  1  0  0  1  1  1 14  1  2  2  2  1 18  0  2  2  2 12  1  1  1\n",
      "  1  1  2  2  0  2  1  2  1  0  1  1  2  2  2  5  0  3  2  3  0  1  2  1\n",
      "  2  1  2  1  2  3  0  8  0  0  2  2  2  0  2  0  0  1  0  1  2  7  0  0\n",
      "  2  1  2  1  0  1  6  2  0  2  1  0  2 11  1  1  0  0  1  2 18  2  1  0\n",
      "  2  2  0  2 20  9  0  2  0  1  0  1  1  1  2  2  2  1  0  7  1  1  0  0\n",
      "  1  1  0  0  2  1  2  0  1  2  1  0  1  0  2  3  0  1  2  2  1 14  1  0\n",
      "  2  2  1  2  1  2  2  2  0  0  0  0  1  2 11  1  1  1  2  8  2  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  7  2  1  2  2  0  1  0  2  5  2 14\n",
      "  2  2  0  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  2  1  0  0 14  2\n",
      "  0  1  1  0 14  0  2  3  2 13  0  2  8  2  0  0  1  1  2  1  5  0  1  1\n",
      "  1  2  0  0  2  1  2  1  2  0  2  0  0  0  1  0  2  2  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1  2  2  0  3  1  1  2  0  1  8 12  1  1  2  2  1\n",
      "  1  2  1  1  1  0  2  1  0  1  2  2  0 10  5  1  1  5  0  1  1  0  4  1\n",
      "  8  0 14  2  3  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  2  2  2\n",
      "  0  1  0  2  2  0  2  1  1  0  1  2  0  1  1  3  1  0  2  2  0 12  2  1\n",
      "  2  2  1  0  1  0  0  2  0  1  1  1  2  1 14 11  2  1  2  1  1  2  1  1\n",
      "  2  0 12  2  1  0  2  3  1  0  1  0  2  0  2  0  2  1  0  0 16  1  2  1\n",
      "  1  2  2  0  2  1  1  1  0 10  1 14  1  2  2 11  1  0  0  1  1  0  2  1\n",
      "  8 12  2  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 10, number of clusters 20, cluster indices [0, 1, 2, 3, 5, 14, 11, 8, 12, 7, 13, 16, 18, 9, 10, 4, 6, 15, 17, 19]\n",
      "[ 1 12  2  1  0  0  1 15  1 14  1  2  2  2  1  1  0  2  2  2  3  1 18  1\n",
      "  1  1  2  2  0  2  2  2  1  0  1  1  0  2  2  3  3  3  2  3  0  1  2  1\n",
      "  2  1  2  0  2  3  0  8  0  0  2  2  2  0  1  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  0  1  4  2  0  2  1  0  2  3  1  1  0  0  1  2  1  2  1  0\n",
      "  2  2  0  2  7 14  0  2  0  1  0  1  1  1  0  2  2  1  0  1  1  1  0  0\n",
      "  1  1  0  0  2  1 10  0  2  2  1  0  1  5  2  3  0  1  2  2  1 14  1  0\n",
      "  2  2  1  2  1  2  2  2  1  0  0  0  1  2  3  1  1  1  8  8  2  0  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  1  2  0  2  2  0  1  0  2 19  2  1\n",
      "  0  2  0  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  2  1  0  0  9  2\n",
      "  0  1  1  0 14  2  2  3  2  1  0  2  8  2  0  0  1  1  2  1  5  0  1 16\n",
      " 10  5  0  1  2  1  2  1  2  0  2  0  0  0  1  0  2  2  1  6 11  2  7  2\n",
      "  0  2  2  0  2  2  0  1  2  2  0  3  1  1  2  0  1  8 17  1  1  0  2  1\n",
      "  1  2  2  1  1  0  7  1  0 13  2  2  0  8  0  1  1  5  0  1  1  0 10  1\n",
      "  3  0 14  2  3  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  0  2  2  0  2  1  1  0  1  2  0  1  2  3  1  0  2  2  0  2  2  1\n",
      " 16  2  1  0  1  0  0  2  0  1  1  1  1  2  1  1  2  1  2  1  1  2  1  1\n",
      "  2  0 12  2  1  0  2  0  1  0  1  0  2  0  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2  0  0  2  1  1  1  0 12  1 14  1  2  2  2  1  0  0  1  1  3  2  1\n",
      "  8  0  0  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 11, number of clusters 16, cluster indices [0, 1, 2, 3, 5, 14, 11, 8, 12, 16, 4, 17, 15, 20, 9, 6]\n",
      "[ 1 12  2  1  0  0  1  1  1 14  1  2  2  2  1  1  0  2  2  2  3  1  0  1\n",
      "  1  1  2  2  0  2  2  2  1  0  1  1  0  2  6  5  0  3  2  3  0  1  2  2\n",
      "  2  1  2  0  2  3  0  8  0  0  2  2  2  0  2  2  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  0  1 20  2  0  2  1  0  2 11  1  1  0  0  1  2  1  2  1  0\n",
      "  2  2  0  2  1  1  0  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  4\n",
      "  1  1  0  0  2  1  8  0  2  2  1  0  1  9  2  3  0  1  2  2  1 14  1  0\n",
      "  2  2  1  2  1  2  2  2  1  0  0  0  1  2 11  1  1  1 15  8  2  1  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  1  2  1  2  2  0  1  0  2  9  2 14\n",
      "  0  2  0  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  2  1  0  0 14  2\n",
      "  0  1  1  0 14  2  2  3  2 12  0  2  8  2  0  0  1  1  2  1  5  0  1  1\n",
      "  0  2  0  1  2  1  2  1  2  0  2  0  0  0  1  0  2  2  1  5 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  3  1  1  2  0  1  8  0  1  1  0  2  1\n",
      "  1  2  2  1  1  0  8  1  0  2  0  2  0  0  0  1  1  0  0  1  1  0  8  1\n",
      "  3  0 14  2  3  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  0  1  0  2  2  0  2  1  1  0  1  2  0  0 11  3  1  0  1  2  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  0  1  1  1  1  2  1 16  2  1  2  1  1  0  1  1\n",
      "  2  0 12  2  1  0  0  3  1  0  1  0  2  3  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2  0  0  2  1  1  1 12 17  1 14  1  2  2  2  1  0  0  1  1  3  2  1\n",
      "  8  0  0  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 12, number of clusters 17, cluster indices [0, 1, 2, 3, 5, 14, 11, 8, 12, 17, 15, 9, 6, 7, 10, 4, 13]\n",
      "[ 0 12  2  8  0  0  1  2  1 14  1  2  2  2  1  1  0  2  2  2  3 12  1  1\n",
      "  1  1  2  2  0  2  2  2  1  0  1  1  0  2  7  8  0  3  2  3  0  1  2  2\n",
      "  2  1  2  0  2  3  0  8  0  0  2  2  2  0  2  2  0  2  0  1  2  1  0  0\n",
      "  0  1  2  2  2  1  2  2  0  2  1  0  2 11  1  1 12  0  1  2  1  2  1  0\n",
      "  2  2  0  2  2 14  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  6\n",
      "  1  1  0  0  2  1 15  0  2  2  1  0  1 10  2  3  0  1  2  2  1 14  1  0\n",
      "  2  2  1  2  1  2  2  2  0  0  0  0  1  2 13  2  1  1 15  8  2  1  2  0\n",
      "  2  0  1  1  2 14  2  1  2  3  1  1  1  2  0  2  2  0  1  0  2 10  2  1\n",
      "  0  2  0  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  2  1  2  0  2  2\n",
      "  0  1  1  0 14  0  2  0  2 12  0  2  8 11  0  0  1  1  2  1  5  0 11  2\n",
      "  0  9 12  1  2  1  2  1  2  0  2  0  0  0  1  0  2  2  1  2 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  3  1  1  2  0  1  8 17  1  2  0  2  1\n",
      "  1  2 11  1  1  0  4  1  0 12 12  2  0  0  0  1  1  5  0  1  1  0 12  1\n",
      "  5  0 14  2  3  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  2  2  2\n",
      "  1  1  0  2  2  0  2  1 12  0  1  2  0  0 11  3  1  0  2  2  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  0  1  1  1  2  2  1 17  2  1  2  1  1  0  1  1\n",
      "  2  0 12  2  1  0  0  3  1  0  1  0  2  3  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1 12 12  1 14  1  2  2  2  1  3  0  1  1  3  2  1\n",
      "  8  6  0  1  2  1  2  2  0  0  1 11  0  2  1  2  2 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 13, number of clusters 17, cluster indices [0, 1, 2, 3, 5, 14, 11, 8, 12, 17, 15, 6, 16, 18, 7, 4, 10]\n",
      "[12  4  2 16  0  0  1  2  1 14  1  2  2  2  1  1  0  2  2  2  6 12 11  1\n",
      "  1  1  2  2  0  2  2  2  1  0  1  1  0  0 18  8  3  3  2  3  0  1  2  1\n",
      "  2  1  0  0  2  3  0  8  0  0  2  2  2  0  2  0  0  2  0  1  2  1  0  2\n",
      "  0  1  2  1  5  1  3  2  0  2  1  0  0 11  1 10  0  0  1  2  1  2  1  0\n",
      "  2  2  0  2  2  7  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0 16\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1 17 15  3  0  6  2  2  1 14  1  0\n",
      "  2  2  1  2  1  2  2  2  0  0  0  0  1  2 11  1  1 12 15  8  2  1  2  0\n",
      "  2  0  1  1  2  1  2  1  2  3  1  1  1  2  0  2  2  0  1  0  2 12  2 14\n",
      "  0  2  2  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  2  2  0  0 14  2\n",
      "  0  1  1  0 14  2  2  3  2 12  0  2  8  2  0  0  1  1  2  1  5  0  1  2\n",
      "  0  2  0  1  2  1  2  1  2  0  2  0  0  0  1  0  2  8  1  1 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  3  1  1  2  0  1  8  6  1  1  0  2  1\n",
      "  1  2 17  1  2  0  8  1  0 17  0  2 18  0  0  1  1  5  0  1  1  0 15  1\n",
      "  8  0 14  2  3  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1 12  0  2  2  0  2  1 12  0  1  2  0  0 11  3  1 12  2  2  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  0  1  1  1  2  2  1 11  2  1  2  1  1 15  1  1\n",
      "  2  0  6  2  1  0  0  3  1  0 12  0  2  2  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1 12  6  1 14  1  2  2 11  1  2  0  1  1  3  2  1\n",
      "  8  0  0  1  2  1  2  2  0  0  1  0  0  1  1  2  2 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 14, number of clusters 20, cluster indices [0, 1, 2, 3, 5, 14, 11, 8, 12, 17, 15, 6, 16, 7, 4, 9, 10, 13, 18, 19]\n",
      "[12  6  2  8  0  0  1  2  1 14  1  2  2  2  1  1  0  2  2  2  3 12 15  7\n",
      "  1  1  2  2  0  2  2  2  1  0  1  1  0  2  0  8  3  3  2  9  0  1  2  1\n",
      "  2  1  2  0  2  9  3  8  0  8  2  2  2  0  1  0  0  2  0  1  2  1  0  0\n",
      "  0  1  2  1 19  1 18  2  0  2  1  0  0  0  1  1  0  8  1  2  1  2  1  0\n",
      "  2  2  0  2  2 14  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  2\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  4 17  3  0  1  2  2  1 14  1  3\n",
      "  2  2  1  2  1  2  2  2  0  0  0  0  1  2  0  1  1 12 15  8  2  1  2  0\n",
      "  2  0  1  1  2  0  2  1  2  3  1  1  1  2  0  2  2  0  1  0  2  4  2 16\n",
      "  0  2  2  1  1  2  5  1  8  1  1  1  2  1  2  0  0  1  2  1  0  0  2  2\n",
      "  0  1  1  0 14  2  2  3  2 12  0 11  8  3  0  0  1  1  2  1  5  0  1  1\n",
      "  0  2  0  1  2  1  2  1 13  0  2  0  0  0  1  0  2  8  1 10 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  3  1  1  2  0  1 15  8  1  2  0  2  1\n",
      "  1  2  0  1  2  0 15  1  0  7 12  2  0  0  0  1  1  5  0  1  1  3  8  1\n",
      "  3  0 14  2  3  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  8  2  2  0  2  1 12  0  1  2  0  0  0  9  1 12  2  0  0  2  2  1\n",
      "  2  2  1  0  1  0  0  2  0  1  1  1  2  2  1  0  2  1  2  1  1  8  1  1\n",
      "  2  0  6  2  1  0  2  0  1  0 12  0  2  2  2  0  2  1  3  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0  8  1 14  1  2  2  2  1  3  0  1  1  3  2  1\n",
      "  8  0  0  1  2  1  2  2  0  0  1  0  0  2  1  2  2 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 15, number of clusters 20, cluster indices [0, 1, 2, 3, 5, 14, 11, 8, 12, 15, 6, 7, 10, 18, 20, 21, 13, 4, 17, 16]\n",
      "[ 4  6  2 20  0  0  1  2  1 14  1  2  2  2  1  1  0  2  2  2  6 12  0  1\n",
      "  1  1  2  2  0  2  8  2  1  0  1  3  0  2 16  5  3  3  2  0  0  1  2  1\n",
      "  2  1  2  0  2  3  0  8  0  8  2  2  2  0  2  2  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  3  1  2  2  0  2  1  0  2  0  1  1  0  0  1  2  0  2  1  0\n",
      "  2  2  0  2  2  7  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  6\n",
      "  1  1  0  0  2  1 15  0  1  2 10  0  1 13 15 10  0  1  2  2  1  7  1  3\n",
      "  2  2  2  2  1  2  2  2  0  0  0  0  1  2  0  1  1  3 15  8  2  1  2  0\n",
      "  2  0  1  1  2  1  2  1  8 10  1  1  1  2  0  2  2  0  1  0  2 12  2  1\n",
      "  0  2  2  3  1  2 15  1  8  1  1  1  2  1  2  0  0  1  2  1  0  0  2  2\n",
      "  0  1  1  0 14  2  2  3  2 12  0  4  8 17  0  0  1  1  2  1 15  0  1  1\n",
      "  0  2  0  1  2  1  2  1 21  0  2  0  0  0  1  0  2  8  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 10  1  1  2  0  1 15  6  1  1  0  2  1\n",
      "  1  2  0  1  1  0  8  1  0 12  2  2  0  0  0  1  1  3  0  1  1  3  8  1\n",
      "  3  0 14  2  3  2  2  2 11  8  2  0  2  2  1  2  1  1  1  1  1  4  2  2\n",
      "  1  1  8  2  2  0  2  1 12  0  1  2  0  0 11 10  1  0  2  2  0  6  2  1\n",
      "  2 15  1  0  1  0  0  2  0  1  1  1  2  1  1  0  2  1  2  1  1  0  1  2\n",
      "  2  0  6  2  1  0  0  0  1  0  4  0  2 17  2  0  2  1  3  0  1  1  2  1\n",
      "  1  2 15  0  2  1 18  1  0  8  1 14  1  2  2  2  1  3  0  1  1  3  2  1\n",
      "  8  0  0  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 16, number of clusters 22, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 7, 10, 21, 4, 17, 16, 9, 13, 18, 19, 22, 20]\n",
      "[19  6  2  8  0  0  1 13  1 14  1  2  2  2  1  1  0  2  2  2  6 12  0  1\n",
      "  1  1  2  2  0  2  2  2  1  0  1  1  0  2 20 21  3  3  2  0  0  1  2  1\n",
      "  2  1  0  0  2  9  0  0  0  0  2  2  2  0 10  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  3  1  2  2  0  2  1  0  2  0  1  1  0  0  1  2  0  2  1  0\n",
      "  2  2  0  2 19 15  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0 17\n",
      "  1  1  0  0  2  1 15  0  1  2  8  0  1  0  2 12  0  1  2  2  1 14  1  3\n",
      "  2  2  2  2  1  2  2  2  0  0  0  0  1  2  0  1  1 12 15  8  2  1  2  0\n",
      "  2  0  1  1  2 16  2  1  2 10  1  1  1  2  0  2  2  0  1  0  2  2  2  7\n",
      "  0  2  0 10  1  2  3  1  0  1  1  1  2  1  2  0  0  1  2  1  0  0  2  2\n",
      "  0  1  1  0 14  2  2 12  2  0  0 11  8  4  0  0  1  1  2  1 15  0  1  1\n",
      "  0  2  0  1  0  1  2  1  2  0  2  0  0  0  1  0  2  8  1 18 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  6  1  1  2  0  1 15  8  1  1  0  2  8\n",
      "  1  2  0  1  2  0 15  1  0 22  0 21  0  0  0  1  1  3  0  1  1  0  8  1\n",
      "  3  0 14  2  3  2  2  2 11  4  2  0  2  2 19  2  1  1  1  1  1  1  2  2\n",
      "  1  1  0  2  2  0  2  1 12  0  1  2  0  0  0  6  1  0  2  2  0  8  2  1\n",
      "  2 10  1  0  1  0  0  2  0  1  1  1  1  1  1  0  0  1  2  1  1  0 15  1\n",
      "  2  0  6  2  1  0  0  0  1  0  2  0  2  3  2  0  2  1  3  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0  6  1 14  1  2  8  2  1  3  0  1  1  3  2  1\n",
      "  8  0  0  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 17, number of clusters 19, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 10, 9, 18, 22, 5, 16, 7, 13, 4]\n",
      "[13  6  2  8  0  0  1  2  1 14  1  2  2  2  1  1  0  2  2  2 10 12  0  1\n",
      "  1  1  2  2  0  2  6  2  1  0  1  1  0  2  5  8  3  3  2  0  0  1  2  1\n",
      "  2  1  0  0  2 11  0  8  0  0  2  2  2  0 16  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  3  1 22  2  0  2  1  0  2  0  1  1  0  0  1  2  0  2  1  0\n",
      "  2  2  0  2  4 13  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0 18\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  0 15 10  0  1  2  2  1 13  1  0\n",
      "  2  2  1  2  1  2  2  2  0  0  0  0  1  2  0  1  1 12 15  8  2 14  2  0\n",
      "  2  0  1  1  2  5  2  1  2 10  1  1  1  2  0  2  2  0  1  0  2 12  2 14\n",
      "  0  6  0  1  1  2 15  1  0  1  1  1  2  1  2  0  0  1  2  1  0  0  2  8\n",
      "  0  1  1  0 14  0  2  0  2 12  0  2  8 11  0  0  1  1  2  1 15  0  1  1\n",
      "  0  2  0  1  0  1  2  1  2  0  2  0  0  0  1  0  2  8  1  1 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  3  1  1  2  0  1 15  0  1  1  0  2  1\n",
      "  1  2  0  1  1  0  8  1  6  2  7  8  0  0  0  1  1  3  0  1  1  0 15  1\n",
      "  3  0 14  2  3  2  2  2 11  8  2  0  2  2  1  2  1  1  9  1  1 11  2  2\n",
      " 14  1  0  2  2  0  2  1 12  0  1  2  0  0  0 10  1  0  2  2  0  6  2  1\n",
      "  2 15  1  0  1  0  0  2  0  1  1  1  2  2  1  0  8  1  2  1  1  0 15  1\n",
      "  2  0  6  2  1  0  0 10  1  0  2  0  2 18  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0  6  1 14  1  2  8  2  1 11  0  1  1  3  2  1\n",
      "  0  0  0  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 18, number of clusters 16, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 10, 22, 4, 7, 9, 5]\n",
      "[ 2  6  2 14  0  0  1  1  1 14  1  2  2  2  1  1  0  2  2  2  6 12  0  1\n",
      "  1  1  2  2  0  2  0  2  1  0  1  1  0  2 14  8  0 11  2  0  0  1  2  1\n",
      "  1  1  0  0  2 11  0  8  0  0  2  2  2  0  1  0  0  1  0  1  2  1  0  0\n",
      "  2  1  2  2  3  1  7  2  0  2  1  0  2  0  1  1  0  0  1  2  1  2  1  0\n",
      "  2  2  0  2  2 14  2  2  0  1  0  1  1  1  0  2  2  1  0  1  1  1  0  6\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  0  2 10  0  1  2  2  1 14  1  0\n",
      " 15  2  1  2  1  2  2  2  0  0  0  2  1  2  0  1  1 10 15  8  2  1  2  0\n",
      "  2  0  1  1  2 14  2  1  2 10  1  1  1  2  0  2  2  0  1  0  2  2  2  1\n",
      "  0 10  0  1  1  2 15  1  8  1  1  1  2  1  2  0  0  1  2  1  0  0  2  8\n",
      "  0  1  1  0 14  0  2  0  2  2  0 11  8 11  0  0  1  1  2  1 15  0  1  2\n",
      "  0  2  0  1  2  1  2  1  2  0  2  0  0  0  1  0  2  8  1  2  0  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  6  1  1  2  0  1 15  6  1  1  0  2  1\n",
      "  1  2  0  1  1  0  4  1  0  2  2 15  0  0  0  1  1  3  0  1  1  0  6  1\n",
      "  3  0 14  2 11  2  2 14 11  8  2  0  2  2 11  2  1  1  1  1  1 11  2  2\n",
      "  1  1  0  2  2  0  2  1 12  0  1  2  0  0  0 10  1  0  2  2  0  6  2  1\n",
      "  9 15  1  0  1  0  0  2  0  1  1  1  1  2  1  0 22  1  2  1  1  0  1  1\n",
      "  2  0  6  2  1  0  0 10  1  0  2  0  2 11  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0  5  1 14  1  2  8  2  1  3  0  1  1  3  2  1\n",
      "  8  0 15  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 19, number of clusters 17, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 10, 22, 4, 5, 9, 13, 7]\n",
      "[ 1  6  2  8  0  0  1  1  1 14  1  2  9  2  1  1  0  2  2  2 10 12  0  1\n",
      "  1  1  2  2  0  2  8  2  1  0  1  1  0  2  0  8  0 11  2  0  0  1  2  2\n",
      "  2  1  0  0  2 11  0  8  0  0  2  2  2  0  1  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  3  1 22  2  0  2  1  0  2  0  1  1  0  0  1 11  1  2  1  0\n",
      "  2  2  0  2 15  7  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  8\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  9 15 10  0  1  2  2  1 14  1  0\n",
      " 15  2  1  8  1  2  2  2  0  0  0  2  1  2  0  1  1  1 15 15  2  0  2  0\n",
      "  2  0  1  1  2  0  2  1  4 10  1  1  1  2  0  2  2  0  2  0  2  2  2  1\n",
      "  0  8  0  1  1  2 15  1  8  1  1  1  2  1  2  0  0  1  2  1  0  0 14  8\n",
      "  0  1  1  0  1  2  2  0  2  2  0  2  8 11  0  0  1  1  2  1 15  0  1  1\n",
      "  0  2  0  1  2  1  2  1  2  0  2  0  0  0  1  0  2  8  1  9 11  2  1  2\n",
      "  2  2  2  0  2  2  0  1 15  2  0 13  1  1  2  0  1 15  5  1  1  0  2  8\n",
      "  1  2  0  1  1  0 15  1  0  2  2  8  0  0  0  1  1  3  0  1  1  0  5  1\n",
      "  3  0 14  2 11  2  2 14 11  8  2  0  2  2 11  2  1  1  1  1  1  2  2  2\n",
      "  1  1  0  1  2  0  2  1 12  0  1  2  0  0  0 10  1  0  2  2  0  8  2  1\n",
      "  2 15  1  0  1  0  0  2  0  1  1  1  2  1  1  0  8  1  2  1  1 15  1  1\n",
      "  2  0  6  9  1  0  0 10  1  0  1  0  2  6  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2  0  0  2  1  1  1  0  6  1  1  2  2  8  2  1  6  0  1  1  3  2  1\n",
      "  0  0 15  1  2  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 20, number of clusters 17, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 10, 5, 9, 13, 7, 16, 4]\n",
      "[ 3  6  2  8  0  0 16 15  1 14  1  2  2  2  1  1  0  2  2  2  6 12  0  1\n",
      "  1  1  2  2  0  2  5  0  1  0  1  1  0  2  7  8  0 11  2  0  0  1  2  2\n",
      "  2  1  2  0  2 11  0  0  0  0  2  2  2  0  1  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  3  1 15  2  0  2  1  0  2  0  1  1  0  0  1  3  0  2  1  0\n",
      "  2  2  0  2  4 14  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  6\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  0  2 15  0  1  2  2  1 14  1  0\n",
      " 15  2  1 15  1  2  2  2  0  0  0  0  1  2  0  1  1  1 15  8  2  1  2  0\n",
      "  2  0  1  1  2  7  2  1  2  1  1  1  1  2  0  2  2  0 12  0  2  9  2  1\n",
      "  0 16  0  1  1  2 15  1  8  1  1  1  2  1 15  0  0  1  2  1  2  0  2  8\n",
      "  0  1  1  0  1  2  2  0  2  2  3 11 15 11  0  0  1  1  2  1  3  0  1  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  0  0  0  1  0  2  8  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 10  1  1  2  0  1 15  6  1  1  0  2  8\n",
      "  1  2  0  1  1  0 13  1 13  2  2 15  0  0  0  1  1  3  0  1  1  0 15  1\n",
      "  3  0 14  2 11  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1 12  0  2  2  0  2  1 12  0  1  2  0  0  0 10  1  0  2  2 13 16  2  1\n",
      "  2 15  1  0  1  0  0  2  0  1  1  1  2  1  1  0 15  1  2  1  1 15  1  1\n",
      "  2  0  6  2 16  0  0  0  1  0  1  0  2  6  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0  5  1  1  1  2  8  2  1  6  0  1  1  3  2  1\n",
      "  8  0 15  1  2  1  2  2  0  0  1  0  0  1  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 21, number of clusters 17, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 10, 13, 16, 4, 17, 18, 7]\n",
      "[ 0  6  2  8  0  0 16  0  1 14  1  2  2  2  1  1  0  2  2  2 10 12  0  1\n",
      "  1  1  2  2  0  2 15 15  1  0  1  1  0  2  0  8  3 11  2  0  0  1  2  2\n",
      "  2  1  0  0  2 11  0  0  0  0  2  2  2  0  8  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  3  1 18  2  8  2  1  0 15  0  1  1  0  0  1  2  0  2  1  0\n",
      "  2  2  0  2 13  1  0  2  0  1  0  1  1  1  0  2  2  1  0  1  1  1  0  1\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  2 15 16  8  1  2  2  1  1  1  0\n",
      "  7  2  1 15  1  2  2  2  0  0  0  0  1  2  0  1  1  1 15 15  2  1  2  0\n",
      "  2  0  1  1  2  1  2  1  2 13  1  1  1  2  0  2  2  0  2  0  2 12  2  1\n",
      "  0 16  0  1  1  2  3  1  0  1  1  1  2  1  2  0  0  1  2 14  0  3 14  8\n",
      "  0  1  1  0  1  0  2 13  2  2  3  0 15 11  0  0  1  1  2  1  3  0  1  1\n",
      "  0  2  0  2  2  1  2  1  8  0  2  0  0  0  1  0  2  8  1  1  0  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 16  1  1  2  0  1 13  6  1  1  0  2  8\n",
      "  1  2  0  1  2  0 13  1 13  2  2 15  0  0  0  1  1  3  0  1  1  0 13  1\n",
      " 13  0 14  2 11  2  2 14 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  0  2  2  0  2  1 12  0  1  2  0  0  0 17  1  0  2  2 13  8  2  1\n",
      "  2 15  1  0  4  0  0  2  0  1  1  1  1  2  1  0 18  1  2  1  1 15  1  1\n",
      "  2  0  6  2 16  0  0  0  1  0  1  0  2  6  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  0  0  6  1  1  1  2  8  2  1  6  0  1  1  3  2  1\n",
      "  0  0 15  1 15  1  2  2  0  0  1  0  0  2  1  2  1 14]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 22, number of clusters 18, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 13, 16, 5, 9, 10, 17, 4, 7]\n",
      "[ 3  6 11  8  0  0 16  1  1  1  1  2  2  2  1  1  0  2  2  2  6 12  7  1\n",
      "  1  1  2  2  0  2  6 15  1  0  1  1  0  2 14  8  1 11  2  0  0  1 11  2\n",
      "  2  1  0  0  2 11  0  8  0  5  2  2  2  0 17  0  0  1  0  1  2 13  0  0\n",
      "  0  1  2  1  3  1  2  2  0  2  1  0  9  0  1  1  0  0  1  2  1  2  1  0\n",
      "  2  2  0  2 13  1  0  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  6\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  2 11 16  0  1  2  2  1  1  1  0\n",
      " 16  2  1 15  1  2  2  2  0  0  0  0  1  2 11  1  1  1 15 15  2  1  2  0\n",
      "  2  0  1  1  2 14  2 16  2 16  1  1  1  2  0  2  2  0  1  0  2  4  2  1\n",
      "  0 16  0  1  1  2  3  1  8  1 16  1  2  1  2  0  0  1  2  2  0  3  3  8\n",
      "  0  1  1  2  1  0  2 16  2  2  3  5 15 11  0  0  1  1  2  1  3  0  1  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  0  0  0  1  0  2 15  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  3  1  1  2  0  1 13 10  1  1  0  2  8\n",
      "  1  2 11  1  1  0 13  1 13  2  2 15  0  0  0  1  1  3  0  1  1  0  6  1\n",
      "  3  0  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  5  1  2  0  2  1 12  0  1  2  0  0  0 16  1  0  1  2 13 16  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  1  2  1  0  0  1  2  1  1 15  1  1\n",
      "  2  0  6  2 16  0  0  0  1  0  1  0  2 16  2  2  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0  6  1  1  2  2  8  2  1 11  0  1  1  3  2  1\n",
      "  8  0 15  1  2  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAL0AAAC7CAYAAAAwjp8tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAADkRJREFUeJztnVGsHOV1x3//GAhV04gQDLIwrpGKKsxDiLyiSH5JSakoiUIekgiMUB6Q/JJIRIkUSF+4SK1EXpK85MUKUf1gFxBJVISipoiAokoV4t4QtTFuahdZiWULQwQK7UORk9OHnes7a2bufLM7szu73/8nXe3ON99835m9Z8+cOXO+s4oIjMmJDyxaAGPmjZXeZIeV3mSHld5kh5XeZIeV3mSHld5kx0xKL+kuSb+SdErSI10JZUyfaNqHU5J2AP8F3AmcAV4B7ouI17oTz5jumcXS3waciojXI+I94Engnm7EMqY/Lpvh2OuB35S2zwB/sd0B11xzTezduxeAjY3ynq2N/fv3v++4jY3t96f3qZZrs3t5f80QZsCcPn2at956S039ZlH6qsHf5ytJOgQcAtizZw/r6+tFe/VQm/svGWPb/el9KpvZ7F7eXzOEGTCj0Sip3yxKfwa4obS9Gzh7aaeIOAwcLoSKrfZSp2Pl91uad+zv922OUbmfg1vtR/ftaxQ4jlYfWymTWVlm8elfAW6SdKOkK4B7gWe7EcuY/pja0kfEBUlfBn4C7AC+HxHHO5PMmJ6Yxb0hIn4M/Hi6o7d3NQAOHix6lpztlBDrsVtu2RrjeOl7WDNPJ5RvCGpk3JRrQiYzd/xE1mSHld5kx0zuzUyo/HZ796XOpWnr9iyaSrcmwS1qRdfjrSC29CY7rPQmOxbn3pQuvdNehFNcmrqnsJ1f+acdsGtB7NI0YktvsmNxlr6WLdMsja1WvfEqm/F2Fs73e/liS2+yw0pvsmMY7s2x6rvNJrejLj154gZ3rdR+U6n3/RMzNYpoVgdbepMdVnqTHXN1bzY26uLmpZh9C5emPk5fcmkerelzf3XzPFjmyNGm7MsmdxlbepMdVnqTHYOI3rSL0iSNOK0oc2GZXYNlln0TW3qTHVZ6kx1zdW/276+pJ1PzcGpzTetkuZDqfJu2C0raXKaXbbFKFatwDl3RaOklfV/SeUm/LLVdLel5SSeL14/0K6Yx3ZHi3vwDcNclbY8AL0TETcALxbYxS0GjexMRP5O095Lme4BPFO+PAC8BDzeN1fbhVGXJjJoyHikVztosKEl7CNaOeZcAKZdCyd2lKTPtjex1EXEOoHi9tjuRjOmX3qM3kg5JWpe0Dm/2PZ0xjUwbvXlD0q6IOCdpF3C+rmO5gKs0qrzG1l1527gBdX3r8lzaPPCqi3y0XX978rUvpE/aAa6kVs20lv5Z4IvF+y8C/9SNOMb0T0rI8h+BfwP+XNIZSQ8CjwN3SjrJ+Od3Hu9XTGO6IyV6c1/Nrk+2nWzy4VTdou42i70TxlirPrIpmjG5u6bvWnUKsx6rbn80Ht12zhTqxjbpOA3BZIeV3mTHAlOL6y7NbS7ZKWNMXxunC7p2R1bBpVm0i2ZLb7JjEItI5kUXa1MnYvNrze2slSaa/T52JVj01cqW3mSHld5kx8q7NxPx87Xt+7Z2f0quS21cf84uTd/lRVwCxJglxEpvsmPl3ZuJmPDRUnvFIpGUS/ZEpuZjNet1a+LQm+19Ri/6djtarS0eaMqELb3JDiu9yY6Vd28mKK2vjYOzDzekS/YQGernY0tvssNKb7Jj0O5N84OQ6audtSsKWz1PnVwTEZsB1qLPvdqZLb3JDiu9yY5BuzfNV97qDtO6NHUlQurKfrR3XYbhSuTo0pRJqYZwg6QXJZ2QdFzSQ0W7i7iapSTFvbkAfC0ibgZuB74kaR8u4mqWlJQSIOeAzbqV70o6AVzPlEVc5820RVv7ZHKV1fT5KVW5LUOMFg2NVjeyRfXijwMvk1jEtVzL8s03XcvSLJ5kpZf0IeAHwFci4nepx0XE4YgYRcRo586d08hoTKckRW8kXc5Y4Y9GxA+L5uQirkOhPjrz/v0pY5SpcyuaxqyrjJbSZ6i5LUMnJXoj4AngRER8q7TLRVzNUpJi6Q8ADwD/IekXRdvfMi7a+nRR0PXXwOf7EdGYbkmJ3vwrk8knZVoXcZ03KTXpp13sXDte2U2pWDyekrNTR12frTnn4wots5vlNASTHYNOQ+iaJkveNsZd36e6BEjTGtmUG9k29LoWd8msexlbepMdVnqTHVm5N030Xj6jIVWgLh7fVFJkYuw1tt1vbOlNhljpTXbYvVkASZGhlhGe6rHt0lRhS2+yw0pvssPuTc9UuSNtsyabXJo+WOXIjy29yQ4rvckOuzc90PQwqW2FsRR3qOva912MM1QXyZbeZIeV3mTHINybVStb0eiO1PziYJscm+3mbEOfLsiQXJoytvQmO6z0Jjsa3RtJVwI/Az5Y9H8mIh6VdCPwJHA18HPggYh4L3nmFa6R3iY6k9J3ajchwW8cqgvSJymW/v+AOyLiY8CtwF2Sbge+CXy7qGX5NvBgf2Ia0x2NSh9j/qfYvLz4C+AO4Jmi/Qjw2V4kNKZjUiuc7QA2gD8Dvgv8N/BORFwoupxhXNQ1nTm5NIt4QDJRz75h/nJL17KWV1HV/4JRiwq3dPQQrINw3SyfVdKNbET8PiJuBXYDtwE3V3WrFM4FXM3AaBWnj4h3JL3EuE79VZIuK6z9buBszTGHgcMAo9Go7otR7t9GpGaZu7CYLeVrZYVK49X1TMmybJqnVqaUalhV87S8QrTukyrHFKTUstwp6ari/R8BfwWcAF4EPld0cy1LszSkWPpdwJHCr/8A8HREPCfpNeBJSX8HvMq4yKsxgyelluW/M/4hhkvbX2fs38/M0OP0M8nX4qZtlpuzqhvMpDFSfn6l6RwG/v+7FD+RNdlhpTfZMYgsy1Vjwq0oZVQ+psfGTVGdZtlJBmXPUZWtc2j4WZc+mOXcStjSm+yw0pvsWDr3ZvPyCvVuwlDpU96L0Zu6FIi6AxvSDS4dk+Lzr91fN0YXKSAduU629CY7rPQmO5bCvSnftK/NaZ4+ghBN0ZtZqMzgXCtttJ2yfOya3rcj6vZHy4djC8CW3mSHld5kh+aZ9zIajWJ9fb2YuYOFBCu8zrZr2kZSJvqvlXYUn3PteAus5zIajVhfX29MJrKlN9lhpTfZsbjoTRerZ1bYpVl08dO6/KHK/SWS1uUuGFt6kx1WepMdi4vemG4poibl0MVEHo5q8nC6ThF29MaY4WGlN9mRrPSSdkh6VdJzxfaNkl6WdFLSU5Ku6E9Mk0pEXPxDuvg30V6m1IeIrb+6PlU07R8YbSz9Q4zr3WziAq5mKUlSekm7gU8B3yu2hQu4Dp86yz0v6q4A5fYurhKbx29sJHVPtfTfAb4O/KHY/iizFnA1ZkGklPX7NHA+Ispfo6qvpQu4mqUgJQ3hAPAZSXcDVwIfZmz5OyvgmhuzZIc2FmJtGyfvOh7ftr08ZNW5lY6bSHHos4BrRHwjInZHxF7gXuCnEXE/LuBqlpRZ4vQPA1+VdIqxj+8CrmYpmGsagqSLk61yhuRCaIp+pHzeCS7IxXW+5UW0dWOnuFodpi04DcGYGqz0JjucZWlmZxY3punYFu6P3RtjarDSm+xYigpnZoBsl3G5SVQ8PGtLy18/TMGW3mSHld5kh92bTFh0rfhjt9xy8f3B48crx2mSsTH3ZjRKksWW3mSHld5kh92bTOi6Slpbd2nCpZmS8jyzuGu29CY7rPQmO4bn3iywQtZCWNLzTXIpWv7YcWWd+5qHUxMjtPxpIVt6kx1WepMdw3NvlugSn0ST+7Ks55viusyyWqrHz8qW3mTH8Cz9stJ1duG8mPZGOsWK1/Vvs163h7W1SUov6TTwLvB74EJEjCRdDTwF7AVOA1+IiLdbzW7MAmjj3vxlRNwaEZtZPY8ALxQFXF8oto0ZPLO4N/cAnyjeHwFeYlwLxywTHZUG6W28FFouLkm19AH8i6QNSYeKtusi4hxA8Xptq5mNWRCplv5ARJyVdC3wvKT/TJ2g+JIcAtizZ88UIhrTLUmWPiLOFq/ngR8BtwFvSNoFULyerzn2cESMImK0c+fObqQeIouuBd8nTeeWUoe+brymPnXjzfB5p5Tq/mNJf7L5Hvhr4JfAs4wLt4ILuJolIsW9uQ74UVFe+jLgWET8s6RXgKclPQj8Gvh8f2Ia0x2NSh8RrwMfq2j/LfDJPoQyS0bNOteJGq8tj21aI0u57+YYXiNrTDVWepMd865P/ybwv8Bbc5t0cVzD6p/n0M7xTyOiMUQ4V6UHkLReSmVYWXI4z2U9R7s3Jjus9CY7FqH0hxcw5yLI4TyX8hzn7tMbs2js3pjsmKvSS7pL0q8knZK0EotOJN0g6UVJJyQdl/RQ0X61pOclnSxeP7JoWWdF0g5Jr0p6rti+UdLLxTk+JemKRcuYwtyUXtIO4LvA3wD7gPsk7ZvX/D1yAfhaRNwM3A58qTivVVxZ9hBworT9TeDbxTm+DTy4EKlaMk9LfxtwKiJej4j3gCcZr75aaiLiXET8vHj/LmOluJ7xuR0puh0BPrsYCbtB0m7gU8D3im0BdwDPFF2W5hznqfTXA78pbZ8p2lYGSXuBjwMvs3ory74DfB34Q7H9UeCdiLhQbC/N/3OeSl+1kHFlQkeSPgT8APhKRPxu0fJ0iaRPA+cjYqPcXNF1Kf6f86x7cwa4obS9Gzg7x/l7Q9LljBX+aET8sGh+Q9KuiDi33cqyJeEA8BlJdwNXAh9mbPmvknRZYe2X5v85T0v/CnBTccd/BXAv49VXS03h2z4BnIiIb5V2rczKsoj4RkTsjoi9jP9vP42I+4EXgc8V3ZbmHOem9IU1+DLwE8Y3e09HxOw/T7F4DgAPAHdI+kXxdzfwOHCnpJPAncX2qvEw8FVJpxj7+E8sWJ4k/ETWZIefyJrssNKb7LDSm+yw0pvssNKb7LDSm+yw0pvssNKb7Ph/6MD1RMhxJZYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 23, number of clusters 17, cluster indices [0, 1, 2, 3, 14, 11, 8, 12, 15, 6, 13, 16, 5, 10, 17, 18, 4]\n",
      "[ 0  6 11  8  0  0 16 14  1  1  1  2  2  2  1  1  0  2  2  2  6 12 18  1\n",
      "  1  1  2  2  0  2 15 15  1  0  1  1  0  2  0  8  3 11  2  0  0  1 11  2\n",
      "  2  1  0  0  2 11  0  0  0  5  2  2  2  0 16  0  0  1  0  1  2  1  0  0\n",
      "  2  1  2  2  3  1 15  2  0  2  1  0  0  0  1  1  0  0  1  2  0  2  1  3\n",
      "  2  2  0  2 18  1  2  2  0  1  0  1  1  2  0  2  2  1 16  1  1  1  0  4\n",
      "  1  1  8  0  2  1 15  0  2  2  1  0  1  2 11 16  0  1  2  2  1  1  1  0\n",
      " 16  2  1 15  1  2  2 17  0  0  0  0  1  2  8  1  1  1 15 15  2  0 11  0\n",
      "  2  0  1  1  2  0  2 16  2 16  1  1  1  2  0  2  2  0  1  2  2  2  2  1\n",
      "  0 16  0  1  1  2  3  1  0  1 16  1  2  1  2  3  0  1  2  2  0  0 18  8\n",
      "  0  1  1  2  1  0  2 16  2  2  3  2 15 11  0  0  1  1  2  1 13  0  1  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  0  8  0  1  0  2 15  1  1 11  2  1  2\n",
      "  2  2  2  0  2  2  0  1 15  2  0 16  1  1  2  0  1 13  6  1  1  0  2  8\n",
      "  1  2 11  1  1  0 13  1 13  2  2 15  0  0  0  1  1  3  0  1  1  0 10  1\n",
      " 13  0  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1 12  0  1  2  0  3  0  6  1  0  1 15 13 16  2  1\n",
      "  2 16  1  0 15  0  0  2  0  1  1  1  1  1  1  5 15  1  2  1  1 15 15  1\n",
      "  2  0  6  2 16  0  0  0  1  0  2  0  2  6  2  2  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0 10  1  1  1  2  8 11  1  3  0  1  1  3  2  1\n",
      "  8  0 15  2  8  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAL0AAAC7CAYAAAAwjp8tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAADtlJREFUeJztnUGsHdV5x3//GChV04g4GGRhXLuqVWEvQvSuKJI3KSkVJVFgkVRgC2WB5E0iESVSIN34WcqCbJJssrFCVC9wwSKJilDUFBFQVKlCvBeiNs9uYgdZiWULQwMK7aLIydfFnfvuue/NeXPm3Zl7Z+75ftLozpx75pxv7vveN9+c851vZGY4Tk58YN4COM6scaV3ssOV3skOV3onO1zpnexwpXeyw5XeyY6plF7SfZJ+IemCpCeaEspx2kTbnZyStAP4JXAvcAl4DXjYzM42J57jNM80lv4u4IKZvWFm7wPPAA80I5bjtMd1U5x7G/Cb4PgS8FdbnXDzzTfbvn37AFhdHZfvv3FtfX/noUObzlsNKi8tLZW2/du1rdvY2GfIqMnw+0g3Toe5ePEib7/9tqrqTaP0ZY1v8pUkHQOOAezdu5eVlZWifFzna38+VtIjxfcb2ljfXyn5HuD0oa3b2NhnyKh6+H2kCafDDAaDpHrTKP0l4PbgeA9weWMlMzsJnCyEsnF5WCt8DBhr3kgJJ587Qs0dlx89O27jSERgs/Jzy2VyFpVpfPrXgAOS9ku6AXgIeL4ZsRynPbZt6c3smqQvAD8CdgDfNbO1itMcZ+5M495gZj8Efridc0/oxPr+8YhfMSoOffqUIdbJto+HLdYTsgYpMo7q+BqG+eIzsk52uNI72TGVezMNx1le3z996Mz6/pG1zY8FMXcgHKact8uQ0nupjOE4aRPX0HR7C4hbeic7XOmd7JibexPeemOTSVWUuUIbic3CNn7n326DTQviLk0lbumd7JifpY8QjrEvFw+7sYfU+Hh8OcuM60vj+m4c88ItvZMdrvROdnTEvRk/bR4PXI3jFaPfBw6Ox/djY/1PHwxCjtfGkZgHgvrgIUM54ZbeyQ5Xeic7ZurerK7Gxs3HbkzVSMrkLHu5W5IS8Xg0WLey3XmC7dLnSIHxwp75yjENbumd7HCld7KjE6M3dVyatPa6fe/tuHhb0mfZR7ild7LDld7Jjpkq/dLS8Pa4cRtOTpVtQybra30LkbS+pTDZ/9bUbbuLLMI1NEWl0kv6rqSrkn4elO2U9KKk88Xnh9sV03GaI8XS/yNw34ayJ4CXzOwA8FJx7Di9oFLpzewnwG83FD8AnCr2TwEPpnQ2mpzaahtOVA23cdmERMFWj/AWP9lnWd3xZmbr2zTM2r1o4xoWge369Lea2RWA4vOW5kRynHZp/UFW0jFJK5JW4K22u3OcSrY7OfWmpN1mdkXSbuBqrGKYwFUalN5bY7fcOnfiumlCTk+k8946tDgWyzOz9bfbpCtydI3tWvrngc8V+58D/rkZcRynfVKGLP8J+HfgLyVdkvQo8CRwr6TzDF+/82S7YjpOc1S6N2b2cOSrT9TtbGlp/LKDcFE3jBdp11nsHas7sbh8eRxDfH6i/tYuzaRrEPETlgO3J1jypRPl5U2MnMTadtLxMAQnO1zpnezY9is1t8NgMLDYO6Paok13INZ2WB7i7khBS0vHBoMBKysrlbN/bumd7MhK6atCD6Zpo4m2s6FOiGsLZKX0jgOu9E6GdGKNbJtMPGAub1239vPVcix1yfweWNtOL+IpQBynh7jSO9mx8O7NxFh6cEuWNrsmKbfsyTrhUE11GMKovM3x+rbdjjrtdzVkwi29kx2u9E52LLx7M0l6oti67Tmb6ZJLE+KW3skOV3onOzrt3lRNhMQWkaTkp68XI1M+ShOTa2LEpqyf6hchtkrK77PIuKV3ssOV3smOTrs3VXfe2Bra7bo0k5NX5e2F5XXjXLriSnRFjnmRkg3hdkkvSzonaU3SY0W5J3F1ekmKe3MN+LKZ3QHcDXxe0kE8iavTU1JSgFwBRnkr35N0DriNYRLXjxfVTgGvAI+3IuUUTK5X3XrkpZVVT8ub+5/oJ5JGJIWy2JY+v7lwVtR6kJW0D/gY8CqJSVzDXJZvveW5LJ35k6z0kj4IfA/4opn9LvU8MztpZgMzG+zatWs7MjpOoySN3ki6nqHCP21m3y+Kk5O4doX46Mzm71PaCIm5FVUuS93UIV0N1+0TKaM3Ap4CzpnZN4KvPImr00tSLP1h4BHgPyX9rCj7B4ZJW88UCV1/DXy2HREdp1lSRm/+jcngk5DaSVxnTWUcDNUxPjGi7YVuSsni8ZSYnRixOuM+Z+MK9dnN8jAEJzs6HYbQNFWWvH5YQfSb8W4QKVG1RrbpHJitrsXtmXUPcUvvZIcrvZMdWbk3VbSePqMiVCA2Hp+SEny97WW2/N5xS+9kiCu9kx3u3syBpJGhmiM85W27S1OGW3onO1zpnexw96ZlytyRulGTVS5NGyzyyI9beic7XOmd7HD3pgWqJpNi6UVipLhDTee+b6KdrrpIbumd7HCld7KjI+5NeZqOvlLpjkQSuNaJsdmqz3q099t3yaUJcUvvZIcrvZMdle6NpBuBnwB/VNR/zsyOS9oPPAPsBH4KPGJm7yf3vMCpuOqMWqTkit+um5Amx2L99imkWPr/A+4xs48CdwL3Sbob+DrwzSKX5TvAo+2J6TjNUan0NuR/isPri82Ae4DnivJTwIOtSOg4DZOa4WwHsAr8BfBt4FfAu2Z2rahyiWFS13Rm5NLMY4KkzshLWDKXyZxYHpPY3+d0Uf/IFPI14dpO0UbSg6yZ/d7M7gT2AHcBd5RVK5fNE7g63aLWOL2ZvSvpFYZ56m+SdF1h7fcAlyPnnAROAgwGg1gmyPCMOiJV0sh0es0Xk9Wy2EF7sZopUZaVD8wxmWIJPmOMLHzdO0TdOi22kZLLcpekm4r9Pwb+BjgHvAx8pqjmuSyd3pBi6XcDpwq//gPAGTN7QdJZ4BlJXwNeZ5jk1XE6T0ouy/9g+CKGjeVvMPTvG6DbY8XTvJis1pj9FA+yZVGWSW2kuDRVD409m2fxGVknO1zpnezoSJTlYhF1KyrGuBuJoGx7VGV0DVWvdWmDaa4twC29kx2u9E529M+9OR3c4qaZCp8HLcq7PnoT3OonRoNiJ0Zcg4lzwzqj3/90tUvTeFhFQ66TW3onO1zpnezohXszEfvydJv9jPdbGYRoIkIxQmkE53JwEFmXm0T4w4x+/6PB90fLfzhfI+s4HcGV3smO+bk3NXyJWOzLCZ1Y3z9u09y/k8SYnlmPNoUjOTXDo6OMriHi0nQ1q1mIW3onO1zpneyYn3vTgC/RhEvTVRp3E5brVZ/os+xn7lk4cYhbeic7XOmd7NBUT/I1GQwGtrKyMrP+sqIs7DY2ehOpM+/UHNMyGAxYWVmpXArmlt7JDld6JzuSlV7SDkmvS3qhON4v6VVJ5yU9K+mG9sR0kjEbb9L6Zmbr2wRBnYlzY3XKqPq+Y9Sx9I8xzHczwhO4Or0kSekl7QE+CXynOBaewLX7xCz3rIjdAcLyJu4So/NXV5Oqp1r6bwFfAf5QHH+EaRO4Os6cSEnr9yngqpmF/0Zl/5aewNXpBSlhCIeBT0u6H7gR+BBDy99gAte8mGYouzIRa93GGx6Pj8qXIFfpucF5Wmbz99sg5aUMXzWzPWa2D3gI+LGZHcUTuDo9ZZpx+seBL0m6wNDH9wSuTi+YaRiCpPXOZtlvHgzdgOAnxqjp6iS4IKcPHQLgyNmz1W2nuFoNhi14GILjRHCld7JjpotIlpaW8CjLthi6BpMeQk13IcG9OLK2trkw5qI0EcHZQtSmW3onO1zpnezoRYYzp4NsFXE5Iubq1KHu2w8TcEvvZIcrvZMd7t5kwrxzxUczrNXIjlYZezMYJMnilt7JDld6JzvcvcmEppOp1nWXmoi1CvuZxl1zS+9khyu9kx0ddG/CCYgMwo/nmBFsGpJcipovOy5daRWZnJpooWYeX7f0Tna40jvZ0UH3pj+3+BQqX3vTI5dmghTXZZrVUi3+Vm7pnezooKXvKRHr1fm1wNt9kE6x4rH6ddbrtrC2NknpJV0E3gN+D1wzs4GkncCzwD7gIvD3ZvZOrd4dZw7UcW/+2szuNLNRVM8TwEtFAteXimPH6TzTuDcPAB8v9k8BrzDMheP0iYZSg7TWXgo1F5ekWnoD/lXSqqRjRdmtZnYFoPi8pVbPjjMnUi39YTO7LOkW4EVJ/5XaQfFPcgxg79692xDRcZolydKb2eXi8yrwA+Au4E1JuwGKz6uRc0+a2cDMBrt27WpG6i4y71zwbVJ1bSl56GPtVdWJtTfF752SqvtPJP3paB/4W+DnwPMME7eCJ3B1ekSKe3Mr8INiZvE64LSZ/Yuk14Azkh4Ffg18tj0xHac5KpXezN4APlpS/t/AJ9oQyukZkXWuLAdVap5btUaWEyXhHb5G1nHKcaV3smPW+enfAv4XeHtmnc6Pm1n86+zaNf6ZmVUOEc5U6QEkrQShDAtLDtfZ12t098bJDld6JzvmofQn59DnPMjhOnt5jTP36R1n3rh742THTJVe0n2SfiHpgqSFWHQi6XZJL0s6J2lN0mNF+U5JL0o6X3x+eN6yToukHZJel/RCcbxf0qvFNT4r6YZ5y5jCzJRe0g7g28DfAQeBhyUdnFX/LXIN+LKZ3QHcDXy+uK5FXFn2GHAuOP468M3iGt8BHp2LVDWZpaW/C7hgZm+Y2fvAMwxXX/UaM7tiZj8t9t9jqBS3Mby2U0W1U8CD85GwGSTtAT4JfKc4FnAP8FxRpTfXOEulvw34TXB8qShbGCTtAz4GvMrirSz7FvAV4A/F8UeAd83sWnHcm7/nLJW+bCHjwgwdSfog8D3gi2b2u3nL0ySSPgVcNbPVsLikai/+nrPMe3MJuD043gNcnmH/rSHpeoYK/7SZfb8oflPSbjO7stXKsp5wGPi0pPuBG4EPMbT8N0m6rrD2vfl7ztLSvwYcKJ74bwAeYrj6qtcUvu1TwDkz+0bw1cKsLDOzr5rZHjPbx/Dv9mMzOwq8DHymqNaba5yZ0hfW4AvAjxg+7J0xs5J3rveOw8AjwD2SflZs9wNPAvdKOg/cWxwvGo8DX5J0gaGP/9Sc5UnCZ2Sd7PAZWSc7XOmd7HCld7LDld7JDld6Jztc6Z3scKV3ssOV3smO/wdV2xTipXhgHwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 24, number of clusters 17, cluster indices [0, 1, 2, 3, 11, 8, 12, 15, 6, 13, 16, 5, 17, 18, 4, 7, 9]\n",
      "[ 3  6 11  8  0  0 16 13  1  1  1 17  2  2  1  1  0  2  2  2  6  1  7  1\n",
      "  1  1  2  2  4  2 15 15  1  0  1  1  0  2  9  8 13 11  2  0  0  1 11  2\n",
      "  2  1  2  0  2 11  0  8  0  5  2  2  2  0 16  0  0  1  0  1  1  1  0  0\n",
      "  0  1  2  1  3  1  2  2  0 18  1  0 15  0  1  1  0  0  1  2  1  2  1  4\n",
      "  2  2  0  2 17  1  0  2  0  1  0  1  1  1  0  2  2  1  0  1  1  1  0  6\n",
      "  1  1  0  0  2  1 15  0  1  2  1  0  1  2 11 16  8  1  2  2  1  1  1  0\n",
      " 16  2  1 15  1  2  2 16  0  0  0  0  1  2 18  1  1 12 15 15  2  1  2  0\n",
      "  2  8  1  1  2  1  2 16 15 16  1  1  1  2  0  2  2  0  1  0  2 12  2  1\n",
      "  0 16  0 16  1  2  3  1  8  1 16  1  2  1  2  0  0  1  2  1  0  3  2  8\n",
      "  0  1  1  0  1  2  2 13  2  2  3  5 15 11  0  0  1  1  2  1  3  0  1  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  0  8  0  1  0  2  8  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 16  1  1  2  0  1 13 15  1  1  0  2  8\n",
      "  1  2 11  1  2  0 15  1 13  2  2 15  0  0  7  1  1  3  0  1  1  0 15  1\n",
      " 13  0  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0  0  5  6  1  0  2  0 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  1  1  5  8  1  2  1  1 15 16  1\n",
      "  2  5  6  2 16  0  0  6  1  0  1  0  2  4  2  2  2  1  8  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0 15  1  1  1  2  8  2  1  8  0  1  1 13  2  1\n",
      "  8  0  3  1 15  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 25, number of clusters 21, cluster indices [0, 1, 2, 3, 11, 8, 15, 6, 13, 16, 5, 18, 4, 10, 12, 14, 19, 7, 9, 20, 21]\n",
      "[19  0 11 15  0  0 16  1  1  1  1 18  2  2  1  1  8  2  2  2  6  1  1  1\n",
      "  1  1  2  2  4  2 15 15  1  0  1  1  0  2  0  8 12 11  2  0  0  1 11  2\n",
      "  2  1  0  0  2 11  0  8  0  5  2  2  2  0 16  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1 11  1  2  2  0 18  1  0 21  0  1  1  0  0  1  2  1  2  1 16\n",
      "  2  2  0  2 15  1  0  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  7\n",
      "  1  1  0  0  2  1 15  0  1  2 10  0  1  2 11  6  8  1  2  2 19  1  1  8\n",
      " 15  2  2 15  1  2  2 14  0  0  0  0  1  2 11  1  1  6 15 15  2  1  2  0\n",
      "  2  0  5  1  2  0  2 16 15 16  1  1  1  2  0  2  2  0  1  2  2  2  2  1\n",
      "  0 16  0 16  1  2  3  1  8  1 16  1  2  1  2  0  0  1  2  1  0  3 18  8\n",
      "  0  1  1  0  1  2  2  6  2  2  3  1 15 11  0  0  1  1  2  1  3  0  5  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  0  8  0  1  0  2  8  1  1 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 13  1  1  2  0  1 15  0  1  1  0  2 10\n",
      "  1  2 11  1  1  0 15  1 13  2  2 15  0  0  0  1  1  3  0  1  1  0 15  1\n",
      "  3  5  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1 19  1  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0  0  5  6  1  0  2  2 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  1  1  5  3  1  9  1  1 15 20  1\n",
      "  2 13  0  2 16  0  2  0  1  0  1  0  2  8  1  2  2  1  8  0  1  1  9  1\n",
      "  1  2 15  0  2  1  1  1  0 15  1  1  1  2  8 11  1  8  0  1  1 12  2  1\n",
      "  8  0 15  1  8  1  2  2  0  0  1  0  0  1  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 26, number of clusters 19, cluster indices [0, 1, 2, 3, 11, 8, 15, 6, 13, 16, 5, 4, 20, 21, 17, 22, 7, 23, 14]\n",
      "[ 1  0 11  7  0  0 16 20  1  1  1  2  2  2  1  1  0  2  2  2  4  1 14  1\n",
      "  1  1  2  2  4  2 15 21  1  0  1  1  0  2 17  8  8 11  2  0  0  1 11  1\n",
      "  2  1  2  0  2 11  0  5  0  5  2  2  2  0 16  0  0  1  0  1  2  1  0  0\n",
      "  0  1  2  1  3  1  2  2  0  2  1  0  0  0  1  1  0  0  1  2  1  2  1  4\n",
      "  2  2  0  2 22  1  2  2  0  1  0  1  1  2  0  2  2  1  0  1  1  1  0  8\n",
      "  1  6  0  0  2  1 15  0  1  2 16  0  1  2 11 16  8  1  2  2  1  1  1  8\n",
      " 16  2  2 15  1  2  2 22  0  0  0  0  1  2  1  1  1 16 15 15  2  1  2  0\n",
      "  2  8  5  1  2 17  2 16  2 16  1  1  1  2  0  2  2  0  1  0  2  1  2  1\n",
      "  0 16  0 16  1  2  3  1  5  1 16  1  2  1  2  0  0  1  2  1  0  3 23  8\n",
      "  0  1  1  0  1  0  2 13  2  2  3  1 15 11  0  0  1  1  2  1  3  5  1  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  0  0  0  1  0  2 15  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0  4  1  1  2  0  1  3  0  1  1  0  2 16\n",
      "  1  2 11  1  2  0 22  1 13  2  2 15  0  0  0  1  2  3  0  1  1  0 15  1\n",
      "  8  5  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0  0  5  4  1  0  2  2 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  5  1  1  1  1  5  7  1  2  1  1 15 16  1\n",
      "  2  5  0  2 16  0  0  6  1  0  1  0  2  8  2  0  2  1  8  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0 15  1  1  1  2  8  1  1  8  0  1  1  8  2  2\n",
      "  8  0 15  1 15  1  2  2  0  0 23  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 27, number of clusters 18, cluster indices [0, 1, 2, 3, 11, 8, 15, 6, 13, 16, 5, 4, 20, 17, 22, 9, 10, 12]\n",
      "[ 0  0 11 22  0  0 16  1  1  1  1  2  2  2  1  1  0  2  2  2 20  1 10  1\n",
      "  1  1  2  2  4  2 22  0  1  0  1  1  0  2  2  8  8 11  2  6  0  1 11  1\n",
      "  2  1  2  0  2 11  0  5  0  5  2  2  2  0 16  0  0  1  0  1  2  1  0  0\n",
      "  2  1  2  1 11  1  2  2  0  2  1  0  0 11  1  1  0  5  1  2  1  2  1  4\n",
      "  2  2  0  2  2  1  0  2  0  1  0  1  1  1  0  2  2  1 16  1  1  1  0  4\n",
      "  1  2  0  0  2  1 15  0  1  2 16  0  1  2 11 20  8  1  2  2  1  1  1  0\n",
      " 16  2  2 15  1  2  2  2  0  0  0  0  1  2 11  1  1 20 15 15  2  1  2  0\n",
      "  2  8  1  1  2  0  2 16 22 16  1  1  1  2  0  2  2  0  1  2  2 20  2  1\n",
      " 17 16  0 16  1  2  3  1  5  1 16  1  2  1  2  0  0  1  2  1  0  3  9  8\n",
      "  0  1  1  2  1  2  2 20  2  2  3  5 15 11  0  0  1  1  2  1  3 20  1  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  0  8  0  1  0  2 15  1  1 11  2  1  2\n",
      "  2  2  2  0  2  2  0  1 15  2  0 20  2  1  2  0  1 15  0  1  1  0  2 16\n",
      "  1  2 11  1  2  0 15  1 13  2  2 15  0  0  0  1  1  3  4  1  1  0 15  1\n",
      "  8  5  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0  0  5  6  1  0  2  2 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  1  1  5 12  1  2  1  1 15 15  1\n",
      "  2 13  0  2 16  0  0  0  1  0  1  0  2  8  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0 22  1  1  1  2  8  6  1  8  0  1  1  8  2  2\n",
      "  5  0 15  1  0  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 28, number of clusters 17, cluster indices [0, 1, 2, 3, 11, 8, 15, 6, 13, 16, 5, 4, 20, 22, 7, 10, 12]\n",
      "[ 3  0 11  5  0  0 16  0  1  1  1  2  2  2  1  1  0  2  2  2  0  1  0  1\n",
      "  1  1  2  2  4  2 22  0  1  0  1  1  0  0  2  8 20 11  2 11  0  1 11  1\n",
      "  2  1  2  0  2 11  0  5  0  5  2  2  2  0 16  0  0  2  0  1  2  1  0  0\n",
      "  0  1  2  1 22  1  2  2  8  2  1  0  0  6  1  1  0 20  1  2  0  2  1 16\n",
      "  2  2  0  2  2  1  2  2  0  1  0  1  1  2  2  2  2  1  0  1  1  1  0  1\n",
      "  1  2  0  0  2  1 15  0  2  2 16  0  1  2 11 16  8  1  2  2  1  1  1  0\n",
      "  0  2  2  0  1  2  2  2  0  0  0  0  1  2  6  1  1 16 15 15  1  1  2  0\n",
      "  2  8  1  1  2  2  2 16  2 16  1  1  1  2  0  2  2  0  1  2  2  1  2  1\n",
      "  0 16  0 16  1  2  3  1  5  1 16  1  2  1  2  0  0  1  2  1  0  3  1  8\n",
      "  0  1  1  0  1  2  2 20  2  2  3  7 15 11  0  0  1  1  2  1  3 20  1  7\n",
      "  0  2  0  1  2  1  2  1  0  0  2  0  8  0  1  0  2 15  1  1 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 20  2  1  2  0  1  3  0  1  2  0  2 16\n",
      "  1  2 11  1  1  0 22  1 13  2  2 15  2  0  0  1  2  3  0  1  1  0 22  1\n",
      "  8  5  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1 11  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0 22  5  0  1  0  2  2 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1  5 10  1  2  1  1 15 16  1\n",
      "  2  5  0  1 16  0  0  0  1  0  1  0  2  4  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  1  1  1  0 22  1  1  1  2  8 12  1  8  0  1  1  8  2  2\n",
      "  5  0 15  1  0  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 29, number of clusters 18, cluster indices [0, 1, 2, 3, 11, 8, 15, 6, 13, 16, 5, 4, 20, 22, 7, 12, 9, 10]\n",
      "[ 8  0 11 15  0  0 16  7  1  1  2  2  2  2  1  1  0  2  2  2  0  1  0  1\n",
      "  1  1  2  2  4  2 22  0  1  0  1  1  0  0  2  8 13 11  2  6  0  2 11  2\n",
      "  2  1  0  0  2 11  0  5  0  5  2  2  2  0 16  0  0  2  0  1  2  1  8  0\n",
      "  0  1  2  1  8  1  2  2  8  2  1  0  0 12  1  1  0  5  1  2  1  2  1  4\n",
      "  2  2  0  2  2  1  0  2  0  1  0  1  1  2  2  2  2  1  0  1  1  1  0  0\n",
      "  1  2  0  0  2  1 15  0  2  2 16  2  1  2 11  0  8  1  2  2  1  1  1  0\n",
      "  0  2  2  0  1  2  2  2  0  0  0  0  1  2 11  2  1  9 15 22  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16 22 16  1  1  1  2  0  2  2  0  2  0  2  1  2  1\n",
      "  2 16  0 16  1  2  3  1  5  1 16  1  2  1  2  0  0  1  2  1  0  3  2  8\n",
      "  0  1  1  0  1  0  2  8  2  2  3 10 22  7  0  0  1  1  2  1  3 20  1  1\n",
      "  0  2  0  1  2  1  2  1  8  0  2  8  0  0  1  0  2 15  1  1 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 13  2  1  2  0  1 22  0  1  2  2  2 16\n",
      "  1  2 11  1  2  0 22  1 13  2  2 15  2 22  0  1  2  3  0  1  1  0 22  1\n",
      "  8  5  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1 11  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0 22  5  0  1  8  2  0 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  1  2  1  5  2  1  2  1  1 15 15  1\n",
      "  2 13  0  2 16  0  0  0  1  0  1  0  2  8  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 22  1  1  1  2  8  6  1 11  0  1 11  8  2  2\n",
      "  5  0  3  1  0  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 30, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 6, 13, 16, 5, 20, 22, 12, 9, 14, 7, 10, 17, 4]\n",
      "[14  0 11  5  0  0 16  0  1  1  2  2  2  4  1  1  0  2  2  2  0  1  0  1\n",
      "  1  1  2  2  8  2 22  0  1  0  1  1  0  2  2  8  8 11  6  0  0  2 11  1\n",
      "  2  1  0  0  2 11  0  5  0  5 17  2  2  0 16  0  0  2  0  1  2  1 22  0\n",
      "  2  1  2  2  8  1  2  2  0  2  1  0  0  6  1  1  0  8  1  2  0  2  1 16\n",
      "  2  2  0  2  2  1  0  2  0  1  0  1  1  2  2  2  2  1  0  1  1  1  0  2\n",
      "  1  2  0  0  2  1 15  0  2  2 16  2  1  0 11  0  8  1  2  2  1  1  1  0\n",
      "  0  0  2  0  1  2  2  2  0  7  0  0  1  2  6  1  1 20 15 22  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16 22 16  1  1  1  2  0 17  2  0  2  0  2  9  2  1\n",
      "  2 16  0 16  1  2  3  1  5  1 16  1  2  1  2  8  0  1  2  1  0  8 10  8\n",
      "  0  1  1  0  1  0  2  8  2  2  3 12 22 11  0  0  1  1  2  1  3  8  1  1\n",
      "  0  2  0  1  2  2  2  1  0  0  2  8  0  0  1  0  2 15  1  2 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15  2  0 11  2  1  2  0  1 22  0  1  2  2  2 16\n",
      "  2  2 11  1  2  0 22  2 13  2  2 15  2  5 14  1  2 22  0  1  1  0 22  1\n",
      "  8  5  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1 11  1  1  1  2  2\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0 22  5  0  1  8  2  0 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1  5  5  1  2  1  1 15  1  1\n",
      "  2  5  0  2 16  0  0  0  1  2  1  0  2 11  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 22  1  1  1  2  8  6  1  7  0  1 11  8  2  2\n",
      "  5  0 15  1  0  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 31, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 13, 16, 5, 22, 9, 14, 17, 18, 19, 10, 4, 12, 6]\n",
      "[ 8  0 11 16  0  0 16  2  1  1  2  2  2  2  1 16  0  2  2  2 11  1  0  1\n",
      "  1  1  2  2  3  2 22  0  1  0  1  1  0  2  2  3 13 11  2  0  0  1 11  1\n",
      "  2  1  2  0  2 11  0  5  0  5 17  2  2  0 16  0  0  2  0  1  2  1 14  0\n",
      "  0  1  2  1  8  1  2  2  8  2  1  0  0  0  1  1  0  8  1  2  1  2  1  3\n",
      "  2  2  0  2  2  1  2  2  0  1  0  1  1  1  2 17  2  1  3 18  1  1  0 18\n",
      "  1  2  0  0  2  1 15  0  2  0 16  2  1  9 11  0  8  1  2  2  1  1  1  0\n",
      "  0  2  2  3  1  2  2  2  0  0  0  0  1  2  0  2  1  9 15 22  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16  2 16  1  1  1  2  0  2  2  0  2  0  2  9  2  1\n",
      "  2 16  0 16  1  2 22  1  5  1 16  1 11  1  2  3  0  1  2  1  0 14  2  8\n",
      "  0  1  1  0  1  0  2  8  2  2  3 11 22 11  0  0  1  1  2  1 22  8  1  1\n",
      "  0  2  0  1  2  1  2  1  0  0  2  8  0  0  1  0  2 15  1  4 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15 19  0 10  2  1  2  0  1 22  0  1  2  2  2 16\n",
      "  1  2 11  1  2 12  6  1 13  2  2 15  2 22 22  1  2 22  0  1  1  0 22  1\n",
      "  8  5  1  2 11  2  2  2 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2 19\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0 22  5  0  1  8  2  0 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1  5 15  1  2  1  2 15 15  2\n",
      "  2 13  0  2 16  0  0  0  1  0  1  0  2 11  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 22  1  1  1  2  8  2  1 11  0  1  1 13  2  2\n",
      "  5  0  3  1  0  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 32, number of clusters 19, cluster indices [0, 1, 2, 3, 11, 8, 15, 13, 16, 5, 22, 9, 14, 19, 10, 12, 7, 4, 6]\n",
      "[ 5  0 11  6  0  0 16  1  1  1  2 10  2  2  1  1  0  2  2  2  8  9  0  1\n",
      "  1  1  2  2 14  2 22  3  2  0  1  1  0  0  2  8 12 11  1  0  0  2 11  1\n",
      "  2  1  0  0 12 11  0  5  0  5  2  0  2  0 16  0  0  2  0  1  2  1  3  8\n",
      "  0  1  7  2 19  1  2  2  0  2  1  0  0  0  1  1  0  3  1  2  1  2  1  3\n",
      "  2  2 12  2  2  1  2  2  0  0  0  1  1  2  2  2  2  1  3  1  1  1  0  8\n",
      "  1  2  0  0  2  1 15  0  2  2 16  0  1  0 11  0  8  1  2  2  1  1  1  0\n",
      "  0  0  2  3  1  2  2  2  0  2  0  0  1  2 11  2  1  1 15 22  2  1  2  0\n",
      "  2  8  1  1 19  2  2 16  2 16  1  1  1  2  0  2  2  0  1  0  2  9  2  1\n",
      "  2 16  0 16  1  2 22  1  5  1 16  1 11  1  2  3  0  1  2  1 12  3  2  8\n",
      "  0  1  1  0  1  0  2  8  2  4  3 11 15 13  0  0  1  1  2  1 22  8  1  1\n",
      "  0  2  0  1  2  2  2  1  8  0  2  8  0  0  1  0  2 15  1 15 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  7  0  8  2  1  2  0  1 22  0  1  2  2  2 16\n",
      "  2  2 11  1  2 12 15  2 13  9  2 15  2  5  3  1  1  8  0  1  1  0 22  1\n",
      "  8  8  1  2 11  2  2  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  2  7\n",
      "  1  1  5  2  2  0  2  1  1  0  1  2  0 22 13  0  1  8  2  2 13  8  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1 13  6  1  2  1  2 15 16  1\n",
      "  2  5  0  2 16  0  0  0  1  0  1  0  2 19  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 22  1  1  1  2  8 11  1 19  0  1  1 13  2  2\n",
      "  5  0  3  1  8  1  2  2  0  0  1  0  0  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 33, number of clusters 25, cluster indices [0, 1, 2, 3, 11, 8, 15, 13, 16, 5, 22, 9, 14, 10, 12, 7, 4, 6, 17, 18, 20, 19, 21, 23, 24]\n",
      "[24  0 11  5  0  0 16  0  1  1  2  2  2  2  1  1  0  2  2  2  0  9  0  1\n",
      "  1  1  2  2  3  2 15  3  1  0  1  1  0  2  2  8 13 11  2  0  0  2 11  2\n",
      "  2  1  0 11  2 11 19  5  0  5  2 16  2  0 16  0  0  1  0  1  2  1  5  0\n",
      "  0  1  7  2 11  1  2  2  8 10  1  0  3  0  1  1  0  8  1  2  1  2  1  3\n",
      "  2  2 12  2  2  1  2  0  0  1  0  1  1  2  2  2  2  1  0  1  1  1  0 17\n",
      "  1  2  0  0  2  1 15  0  2  2  1  0  1  0 11  0  8 18  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0 12  0  0  1  2  0  2  1  8 15 15  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16 22 16  1  1  1  2  0  2  2  0  1  0  2  0  2  1\n",
      "  2 16  0 16  1  2 22  1  5 14 16  1  2  1  2  3  0  1  2  1 17  3 10  8\n",
      "  0  1  1  0  1  2  2  8  2  9  3 23 15 11  0  0  1  1  2  1 22  8  1  1\n",
      "  0  2  0  1  2  2  2  1  8  0  2  8  0  0  1  0  2 15  1  5 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15 10  0  8  2  1  2  0  1 22  0  1  2  0  2  6\n",
      "  2  2 11  1  2 17 22  2 11  4  9 15  2 21 22  1  2 22 16  1  1  0 22  1\n",
      "  8  5  1  2 11  2  0  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  0  7\n",
      "  1  6  5  2  2  0 10  1  9  0  1  2  0  5 11  0  1  8  2  2 13 20  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11 15  1  2  1  2 15 15  2\n",
      "  2 11  0  2 16  0  0  0  1  0  1  0  2 20  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  7  2  1  1  0 22  1  1  1  2  8  0  1 11  0  1  1 11  2  2\n",
      "  5  0  3  2 15  1  2  2  0  0  1  0 16  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 34, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 22, 9, 10, 6, 17, 19, 24, 13, 18, 25, 26]\n",
      "[25  0 11  6  0  0 16  2  1  1  2  2  2 25  1  1  0  2  2  2  0  9  0  1\n",
      "  1  1  2  2  3  2 15  3  2  0  1  1  0  2  2  8  8 11  2  0  0  2 11 25\n",
      "  2  1  0 11  2 11  0  5  0  5  2  0  2  0 16  0  0  2  0  1  2  1 24  0\n",
      "  0  1  6  1  3  1  2  2  3  2  1  0  3  0  1  1  0  8  1  2  0  2  1  3\n",
      "  2  2  0  2  2  1  2  5  0  1  0  1  1 13  2  2  2  1  3  1  1  1  0  3\n",
      "  1  2  0  0  2  1 15  0  1  2  1  0  1  0 11  0  8 15  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0  2  8  0  1 19  0  1  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16  2  0  1  1  1  2  0  2  2  0  1  0  2  9  2  1\n",
      "  2 16  0  1  1  2 22  1  5  1 16  1  2  1  2  8  0  1  2  1  0  3 26  8\n",
      "  0  1  1  0  1  0  2  8  2  6 15 11 15 11  0  0  1  1  2  1 22  8  1  1\n",
      "  6  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2 15  1 25 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15 17  0 11  1  1  2  0  1 22  0  1  2  0  2 18\n",
      "  2  2 11  1  1  0 22  2 11  9  2 15  2  6  5  1  2  8 15  1  1  0 22  1\n",
      "  8  8  1  2 11  2  0  1 11  8  2  0  2  2  1  2  1  1  1  1  1  1  0 17\n",
      "  1  6  5  2  2  0  2  1  9  0  1  2  0  5 22  0  1 22  2  2 11  3  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  1  1 11  2  1  2  1  2 15 16  1\n",
      "  2 11  0  2  3  0  0  0  1  0  9  0  2  3  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  6  2  1  1  0 15  1  1  1  2  8 25  1  3  0  1  1 11 24  2\n",
      "  5  0 10  1 15  1  2  2  0  0  1  0  3  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 35, number of clusters 18, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 22, 9, 6, 17, 24, 13, 4, 7, 10]\n",
      "[ 5  0 11  6  0  0 16  4  1  1  2  2  2  2  1  1  0  2  2  2  0  9  0  1\n",
      "  1  1  2  2  3  2 15  8  1  0  1  1  0  2  2  8 11 11  2 11  0  2 11  2\n",
      "  2  1  2 11  2 11  0  5  0  5  2 16  2  0  3  4  0  1  0  1  2  1  5  0\n",
      "  0  1  6  1  5  1  2  6  0  2  1  0  3  0  1  1  0  8  1  2  0  2  1 16\n",
      "  2  2  0  2  2  1  0 10  0  1  0  1  1  2  2  2  2  1 16  1  1  1  0  0\n",
      "  1  2  0  0  2  1 15  0  1  2  1  0  1  0 11 16  8  2  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0  9  0  0  1  2  0  1  1 22 15 15  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16  2  0  1  1  1  2  0  2  2  0  1  0  2 13  2  1\n",
      "  2 16  0  1  1  2 22  1  5  0 16  1  2  1  2  8  0  1  2  1  0  5 17  8\n",
      "  0  1  1  0  1  0  2  8  2  9  3  2 15 11  0  0  1  1  2  1 22  8  1  1\n",
      " 11  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2 15  1  2 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15  4  0  7  1  1  2  0  1 22  0  1  1  0  2  6\n",
      "  0  2 11  1  1  0 15  2 11  6  0 15  2  6  8  1  2 22 16  1  1  0 15  1\n",
      "  8  5  1  2 11  2  0  1 11  8  2  0  2  2  1  2  1  1  1  1  1 11  0  6\n",
      "  1  1  5  4  2  0  2  1  9  0  1  2  0  5 11  0  1  8  2  2 11  3  2  1\n",
      "  2 16  1  0 24  0  0  2  0  1  1  1  2  2  1 11  5  1  2  1  2 15  1  2\n",
      "  2 11  0  2 16  0  0  0  1  0 13  0  2  0  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 15  1  1  2  2  8  0  1  0  0  1  1 11  2  2\n",
      "  5  0 22  1  2  1  2  2  0  0  1  0  3  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 36, number of clusters 19, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 22, 9, 6, 24, 13, 4, 7, 10, 12, 14]\n",
      "[ 8  3 11  6  0  0 16  2  1  1  2  2  2 11  1  1  0  2  2  2  0  9  0  1\n",
      "  1  1  2  2  3  2 15  8  1  0  1  1  0  2  2  8  7 11  2 11  0  2 11  1\n",
      "  0  1  2 11  2 11  0  5  0  5  2 16  2  0 16  0  0  1  0  1  9  1  8  0\n",
      "  0  1  4  1  3  1  2  6  0  2  1  0  8  0  1  1  0  8  1  0  0  2  1  3\n",
      "  2  2  0  6  2  8  2  7  0  1  0  1  1  2  2  3  9  1  8  1  1  1  0  3\n",
      "  1  4  0  0  2  1 15  0  1  2  1  0  1  0 11  7  8 15  2  2  9  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0  0  0  0  1  2  0  1  1 13 15 15  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16  2 13  1  1  1  6  0  2  2  0  1  0  2  9  2  1\n",
      "  2 16  0 13  1  9 22  1  5 24 16  1  2  1  4  8  0  1  2  1  0  3  8 16\n",
      "  0  1  1  0  1  0  2  8  2 12 15 11 15 11  0  0  1  1  2  1 22  8  1  1\n",
      " 11  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2 15  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  8  0  7  1  1  2  0  1  8  0  1  2  0  2  3\n",
      "  2  2 11  1  1  0  6  2 11  1 10 15  2 14  3  1  2  8 16  2  1  0 15  1\n",
      "  8  5  1  2 11  2  2  1 11  8  2  0  2  2 11  2  1  1  1  1  1 11  0  4\n",
      "  1  1  5  2  2  8  2  1  9  0  1  2  0  5  5  0  1  8  2  2  7  3  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  1  1 11  6  1  2  1  2 15 15  1\n",
      "  2 11  0  2 16  0  0  0  1  0  9  0  2  3  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 15  1  1  1  2 16  0  1  0  0  1  1 11  2  2\n",
      "  5  0 15  1 16  1  2  2  0  3  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 37, number of clusters 16, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 22, 9, 6, 13, 7, 17, 14]\n",
      "[17  3 11  5  0  0 16  1  1  1  2  2  2  2  1  1  0  2  2  2  0  9  0  1\n",
      "  1  1  2  2  3  2 15  8  1  0  1 17  0  2  2  8  7 11  0 11  0  2 11  2\n",
      "  2  1  0 11  2 11  0  5  0  5  2 16  2  0 13  0  0  2  0  1 11  1  3  0\n",
      "  0  1  6  1  3  1  2  6  0  2  1  0  8  0  1  1  0  8  1 11  0  2  1  8\n",
      "  2  2  0  6  2  1  0 16  0  1  0  1  1  2  2  3  9  1  8  1  1  1  0  1\n",
      "  1 14  0  0  2  1 15  0  2  2 13  0  1  0 11  7  8  1  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0  9  0  0  1  2  0  6  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2  2  2 16  2  0  1  1  1  6  0  2  2  0  0  0  2  2  2  1\n",
      "  2 16  0 13  1  2 22  1  5  1 16  1  2  1  2  8  0  1  2  1  0  3  8 16\n",
      "  0  1  1  0  1  0  2  8  2  0 22  2  6 11  0  0  1  1  6  1 22  8  1  1\n",
      " 11  2  0  1  2  2  2  1  8  0  2  8  0  0  1  0  2 15  1  2 11  2  2  2\n",
      "  0  2  2  0  2  2  0  1 15  1  0  7 13  1  2  0  1 22 15  1  2  0  2 13\n",
      "  2  2 11  1  1  0 15  2 11  9  2 16  2  6 17  1  2  8 16  1  1  0 15  1\n",
      "  8  0  1  6 11  6  2  1 11  8  2  0  6  2 11  2  1  1  1  1  1 11  0  6\n",
      "  1  1  5  2  2  0  2  1  9  0  1  2  0 15  0  0  1  8  2  3 11  3  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11  5  7  2  1  2 15 16  2\n",
      "  2 11  3  2 16  0  0  0  1 14  9  0  2  3  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 15  1  1  2  2 16  0  1  3  0  1  1 11  2  2\n",
      "  5  0 22  1 16  1  2  2  0  0  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAL0AAAC7CAYAAAAwjp8tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAADohJREFUeJztnV2MXVUVx3/LAmL8CCKFVEoticRQTMT0BjF9QRRFJMIDGCghPpD0RROMJgL60GliIr6AL740YuxDKxLUSIhRCUKMiSHcUaMOFamk0dqGgoGIPkiqy4d7Zubc9uw5+9x7Pu/+/5LJnI999t7nzpp1/2ftvdcxd0eIlHhT1x0Qom1k9CI5ZPQiOWT0Ijlk9CI5ZPQiOWT0IjnmMnozu97MnjezI2Z2b12dEqJJbNbBKTPbBPwZuA44BjwL3O7uz9XXPSHqZx5PfxVwxN1fdPc3gIeBm+rplhDNcdYc114M/C23fwz40EYXXHDBBb59+3YAji8fXzt+ghNr2zt37jzjuuXl5Q3Pn17fu3e+u7RMntXyuWYINCN6zNGjR3nllVesrNw8Rl9U+Rlaycz2AHsAtm3bxng8BmCf7Vsrs8TS2vbq+dPq2PD86fXtHe8tLZNntXyuGQLNiB4zGo2iys2j6T8MLLn7J7L9+wDc/esbdMqLjPbQFVcUlr/jucnjQb6P+bK7V1by/VnbDt1TTBkxXEajEePxuNTTz6PpnwUuM7NLzewc4DbgsTnqE6IVZpY37n7KzD4P/AzYBHzH3VdKLhOic+bR9Lj7T4CfzHLtlAYP/K/sXtta/8bavVIsS/LPBSEZ06SkiZNOq2UkrbpEI7IiOWT0Ijlmjt7MQj56k5cDear0p0/RmJn7kv8c6riHuusbEG1Eb4QYJDJ6kRxzRW/moQ45ElNHQEXV/s0/8/3U35F661tA5OlFcnTm6YMcWnfN++5YAmCvF8+lyZdld7GH84Nn1gewL3dpsH6xkMjTi+SQ0Yvk6IW8mZo5+dX141Vkx/S0BskVEUaeXiSHjF4kR6vTEMxGDkVLkqoM4ecD76HFIrkSgeq6HK3v0/SJ6vR3pqimIQgRQEYvkqMX0Ztqkiamvtn70gbDkzR5htz3CfL0Ijlk9CI5WpU3O3cW55PJRzMO7tixtr2e4iOUAqS4jhj5UEVhhNKODIlFuIe6KPX0ZvYdMztpZn/MHTvfzJ4wsxey3+9stptC1EeMvPkucP1px+4FnnT3y4Ans30hBkGpvHH3X5rZ9tMO3wRck20fAJ4G7imra3l5eU2G5FN2FEuadcmSlyuhr+Z8fSGCaf0K5uo0MYBUdD9NMuxBsOaY9UH2Inc/AZD9vrC+LgnRLI0/yOYTuArRB6Lm3mTy5nF3f3+2/zxwjbufMLMtwNPu/r7yeorn3jT5zRuaY1M29yZ//uCOUNLY4jbD833alTep0fTcm8eAz2bbnwV+PGM9QrROTMjye8CvgfeZ2TEzuwu4H7jOzF5g8vqd+5vtphD1ERO9uT1w6qNVG8sPToWkQcxi77KyU1GLg+SO55O5blz19PnywZwY6VSLrKny+YhCNA1BJIeMXiRHhxnOAieqfGWHct3kKrd9+QGa+KqjWAqs4sodz4+H+d4aOrAAkmbqb1LHZ1IReXqRHJ2l6m6LunNZBh/AI1B4vlm0RlaIADJ6kRy9WCPbJKH4eRHzpAXpi3RpOrXJav19ud9ZkKcXySGjF8mx8PImHxPOLzQpyk8f85U9LZeK4/ShOHQbsyyblh2V3iHXcTw+hDy9SA4ZvUiOhZc3eerOW19ZpizV2nzv6ZOkySNPL5JDRi+So9fypizaEUpxMT1AU57DPqInue3yhSjTEZuCAl2/HajL5Pw9QJ5eJIeMXiRHr+VNWXQkdD7mlTtl14UkUv54VZXQm9QffelHR8RkQ7jEzJ4ys8NmtmJmd2fHlcRVDJIYeXMK+JK7Xw5cDXzOzHagJK5ioMSkADkBrOatfN3MDgMXM2MS17apsnJqnlVRVahrNVfR3JbEAzNRVHqQzdL7fRB4hsgkrma2x8zGZjZ++eWX5+utEDUQbfRm9jbgB8AX3P2fsde5+353H7n7aPPmzbP0UYhaiYremNnZTAz+oLv/MDv8kpltySVxPdlUJ+uiLAtZ1anFecJZzQoKLwWmHu8r1j2hMn2d29J3YqI3BjwEHHb3B3KnlMRVDJIYT78LuBP4g5n9Ljv2FSZJWx/JErr+Fbi1mS4KUS8x0ZtfEX57ceUkrm0TszB81sXOwfryMmXpzLk6IVkSI1dCZdbbbEcKDVlmaRqCSI5eT0OomzJPXn1aQfDM+ubUjMozvfHU2YgH2So06YGH5t3zyNOL5JDRi+RISt6U0fywfTZVIPAQGIrHB1OKFE1DaDJN+IIgTy+SQ0YvkkPypgPqiccHrtPUylLk6UVyyOhFckjeNEyRHClK6gobpCspkTRNMORpBmXI04vkkNGL5JC8aYCywaTpd9tWi+SU1V2XFKmjnr5KJHl6kRwyepEcvZA3MRGMITHrmtYqc2w2arMKTUqQPkmaPPL0Ijlk9CI5SuWNmZ0L/BJ4c1b+UXffa2aXAg8D5wO/Ae509zdiG140SZOnkqSJ+Bxmli4N1j1kYjz9f4Br3f0DwJXA9WZ2NfAN4MEsl+WrwF3NdVOI+ig1ep/wr2z37OzHgWuBR7PjB4CbG+mhEDUTm+FsE7AMvBf4FvAX4DV3P5UVOcYkqWs0bUmaLhKaVou8NJiyY6m8SNXF6LOmSymsZI6K5vmsoh5k3f2/7n4lsBW4Cri8qFhh55TAVfSMSnF6d3/NzJ5mkqf+PDM7K/P2W4HjgWv2A/sBRqNR4T/GoSuuWNvevbJSpUsRfZ6/jqr9q+KFpvs3+yzLWccAYnJpTrXjBfUtFRSIqWQO5vkmjMlludnMzsu23wJ8DDgMPAXckhVTLksxGGI8/RbgQKbr3wQ84u6Pm9lzwMNm9jXgt0ySvArRe2JyWf6eyYsYTj/+IhN9Pzd1S5q6mad/dUxDqNJOSLrEtBksUxbvH9g4i0ZkRXLI6EVyWJtTAEajkY/H49ba6x8bJ3Ctp4l8DDx/IhR3j5gOUlhnyWtdmqDkDXWj0YjxeFyq1+TpRXLI6EVy9GIRSTXy317Diho02d+16M2UvIgZbCofECuqM2adb+3TKmqSTvL0Ijlk9CI5BiJv2snw1XwQornoTaOLQQqiN1PzbZaKP7i+LlCRpxfJIaMXydGZvKm2RnaOgZUKtPX6nfaoZ4HKdLzszNf8zJrypCvk6UVyyOhFcnQmb+qRI/38+qyDrmXCVJt7S84PDHl6kRwyepEcAxmcSo+ZV04t5SuJmSo8rMhLHcjTi+SQ0YvkiJY3WTaEMfB3d79x3gSuol7WZE1e0oQG7/JzZSqvnCook68jf7wg6tMHqnj6u5nku1lFCVzFIIkyejPbCnwK+Ha2byiBa/9xX//pArP1n9DxUJlZ2llejioe6+m/CXwZ+F+2/y7mTOAqRFfEpPW7ETjp7vl/o6J/SyVwFYMg5kF2F/BpM7sBOBd4BxPPX1sC19SoI5PZGdeuZVatuBJmVukTaqfq8akqCx6qS9J+zELMSxnuc/et7r4duA34hbvfgRK4ioEyT5z+HuCLZnaEicZXAlcxCFrNcGY28kmof3A5P/tPJgMs92jlVJQ6URKkQt0xUqvGhcnKcCZEABm9SI5WjX7nzm7HShaa7IOdGo+qOjgVUb6w7tAAU0yZMuoYvDoNeXqRHDJ6kRxaRCJmIyQ3QtGYWTVtaIBrDuTpRXLI6EVySN4kQte54vfZvrXtvZ5bXTLrGt2i9kejqL7I04vkkNGL5JC8SYS6U3pUlUtTkmZWakpXIk8vkkNGL5Kjf/LmUG4AYncCk3TaevFwzURJioqrnqbqXJvDHEhpkr+wonKSpxfJIaMXydE/ebNokqZMvgxI0kwRI13mWS3V4GclTy+So3+efqjUPbuwLWZ9kI7x4qHyVdbrNrC2Nsrozewo8DrwX+CUu4/M7Hzg+8B24CjwGXd/tVLrQnRAFXnzEXe/0t1XZ/XcCzyZJXB9MtsXovfMI29uAq7Jtg8ATzPJhSOGRE2pQRqrL4aKi0tiPb0DPzezZTPbkx27yN1PAGS/L6zUshAdEevpd7n7cTO7EHjCzP4U20D2T7IHYNu2bTN0UYh6ifL07n48+30S+BFwFfCSmW0ByH6fDFy7391H7j7avHlzPb3uI13ngm+SsnuLyUMfqq+sTKi+OT7vmFTdbzWzt69uAx8H/gg8xiRxKyiBqxgQMfLmIuBHWRrls4BD7v5TM3sWeMTM7gL+CtzaXDeFqI9So3f3F4EPFBz/B/DRJjolBkZ+cUcgkhIUIYFrixILT9VcVFZrZIUoRkYvkqPl/PT2MvBv4JXWGu2OC1j8++zbPb7H3UtDhK0aPYCZjXNTGRaWFO5zqPcoeSOSQ0YvkqMLo9/fQZtdkMJ9DvIeW9f0QnSN5I1IjlaN3syuN7PnzeyImS3EohMzu8TMnjKzw2a2YmZ3Z8fPN7MnzOyF7Pc7u+7rvJjZJjP7rZk9nu1fambPZPf4fTM7p+s+xtCa0ZvZJuBbwCeBHcDtZrajrfYb5BTwJXe/HLga+Fx2X4u4suxu4HBu/xvAg9k9vgrc1UmvKtKmp78KOOLuL7r7G8DDTFZfDRp3P+Huv8m2X2diFBczubcDWbEDwM3d9LAezGwr8Cng29m+AdcCj2ZFBnOPbRr9xcDfcvvHsmMLg5ltBz4IPMPirSz7JvBl4H/Z/ruA19z9VLY/mL9nm0ZfNP1uYUJHZvY24AfAF9z9n133p07M7EbgpLsv5w8XFB3E37PNvDfHgEty+1uB4y223xhmdjYTgz/o7j/MDr9kZlvc/cRGK8sGwi7g02Z2A3Au8A4mnv88Mzsr8/aD+Xu26emfBS7LnvjPAW5jsvpq0GTa9iHgsLs/kDu1MCvL3P0+d9/q7tuZ/N1+4e53AE8Bt2TFBnOPrRl95g0+D/yMycPeI+6+0lb7DbILuBO41sx+l/3cANwPXGdmLwDXZfuLxj3AF83sCBON/1DH/YlCI7IiOTQiK5JDRi+SQ0YvkkNGL5JDRi+SQ0YvkkNGL5JDRi+S4//1tSh/mbnIuQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 38, number of clusters 21, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 22, 9, 6, 13, 7, 4, 10, 12, 18, 19, 20, 17]\n",
      "[ 9  3 11  7  0  0 16  6  1  1  2  2  2  2  1  1  0  2  2  2  0  9  0  1\n",
      "  1  1  2  2 18  2 15  8  1  0  1  1  0  2 13  8  7 11  2 11  0  2 11  2\n",
      "  2  1  0 11  2 11  0  5  0  5  2 16  2  0 13  0  0  2  0  1 11  1  3  0\n",
      "  0  1  1  6  3  1  2  6  0 22  1  0  8  0  1  1  0  8  1  2  1  2  1  8\n",
      "  2  2  0  6  2  1  0  2  0  1  0  1  1  2  0  3  2  1  8  1  1  1  0  3\n",
      "  1  1  0  0  2  1 15  3 20  2 13  0  1  0 11  0  8 10  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0  0  0  0  1  2 11  1  1  8 15 15 20  1  2  0\n",
      "  2  8  1  1 12  2  2 16  2  0  1  1  1  6  0  2  2  0  0  0  2  9  2  1\n",
      "  0 16  0 13  1  9 22  1  5 19 16  1 11  1  2  8  0  1  2  1  0  8 22 16\n",
      "  0  1  1  0  1  0  2  8  2  0 18 11 15 11  0  0  1  1  6  1 22  0  1  1\n",
      " 11  2  3  1  2  2  2  1  8  0  2  8  0  0  1  0  2  5  1  2 11  2  1  2\n",
      "  0  2  2  0  2  2  0  1 15  6  0  7 20  1  2  0  1 22  3  1  2  0  2 13\n",
      "  2  2 11  1  1  0 22  2 11 12 17 16  0  5  8  1  2  8 16  2  1  0 15  1\n",
      "  8  0  1  6 11  6  2  1 11  8  2  0  6  2 11  2  1  1  1  1  1  1 16  6\n",
      "  1  1  5  2  2  0  6  1  9  0  1  2  0  5 11  0  1  8  2  2 11 18  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11 15  7  2  1  2 15  4  2\n",
      "  2 11  3  2 16  0  0  0  1  0  9  0 13  3  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 15  1  1  2  2 16  0  1  3  0  1  1 11  2  2\n",
      "  5  0 18  1 16  1  2  2  0  0  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAL0AAAC7CAYAAAAwjp8tAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAADqRJREFUeJztnV2MJVW1x39/BxDjRxAZyMgwDonEMJqI6RMuN/OCKIpIhAc1fIT4QDIvmmA0Ebgv052YiC/qiy8TMc4DiAQ1EmJUghBjYgin0Xgd5uKMZKLjTBgwELnX5JLR5cOp7q7TXbtr1zn1efb6JZVTtWvX3rv6rF61zt5rrZKZ4Tgp8aauB+A4beNC7ySHC72THC70TnK40DvJ4ULvJIcLvZMccwm9pBskvSDpuKR76xqU4zSJZl2ckrQD+CNwPXASeBa4zcyer294jlM/82j6q4HjZvaimb0BPAzcXM+wHKc5zpnj2kuBv+SOTwL/sd0FF110ke3duxeA1dX8mY2DpaWlLdetrm5/HuDU6qn1/Xcvvbu0Tp61+vkxBbpxesyJEyd45ZVXVFZvHqEvanyLrSTpAHAAYM+ePYzH46y8uKm185va2PY8wIpW1vcPjg+W1smzVj8/pkA3To8ZjUZR9eax6f8TWDazj2fH9wGY2de2GZSVCXUR+THm68aUh/pxR7vFYzQaMR6PSzX9PDb9s8AVki6XdB5wK/DYHO05TivMbN6Y2VlJXwB+DuwAvmtmR2obmeM0xDw2PWb2U+Cns1ybt6/LTY1ys2SZ5Y3aATOmSZMmznRaq+OmVZf4iqyTHC70TnLMPHszC/nZm9CMTZXx9Gk2Zuax5P8OddxD3e0NiDZmbxxnkLjQO8kx1+zNPNRhjsS0EVr3qvvJP/P91D+QettbQFzTO8nRmaYP8tCGal65YxmAg1bsS5Ovy+3FGs4e3NoewEru0mD7zkLimt5JDhd6Jzl6Yd5MzXE/uFFexeyYci12c8XZBtf0TnK40DvJ0aobgjQyKApJqrKEn594DwWL5GoEmutytX7YngL99RR1NwTHCeBC7yRHL2Zvqpk0Me3NPpY26Pv4tmfQgwdc0zsJ4kLvJEerQr+0NHm0b94kFW4b2Pomsb7lKb4uTL7/Mqq23UcW4R7qolToJX1X0hlJf8iVXSjpCUnHss93NjtMx6mPGE3/PeCGTWX3Ak+a2RXAk9mx4wyC0tkbM/uVpL2bim8Grs32DwNPA/eUtXVq9VQwtV6uv/X9tbp5X5qYFCAhgmn9Cnx1moi/XWuzrQXBPsUQ94lZbfpLzOw0QPZ5cX1DcpxmafyHrKQDksaSxv/gH0135zilRPneZObN42b2gez4BeBaMzstaRfwtJm9r7ydYt+bJp+8IT+XMv+XmBQlVeNv2zZvUqNp35vHgM9l+58DfjJjO47TOjFTlt8HfgO8T9JJSXcB9wPXSzrG5PU79zc7TMepj5jZm9sCpz5StbOlpY2XHQTXSCKCvcvqTpkuucBwKZ/Mdfumq5ogU9UD46rDrFEuot0Oupk0C+6G4CSHC72THB1mOAucKDNpIupOzdKsFJfXwnJxFJeO5YprNkcWwaTp2kRzTe8kR2eputui7lyW8zgp+vR8s3iMrOMEcKF3kqMXMbJNEnI9KGKe1Bx9MV2aTi+y1n5f7ncWXNM7yeFC7yTHwps3+TnhfKBJUX76mEf29BpAYJ4+MA/dhpdl02ZHpXfI9dRlwjW9kxwu9E5yLLx5k6fuvPWVH9nLtXbfe/pk0uRxTe8khwu9kxy9Nm/KZjtCKS6mF2jKc9gHeg+UlweiTM/YFFTo+O1AqacGcU3vJIcLvZMcvTZvyh69ofMxr9wpuy4UT5tvo6qfS19Mib6MoytisiFcJukpSUclHZF0d1buSVydQRJj3pwFvmxmVwLXAJ+XtA9P4uoMlFKhN7PTZvZctv86cBS4lEkS18NZtcPALU0Nch5C+eyL8+QX121yTPP0WZRzvq17GDKVfshm6f0+BDxDZBLXfC7Ll19+eb7ROk4NRAu9pLcBPwS+aGZ/j73OzA6Z2cjMRjt37pxljI5TK1GzN5LOZSLwD5rZj7LilyTtyiVxPdPUIOuiLIFrVdfiPKGZnPJMark2VoptkqnFrpW0F5bqIGb2RsADwFEz+0bulCdxdQZJjKbfD9wJ/Lek32Vl/8UkaesjWULXPwOfaWaIjlMvMQlcf03YEaVyEte2iQkMnzXYOdhe3kxZ3rrIFeOzEyJUZ6PP8giuOuhrVFQM7obgJEev3RDqpkyTV3crCJ7Z2M15VK5px3Jtvam1GTVpkxp4aNo9j2t6Jzlc6J3kSMq8KaPx9BmZSRCc0w/MxwdTihSUa3mjvdSDRUK4pneSw4XeSQ43bzogamao4gxPcdtu0hThmt5JDhd6JzncvGmYaZeEycd0upLyGZYyk6YJhuxmUIZreic5XOid5HDzpgGCi0mZs2rVRaOyBal8eV2mSB3t9HVxzDW9kxwu9E5y9MK86etjcFZKzZFAAtcqPjbb9VmFJmdp+vpduqZ3ksOF3kmOUvNG0vnAr4A3Z/UfNbODki4HHgYuBJ4D7jSzN2I7XtHK+n5fH4OzUsVci6k7s+nSYNtDJkbT/z9wnZl9ELgKuEHSNcDXgW9muSxfBe5qbpiOUx8xuSzNzP43Ozw32wy4Dng0K+9tLkvH2UxshrMdwCrwXuDbwJ+A18zsbFblJJOkrtHU/aa/EF34kEz51pT232DKjuXyKlWD0WdNlxLqs4tZp6gfsmb2TzO7CtgNXA1cWVStcHCewNXpGZXm6c3sNUlPM8lTf4GkczJtvxs4FbjmEHAIYDQahf4x8vWrDKl8zB0sp1fRQjFBHzFelmX9hMYUWg8I9mMFdZfz52dPWFWFedqIyWW5U9IF2f5bgI8yyVH/FPDprJrnsnQGQ4ym3wUczuz6NwGPmNnjkp4HHpb0VeC3TJK8Ok7vicll+XsmL2LYXP4iE/t+bvo+Tz/P+KqYOvP8OCvysoxpIypApST1W9+/v834iqyTHC70TnKozUfTaDSy8XjcWn/9Y2s67fq7yJsi+RMVX6FS2mbJa12aIJQbPetzNBoxHo9L7TXX9E5yuNA7ydGLIJJq5J9ew5o1aHK867M3U+ZFxMxM0BwJmSya6m9rn1vHBDW5VdRkOrmmd5LDhd5JjoGYN21l+GradGpu9qbQfKjrdgpmb2w5d3652BTqa4CKa3onOVzoneTozLyp5q4bCmio2yW56cdxy4/7SsEs0+RCmKebZO01P4F0JQNI5+Ka3kkOF3onOTozb+p49PX18VkHXeeHn+qzIJy5rzMzMbimd5LDhd5JjoEsTqXHzJFTy/lGqrkT12FSDcHocU3vJIcLvZMc0eZNlg1hDPzVzG6aN4GrUy/rZk3epAktFOV9ZWIWk8oio0IRTT2liqa/m0m+mzU8gaszSKKEXtJu4JPAd7Jj4Qlc+4/ZxtYF0sYWKg/VmaWf1dWo6rGa/lvAV4B/ZcfvYs4Ero7TFTFp/W4CzphZ/t+o6N/SE7g6gyDmh+x+4FOSbgTOB97BRPPXlsA1NerIZLbl2vXMqhXTccxq+oT6qVqe46H3v399//YjR7Zel2cOky3mpQz3mdluM9sL3Ar80szuwBO4OgNlnnn6e4AvSTrOxMb3BK7OIGg1w5k0sslUf3cTCgtL2exHTF79qAAQTX1s23bl7GnzCYVnOHOcAC70TnK0KvRLS92ulSw0+YWooi2qCVvftqk12fJthxaYYuqUUcfi1SZc0zvJ4ULvJIcHkTizETI3QrMxs9q0oQWuOXBN7ySHC72THG7eJELtmccqthHsP7df6HuT46F9+7Y9z2gUNRbX9E5yuNA7yeHmTSLU7WNV1T06pv87Pvv8+v7tBefzJs087tmu6Z3kcKF3kqN35s0Q8pvXSlsvHq6ZKJOiYtTTVJtr14ZSmuQvLEgwux2u6Z3kcKF3kqN35s2imTRlCy5DMmmmiDFd5omWKir3lyc7zmz0TtMPloD2KtTufWLWH9IxWjxUP6afoh+yoX4qPgGihF7SCeB14J/AWTMbSboQ+AGwFzgBfNbMXq3Uu+N0QBXz5sNmdpWZrXn13As8mSVwfTI7dpzeM495czNwbbZ/GHiaSS4cZ0hUMTVi6tfdXgwVg0tiNb0Bv5C0KulAVnaJmZ0GyD4vrtSz43RErKbfb2anJF0MPCHpf2I7yP5JDgDs2bNnhiE6Tr1EaXozO5V9ngF+DFwNvCRpF0D2eSZw7SEzG5nZaOfOnfWMuo90nQu+ScruLSYPfai9sjqh9ub4e8ek6n6rpLev7QMfA/4APMYkcSt4AldnQMSYN5cAP84cwc4BHjKzn0l6FnhE0l3An4HPNDdMx6mPUqE3sxeBDxaU/w34SBODcgZGIM6VfExrxWtLY2SLZoA8RtZxinGhd5Kj5fz0ehn4P+CV1jrtjotY/Pvs2z2+x8xKpwhbFXoASeOcK8PCksJ9DvUe3bxxksOF3kmOLoT+UAd9dkEK9znIe2zdpnecrnHzxkmOVoVe0g2SXpB0XNJCBJ1IukzSU5KOSjoi6e6s/EJJT0g6ln2+s+uxzoukHZJ+K+nx7PhySc9k9/gDSed1PcYYWhN6STuAbwOfAPYBt0nat/1Vg+As8GUzuxK4Bvh8dl+LGFl2N3A0d/x14JvZPb4K3NXJqCrSpqa/GjhuZi+a2RvAw0yirwaNmZ02s+ey/deZCMWlTO7tcFbtMHBLNyOsB0m7gU8C38mOBVwHPJpVGcw9tin0lwJ/yR2fzMoWBkl7gQ8Bz7B4kWXfAr4C/Cs7fhfwmpmdzY4H8322KfRFgYwLM3Uk6W3AD4Evmtnfux5PnUi6CThjZqv54oKqg/g+28x7cxK4LHe8GzjVYv+NIelcJgL/oJn9KCt+SdIuMzu9XWTZQNgPfErSjcD5wDuYaP4LJJ2TafvBfJ9tavpngSuyX/znAbcyib4aNJlt+wBw1My+kTu1MJFlZnafme02s71MvrdfmtkdwFPAp7Nqg7nH1oQ+0wZfAH7O5MfeI2bW8/RfUewH7gSuk/S7bLsRuB+4XtIx4PrseNG4B/iSpONMbPwHOh5PFL4i6ySHr8g6yeFC7ySHC72THC70TnK40DvJ4ULvJIcLvZMcLvROcvwb7i84N2OWEmUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 39, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 22, 9, 6, 13, 7, 4, 18, 14, 19, 10, 12]\n",
      "[14  3 11  5  0  0 16  2  1  1  2  2  2 11  1  7  0  2  2  2  0  9  0  4\n",
      "  1  1  2  2  8  2 15  8  1  0  1 11  0  2  2  8  7 11  2 11  0  2 11  2\n",
      "  2  1  0 11  2 11  0  5  0  5  2 16  1  0 13  0  0  2  0  1  2  1  3  0\n",
      "  0  1 12  1  3  1  2  6  8  6  1  0  8  0  1  1  0  8  1  2  0  2  1  8\n",
      "  2  2  0  6  2  1  2  2  0  1  0  1  1  2  0  2  2  1  8  1  1  1  0  1\n",
      "  1 13  0  0  2  1 15  0 10  2 13  0  1 18 11  0  8 19  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0  0  0  0  1  2  0  6  1  8 15 15 10  1  2  0\n",
      "  2  8  1  1  2  6  2 16  2  0  1  1  1  6  0  2  2  0  0  0  2  8  2  1\n",
      "  0 16  0 10  1  2 22  1  5  0 16  1 11  1  2  8  0  1  2  6  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  2  0  8 11 15 11  0  0  1  1  6  1 22  8  1  1\n",
      " 11  2  0  1  2  2  2  1  8  0  2  8  0  0  1  0  2 16  1  2 11  2  1  2\n",
      "  0  2  2  0  2  6  0  1 15  6 11  8 13  1  2  0  1 22  3  1  2  0  2 13\n",
      "  2  2 11  1  1  0 22  2 11  9  2 16  0 15  0  1  2  8 16  2  1  0 15  1\n",
      "  8  8  1  6 11  6  2  6 11  8  2  0  6  2  1  2  1  1  1  1  1  1  2  6\n",
      "  1  7  5  2  2  0  6  1  9  0  1  2  0 15 11  0  1  8  2  2 11 16  2  1\n",
      "  2 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11  5  7  2  1  2 15 16  2\n",
      "  2 11  3  2 16  0  2  0  1  0  1  0  2  3  2  0  2  1  0  0  1  1  2  1\n",
      "  1  2 15  0  2  2  1  1  0 15  1  1  2  2 16  0  1  3  0  1  1 11  2  2\n",
      "  5  0  8  6 16  1  2  2  0  0  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 40, number of clusters 23, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 22, 9, 6, 13, 7, 14, 17, 20, 21, 4, 19, 23, 18, 12]\n",
      "[23  3 11  7  0  0 16  0  1  1  2  2  2  2  1 13  0  2  2  2  0  9  0  9\n",
      "  1  1  2  2 12  2 15  8  1  0  1  1  0  2  2  8 11 11  2  0  0  2 11  2\n",
      "  0  1  0 15  2 11  0  5  0  5  2  2 19  0 13  0  0  1  0  1  2  1  0  0\n",
      "  0  1 17  6 14  1  2  6  0 22  1  0  8  0  1  1  0  8  1 20  1  2  1  8\n",
      "  2  2  0  6  2  1  0  2  0  1  0  1  1  2  2  2  2  1  8  1  1  1  0  4\n",
      "  1 21  0  0  2  1 15  0  2  2  7  0  1  0 11  0  8  1  2  2  1  1  1  0\n",
      "  0  2  2  8  1  2  2  2  0  0  0  0  1  2  0  6  1  8 15 15  2  1  2  0\n",
      "  2  8  1  1  2 19  2 16  2  0  1  1  1  6  0  2  2  0  0  0  2  9  2  1\n",
      "  0 16  0  8  1  2  8  1  5  2 16  1 11  1  2  8  0  1  2  1  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  2  0  8 11 15 11  0  0  1  1  6  1  8  8  1  1\n",
      " 11  2  4  1  2  2  2  1  8  0  2  0  0  0  1  0  2  5  1  2 11  2  2  2\n",
      "  0  2  2  0  2 17  0  1 15  6 20  3 21  1  2  0  1 15  3  1  1  0  2 13\n",
      "  2  2 11  1  6  0 22  2 11  7  0 16  2 11  0  1  2  8 16  2  1  0 15  1\n",
      "  8  5  1  6 11  6  2  6 11  8  2  0  6  2  1  2  1  1  1  1  1 11  2  6\n",
      "  1  7  5  2  2  0  6  1  9  0  1  2  0  5 20  0  1  8  2  2 11  3  2  1\n",
      "  6 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11 15  7  2  1  2 15  9  2\n",
      "  2 11  3  2 16  0  0  0  1  0 21  0  2  3  2  0  2  2  0  0  1  1  2  1\n",
      "  9  2 15  0 17  2  1  2  0 15  1  1  2  2 16  0 18  3 20  1  1 11  2  2\n",
      "  5  0  8  6 16  1  2  2  0  0  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 41, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 9, 6, 13, 7, 14, 17, 20, 21, 23, 10, 18]\n",
      "[23  3 11  7  0  0 16 18  1  1  2  2  2  2  1  7  0  2  2  2  3  9  0  9\n",
      "  1  1  2  2  3  2 15  8  1  0  1  1  0  2  2  8  3 11  0  0  0  2 11  0\n",
      "  2  1  0 11  2 11 13  5  0  5  2 16  1  0 16  0  0  2  0  1 11  1 23  0\n",
      "  0  1  6  6 21  1  2  6  0  2  1  0  8  0  1  1  0  8  1  2  1  2  1  8\n",
      "  2  2  0  6  2  1  0  2  0  0  0  1  1  2  0  2  2  1  8  1  1  1  0  3\n",
      "  1  2  0  0  2  1 15  0  2  0  7  0  1  0 11  0  8 15  2  2  1  1  1  0\n",
      "  2  2  2  8  1  2  2  2  0  7  0  0  1  2  0  6  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2  6  2 16  2  7  1  1  1  6  0  2  2  0  0  0  2 21  2  1\n",
      "  2 16  0  8  1  2  8  1  5  2 16  1 11  1  2  0  0  1  2  6  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  2  0  8 11 15 11  0  0  1  1  6  1  8  7  1  1\n",
      " 11  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2  8  1  2 11  2  2  7\n",
      "  0  2  2  0  2  6  0  1 15  6 20  3  8  1  2  0  1  8  3  1  2  0  2 13\n",
      "  2  2 11  1 10  0 15  2 11  9  2 16  2  5  0  1  2  8 16  2  1  0 15  1\n",
      "  8 20  1  6 11  6  2  6 11  8  2  0  6  2  1  2  2  1  1  1  1  1  2  6\n",
      "  1 13  5  2  2  0  6  1  9  0  1  2  0 14 11  0  1  8  2  2 11 16  2  1\n",
      " 17 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11  7 13  2  1  2 15 16  2\n",
      "  2  5  3  2 16  0  0  0  1  0 23  0  2  3  2  0  2  2  0  0  1  1  2  1\n",
      "  1  2 15  0 17  2  1  1  0 15  1  1 17  2 16  0  1  3  1  1  1 11  2  2\n",
      "  5  0  8  1 16  1  2  2  0  0  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 42, number of clusters 23, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 9, 6, 13, 7, 14, 17, 10, 18, 4, 12, 19, 20, 21, 22]\n",
      "[ 0  3 11  7  0  0 16  1  1  1  2  2  2  2  1  7  0  2  2  2  3  9  0  6\n",
      "  1  1  2  2 12  2 15  8  1  0  1  1  0  2  2  8 20 11  0  0  0  2 11  2\n",
      "  0  1  2 19  2 11  0  5  0  5  2 16 10  0  3  0  0  1  0  1  2  1  0  0\n",
      "  0  1 22  1 14  1  2  6  0  6  1  0  8  0  1  1  0  8  1 11  0  2  1  8\n",
      "  2  2  0  6  2  6  2  2  0  2  0  1  1  2  2  2  2  1  8  1  1  1  0  1\n",
      "  1  2  0  0  2  1 15  0  8  2  0  0  1  0 11  3  8 21  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0  7  0  0  1 13  0  6  1  4 15 15  2  1  2  0\n",
      "  2  8  1  1  2 10  2 16 15  0  1  1  1  6  8  2  2  0  0  0  2  9  2  1\n",
      "  2 16  0  8  1  2  8  1  5  2 16  1 11  1  2  8  0  1  2  6  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  2  0  8 11 15 11  0  0  1  1  6  1  8  8  1  1\n",
      " 15  2  0  1  2  2  2  1  8  0  2  0  0  0  1  0  2  8  1  2 11  2  1  2\n",
      " 18  2  2  0  2 17  0  1 15  6 11  3  2  1  2  0  1 15  3  1  1  0  2  7\n",
      "  2  2 11  1 10  0 15  2 11  6  0 16  2 15  0  1  2  8 16  2  1  0 15  1\n",
      "  8  5  1  6 11  6  0  1 11  8  2  0  6  2  1  2  2  1  1  1  1  1  0  6\n",
      "  1  7  5  2  2  0  6  1  9  0  2  2  0  8 11  3  1  8  2  2 11  3  2  1\n",
      " 10 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11  5 13  2  1  2  8 15  2\n",
      "  2 11  3  2 16  0  0  0  1  0  9  0  2  3  2  0  2  2  0  0  1  1  2  2\n",
      "  1  2 15  0 10  2  1  2  0 15  1  1  4  2 16 11  1  3 11  1  1 11  2  2\n",
      "  5  0  8  6 16  1  2  2  0  0  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 43, number of clusters 22, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 9, 6, 7, 14, 17, 10, 18, 19, 23, 24, 25, 26, 4]\n",
      "[ 8  3 11  5  0  0 16 23  1  1  2  2  2  2  4  7  0  2  2  2 26  9  0  9\n",
      "  1  1  2  2  8  2 15  8  1  0  1  1  0  2  2  8  8 11  2  0  0  2 11  2\n",
      "  2  1  0 15  2 11  0  5  0  5  2 16  9  0 24  3  0  2  0  1  2  1 14  0\n",
      "  0  1 10  6 11  1  2  6  0 15  1  0  8  0  1  1  0  8  1 11  1  2  1  8\n",
      "  2  2  0  6  2  6  0 10  0  1  0  1  1  2  2  2  2  1  8  1  1  1  0 16\n",
      "  1  2  0  0  2  1 15  0  2  2  7  0  1  0 11 25  8  1  2  2  1  1  1  0\n",
      "  0  0  2  8  1  2  2  2  0 11  0  0  1  2  0  6  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2 19  2 16 15  0  1  1  1  6  8  2  2  0  0  0  2 10  2  1\n",
      "  2 16  0  9  1  2  8  1  8  2 16  1 11  1  2  8  0  1  2  1  0  8  6 16\n",
      "  0  1  1  0  1  2  2  8  2  0  8 11 15 11  0  0  1  1  6  1  8  8  1  1\n",
      " 15  2  0  1  2  2  2  1  8  0  2  8  0  0  1  0  2  8  1  2 11  2  2  2\n",
      "  0  2  2  0  2  6  0  1 15  6 11  8  2  1  2  0  1  8  3  1  2  0  2 24\n",
      "  2  2 11  1  1  0 15  2 11  9 18 16 19 15  0  1  2  8 16  2  1  0 15  1\n",
      "  8  0  1  6 11  6  2  6 11  8  2  0  6  2  1  2  1  1  1  1  1  1  0  6\n",
      "  1  9  5  2  2  0  6  1  9  0  1  2  0  8 11 26  1  0  2  2 11 16  2  1\n",
      " 17 16  1  0  2 17  0  2  8  1  1  1  2  2  1 11 15  7  2  1  2 15 16  1\n",
      "  2 19  3  2 16  0  2  0  1  0 10  0  2  3  2  0  2  2  0  0  1  1  2  1\n",
      "  1  2 15  0 17  2  1  1  0 15  1  1  1  2 16  0  1  3 11  1  1 11  2  2\n",
      "  8  0  8  1 16  1  2  2  0  0  1  0  8  2  1  2  1  1]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 44, number of clusters 21, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 9, 6, 7, 17, 10, 18, 19, 23, 26, 12, 4, 13]\n",
      "[ 2  3 11  7  0  0 16  4  1  6  2  2  2  2  1  7  0  2  2  2  8  9  0  6\n",
      "  1  1  2  2  8  2 15  8  1  0  1  1  0  2  2  8  7 11  2  0  0  2 11  2\n",
      "  2  1  2 11  2 11  0  5  0  5  2 16 10 17  9  0  7  1  0  1  2  1  0  0\n",
      "  0  1 17  1 18  1  2  6  0  6  1  0  8  0  1  1  0  8  1 11  1  2  1  8\n",
      "  2  1  0  6  2  6  2  2  0  2  0  1  1  2  0  2  2  1  8  1  1  1  0  8\n",
      "  1  2  0  0  2  1 15  0  2  2 13  0  1  0 11 26  8  2  2  2  1  6  1  0\n",
      "  0  0  2  8  1  2  2  2  0  7  0  0  1  2  0  1  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2 13  2 16 15  0  1 23  1  6  0  2  2  0  0  0  2  9  2  1\n",
      " 19 16  0  9  1  2  8  1  5  2 16  1 11  1  2  8 17  1  2  6  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  2  0  8 11 15 11  0  0  1  1  6  1 15  8  1  1\n",
      " 15  2  8  1  0  2  2  1  8  0  2  8  0  0  1  0  2  8  1  2 11  2  2  2\n",
      "  0  2  2  0  2  6  0  1 15  6 11  8  2  1  2  0  1 15  3  1  2  0  2  7\n",
      "  2  2 11  1  1  0 15  2 11  7  2 16  2 15  0  1  2  8 16  2  1  0 15  1\n",
      "  8 11  6  6 11  6  2  6 11  8  2  0  6  2  1  2  2  1  1  1  1  1  0  6\n",
      "  1  7  5  2  2  0  6  1  9  0  1  2  0  8 11 15  1  0  2  2 11 16  2  1\n",
      "  6 16  1  0  1 17  0  2  0  1  1  1  2  2  1 11  8  7  2  1  2  8  1  1\n",
      "  2 11  3  2 16  0  0  0  1  0 12  0  2  3  2  0  2  2  0  0  1  1  2  1\n",
      "  1  2 15  0  6  2  1  1  0 15  1  1  1  2 16  0  1  3  1  1  1 11  2  2\n",
      "  8  0  8  6 16  1  2  2  0  0  1  0  8  2  1  2  1  6]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 45, number of clusters 19, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 9, 6, 7, 10, 18, 19, 23, 4, 13, 12]\n",
      "[ 2  3 11  5  0  0 16  0  1  6  2  2  2  2  1  9  0  2  2  2  8  9  0  9\n",
      "  1  1  2  2  8  2 15  8  1  0  1  1  0  2  2  8  1 11  0  0 23  2 11  2\n",
      "  0  1  0 11  2 11  0  8  0 11 11  0 13  0  7  0  0  2  0  1  2  1  8  0\n",
      "  0  6  6  6  2  1  2  6  8  4  1  0  8  0  1  1  0  8  1 11  1 10  1  8\n",
      "  2  1  0  6  2  6  0 16  0  2  0  1  1  2  0  2  2  1  8  1  1  1  0  3\n",
      "  1  2  0  0  2  1 15  0  2  2  7  0  1  0 11  8  8 15  2  2  1  6  6  0\n",
      "  8  0  2  8  1  2  2  2  0 12  0  0  1  2  0  6  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2  6  2 16  2  0  1  1  1  6  0  2  2  0  9  0  2  2  2  1\n",
      " 19 16  0  9  1  2 15  1  8  2 16  1 11  1  2  8  0  1  2  1  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  2  9  8 11 15 11  0  0  1  1  6  6 15  8  1  1\n",
      " 15  2  0  1  2  2  2  1  8  0  2  8  0  0  1  0  2  8  1  2 11  2  2  2\n",
      "  0  2  2  0  2  6  0  1 15  6 11  8  2  1  2  0  1 15  3  1  2  0  2  7\n",
      "  2  2 11  1  6  0  4  2 11  2  0 16 19 15 18  1  2  8 16  1  1  0 15  1\n",
      "  8 11  6  6 11  6  2  6 11  8  2  0  6  2  1  2  1  1  1  1  1  1  0  6\n",
      "  1  9 11  2  2  0  6  1  9  0  1  1  0  8 11 23  1  8  2  2 11  7  2  1\n",
      "  6 16  1  0  2  0  0  2  0  1  1  6  2  2 23 11  2  5  2  1  2 15 16  1\n",
      "  2 11  3  2 16  0  0  0  1  0  2  0  2  3  2  0  2  2  0  0  1  1  2  2\n",
      "  1  2 15  0  6  2  1  2  0 15  1  1  1  2 16  0  1  3  1  1  1 11  2  2\n",
      "  8  3  8  6 16  1  2  2  0  0  1  0  8  2  1  2  1  6]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 46, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 5, 9, 6, 7, 18, 19, 23, 13, 12, 14, 17, 4]\n",
      "[ 2  3 11  9  0  0 16  1  1  6  2  2  2  2  1  5  0  2  2  2  8  9  0 14\n",
      "  1  1  2  2  8  2 15  8  1 23  1 11  0  2 19  8  5 11  2  0  0  2 11  2\n",
      "  2  1  0 11  2 11  0  8  0 11  2  0  2  0 16  0  0  2 23  1  2  1 18  0\n",
      "  0  6  6  1  3  1  2  6  8 17  1  0  8  0  1  1  0  8  1  2  1  2  1  8\n",
      "  2  2  5  6  2  6  0 13  0  2  0  1  1  2  0  2 12  1  8  1  1  1  0  7\n",
      "  1  2  0  0  2  1 15  0  6  2  9  0  1  0 11  8  8  1  2  2  1  6  1  0\n",
      "  8  0  2  8  1  2  2  2  0  2  0  0  1  2  0  6  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2 19  2 16  2  8  1  1  1  6  0  2  2  0 14  0  2  2  2  1\n",
      " 19 16  0  0  1  2 15  1  8  2 16  1 11  1  2  8  0  1  2  6  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  0 14  8 11 15 11  0  0  1  1  6  6 15  8  1  1\n",
      " 15  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2  8  1  1 11  2  2  2\n",
      "  0  2  2  0  2  6  0  1 15  6 11  8  2  1  2  0  1 15  3  1  2  0  2  7\n",
      "  2  2 11  1  1  0 15  2 11  9  2 16 19 23  4  1  2  8 16  1  1  0 15  1\n",
      "  8  8  6  6 11  6  2  6 11  8  2  0  6  2  1  2  2  1  1  1  1  1  0  6\n",
      "  1  9 11  2  2  0  6  1  9  0  1  2  0  8 11  8  1  8  2  2 11  7  2  1\n",
      "  6 16  1  0  1  0  0  2  0  1  1  6  2  2  1 11  8  5  0  1  2 15  6  2\n",
      "  2 11  3  2 16  0  0  0  1  0  2  0  2  3  2  0  2  1  0  0  1  1  0  1\n",
      "  6  2 15  0  6  2  1  1  0 15  1  1  1  2 16  0  1  3  1  1  1 11  2  2\n",
      "  8  0  8  1 16  1  2  2  0  0  1  0  8  2  1  2  1  6]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 47, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 9, 6, 18, 19, 23, 14, 17, 10, 20, 13, 21, 4]\n",
      "[18  3 11 13  0  0 16  2  1  6  2  2  2  2  1 13  0  2  2  2  8  9  0  9\n",
      "  1  1  2  2  8  2 15  0  1 23  1  1  0  2 14  8 11  1  0  0 23  2 11  2\n",
      "  2  1  0 10  2 11  0  8  0 11  2 16  6  0 16  0  0  2 23  1  2  2  0  0\n",
      "  0  6  6  1 18  1  2  6  8  6  1  0  2  0  1  1  0  8  1  2  1  2  1  8\n",
      "  2  2  0  6  2  6  2  8  0  1  0  1  1  2  0  2  2  1  8  1  1  1  0 16\n",
      "  1  2  0  0  2  1 15  0  2  2  9  0  1  0 11  8  8 21  2  2  1  6  6  0\n",
      "  8  0  2  8  1  2  2  2  0  2  0  0  1  0  0  1  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2  6  2 16  2  8  1  1  1  6  0  2  2  0  9  0  2  2  2  1\n",
      " 19 16  0 16  1  2 15  1  8  2 16  1  2  1  2  8  0  1  2  6  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  0 20  8 11 15 11  0  0  1  1  6  1  8  8  1  1\n",
      "  8  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2  8  1  0 11  2  2  2\n",
      "  0  2  2  0  2  4  0  1 15  6  2  8  2  1  2  0  1 15  3  1  1  0  2  9\n",
      "  2  2 11  1  1  0 17  2 11 20  0 16 19  8 18  1  2  8 16  0  1  0 15  1\n",
      "  8 11  6  4 11  6  2  6 11  8  2  0  6  2  1  2  1  1  1  1  1  1 16  6\n",
      "  1  9 11  2  2  0  6  1  9  0  1  2  0  8 11  8  1  8  2 10 11 16  2  1\n",
      "  1 16  1  0  1  0  0  2  0  1  1  1  2  2  1 11 15  9  0  1  2 15 16  2\n",
      "  2 11  3  0 16  0  0  8  1  0  2  0  2 11  2  0  2  1  0  0  1  1  0  1\n",
      "  1  2 15  0  1  2  1  1  0 15  1  1 14  2 16  0  1 11  2  1  1 11  2  2\n",
      "  8  0  8  1 16  1  0  2  0  0  1  0  8  2  1  2  1  6]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 48, number of clusters 18, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 9, 6, 19, 23, 10, 20, 21, 5, 7, 4]\n",
      "[ 8  3 11 15  0  0 16 10  1  6  2  2  0  2  1  9  0  2  2  2  8  9  0 20\n",
      "  1  1  2  2  8  2 15  2  1  8  1  1  0  2  6  8  9  1  2  0  8  2 11  2\n",
      "  2  1  2 15  2 11  0  8  0 11  2  0  1  0 16  0  0  2  0  1  2  2  0  0\n",
      "  0  6  6  1  0  1  2  6  0  2  1  0 10  0  1  1  0  8  1  2  2  2  1  8\n",
      "  2  2  0  6  2  6  2  2  0  1  0  1  1  2  0  2  2  1  8  1  1  1  0  3\n",
      "  1  2  0  0  2  1 15  0  2  2 16  0  1  0 11  8  8  1  2  2  1  6  1  0\n",
      "  0  0  2  8  1  2  2  2  0  2  0  0  1  2  0  1  1  9 15 15  2  1  2  0\n",
      "  2  8  1  1  2  6  2 16  2  8  1  1 21  6  0  2  2  0  2  0  2  2  2  1\n",
      " 19 16  0 16  1  2  8  1  8  2 16  1 11  1  2  8  0  1  2  1  0  8  6 16\n",
      "  0  1  1  0  1  0  2  8  0  9  8 11 15 11  0  0  1  1  6  6  8  8  1  1\n",
      " 23  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2  8  1  0 11  2  2  2\n",
      "  0  2  2  0  2  6  0  1 15  6  7  8  2  1  2  0  1  8 15  1  2  0  2 16\n",
      "  2  2 11  1  1  0 15  2 11  2  4 16 19  8  0  1  2  8 16  0  1  0 15  1\n",
      "  8 11  6  6  1  6  0  6 11  8  2  0  6  2  1  2  1  1  1  1  1  1  0  6\n",
      "  1  9 11  2  2  0  6  1  9  0  1  2  0  8 11  8  1  8  2 21 11 16  2  1\n",
      "  1 15  1  0  1  0  0  2  0  1  1  6  2  2  1 11  2  9  0  1  2 15 20  2\n",
      "  2 10  3  0 16  0  0  8  1  0  2  0  2  0  2  0  2  1  0  0  1  1  0  1\n",
      "  1  2 15  0  1  2  1  2  0 15  1  1  5  2 16  0  1  0  2  1  1 11  2  2\n",
      "  8  0 15  1 16  1  0  2  0  0  1  0  8  2  1  0  1  6]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train | iter 49, number of clusters 20, cluster indices [0, 1, 2, 3, 11, 8, 15, 16, 9, 6, 19, 23, 10, 21, 12, 5, 7, 13, 14, 17]\n",
      "[17  0 11  9  0  0 16  2  1  6  2  2  0  2  1 13  0  2  2  2  8  2  0  9\n",
      "  1  1  2  5  8  2 15  8  1  8  1  1  0  2 19  8 11  1  2  0  0  2 11  2\n",
      "  0  1  0 10  2 14  0  8  0 11  2 16  6  0 16  0  0  2  8  1  2  2  0  0\n",
      "  0  6  6  1  0  1  2  6  8  3  1  0 12  0  1  1  0  8  1  2  2  2  1  8\n",
      "  2  2  2  6  2  6  0  2  0  1  2  1  1  2  0  2  2  1  8  1  1  1  0 21\n",
      "  1  2  0  0  2  1 23  0  2  2 16  0  1  0 11  8  8 15  2  2  1  6  6  0\n",
      "  8  0  2  8  1  2  2  2  0  2  0  0  1  2  0  1  1  8 15 15  2  1  2  0\n",
      "  2  8  1  1  2  6  2 16  2 16  1  1  2  6  0  2  2  0  2  0  2  2  2  1\n",
      " 19 16  0 16  1  2  8  1  8  2 16  1 11  1  2  8  0  1  2  6 12  8  6 16\n",
      "  0  1  1  0  1  0  2  8  0  0  8 11 15 11  0  0  1  1  6  6  8  8 11  1\n",
      " 10  2  0  1  0  2  2  1  8  0  2  8  0  0  1  0  2  8  1  0 11  2  2  2\n",
      "  0  2  2  0  2  1  0  1 15  6 11  8  1  1  2  0  1  7  0  1  2  0  2 16\n",
      "  2  2 11  1  1  0 23  2 11  2  0 16 19  8  0  1  2  8 16  1  1  0 15  1\n",
      "  8 11  6  5  1  6  0  6 11  8  2  0  6  2  1  2  1  1  1  1  1 11  2  6\n",
      "  1  9 11  2  2  0  6  1  9  0  1  2  0  8 11  8  1  8  2 12 11  3  2  1\n",
      "  1 16  1  0  1  0  0  2  0  1 11  6  2  2  1 11  9  9  0  1  2 15 16  2\n",
      "  2 11  3  0 16  0  0  8  1  0  2  0  2  0  2  0  2  2  0  0  1  1  0  1\n",
      "  6  2 23  0  1  2  1  2  0 15  1  1  1  2 16 14  1  0  2  1  1 11  2  2\n",
      "  8  0  8  1 16  1  0  2  0  0  1  0  8  2  1  0  1  6]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "run time =  1305.478440284729\n"
     ]
    }
   ],
   "source": [
    "train = {\n",
    "    'n_burnin': 5,\n",
    "    'n_iter': 50,\n",
    "}\n",
    "start = time.time()\n",
    "state,sample_params,train = mdnd(4,np.array(links_train),train)\n",
    "stop = time.time()\n",
    "print('run time = ',stop-start)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "clusters = sample_params['assignments'][49]\n",
    "edges = links_train\n",
    "plt.set_cmap('gist_rainbow')\n",
    "g = nx.MultiDiGraph()\n",
    "unique_edges = []\n",
    "unique_clusters = []\n",
    "unique_index = []\n",
    "for i,e in enumerate(edges):\n",
    "    if node_clusters[e[0]] == node_clusters[e[1]]:\n",
    "        g.add_edge(e[0], e[1], weight=100)\n",
    "    else:\n",
    "        g.add_edge(e[0], e[1], weight=1)\n",
    "    unique_edges += [(e[0], e[1])]\n",
    "    unique_clusters += [clusters[i]]\n",
    "    unique_index += [i]\n",
    "\n",
    "pos = nx.nx_agraph.pygraphviz_layout(g, prog='neato')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "node = sorted([i for i in pos.keys()])\n",
    "cluster_color = get_color(unique_clusters)\n",
    "nx.draw_networkx_nodes(g, pos, nodelist=node, node_size=50, alpha=0.7)\n",
    "nx.draw_networkx_edges(g, pos, edgelist=unique_edges, edge_color=cluster_color, width=0.5,alpha=0.5)\n",
    "plt.savefig('gibbs_viz',format='pdf')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sequential variational inference\n",
      "average score by Lin  1.0362843639650923\n",
      "merged  0\n",
      "pruned  0\n",
      "9\n",
      "average score by Lin  1.147042040671941\n",
      "merged  0\n",
      "pruned  0\n",
      "9\n",
      "average score by Lin  1.165095224755577\n",
      "merged  0\n",
      "pruned  0\n",
      "9\n",
      "self.train | iter 0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-9.139545083817342, -8.573624854756828, -8.312684367965769, -8.057842135133665, -7.995528492496669, -7.911049646682931, -7.948033879203086, -7.96244968024801, -7.789172292433984]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "run time =  1.792999505996704\n"
     ]
    }
   ],
   "source": [
    "from sequentialDN_future import sequentialDN\n",
    "DN = sequentialDN(links_train, np.array(nodes), links_test)\n",
    "start = time.time()\n",
    "DN.vi_train_stochastic()\n",
    "stop = time.time()\n",
    "print('run time = ',stop - start)\n",
    "sequential_clusters = DN.get_map()\n",
    "# edges, clusters = DN.generate(100)\n",
    "# display_adjacency(edges, clusters)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "unique_clusters = np.array(sequential_clusters)[unique_index]\n",
    "cluster_color = get_color(unique_clusters)\n",
    "node = sorted([i for i in pos.keys()])\n",
    "nx.draw_networkx_nodes(g, pos, nodelist=node, node_size=50, alpha=0.7)\n",
    "nx.draw_networkx_edges(g, pos, edgelist=unique_edges, edge_color=cluster_color, width=0.5,alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "stochastic variational inference\n",
      "number of clusters  31\n",
      "train | iter 0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 1\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 2\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 3\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 4\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 5\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 6\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 7\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 8\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of clusters  33\n",
      "train | iter 9\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP4AAAD8CAYAAABXXhlaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAEgNJREFUeJzt3V+sHOV5x/HvL05o0iYSEA7UxaSmklWBpQbkFSWiF5SEhkZp4CKJglFFJUvuRSsRJVWAVqqPq1YlN0luWjWWiOILHEiaRCAUNbUoqKpUAWv+BeMSk8hNLFv40IJCb6KYPL3YcXJ2ds+ZmZ2/6/f3kVbnzJz585yzfvzuM/PO+yoiMLO0vK3vAMyse058swQ58c0S5MQ3S5AT3yxBTnyzBDnxzRLkxDdLUK3El3SLpJclvSLpnqaCMrN2adGee5K2AN8HbgZOAk8Dt0fESxvtc8kll8T27dt/ueLIkc1PsmtXcSD5YxTtU3H7mc0psX/VmMwacuLECV577TUVbVcn8T8ArEbEh7PlewEi4u832mc0GsV4PF5/kM1PUia2/DGK9qm4/czmlNi/akxmDRmNRozH48LEr/NR/3Lgx+uWT2brpkjaK2ksaby2tlbjdGbWlDqJP+9/lZmmLSIORMQoIkYrKys1TmdmTXl7jX1PAlesW94GnKp0hPxH4PxH5EPTy7pj3iEKjlHzY3Y8kDve7hLH80d7G7g6Lf7TwA5JV0q6APgU8EgzYZlZmxZu8SPirKQ/B74LbAG+EhFHG4vMzFpT56M+EfEd4DsNxWJmHamV+LVVrMdjd/1jzO4+vf/MNYMyNX3LCmMs2L7MPpYWd9k1S5AT3yxBTnyzBPVb4+ctUq9XrHfzte75WPmWqeerXjdowqGdO6eWdx/1TaC+uMU3S5AT3yxBTnyzBPVb43dQV1auXWeeD8hdE+jhokAb9Xcf9/Vd0w+HW3yzBDnxzRLkxDdLkBPfLEHD6sCTt8hAHDl1O6rMDMRxKLfBAB7iMavKLb5Zgpz4Zgly4pslaFiJL02/ciJi5tX0OcxSMKzEN7NOOPHNEuTEN0vQwnPnLXQyaepkhWdetIavcIzK9/mXdF68PgbeGJoUBiHtYu48M1tSTnyzBDnxzRLUaeLvYlLXn3vNiJh+lVHzvnztfgFLIpXfczON9AM5T7jFN0uQE98sQU58swR1m/i7ds3W8etfuXpduddcVa8L1O2rv8h1iJkQSvxeCfLfpTtu8c0S5MQ3S1Bh4kv6iqQzkl5ct+5iSYclHc++XtRumGbWpDIt/leBW3Lr7gEei4gdwGPZcqEjR45M13AVn7/P14Bz68AGavD1ZkJc4BpBfpe27yWX+jsNQD5G32PvTmHiR8S/A/+bW30rcDD7/iBwW8NxmVmLFh1l97KIOA0QEaclXbrRhpL2AnsXPI+ZtaD14bUj4gBwAGYfyzWzfiya+K9K2pq19luBM4scJF95FtV1jdR9NZ/Pn3kcv9RJc/MDzDzSnz/H5jX5+VL/ni+/xzJa9HbeI8Cd2fd3Ag83E46ZdaHM7byvAf8J/Lakk5L2APcBN0s6DtycLZvZkij8qB8Rt2/wow82HIuZdaTTufN27drFeDz+xXLh/eXc3HkLzVNXdIzcz+OBzfdfpC6tW8sWXQPI/7yL2tlj+C03d9k1S5AT3yxBTnyzBHVa4+cV1oVNzD1fdIzd1ernLhSdcvbnMz0iGoxmoxhc088zhH8/ZbjFN0uQE98sQU58swQ58c0S1OvFvSHKd+DpYo7M4nMMcyANmzXUi3l5bvHNEuTEN0uQE98sQa7x8/Idfu6ofoi61wXyAxUtSdlYqIvrJVXO30cMQ+EW3yxBTnyzBDnxzRLkGj8vP3AH+Yd4cj+dUyNWrRvz2xfPf1EU4/zJSfrWdwhNnF/7cw/h7Ov/77oIt/hmCXLimyXIiW+WINf4BfqoS/uuhW1jy1rT57nFN0uQE98sQU58swQtVY2/yP3p4kkwqw2OWDgJSAOK77lXm2RzWS3LwJXLyC2+WYKc+GYJcuKbJWipavw2Jqwsruln9shvUXjMov79+Z9Xf279/Hx+3zV9e9zimyXIiW+WoMLEl3SFpMclHZN0VNJd2fqLJR2WdDz7elH74ZpZE8rU+GeBz0bEM5LeAxyRdBj4E+CxiLhP0j3APcDd7YXakYLn8WfLztk6dIi31Q/t3Dm1fMdLR6eW65bTZe6552PYffTozDbWjcIWPyJOR8Qz2fdvAseAy4FbgYPZZgeB29oK0syaVanGl7QduBZ4ErgsIk7D5D8H4NIN9tkraSxpvLa2Vi9aM2tE6cSX9G7gm8CnI+InZfeLiAMRMYqI0crKyiIxmlnDSt3Hl/QOJkn/QER8K1v9qqStEXFa0lbgTFtB9ilfqpapU+uOuZdXdM5F6vPZvgTTNfoDV1+96Tndj365lbmqL+B+4FhEfGHdjx4B7sy+vxN4uPnwzKwNZVr8G4A/Br4n6bls3V8C9wFfl7QH+BHwiXZCNLOmFSZ+RPwHG8/T/MFmwzGzLixVX/1O7N68nz1M17q75/6fWLfenT7m7pnLCEUdBWbPX3TPvGqNXviMw/7ZGB/g6vxW+aNWiqEZQ4ihe+6ya5YgJ75Zgpz4Zgly4pslyBf3ClQdBAOaeBil3gWmJgYlLerAU2h1zrr8tb1BXEgbQgzdc4tvliAnvlmCnPhmCXKN34K+B5hYZBCM/D757Yv2X2QgDj/o0x+3+GYJcuKbJciJb5Yg1/iJqHrdoWj7RSbqzB9zd+UjWFPc4pslyIlvliAnvlmCXOOfF6br7Xnl9+xEntXuoS9S03fN/QLKc4tvliAnvlmCnPhmCXKNv5SKBoicrcfr1r9VrxHMG2wz9rVbc3dR058v1xHc4pslyIlvliAnvlmChlXj5+8VL2n91L569+TnKazZK56jjXp+CPX1stb0eW7xzRLkxDdLkBPfLEG91vhDqNmWU/s1fV477021CSv976M5bvHNEuTEN0tQYeJLeqekpyQ9L+mopP3Z+islPSnpuKSHJF3Qfrhm1oQyLf5PgZsi4v3ANcAtkq4HPg98MSJ2AK8De6qePCKmXkMgaeo1TDH1aiLmPn5vafpVvL02fS2yf9O6+Ds2cY7CxI+J/8sW35G9ArgJ+Ods/UHgtoUiMLPOlarxJW2R9BxwBjgM/AB4IyLOZpucBC7fYN+9ksaSxmtra03EbGY1lUr8iHgrIq4BtgHXAVfN22yDfQ9ExCgiRisrK4tHamaNqXRVPyLeAJ4ArgculHSuH8A24FSzoZlZWwo78EhaAX4WEW9IehfwISYX9h4HPg48CNwJPFw7mgE8pLMcgznU78BTdZ+qF5HKHD+/yf7JDaNf2Bf7Nj1mMzFNH2O/VjeNYbFzNKuJc5TpubcVOChpC5NPCF+PiEclvQQ8KOlvgWeB+2tHY2adKEz8iHgBuHbO+h8yqffNbMm4555ZgoY1EMdAOvG0rfk6sNrDLnOP0PADU/Pq79ljTi8Xjd1RtaZf5HeqWtMvK7f4Zgly4pslyIlvlqBh1fhWUvUJLusOtNHEdYlDO3dOLR9/6ZNTy03fM19ldWq5zN+lqC9BH4qubSzy3rjFN0uQE98sQU58swSddzW+B/Acxu88d9LMo5tPrLmPzevpot8rf7z89qv7VyvvU7h9yxOBQjvvp1t8swQ58c0S5MQ3S9DS1/hDHBCz7+sMi9zHb1qZ2nd1dXV6Rd1b5rnD7V/dP3ezzeT3mRkToIOavgtu8c0S5MQ3S5AT3yxBSz9pZuF91x7q7b7vo/d9/rLy9XPb79W84830Jdg3HdMQ++43wS2+WYKc+GYJcuKbJUhd1oOj0SjG43Fn57PF9XFtpO/+D/MMMabNjEYjxuNxYecWt/hmCXLimyXIiW+WoKXvq2/tqD2u/pzn8fN96ev2wVi2+ntI3OKbJciJb5YgJ75ZglzjWytmnrVntp97vh980Rj3RdvnLdLPPr9Pfmz+84VbfLMEOfHNElQ68SVtkfSspEez5SslPSnpuKSHJF3QXphm1qQqLf5dwLF1y58HvhgRO4DXgT1NBmZm7Sn1kI6kbcBB4O+AzwB/BKwBvx4RZyV9AFiNiA9vdhw/pJOOuRNqVByosuhCWxuTX+TPmZe/aJk/5tyOSxVjqGp9zF/my5yKU409pPMl4HPAz7Pl9wJvRMTZbPkkcHn5UM2sT4WJL+mjwJmIOLJ+9ZxN5/5XJmmvpLGk8dra2oJhmlmTyrT4NwAfk3QCeBC4ickngAslnesHsA04NW/niDgQEaOIGK2srDQQspnVVWkgDkk3An8RER+V9A3gmxHxoKR/Al6IiH/cbH/X+P05tHPn1PLuo0drHW+RB2Sq7lP3IZwy+9d9EKiN36lokpgSMbQ6EMfdwGckvcKk5r+/xrHMrEOVuuxGxBPAE9n3PwSuaz4kM2ube+6ZJajTwTb1Gwr+9JfL58sEhDar6H72PFX/PVS9T18mpqrHaHr7ujzYppltyIlvliAnvlmCPKGG2TpN35fvor/Deq7xzWxDTnyzBDnxzRLkwTYtaUX94osG7GzjGlnRdYImuMU3S5AT3yxBTnyzBLnGt8HoYxLM2pODFsTcxkSfm+0zGo1KHcMtvlmCnPhmCXLimyXINb4NRpfPjZxTt+bu4zpEE9cJ3OKbJciJb5YgJ75ZggZd4/dxXzcV/ttO1B2rv+rxyvTDd199M2uFE98sQU58swQNusZPte6sLl8DFv/d/LedqFqzF//dqr0XZd6HNt4rt/hmCXLimyXIiW+WICe+WYIGfXHPJoo726Rxoa6NTkdVJ8xgNbf/zCSY9TsE1Z3Eowy3+GYJKtXiSzoBvAm8BZyNiJGki4GHgO3ACeCTEfF6O2GaWZOqtPi/HxHXRMS5Qb3uAR6LiB3AY9mymS2BOjX+rcCN2fcHgSeAu2vGY7ahug/UND2wZVvnrGqRh3jKtvgB/KukI5L2Zusui4jTANnXSyuf3cx6UbbFvyEiTkm6FDgs6b/KniD7j2IvwPve974FQjSzppVq8SPiVPb1DPBt4DrgVUlbAbKvZzbY90BEjCJitLKy0kzUZlZLYYsv6deAt0XEm9n3fwD8DfAIcCdwX/b14TYDTZkfqFnMIn+3okkytT9Xw+/b/CGeqhNuzFN0jEVq/DIf9S8Dvp0d/O3AoYj4F0lPA1+XtAf4EfCJymc3s14UJn5E/BB4/5z1/wN8sI2gzKxd7rlnliD31TdbZ6amL+irX6SNyTA2O4YnzTSzDTnxzRLkxDdLkLq8RyxpDfhv4BLgtc5OvBjH2AzH2IyyMf5mRBT2lOs08X9xUmm87im/QXKMzXCMzWg6Rn/UN0uQE98sQX0l/oGezluFY2yGY2xGozH2UuObWb/8Ud8sQZ0mvqRbJL0s6RVJgxmjT9JXJJ2R9OK6dRdLOizpePb1oh7ju0LS45KOSToq6a4BxvhOSU9Jej6LcX+2/kpJT2YxPiTpgr5iXBfrFknPSnp0wDGekPQ9Sc9JGmfrGnu/O0t8SVuAfwD+ELgauF3S1V2dv8BXgVty64Y0mOhZ4LMRcRVwPfBn2d9uSDH+FLgpIt4PXAPcIul64PPAF7MYXwf29BjjOXcBx9YtDzFGaHOA24jo5AV8APjuuuV7gXu7On+J+LYDL65bfhnYmn2/FXi57xjXxfYwcPNQYwR+FXgG+F0mnU7ePu/fQE+xbcuS5ibgUSbT2w4qxiyOE8AluXWNvd9dftS/HPjxuuWT2bqhGuRgopK2A9cCTzKwGLOP0M8xGYbtMPAD4I2IOJttMoT3/EvA54CfZ8vvZXgxQssD3Hb5WO688YF8S6ECSe8Gvgl8OiJ+ssiQS22KiLeAayRdyGRsxqvmbdZtVL8k6aPAmYg4IunGc6vnbDqEf5cLD3BbRpct/knginXL24BTHZ6/qlKDiXZF0juYJP0DEfGtbPWgYjwnIt5gMs/C9cCFks41MH2/5zcAH8tmhnqQycf9LzGsGIF6A9yW0WXiPw3syK6gXgB8ismAnUN1bjBR6HkwUU2a9vuBYxHxhXU/GlKMK1lLj6R3AR9icgHtceDj2Wa9xhgR90bEtojYzuTf379FxB0MKEaYDHAr6T3nvmcywO2LNPl+d3zB4iPA95nUfn/V9wWUdXF9DTgN/IzJJ5M9TGq/x4Dj2deLe4zv95h8/HwBeC57fWRgMf4O8GwW44vAX2frfwt4CngF+AbwK32/31lcNwKPDjHGLJ7ns9fRc7nS5PvtnntmCXLPPbMEOfHNEuTEN0uQE98sQU58swQ58c0S5MQ3S5AT3yxB/w9JBSUbYOO/BAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-13.074828700467933, -10.763289536163864, -8.765460329708537, -9.267339286620953, -9.04095881817469, -8.883859292910845, -8.778789288811483, -8.917415604013874, -8.946643220838503, -8.960254864125508]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "run time =  17.694042205810547\n"
     ]
    }
   ],
   "source": [
    "from trunfreeDirichletMix import trunfreeInfiniteClusterDirichletMix as trunfreeDirichletMix\n",
    "start = time.time()\n",
    "state, vi, train = trunfreeDirichletMix(2, links_train,np.array(nodes),links_test)\n",
    "stop = time.time()\n",
    "print('run time = ',stop - start)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2 2 6 5 2 2 1 6 0 0 6 5 6 6 1 2 1 2 2 1 2 2 3 2 6 1 3 2 2 2 5 1 6 1 2 2 6\n",
      " 1 3 5 5 2 2 2 1 1 6 2 6 6 6 3 4 2 2 5 2 5 5 5 6 2 5 6 2 1 1 1 6 1 2 1 2 2\n",
      " 4 6 5 0 1 4 1 1 0 2 1 2 5 6 0 5 6 5 2 2 0 1 2 2 2 6 1 1 2 3 2 2 2 1 6 2 2\n",
      " 1 1 2 1 6 2 6 2 6 2 5 2 2 3 5 0 2 2 2 2 2 5 2 6 5 1 5 1 2 6 0 6 2 5 5 1 2\n",
      " 0 2 2 5 3 1 2 2 1 2 2 6 0 2 5 5 2 6 6 2 1 1 5 1 5 6 1 1 5 5 5 5 6 4 3 5 1\n",
      " 2 2 6 2 2 6 0 1 5 2 2 6 1 5 5 5 6 1 5 6 5 6 2 2 1 3 6 1 2 1 5 2 6 5 2 6 6\n",
      " 2 5 6 2 2 5 5 5 2 1 0 5 6 1 5 5 5 5 0 1 2 6 2 1 6 6 1 2 6 1 1 3 5 2 1 5 0\n",
      " 6 5 2 1 2 6 3 2 2 6 4 2 2 2 6 2 5 5 6 3 2 5 5 2 6 1 2 2 5 2 2 0 2 6 2 5 2\n",
      " 5 2 2 2 1 5 3 5 6 5 1 0 6 2 5 6 5 2 0 4 2 4 2 0 5 5 1 2 4 2 2 1 2 6 2 1 5\n",
      " 5 5 4 6 2 5 6 3 2 4 5 2 6 2 2 2 5 2 2 0 2 2 1 5 5 2 0 1 5 1 2 2 2 1 3 1 0\n",
      " 5 6 6 2 5 5 5 2 6 2 2 5 2 1 1 5 2 6 1 2 2 2 2 2 2 2 3 5 3 2 3 2 2 3 2 0 6\n",
      " 2 1 2 0 2 1 1 1 2 0 6 5 6 2 3 5 5 6 4 2 5 5 5 2 1 5 2 2 6 1 5 6 1 2 3 5 2\n",
      " 1 2 2 6 2 0]\n",
      "450\n",
      "450\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "trunfree_clusters = state['assignments'][train['best_validation']]\n",
    "print(trunfree_clusters)\n",
    "print(len(trunfree_clusters))\n",
    "print(len(unique_index))\n",
    "unique_clusters = np.array(trunfree_clusters)[unique_index]\n",
    "cluster_color = get_color(unique_clusters)\n",
    "node = sorted([i for i in pos.keys()])\n",
    "nx.draw_networkx_nodes(g, pos, nodelist=node, node_size=50, alpha=0.7)\n",
    "nx.draw_networkx_edges(g, pos, edgelist=unique_edges, edge_color=cluster_color, width=0.5,alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7fa07f20f048>]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sequential_ll = [-8.593266900610274, -8.366262149187582, -8.377490940758353, -8.180822158554474, -8.143191825671357,-8.143191825671357,-8.143191825671357,-8.143191825671357,-8.143191825671357,-8.143191825671357]\n",
    "truncation_ll = [-11.974750429074787, -9.874790432461978, -9.2317666626395, -9.461626901671144, -9.81523960477634, -9.902510734228745, -10.033196931010492, -9.178954625038891, -10.240917120981955, -9.247384152425981]\n",
    "plt.xticks(np.arange(0,10,1),np.arange(50,550,50))\n",
    "plt.plot(sequential_ll)\n",
    "plt.plot(truncation_ll)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}