analysis-predictions.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analyze predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import ast\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert strings into lists\n",
    "def literal_eval_column(column):\n",
    "    try:\n",
    "        return ast.literal_eval(column)\n",
    "    except (SyntaxError, ValueError):\n",
    "        return column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read distilbert predictions\n",
    "dist_test = pd.read_csv('outputs/distilbert_attributions.csv')\n",
    "dist_test = dist_test.applymap(literal_eval_column).drop(columns=['Unnamed: 0', 'index'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>attributions_pred</th>\n",
       "      <th>label_pred</th>\n",
       "      <th>score</th>\n",
       "      <th>tokens</th>\n",
       "      <th>label_gold</th>\n",
       "      <th>correct</th>\n",
       "      <th>sum_attributions_pred</th>\n",
       "      <th>mean_attributions_pred</th>\n",
       "      <th>attributions_gold</th>\n",
       "      <th>sum_attributions_gold</th>\n",
       "      <th>mean_attributions_gold</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[-0.011615987537462793, 0.003815996398574733, ...</td>\n",
       "      <td>rec.autos</td>\n",
       "      <td>0.801348</td>\n",
       "      <td>[i, am, a, little, confused, on, all, of, the,...</td>\n",
       "      <td>rec.autos</td>\n",
       "      <td>True</td>\n",
       "      <td>2.790767</td>\n",
       "      <td>0.025371</td>\n",
       "      <td>[-0.011615987537462793, 0.003815996398574733, ...</td>\n",
       "      <td>2.790767</td>\n",
       "      <td>0.025371</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[-0.32846125995080416, 0.021676696368081335, -...</td>\n",
       "      <td>comp.windows.x</td>\n",
       "      <td>0.915943</td>\n",
       "      <td>[i, ', m, not, familiar, at, all, with, the, f...</td>\n",
       "      <td>comp.windows.x</td>\n",
       "      <td>True</td>\n",
       "      <td>3.255837</td>\n",
       "      <td>0.018395</td>\n",
       "      <td>[-0.32846125995080416, 0.021676696368081335, -...</td>\n",
       "      <td>3.255837</td>\n",
       "      <td>0.018395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[-0.014673215540331483, 0.1312478060786364, 0....</td>\n",
       "      <td>alt.atheism</td>\n",
       "      <td>0.072379</td>\n",
       "      <td>[in, a, word, ,, yes, .]</td>\n",
       "      <td>alt.atheism</td>\n",
       "      <td>True</td>\n",
       "      <td>1.704919</td>\n",
       "      <td>0.284153</td>\n",
       "      <td>[-0.014673215540331483, 0.1312478060786364, 0....</td>\n",
       "      <td>1.704919</td>\n",
       "      <td>0.284153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[-0.6194980171999851, 0.0018658245313804055, 0...</td>\n",
       "      <td>talk.politics.mideast</td>\n",
       "      <td>0.808439</td>\n",
       "      <td>[they, were, attacking, the, iraqi, ##s, to, d...</td>\n",
       "      <td>talk.politics.mideast</td>\n",
       "      <td>True</td>\n",
       "      <td>4.223599</td>\n",
       "      <td>0.008282</td>\n",
       "      <td>[-0.6194980171999851, 0.0018658245313804055, 0...</td>\n",
       "      <td>4.223599</td>\n",
       "      <td>0.008282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[-0.08235069488343288, 0.03930023488145007, -0...</td>\n",
       "      <td>alt.atheism</td>\n",
       "      <td>0.626613</td>\n",
       "      <td>[i, ', ve, just, spent, two, solid, months, ar...</td>\n",
       "      <td>talk.religion.misc</td>\n",
       "      <td>False</td>\n",
       "      <td>1.368739</td>\n",
       "      <td>0.068437</td>\n",
       "      <td>[-0.09691212604112631, 0.011566433822874374, -...</td>\n",
       "      <td>1.375414</td>\n",
       "      <td>0.068771</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   attributions_pred             label_pred  \\\n",
       "0  [-0.011615987537462793, 0.003815996398574733, ...              rec.autos   \n",
       "1  [-0.32846125995080416, 0.021676696368081335, -...         comp.windows.x   \n",
       "2  [-0.014673215540331483, 0.1312478060786364, 0....            alt.atheism   \n",
       "3  [-0.6194980171999851, 0.0018658245313804055, 0...  talk.politics.mideast   \n",
       "4  [-0.08235069488343288, 0.03930023488145007, -0...            alt.atheism   \n",
       "\n",
       "      score                                             tokens  \\\n",
       "0  0.801348  [i, am, a, little, confused, on, all, of, the,...   \n",
       "1  0.915943  [i, ', m, not, familiar, at, all, with, the, f...   \n",
       "2  0.072379                           [in, a, word, ,, yes, .]   \n",
       "3  0.808439  [they, were, attacking, the, iraqi, ##s, to, d...   \n",
       "4  0.626613  [i, ', ve, just, spent, two, solid, months, ar...   \n",
       "\n",
       "              label_gold  correct  sum_attributions_pred  \\\n",
       "0              rec.autos     True               2.790767   \n",
       "1         comp.windows.x     True               3.255837   \n",
       "2            alt.atheism     True               1.704919   \n",
       "3  talk.politics.mideast     True               4.223599   \n",
       "4     talk.religion.misc    False               1.368739   \n",
       "\n",
       "   mean_attributions_pred                                  attributions_gold  \\\n",
       "0                0.025371  [-0.011615987537462793, 0.003815996398574733, ...   \n",
       "1                0.018395  [-0.32846125995080416, 0.021676696368081335, -...   \n",
       "2                0.284153  [-0.014673215540331483, 0.1312478060786364, 0....   \n",
       "3                0.008282  [-0.6194980171999851, 0.0018658245313804055, 0...   \n",
       "4                0.068437  [-0.09691212604112631, 0.011566433822874374, -...   \n",
       "\n",
       "   sum_attributions_gold  mean_attributions_gold  \n",
       "0               2.790767                0.025371  \n",
       "1               3.255837                0.018395  \n",
       "2               1.704919                0.284153  \n",
       "3               4.223599                0.008282  \n",
       "4               1.375414                0.068771  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dist_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read SVM predictions\n",
    "svm_test = pd.read_csv('outputs/coefs_test.csv')\n",
    "svm_test = svm_test.applymap(literal_eval_column).drop(columns=['Unnamed: 0'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# exclude erroneous indices\n",
    "with open('outputs/nohup-att-dist.out', 'r', encoding='utf-8') as f:\n",
    "    lines = f.readlines()\n",
    "error_indices = [int(l.split(' ')[0]) for l in lines[7:] if l]\n",
    "\n",
    "svm_test = svm_test[~svm_test.index.isin(error_indices)]\n",
    "svm_test = svm_test.reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>true class no</th>\n",
       "      <th>true class name</th>\n",
       "      <th>pred class no</th>\n",
       "      <th>pred class name</th>\n",
       "      <th>feature ind</th>\n",
       "      <th>tfidf</th>\n",
       "      <th>feature names</th>\n",
       "      <th>coef true</th>\n",
       "      <th>coef pred</th>\n",
       "      <th>coef true*tfidf</th>\n",
       "      <th>coef pred*tfidf</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7304</th>\n",
       "      <td>7527</td>\n",
       "      <td>14</td>\n",
       "      <td>sci.space</td>\n",
       "      <td>14</td>\n",
       "      <td>sci.space</td>\n",
       "      <td>[18607, 18295, 18294, 17756, 17294, 17279, 171...</td>\n",
       "      <td>[0.017691756398058683, 0.018650733635436213, 0...</td>\n",
       "      <td>[warm, varies, varied, try, things, thermal, t...</td>\n",
       "      <td>[-0.4585002959556874, -0.4052604615314347, -0....</td>\n",
       "      <td>[-0.4585002959556874, -0.4052604615314347, -0....</td>\n",
       "      <td>[-0.008111675544485832, -0.007558404920996732,...</td>\n",
       "      <td>[-0.008111675544485832, -0.007558404920996732,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7305</th>\n",
       "      <td>7528</td>\n",
       "      <td>4</td>\n",
       "      <td>comp.sys.mac.hardware</td>\n",
       "      <td>4</td>\n",
       "      <td>comp.sys.mac.hardware</td>\n",
       "      <td>[17316, 10666, 9725, 6231, 5108, 1304]</td>\n",
       "      <td>[0.14779138117244353, 0.2403323623065925, 0.20...</td>\n",
       "      <td>[thought, macs, jim, edu, day, actually]</td>\n",
       "      <td>[-0.4994580224796735, -0.15024876263583953, -0...</td>\n",
       "      <td>[-0.4994580224796735, -0.15024876263583953, -0...</td>\n",
       "      <td>[-0.0738155909799283, -0.036109640057913804, -...</td>\n",
       "      <td>[-0.0738155909799283, -0.036109640057913804, -...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7306</th>\n",
       "      <td>7529</td>\n",
       "      <td>9</td>\n",
       "      <td>rec.sport.baseball</td>\n",
       "      <td>9</td>\n",
       "      <td>rec.sport.baseball</td>\n",
       "      <td>[18639, 18501, 18373, 17986, 17316, 17294, 156...</td>\n",
       "      <td>[0.020255864930211472, 0.029267291360587883, 0...</td>\n",
       "      <td>[wasn, void, version, understood, thought, thi...</td>\n",
       "      <td>[-0.40429532821330094, -0.3673492016941159, -0...</td>\n",
       "      <td>[-0.40429532821330094, -0.3673492016941159, -0...</td>\n",
       "      <td>[-0.00818935156020414, -0.010751316117061054, ...</td>\n",
       "      <td>[-0.00818935156020414, -0.010751316117061054, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7307</th>\n",
       "      <td>7530</td>\n",
       "      <td>6</td>\n",
       "      <td>misc.forsale</td>\n",
       "      <td>12</td>\n",
       "      <td>sci.electronics</td>\n",
       "      <td>[18939, 18226, 18177, 18175, 17392, 16625, 166...</td>\n",
       "      <td>[0.01227474898275108, 0.028237249554259176, 0....</td>\n",
       "      <td>[work, uv, used, use, timer, stuff, student, r...</td>\n",
       "      <td>[-0.6091581217981357, -0.4379001261747615, -0....</td>\n",
       "      <td>[-0.31283400980582565, -0.1472211033427624, -0...</td>\n",
       "      <td>[-0.007477263035876224, -0.01236509514263832, ...</td>\n",
       "      <td>[-0.003839958943634, -0.004157119034742961, -0...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7308</th>\n",
       "      <td>7531</td>\n",
       "      <td>15</td>\n",
       "      <td>soc.religion.christian</td>\n",
       "      <td>15</td>\n",
       "      <td>soc.religion.christian</td>\n",
       "      <td>[18939, 18428, 18399, 16809, 16395, 15324, 139...</td>\n",
       "      <td>[0.025833554574752265, 0.05359619747264651, 0....</td>\n",
       "      <td>[work, viewpoint, viable, support, start, scri...</td>\n",
       "      <td>[-0.4113355308195925, -0.4774639320789409, -0....</td>\n",
       "      <td>[-0.4113355308195925, -0.4774639320789409, -0....</td>\n",
       "      <td>[-0.010626258883962635, -0.025590251189769198,...</td>\n",
       "      <td>[-0.010626258883962635, -0.025590251189769198,...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      index  true class no         true class name  pred class no  \\\n",
       "7304   7527             14               sci.space             14   \n",
       "7305   7528              4   comp.sys.mac.hardware              4   \n",
       "7306   7529              9      rec.sport.baseball              9   \n",
       "7307   7530              6            misc.forsale             12   \n",
       "7308   7531             15  soc.religion.christian             15   \n",
       "\n",
       "             pred class name  \\\n",
       "7304               sci.space   \n",
       "7305   comp.sys.mac.hardware   \n",
       "7306      rec.sport.baseball   \n",
       "7307         sci.electronics   \n",
       "7308  soc.religion.christian   \n",
       "\n",
       "                                            feature ind  \\\n",
       "7304  [18607, 18295, 18294, 17756, 17294, 17279, 171...   \n",
       "7305             [17316, 10666, 9725, 6231, 5108, 1304]   \n",
       "7306  [18639, 18501, 18373, 17986, 17316, 17294, 156...   \n",
       "7307  [18939, 18226, 18177, 18175, 17392, 16625, 166...   \n",
       "7308  [18939, 18428, 18399, 16809, 16395, 15324, 139...   \n",
       "\n",
       "                                                  tfidf  \\\n",
       "7304  [0.017691756398058683, 0.018650733635436213, 0...   \n",
       "7305  [0.14779138117244353, 0.2403323623065925, 0.20...   \n",
       "7306  [0.020255864930211472, 0.029267291360587883, 0...   \n",
       "7307  [0.01227474898275108, 0.028237249554259176, 0....   \n",
       "7308  [0.025833554574752265, 0.05359619747264651, 0....   \n",
       "\n",
       "                                          feature names  \\\n",
       "7304  [warm, varies, varied, try, things, thermal, t...   \n",
       "7305           [thought, macs, jim, edu, day, actually]   \n",
       "7306  [wasn, void, version, understood, thought, thi...   \n",
       "7307  [work, uv, used, use, timer, stuff, student, r...   \n",
       "7308  [work, viewpoint, viable, support, start, scri...   \n",
       "\n",
       "                                              coef true  \\\n",
       "7304  [-0.4585002959556874, -0.4052604615314347, -0....   \n",
       "7305  [-0.4994580224796735, -0.15024876263583953, -0...   \n",
       "7306  [-0.40429532821330094, -0.3673492016941159, -0...   \n",
       "7307  [-0.6091581217981357, -0.4379001261747615, -0....   \n",
       "7308  [-0.4113355308195925, -0.4774639320789409, -0....   \n",
       "\n",
       "                                              coef pred  \\\n",
       "7304  [-0.4585002959556874, -0.4052604615314347, -0....   \n",
       "7305  [-0.4994580224796735, -0.15024876263583953, -0...   \n",
       "7306  [-0.40429532821330094, -0.3673492016941159, -0...   \n",
       "7307  [-0.31283400980582565, -0.1472211033427624, -0...   \n",
       "7308  [-0.4113355308195925, -0.4774639320789409, -0....   \n",
       "\n",
       "                                        coef true*tfidf  \\\n",
       "7304  [-0.008111675544485832, -0.007558404920996732,...   \n",
       "7305  [-0.0738155909799283, -0.036109640057913804, -...   \n",
       "7306  [-0.00818935156020414, -0.010751316117061054, ...   \n",
       "7307  [-0.007477263035876224, -0.01236509514263832, ...   \n",
       "7308  [-0.010626258883962635, -0.025590251189769198,...   \n",
       "\n",
       "                                        coef pred*tfidf  \n",
       "7304  [-0.008111675544485832, -0.007558404920996732,...  \n",
       "7305  [-0.0738155909799283, -0.036109640057913804, -...  \n",
       "7306  [-0.00818935156020414, -0.010751316117061054, ...  \n",
       "7307  [-0.003839958943634, -0.004157119034742961, -0...  \n",
       "7308  [-0.010626258883962635, -0.025590251189769198,...  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_test.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "assert len(svm_test) == len(dist_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def merge_subwords(tokens, attributions, aggregate='mean'):\n",
    "    assert len(tokens) == len(attributions)\n",
    "    # skip merged tokens\n",
    "    skip_next_iteration = 0\n",
    "    tokens_merged = []\n",
    "    attributions_merged = []\n",
    "    for i, token in enumerate(tokens):\n",
    "        if skip_next_iteration != 0:\n",
    "            skip_next_iteration -= 1\n",
    "            continue\n",
    "        if token:\n",
    "            word = token\n",
    "            attribution = attributions[i]\n",
    "            # detokenize\n",
    "                # check whether next token starts with ##\n",
    "            if i < len(tokens)-1 and tokens[i+1].startswith('##'):\n",
    "                # check whether further tokens start with ##\n",
    "                to_merge = tokens[i:]\n",
    "                count = 1\n",
    "                iterator = iter(to_merge[1:])\n",
    "                while True:\n",
    "                    try:\n",
    "                        token = next(iterator)\n",
    "                        if token.startswith(\"##\"):\n",
    "                            count += 1\n",
    "                        else:\n",
    "                            break\n",
    "                    except StopIteration:\n",
    "                        break\n",
    "                to_merge = to_merge[:count]\n",
    "                j = len(to_merge)\n",
    "                word = ''.join(tokens[i:i+j]).replace('##', '')\n",
    "                attr = [attributions[i:i+j]]\n",
    "                if aggregate == 'mean':\n",
    "                    attribution = np.mean(attr)\n",
    "                skip_next_iteration = count-1\n",
    "        tokens_merged.append(word)\n",
    "        attributions_merged.append(attribution)\n",
    "    assert len(tokens_merged) == len(attributions_merged)\n",
    "    return tokens_merged, attributions_merged\n",
    "\n",
    "# features (SVM) != tokens (DistilBERT)\n",
    "def compare_pred(i, print_result=True, mode='correl'):\n",
    "    row_svm = svm_test.iloc[i]\n",
    "    row_dist = dist_test.iloc[i]\n",
    "    # initialize lists for common features and their attribution/coeff values\n",
    "    feats = []\n",
    "    val_svm_gold = []\n",
    "    val_dist_gold = []\n",
    "    val_svm_pred = []\n",
    "    val_dist_pred = []\n",
    "    # merge subwords and aggregate subword attributions\n",
    "    tokens, attributions_pred = merge_subwords(row_dist.tokens, row_dist['attributions_pred'])\n",
    "    tokens2, attributions_gold = merge_subwords(row_dist.tokens, row_dist['attributions_gold'])\n",
    "    assert tokens == tokens2  # just in case\n",
    "    for k, f in enumerate(tokens):\n",
    "        # count tokens only once\n",
    "        if f in feats:\n",
    "            continue\n",
    "        # iterate tokens, check if in SVM features\n",
    "        if f in row_svm['feature names']:\n",
    "            feats.append(f)\n",
    "            svm_index = row_svm['feature names'].index(f)\n",
    "            val_svm_gold.append(row_svm['coef true*tfidf'][svm_index])\n",
    "            val_svm_pred.append(row_svm['coef pred*tfidf'][svm_index])\n",
    "            # feature more than once in sentence\n",
    "            if tokens.count(f) > 1:\n",
    "                indices = [j for j, token in enumerate(tokens) if token == f]\n",
    "                val_dist_gold.append(np.mean([attributions_gold[j] for j in indices]))\n",
    "                val_dist_pred.append(np.mean([attributions_pred[j] for j in indices]))\n",
    "            else:\n",
    "                # feature index in distilbert tokens\n",
    "                val_dist_gold.append(attributions_gold[k])\n",
    "                val_dist_pred.append(attributions_pred[k])\n",
    "    assert len(feats) == len(val_dist_gold)\n",
    "    assert len(val_dist_gold) == len(val_dist_pred)\n",
    "    assert len(val_dist_pred) == len(val_svm_gold)\n",
    "    assert len(val_svm_gold) == len(val_svm_pred)\n",
    "    correl_gold = np.corrcoef(val_dist_gold, val_svm_gold)\n",
    "    correl_pred = np.corrcoef(val_dist_pred, val_svm_pred)\n",
    "    if print_result:\n",
    "        print(feats)\n",
    "        print(correl_gold)\n",
    "        print(correl_pred)\n",
    "    if mode == 'correl':  # return correlations information\n",
    "        return len(feats), len(tokens), correl_gold[0][1], correl_pred[0][1], feats\n",
    "    elif mode == 'feats':  # return features, coefficients, attributions\n",
    "        return feats, tokens, val_dist_gold, val_dist_pred, val_svm_gold, val_svm_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n"
     ]
    }
   ],
   "source": [
    "all_dist_gold = []\n",
    "all_dist_pred = []\n",
    "all_svm_gold = []\n",
    "all_svm_pred = []\n",
    "\n",
    "for i in range(len(svm_test)):\n",
    "    if not svm_test.iloc[i]['feature ind']:  # exclude instances wo features\n",
    "        continue\n",
    "    _, _, val_dist_gold, val_dist_pred, val_svm_gold, val_svm_pred = compare_pred(i, print_result=False, mode='feats')\n",
    "    all_dist_gold.extend(val_dist_gold)\n",
    "    all_dist_pred.extend(val_dist_pred)\n",
    "    all_svm_gold.extend(val_svm_gold)\n",
    "    all_svm_pred.extend(val_svm_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_values = pd.DataFrame({'all_dist_gold': all_dist_gold,\n",
    "                           'all_dist_pred': all_dist_pred,\n",
    "                           'all_svm_gold': all_svm_gold,\n",
    "                           'all_svm_pred': all_svm_pred})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>all_dist_gold</th>\n",
       "      <th>all_dist_pred</th>\n",
       "      <th>all_svm_gold</th>\n",
       "      <th>all_svm_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>all_dist_gold</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.795057</td>\n",
       "      <td>0.037089</td>\n",
       "      <td>0.024632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>all_dist_pred</th>\n",
       "      <td>0.795057</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.026069</td>\n",
       "      <td>-0.005021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>all_svm_gold</th>\n",
       "      <td>0.037089</td>\n",
       "      <td>-0.026069</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.945506</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>all_svm_pred</th>\n",
       "      <td>0.024632</td>\n",
       "      <td>-0.005021</td>\n",
       "      <td>0.945506</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               all_dist_gold  all_dist_pred  all_svm_gold  all_svm_pred\n",
       "all_dist_gold       1.000000       0.795057      0.037089      0.024632\n",
       "all_dist_pred       0.795057       1.000000     -0.026069     -0.005021\n",
       "all_svm_gold        0.037089      -0.026069      1.000000      0.945506\n",
       "all_svm_pred        0.024632      -0.005021      0.945506      1.000000"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# analyze all concatenated attributions\n",
    "all_values.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>all_dist_gold</th>\n",
       "      <th>all_dist_pred</th>\n",
       "      <th>all_svm_gold</th>\n",
       "      <th>all_svm_pred</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>303459.000000</td>\n",
       "      <td>303459.000000</td>\n",
       "      <td>303459.000000</td>\n",
       "      <td>303459.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.027565</td>\n",
       "      <td>0.030366</td>\n",
       "      <td>-0.009247</td>\n",
       "      <td>-0.009001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.103340</td>\n",
       "      <td>0.103726</td>\n",
       "      <td>0.012641</td>\n",
       "      <td>0.011655</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-0.997724</td>\n",
       "      <td>-0.996871</td>\n",
       "      <td>-0.556641</td>\n",
       "      <td>-0.398256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.000397</td>\n",
       "      <td>0.000261</td>\n",
       "      <td>-0.010934</td>\n",
       "      <td>-0.010779</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.006847</td>\n",
       "      <td>0.007377</td>\n",
       "      <td>-0.005900</td>\n",
       "      <td>-0.005831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.025827</td>\n",
       "      <td>0.026989</td>\n",
       "      <td>-0.003532</td>\n",
       "      <td>-0.003493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.988448</td>\n",
       "      <td>0.988448</td>\n",
       "      <td>0.345678</td>\n",
       "      <td>0.345678</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       all_dist_gold  all_dist_pred   all_svm_gold   all_svm_pred\n",
       "count  303459.000000  303459.000000  303459.000000  303459.000000\n",
       "mean        0.027565       0.030366      -0.009247      -0.009001\n",
       "std         0.103340       0.103726       0.012641       0.011655\n",
       "min        -0.997724      -0.996871      -0.556641      -0.398256\n",
       "25%        -0.000397       0.000261      -0.010934      -0.010779\n",
       "50%         0.006847       0.007377      -0.005900      -0.005831\n",
       "75%         0.025827       0.026989      -0.003532      -0.003493\n",
       "max         0.988448       0.988448       0.345678       0.345678"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_values.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "classification = {'0': 'both_true',\n",
    " '1': 'svm_true_dist_false',\n",
    " '2': 'dist_true_svm_false',\n",
    " '3': 'both_false_same_label',\n",
    " '4': 'both_false_diff_label'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "classification_results = []\n",
    "\n",
    "for i, row in svm_test.iterrows():\n",
    "    svm_class, dist_class = False, False\n",
    "    if row['true class no'] == row['pred class no']:\n",
    "        svm_class = True\n",
    "    if dist_test.iloc[i].correct:\n",
    "        dist_class = True\n",
    "    if svm_class:\n",
    "        if dist_class:  # both correct\n",
    "            classification_results.append(0)\n",
    "        else:\n",
    "            classification_results.append(1)\n",
    "    elif dist_class and not svm_class:\n",
    "        classification_results.append(2)\n",
    "    else:  # both false\n",
    "        if row['pred class name'] == dist_test.iloc[i]['label_pred']:\n",
    "            classification_results.append(3)\n",
    "        else:  # different labels\n",
    "            classification_results.append(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 4588\n",
      "1 635\n",
      "2 667\n",
      "3 687\n",
      "4 732\n"
     ]
    }
   ],
   "source": [
    "for i in range(5):\n",
    "    print(i, classification_results.count(i))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:518: RuntimeWarning: Mean of empty slice.\n",
      "  avg = a.mean(axis, **keepdims_kw)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\core\\_methods.py:182: RuntimeWarning: invalid value encountered in divide\n",
      "  ret = um.true_divide(\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2845: RuntimeWarning: Degrees of freedom <= 0 for slice\n",
      "  c = cov(x, y, rowvar, dtype=dtype)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: divide by zero encountered in divide\n",
      "  c *= np.true_divide(1, fact)\n",
      "c:\\Users\\lydia\\anaconda3\\lib\\site-packages\\numpy\\lib\\function_base.py:2704: RuntimeWarning: invalid value encountered in multiply\n",
      "  c *= np.true_divide(1, fact)\n"
     ]
    }
   ],
   "source": [
    "correls = pd.DataFrame(columns=['num_feats', 'num_tokens', 'correl_gold', 'correl_pred', 'features'])\n",
    "\n",
    "num_instances = len(svm_test)\n",
    "\n",
    "for i in range(num_instances):\n",
    "    correls.loc[len(correls)] = list(compare_pred(i, print_result=False))\n",
    "\n",
    "correls['classification_result'] = classification_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>num_feats</th>\n",
       "      <th>num_tokens</th>\n",
       "      <th>correl_gold</th>\n",
       "      <th>correl_pred</th>\n",
       "      <th>features</th>\n",
       "      <th>classification_result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5340</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[simple, eh]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6923</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[discussion, tesla]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6876</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[lou, gehrig]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[1993apr21, ca]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6633</th>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[really, 10]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7148</th>\n",
       "      <td>2</td>\n",
       "      <td>14</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[compared, os]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5916</th>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[dear, come]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5737</th>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[sub, says]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5510</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[urban, areas]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5415</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[agree, joe]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5234</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[abolish, cults]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7213</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[answered, question]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5076</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[does, bbs]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5258</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[25mhz, iisi]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1391</th>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[goto, flag]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3713</th>\n",
       "      <td>2</td>\n",
       "      <td>15</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[easy, decision]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3686</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[look, spice]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3668</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[clint, happened]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3338</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[suggest, change]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4000</th>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[ok, posted]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4213</th>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[say, expensive]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4084</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[ego, trip]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5164</th>\n",
       "      <td>2</td>\n",
       "      <td>21</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[course, bypass]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6265</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[test, works]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>672</th>\n",
       "      <td>2</td>\n",
       "      <td>15</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[near, stuff]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4802</th>\n",
       "      <td>2</td>\n",
       "      <td>19</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[certainly, error]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5453</th>\n",
       "      <td>3</td>\n",
       "      <td>13</td>\n",
       "      <td>-0.999540</td>\n",
       "      <td>0.999515</td>\n",
       "      <td>[church, did, better]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6556</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.999390</td>\n",
       "      <td>-0.809133</td>\n",
       "      <td>[mark, ira, writes]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>902</th>\n",
       "      <td>3</td>\n",
       "      <td>20</td>\n",
       "      <td>-0.999312</td>\n",
       "      <td>-0.996260</td>\n",
       "      <td>[interested, send, info]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3235</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>-0.994879</td>\n",
       "      <td>-0.264902</td>\n",
       "      <td>[hopefully, doesnt, personal]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2359</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>-0.991949</td>\n",
       "      <td>0.940144</td>\n",
       "      <td>[did, regularly, michael]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1468</th>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>-0.990922</td>\n",
       "      <td>0.304497</td>\n",
       "      <td>[gee, forgotten, nut]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4393</th>\n",
       "      <td>3</td>\n",
       "      <td>17</td>\n",
       "      <td>-0.990272</td>\n",
       "      <td>0.998406</td>\n",
       "      <td>[davidians, black, question]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6500</th>\n",
       "      <td>3</td>\n",
       "      <td>14</td>\n",
       "      <td>-0.986599</td>\n",
       "      <td>-0.986599</td>\n",
       "      <td>[sc300, inline, correct]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2775</th>\n",
       "      <td>4</td>\n",
       "      <td>22</td>\n",
       "      <td>-0.981637</td>\n",
       "      <td>-0.987305</td>\n",
       "      <td>[called, taking, picture, leaf]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7182</th>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>-0.977233</td>\n",
       "      <td>-0.988311</td>\n",
       "      <td>[currently, 240sx, known]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4699</th>\n",
       "      <td>4</td>\n",
       "      <td>15</td>\n",
       "      <td>-0.969600</td>\n",
       "      <td>-0.926789</td>\n",
       "      <td>[diluted, gallon, swimming, pool]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3075</th>\n",
       "      <td>3</td>\n",
       "      <td>15</td>\n",
       "      <td>-0.967043</td>\n",
       "      <td>-0.590251</td>\n",
       "      <td>[remove, legend, list]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1434</th>\n",
       "      <td>4</td>\n",
       "      <td>19</td>\n",
       "      <td>-0.943217</td>\n",
       "      <td>-0.858022</td>\n",
       "      <td>[use, tool, yes, newer]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5741</th>\n",
       "      <td>4</td>\n",
       "      <td>17</td>\n",
       "      <td>-0.940684</td>\n",
       "      <td>0.850130</td>\n",
       "      <td>[lord, hope, hoover, pro]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      num_feats  num_tokens  correl_gold  correl_pred  \\\n",
       "5340          2           4    -1.000000    -1.000000   \n",
       "6923          2          11    -1.000000    -1.000000   \n",
       "6876          2           4    -1.000000     1.000000   \n",
       "100           2          21    -1.000000     1.000000   \n",
       "6633          2          12    -1.000000     1.000000   \n",
       "7148          2          14    -1.000000     1.000000   \n",
       "5916          2          13    -1.000000     1.000000   \n",
       "5737          2           7    -1.000000    -1.000000   \n",
       "5510          2           6    -1.000000    -1.000000   \n",
       "5415          2          11    -1.000000    -1.000000   \n",
       "5234          2           3    -1.000000     1.000000   \n",
       "7213          2          11    -1.000000     1.000000   \n",
       "5076          2           6    -1.000000    -1.000000   \n",
       "5258          2           5    -1.000000    -1.000000   \n",
       "1391          2           9    -1.000000    -1.000000   \n",
       "3713          2          15    -1.000000     1.000000   \n",
       "3686          2           5    -1.000000    -1.000000   \n",
       "3668          2          11    -1.000000    -1.000000   \n",
       "3338          2           6    -1.000000     1.000000   \n",
       "4000          2           7    -1.000000     1.000000   \n",
       "4213          2           9    -1.000000    -1.000000   \n",
       "4084          2           5    -1.000000    -1.000000   \n",
       "5164          2          21    -1.000000     1.000000   \n",
       "6265          2          11    -1.000000    -1.000000   \n",
       "672           2          15    -1.000000    -1.000000   \n",
       "4802          2          19    -1.000000    -1.000000   \n",
       "5453          3          13    -0.999540     0.999515   \n",
       "6556          3           4    -0.999390    -0.809133   \n",
       "902           3          20    -0.999312    -0.996260   \n",
       "3235          3          10    -0.994879    -0.264902   \n",
       "2359          3          10    -0.991949     0.940144   \n",
       "1468          3          10    -0.990922     0.304497   \n",
       "4393          3          17    -0.990272     0.998406   \n",
       "6500          3          14    -0.986599    -0.986599   \n",
       "2775          4          22    -0.981637    -0.987305   \n",
       "7182          3          15    -0.977233    -0.988311   \n",
       "4699          4          15    -0.969600    -0.926789   \n",
       "3075          3          15    -0.967043    -0.590251   \n",
       "1434          4          19    -0.943217    -0.858022   \n",
       "5741          4          17    -0.940684     0.850130   \n",
       "\n",
       "                               features  classification_result  \n",
       "5340                       [simple, eh]                      3  \n",
       "6923                [discussion, tesla]                      1  \n",
       "6876                      [lou, gehrig]                      1  \n",
       "100                     [1993apr21, ca]                      3  \n",
       "6633                       [really, 10]                      4  \n",
       "7148                     [compared, os]                      2  \n",
       "5916                       [dear, come]                      1  \n",
       "5737                        [sub, says]                      4  \n",
       "5510                     [urban, areas]                      2  \n",
       "5415                       [agree, joe]                      4  \n",
       "5234                   [abolish, cults]                      4  \n",
       "7213               [answered, question]                      4  \n",
       "5076                        [does, bbs]                      4  \n",
       "5258                      [25mhz, iisi]                      1  \n",
       "1391                       [goto, flag]                      4  \n",
       "3713                   [easy, decision]                      1  \n",
       "3686                      [look, spice]                      1  \n",
       "3668                  [clint, happened]                      4  \n",
       "3338                  [suggest, change]                      4  \n",
       "4000                       [ok, posted]                      3  \n",
       "4213                   [say, expensive]                      4  \n",
       "4084                        [ego, trip]                      4  \n",
       "5164                   [course, bypass]                      4  \n",
       "6265                      [test, works]                      4  \n",
       "672                       [near, stuff]                      4  \n",
       "4802                 [certainly, error]                      4  \n",
       "5453              [church, did, better]                      3  \n",
       "6556                [mark, ira, writes]                      2  \n",
       "902            [interested, send, info]                      4  \n",
       "3235      [hopefully, doesnt, personal]                      4  \n",
       "2359          [did, regularly, michael]                      4  \n",
       "1468              [gee, forgotten, nut]                      4  \n",
       "4393       [davidians, black, question]                      1  \n",
       "6500           [sc300, inline, correct]                      0  \n",
       "2775    [called, taking, picture, leaf]                      2  \n",
       "7182          [currently, 240sx, known]                      4  \n",
       "4699  [diluted, gallon, swimming, pool]                      2  \n",
       "3075             [remove, legend, list]                      4  \n",
       "1434            [use, tool, yes, newer]                      4  \n",
       "5741          [lord, hope, hoover, pro]                      4  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correls.sort_values(by='correl_gold').head(n=40)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>num_feats</th>\n",
       "      <th>num_tokens</th>\n",
       "      <th>correl_gold</th>\n",
       "      <th>correl_pred</th>\n",
       "      <th>features</th>\n",
       "      <th>classification_result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2543</th>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>0.995785</td>\n",
       "      <td>0.995785</td>\n",
       "      <td>[works, really, bike, tony]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6024</th>\n",
       "      <td>3</td>\n",
       "      <td>11</td>\n",
       "      <td>0.996526</td>\n",
       "      <td>0.022986</td>\n",
       "      <td>[apparently, second, edition]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5847</th>\n",
       "      <td>4</td>\n",
       "      <td>18</td>\n",
       "      <td>0.999237</td>\n",
       "      <td>0.999237</td>\n",
       "      <td>[isn, point, better, windows]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1332</th>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "      <td>0.999408</td>\n",
       "      <td>0.121503</td>\n",
       "      <td>[works, great, far]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5809</th>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>0.999857</td>\n",
       "      <td>-0.999560</td>\n",
       "      <td>[thanks, bunch, nate]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6565</th>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>0.999980</td>\n",
       "      <td>0.999980</td>\n",
       "      <td>[pierre, turgeon, islanders]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4057</th>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[post, news]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>489</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[stuff, deleted]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6855</th>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[evening, sky]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3851</th>\n",
       "      <td>2</td>\n",
       "      <td>15</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[carried, away]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7079</th>\n",
       "      <td>2</td>\n",
       "      <td>18</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[liar, cheat]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5441</th>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[isbn, 13]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4420</th>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[does, status]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1848</th>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[project, arizona]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6768</th>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[use, instead]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[word, yes]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6264</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[trusted, tools]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5813</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[afford, harley]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[dog, thrust]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3278</th>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[number, legs]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3295</th>\n",
       "      <td>2</td>\n",
       "      <td>95</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[speak, hear]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7235</th>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[conner, meet]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2885</th>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[resource, issue]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3447</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[3do, thanks]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3609</th>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[got, reversed]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1303</th>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[thing, heater]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1671</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[caching, drive]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3127</th>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[compuserve, com]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1147</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[alt, religion]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5498</th>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[okay, forgot]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3747</th>\n",
       "      <td>2</td>\n",
       "      <td>14</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[outrage, dog]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>782</th>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[non, cold]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2460</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[325, 577]</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3302</th>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[easy, poke]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6614</th>\n",
       "      <td>2</td>\n",
       "      <td>14</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[let, clarify]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2778</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>[shut, andi]</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6801</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[opened, distribution]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5082</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[lets, forget]</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>751</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[getting, close]</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>[flat, engine]</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      num_feats  num_tokens  correl_gold  correl_pred  \\\n",
       "2543          4          10     0.995785     0.995785   \n",
       "6024          3          11     0.996526     0.022986   \n",
       "5847          4          18     0.999237     0.999237   \n",
       "1332          3          12     0.999408     0.121503   \n",
       "5809          3           6     0.999857    -0.999560   \n",
       "6565          3           5     0.999980     0.999980   \n",
       "4057          2           7     1.000000    -1.000000   \n",
       "489           2           6     1.000000    -1.000000   \n",
       "6855          2          12     1.000000     1.000000   \n",
       "3851          2          15     1.000000    -1.000000   \n",
       "7079          2          18     1.000000     1.000000   \n",
       "5441          2           8     1.000000    -1.000000   \n",
       "4420          2          10     1.000000    -1.000000   \n",
       "1848          2           7     1.000000     1.000000   \n",
       "6768          2           9     1.000000     1.000000   \n",
       "2             2           6     1.000000    -1.000000   \n",
       "6264          2          11     1.000000     1.000000   \n",
       "5813          2          11     1.000000     1.000000   \n",
       "268           2           6     1.000000     1.000000   \n",
       "3278          2          17     1.000000     1.000000   \n",
       "3295          2          95     1.000000    -1.000000   \n",
       "7235          2           7     1.000000    -1.000000   \n",
       "2885          2           5     1.000000     1.000000   \n",
       "3447          2          11     1.000000     1.000000   \n",
       "3609          2           9     1.000000    -1.000000   \n",
       "1303          2          10     1.000000    -1.000000   \n",
       "1671          2           4     1.000000     1.000000   \n",
       "3127          2           7     1.000000    -1.000000   \n",
       "1147          2           6     1.000000    -1.000000   \n",
       "5498          2          11     1.000000    -1.000000   \n",
       "3747          2          14     1.000000     1.000000   \n",
       "782           2          13     1.000000    -1.000000   \n",
       "2460          2           6     1.000000    -1.000000   \n",
       "3302          2          13     1.000000    -1.000000   \n",
       "6614          2          14     1.000000     1.000000   \n",
       "2778          2           4     1.000000    -1.000000   \n",
       "6801          2           6     1.000000     1.000000   \n",
       "5082          2           3     1.000000     1.000000   \n",
       "751           2           6     1.000000     1.000000   \n",
       "235           2          12     1.000000     1.000000   \n",
       "\n",
       "                           features  classification_result  \n",
       "2543    [works, really, bike, tony]                      0  \n",
       "6024  [apparently, second, edition]                      4  \n",
       "5847  [isn, point, better, windows]                      0  \n",
       "1332            [works, great, far]                      4  \n",
       "5809          [thanks, bunch, nate]                      4  \n",
       "6565   [pierre, turgeon, islanders]                      0  \n",
       "4057                   [post, news]                      4  \n",
       "489                [stuff, deleted]                      4  \n",
       "6855                 [evening, sky]                      0  \n",
       "3851                [carried, away]                      4  \n",
       "7079                  [liar, cheat]                      3  \n",
       "5441                     [isbn, 13]                      4  \n",
       "4420                 [does, status]                      4  \n",
       "1848             [project, arizona]                      0  \n",
       "6768                 [use, instead]                      2  \n",
       "2                       [word, yes]                      2  \n",
       "6264               [trusted, tools]                      3  \n",
       "5813               [afford, harley]                      0  \n",
       "268                   [dog, thrust]                      0  \n",
       "3278                 [number, legs]                      4  \n",
       "3295                  [speak, hear]                      4  \n",
       "7235                 [conner, meet]                      4  \n",
       "2885              [resource, issue]                      0  \n",
       "3447                  [3do, thanks]                      0  \n",
       "3609                [got, reversed]                      4  \n",
       "1303                [thing, heater]                      4  \n",
       "1671               [caching, drive]                      4  \n",
       "3127              [compuserve, com]                      4  \n",
       "1147                [alt, religion]                      4  \n",
       "5498                 [okay, forgot]                      4  \n",
       "3747                 [outrage, dog]                      3  \n",
       "782                     [non, cold]                      4  \n",
       "2460                     [325, 577]                      3  \n",
       "3302                   [easy, poke]                      4  \n",
       "6614                 [let, clarify]                      4  \n",
       "2778                   [shut, andi]                      1  \n",
       "6801         [opened, distribution]                      4  \n",
       "5082                 [lets, forget]                      2  \n",
       "751                [getting, close]                      4  \n",
       "235                  [flat, engine]                      0  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correls.sort_values(by='correl_gold', na_position='first').tail(n=40)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-> highest and lowest correlations for fewest features / shortest sentences!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.2826740357767589, 0.25590147310534667)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# slightly higher correlation of gold class attributions/coefficients\n",
    "np.mean(correls.correl_gold), np.mean(correls.correl_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>num_feats</th>\n",
       "      <th>num_tokens</th>\n",
       "      <th>correl_gold</th>\n",
       "      <th>correl_pred</th>\n",
       "      <th>classification_result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>num_feats</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.944143</td>\n",
       "      <td>-0.017452</td>\n",
       "      <td>0.023974</td>\n",
       "      <td>-0.220609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>num_tokens</th>\n",
       "      <td>0.944143</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.039823</td>\n",
       "      <td>0.004972</td>\n",
       "      <td>-0.206382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>correl_gold</th>\n",
       "      <td>-0.017452</td>\n",
       "      <td>-0.039823</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.620412</td>\n",
       "      <td>-0.376070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>correl_pred</th>\n",
       "      <td>0.023974</td>\n",
       "      <td>0.004972</td>\n",
       "      <td>0.620412</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.380150</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>classification_result</th>\n",
       "      <td>-0.220609</td>\n",
       "      <td>-0.206382</td>\n",
       "      <td>-0.376070</td>\n",
       "      <td>-0.380150</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       num_feats  num_tokens  correl_gold  correl_pred  \\\n",
       "num_feats               1.000000    0.944143    -0.017452     0.023974   \n",
       "num_tokens              0.944143    1.000000    -0.039823     0.004972   \n",
       "correl_gold            -0.017452   -0.039823     1.000000     0.620412   \n",
       "correl_pred             0.023974    0.004972     0.620412     1.000000   \n",
       "classification_result  -0.220609   -0.206382    -0.376070    -0.380150   \n",
       "\n",
       "                       classification_result  \n",
       "num_feats                          -0.220609  \n",
       "num_tokens                         -0.206382  \n",
       "correl_gold                        -0.376070  \n",
       "correl_pred                        -0.380150  \n",
       "classification_result               1.000000  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# correlations not related to numbers of features or tokens in a sentence\n",
    "correls.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "label2id = {\n",
    "    \"alt.atheism\": 0,\n",
    "    \"comp.graphics\": 1,\n",
    "    \"comp.os.ms-windows.misc\": 2,\n",
    "    \"comp.sys.ibm.pc.hardware\": 3,\n",
    "    \"comp.sys.mac.hardware\": 4,\n",
    "    \"comp.windows.x\": 5,\n",
    "    \"misc.forsale\": 6,\n",
    "    \"rec.autos\": 7,\n",
    "    \"rec.motorcycles\": 8,\n",
    "    \"rec.sport.baseball\": 9,\n",
    "    \"rec.sport.hockey\": 10,\n",
    "    \"sci.crypt\": 11,\n",
    "    \"sci.electronics\": 12,\n",
    "    \"sci.med\": 13,\n",
    "    \"sci.space\": 14,\n",
    "    \"soc.religion.christian\": 15,\n",
    "    \"talk.politics.guns\": 16,\n",
    "    \"talk.politics.mideast\": 17,\n",
    "    \"talk.politics.misc\": 18,\n",
    "    \"talk.religion.misc\": 19\n",
    "  }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "dist_label_pred = [label2id[l] for l in dist_test['label_pred']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.        , 0.71812724],\n",
       "       [0.71812724, 1.        ]])"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# correlations\n",
    "# compare predicted labels, have to ignore nan values with ma\n",
    "np.corrcoef(np.ma.masked_invalid(svm_test['pred class no']),\n",
    "            np.ma.masked_invalid(dist_label_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# do some nice visualizations\n",
    "def barplot(feats, coeffs, attributions, outpath):\n",
    "    \n",
    "    # Define the x-axis positions for the tokens\n",
    "    y = np.arange(len(feats))\n",
    "\n",
    "    # Create the barplot\n",
    "    plt.figure(figsize=(10, 6))  # Adjust the figure size as needed\n",
    "    plt.barh(y + 0.175, attributions[::-1], label='DistilBERT attributions', color='red', alpha=0.7, height=0.35)\n",
    "    plt.barh(y - 0.175, coeffs[::-1], label='SVM coefficients', color='blue', alpha=0.7, height=0.35)\n",
    "\n",
    "    # Add labels and title\n",
    "    plt.ylabel('Tokens', fontsize=14)\n",
    "    plt.xlabel('Attributions and Coefficients', fontsize=14)\n",
    "    # plt.title('Attributions Comparison')\n",
    "    plt.yticks(y, feats[::-1], fontsize=14)  # Set token labels on the x-axis\n",
    "    plt.legend(fontsize=14)  # Show the legend\n",
    "    plt.savefig(outpath, bbox_inches='tight')\n",
    "\n",
    "def plot(i, target='gold'):\n",
    "    feats, tokens, val_dist_gold, val_dist_pred, val_svm_gold, val_svm_pred = \\\n",
    "        compare_pred(i, print_result=False, mode='feats')\n",
    "    if target == 'gold':\n",
    "        barplot(feats, val_svm_gold, val_dist_gold, outpath=f'outputs/viz/instance_{i}.png')\n",
    "    elif target == 'pred':\n",
    "        barplot(feats, val_svm_pred, val_dist_pred, outpath=f'outputs/viz/instance_{i}_pred.png')\n",
    "    elif target == 'both':\n",
    "        barplot_both(feats, val_dist_gold, val_dist_pred, val_svm_gold, val_svm_pred, outpath=f'outputs/viz/instance_{i}_both.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def barplot_both(feats, attributions_gold, attributions_pred, coeffs_gold, coeffs_pred, outpath):\n",
    "    \n",
    "    # Define the y-axis positions for the tokens\n",
    "    y = np.arange(len(feats))\n",
    "    height = 0.35/2  # Height of the bars\n",
    "\n",
    "    # Create the horizontal stacked barplot\n",
    "    plt.figure(figsize=(10, 6))  # Adjust the figure size as needed\n",
    "\n",
    "    # Attributions (red)\n",
    "    plt.barh(y + height * 1.5, attributions_gold[::-1], height, label='Gold Attributions', color='red', alpha=0.7)\n",
    "    plt.barh(y + height * 0.5, attributions_pred[::-1], height, label='Pred Attributions', color='lightcoral', alpha=0.7)\n",
    "\n",
    "    # Coefficients (blue)\n",
    "    plt.barh(y - height * 0.5, coeffs_gold[::-1], height, label='Gold Coefficients', color='blue', alpha=0.7)\n",
    "    plt.barh(y - height * 1.5, coeffs_pred[::-1], height, label='Pred Coefficients', color='lightblue', alpha=0.7)\n",
    "\n",
    "\n",
    "    # Add labels and title\n",
    "    plt.xlabel('Attributions and Coefficients', fontsize=14)\n",
    "    plt.ylabel('Tokens', fontsize=14)\n",
    "    plt.yticks(y, feats[::-1], fontsize=14)  # Set token labels on the y-axis\n",
    "    plt.legend(fontsize=14)  # Show the legend\n",
    "    plt.tight_layout()\n",
    "\n",
    "    # Save the plot to the specified output path\n",
    "    plt.savefig(outpath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(2694)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(4013)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(6210, target='gold')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(678, target='both')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(None, None)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA54AAAISCAYAAACkihDwAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAABx00lEQVR4nO3deXxN1/7/8ffJHGRAJMQUpTXPY4tKUNFqil41lZo7o1e1pUqStoYqqpfqoMbWWFJtTKWImZprpiqGmGpKVCoS9u8Pv5yv45yQRHYS9Xo+HnlcZ+219/nsc3bcvq2917IYhmEIAAAAAACTOOV0AQAAAACAfzeCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMJVLTheAB8/Nmzd16tQpeXl5yWKx5HQ5AAAAAHKIYRi6cuWKAgMD5eSU9rgmwRMZdurUKRUvXjynywAAAACQS5w4cULFihVLczvBExnm5eUl6dbF5e3tncPVAAAAAMgpCQkJKl68uDUjpIXgiQxLvb3W29ub4AkAAADgno/gMbkQAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAEzFrLbIFsnJybpx40ZOlwHgX8zZ2Vmurq45XQYAAHCA4AlTJSQk6Pz580pKSsrpUgA8BNzd3eXn58dSTwAA5DIET5gmISFBcXFxypcvn/z8/OTq6nrP9X0AIDMMw1BycrLi4+MVFxcnSYRPAAByEYInTHP+/Hnly5dPxYoVI3ACMJ2np6e8vLx08uRJnT9/nuAJAEAuwuRCMEVycrKSkpLk4+ND6ASQbSwWi3x8fJSUlKTk5OScLgcAAPx/BE+YInUiISb6AJDdUv/eYUIzAAByD4InTMVoJ4Dsxt87AADkPgRPAAAAAICpCJ4AAAAAAFMRPIFcJiYmRhaLRREREaYc32KxKDg42KYtIiJCFotFMTExprwn7s3R99K1a1dZLBbFxsbmSE1mX4sAAODhQfAETBAbGyuLxWLzkydPHgUGBqpJkyYaMmSIjhw5Ysp7BwcHZ9kzbneeg8Vikaenp8qWLau3335bf/31V5rvf7efnTt3Wvunht47P6tKlSpp0KBBSkhIkPR/ISi9P3eGuPt1r3AeFBSkoKCgLH3P7GDGZwUAAHAn1vFEzgkLy+kK0hYdnSWHKV26tDp16iRJSkpK0rlz5/Tbb7/po48+0rBhw/Tuu+9q6NChNkGxTp062r9/v/z8/LKkhjvt379fefLkSXf/ggUL6s0337S+vnDhgmJiYjRmzBj99NNP2rFjh7y8vOz2e/vtt5UvXz6HxyxcuLBd23/+8x9VqlRJknTmzBktWbJEw4YN08KFC/Xbb78pKChI4eHhNvvExsZq2rRpqlq1qlq1amWz7UELgRn9XrKD2dciAAB4eBA8AROVKVPG4W2Ka9eu1UsvvaThw4fL2dlZH330kXVbnjx5VK5cOdNqyuix/fz87M7BMAyFhYVp0aJFmjdvnrp162a3X//+/R0GzLS0adNG7du3t76+du2a6tWrp127dmnmzJnq1q2bXR0xMTGaNm2aqlWr9sDfDmrmd55ZZl+LAADg4UHwBHJAw4YN9csvv6hKlSoaOXKkXn75ZRUvXlzSrTAVEhKi8PBwmzB1+PBhDRs2TDExMTp9+rTy5cunEiVKKCQkRKNHj5Zku4zE7X/u0qWLpk6dam1v1KjRfT3PabFYFBoaqkWLFjm83TYreHh46MUXX9SuXbu0bds2h+H2fsTHx+urr77S4sWLdfjwYZ0/f15+fn5q2rSpwsPDVbp0aWvf4OBgrV69WpIUEhJibS9ZsqRiYmJUqlQpa9vtn3vqd3j7dxoaGqqIiAht3rxZ8fHxMgzDul9a38uNGzc0fPhwTZo0SSdPnlTx4sX18ssv6+2335aT0/89MTF16lR169ZNU6ZMUdeuXW2Oced1lfpaklavXm1Td+r+aV2LkrR37159+OGHWrVqleLj4xUYGKhWrVpp8ODBKlCggE3f1NHnvXv3avDgwZozZ47Onz+vsmXLasiQIWrTpo3ddzN69GjNmzdPx48fl7Ozs4oUKaLHH39cH374ofV3Bf8CufnOl4dBFt3dAwDpQfAEcshjjz2mdu3aafr06VqwYIF69+6dZt9Tp06pTp06unr1qlq0aKF27drp77//1uHDhzVu3Dhr8AwPD9fUqVN17Ngxm9tSq1WrluX1L1++XJJUo0aNLD92qtRQ5uKS9X9V7d+/X0OGDFFISIhat26tvHnz6sCBA5o5c6YWLVqk7du3q2TJkpJkDXGrV69Wly5drEHK19dXvr6+Cg8P19ixYyVJb731lvU97nx2csOGDRo2bJhCQkL08ssv6/jx4+mq9a233tKmTZvUtm1beXh4KCoqSu+++67++OMPff3115k6/9RblyMjI1WyZEmboHqv62XDhg1q1qyZkpKS1KZNGwUFBWnTpk0aO3asFi1apI0bN6pgwYI2+yQnJ6tZs2a6ePGinn/+eSUmJmr27Nlq27atli5dqmbNmkm69Z2HhoZq8+bNql+/vpo3by4nJyfFxsbqxx9/VJcuXQieAAA8gAieQA5q1KiRpk+fri1btty13/z583X58mV9/vnn6tOnj8228+fPW/+cOpJ17NixLLv19Pz58zbHunTpkmJiYrRv3z717dtXTZs2dbjfqFGjHD7j6eHhoQEDBtzzff/55x99//33kqQGDRpkrvi7KF++vE6fPm03Ordq1So1bdpUH3/8sSZOnCjpVvCMjY3V6tWr1bVrV4ezAqeOKN/tc1++fLkmTZqk7t27Z6jWLVu2aNeuXQoMDJQkRUZG6oknntA333yjTp06qWHDhhk6nnQreEZERCgyMtL65/S4efOmunbtqqtXr2rp0qUKDQ21bnv//fc1fPhwvffee/r2229t9jt16pRq166tVatWyc3NTZLUsWNHNW3aVGPGjLEGzz179mjz5s1q3bq1oqKibI6RlJSk5OTkDJ8rAADIeQRPIAelBonbw+PdeHp62rWZPfHLhQsXFBkZadfesGFDm2cy75Q6CnsnHx8fh8Fz3rx5OnDggCTp7NmzWrhwoU6ePKmWLVvq+eefz2T1afPx8XHYHhISoooVK+rXX3/N8vesXr16hkOnJPXp08d6rUhSvnz5NGTIEL3wwguaNm1apoJnZq1fv16HDx/W008/bRM6JWnQoEH65ptvNHPmTE2YMMEaMFN99tlnNm1NmjRRyZIlHf7Di6Nr3d3dXe7u7ll0JgAAIDuxnAqQg1JvJb2XZ599Vnny5NEbb7yhtm3bavLkyTp06JDJ1d1StmxZGYZh/blw4YIWLlyouLg4BQcHa+3atQ73O336tM1+qT+XL1922H/+/PmKjIxUZGSkvvrqK508eVLPP/+8fvzxR5vnGLNSTEyMWrVqpSJFisjV1dW6FMvu3bt16tSpLH+/OnXqZGo/R8Eyte32pWmyw44dOyTZ30YsSXnz5lWtWrX0zz//2F2fvr6+Ns/CpipWrJjNNVG+fHlVrlxZM2fO1JNPPqkxY8Zoy5YtunHjRpaeBwAAyF4ETyAHnT59WpJUqFChu/YrVaqUNm7cqLCwMC1ZskQ9evRQ2bJlVa5cOf3www/ZUapVgQIF1KJFC02cOFFJSUl2S5xk1qxZs2QYhpKTk7Vnzx41b95cUVFRGjJkSJYc/04//PCDGjdurJUrV6pBgwZ66623NGTIEIWHh6tkyZK6fv16lr9nQEBApvbz9/d32Obk5KT4+Pj7LStDUtdVTetcUmcyvrOutEaYXVxcdPPmTZvXK1eu1BtvvKE//vhDb7/9turUqaPChQvrww8/JIACAPCA4lZbIAelzmBau3bte/atUqWK5s+fr+TkZG3btk1LlizR//73P7Vr106BgYGqX7++ydXaSh292759e5Ye18XFRRUrVtSPP/6oypUra+jQoWrdunWWT2IUEREhDw8Pbdu2TY8++qjNttmzZ2fpe6W6febYjDh37pzKli1r13bz5k2bQJc6MpySkmJ3jKwKqN7e3pJu3Q7tSGp7ar/M8PPz0/jx4zVu3DgdOHBAK1eu1Lhx4xQeHi5XV1cNHDgw08cGAAA5gxFPIIccOnRIc+fOlbu7u1q3bp3u/VxdXVWvXj1FRkbqf//7nwzD0MKFC63bnZ2dJcn0kaGLFy9Kks1oVVby8PDQqFGjZBhGuiYjyqgjR46ofPnydqHz1KlTOnLkiF3/e32uzs7Opn3mjm5nTm27fQba/PnzS5Li4uLs+qfeInsnJyenDNVdvXp1SXK47EtiYqK2bt0qT09Pu6CcGRaLReXLl9cbb7xhnUX5559/vu/jAgCA7EfwBHLAunXrFBoaqqSkJA0cOFBFixa9a/8tW7bo3Llzdu2po0u3T8SSOkvryZMns7Bie6nLh5g5sU3Lli1Vo0YNLV++PM1nSTOrZMmS+uOPP2xG7q5du6bXXnvN4YjhvT7XAgUK6Pz587p27VqW1ilJ//vf/2yeOf3777/14YcfSpJeeukla3uNGjVksVg0e/ZsmzoOHz6szz//PM26M3Kt1K9fX6VLl9aSJUvsJmAaPny4zp8/rw4dOthNLJReR48e1b59++zaHV3rAADgwcGttoCJ/vjjD+syFdevX9e5c+e0efNm7dmzR87Ozvrggw/S9QzjjBkzNGHCBAUHB6tMmTLy9vbWvn37tHjxYvn5+dnMlNq4cWPNmzdPL7zwgp555hl5eHiocuXKatGiRabOwdFyKhs3btSWLVvk4+OjTz75xOF+aS2nIkmtWrVK99qiEREReu655zRkyBCtWrUqo+WnqXfv3urdu7eqV6+uNm3aKCUlRcuXL5dhGKpatap27dpl0z8kJEQWi0WDBg3SgQMH5OPjIx8fH7322muSbn3uW7duVVhYmBo2bCg3Nzc1aNAgS5aCqV27tqpWrap27drJ3d1dUVFRio2NVa9evfTkk09a+xUtWlTt2rXT7NmzVbNmTTVv3lznzp3Tjz/+qObNm2v+/Pl2x27cuLHmzp2rNm3aqHr16nJ2dlaLFi1UuXJlh7U4OTlp6tSpCg0N1TPPPKMXXnhBJUuW1ObNm7Vy5UqVLl1aI0aMyPS57tq1S61bt1bt2rVVqVIlFS5cWHFxcVqwYIGcnZ319ttvZ/rYAAAg5xA8ARMdOXLEuhSJp6enfH19Va5cOQ0ePFhdunRR6dKl03WcDh066Nq1a1q/fr22bNmipKQkFStWTG+88Yb69++vYsWKWfv26tVLsbGxmj17toYOHaqUlBR16dIl08HzzuVU3NzcVLRoUb388ssaOHCggoKCHO6X1nIq0q01JNMbPMPCwlSrVi3FxMRo5cqVaty4cUbKT9Mbb7whV1dXjRs3ThMnTpSvr69atGihYcOGqW3btnb9K1SooClTpmj06NH67LPPlJSUpJIlS1qD5+DBg3Xp0iUtXLhQK1eu1M2bNxUeHp4lwXPs2LGaO3euvv32W508eVLFixfXJ5984jCETZo0SYUKFdLcuXP1xRdfqGzZsvrmm28UGBjoMHimjoSuXLlSP/74o27evKnChQunGTylW+uqbtq0SR9++KGWLVum+Ph4BQYGqk+fPho8ePB9LfFTq1YtDRgwQDExMVq0aJEuX76swoULq1mzZnrnnXcyPTMwAADIWRYjves5AP9fQkKCfHx8FB8fn+YEIteuXdPRo0dVqlQpeXh4ZHOFAB5m/P3zAAkLy+kKHm7R0TldAYB/gfRkA4lnPAEAAAAAJiN4AgAAAABMRfB8yEVERMhisThcGgEAAAAAsgLBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIntkkJiZGFotFERER2rhxo0JDQ+Xr6yuLxSJJMgxDkydPVv369eXt7a08efKoVq1amjx5st2xTp06pfDwcNWrV0/+/v5yd3dXUFCQXn/9dZ07d87h+584cUIdOnRQgQIFlC9fPjVq1Ehr1qwx9ZwBAAAAQJJccrqAh82GDRs0bNgwhYSE6OWXX9bx48dlGIY6deqkmTNn6rHHHlPHjh3l5uam5cuXq0ePHtq3b59GjRplPcaaNWs0evRoNWnSRHXr1pWrq6t27NihL7/8Ur/88ou2b98uHx8fa//Tp0/r8ccfV1xcnEJDQ1WjRg3t379fTz31lEJCQnLiYwAAAADwECF4ZrPly5dr0qRJ6t69u7Vt4sSJmjlzpnr06KGvvvpKLi63vpbr16+rTZs2Gj16tDp06KCaNWtKkho3bqwzZ84oX758NseePn26unTpovHjx2vQoEHW9oEDByouLk4ff/yxTfs333yjV155xczTBQAAAAButc1u1atXtwmdkjR+/HjlzZtX48ePt4ZOSXJzc9PQoUMlSbNmzbK2+/v724VOSercubO8vb3166+/WtuuX7+uOXPmyN/fX2+//bZN/549e+qxxx67Z81JSUlKSEiw+QEAAACA9GLEM5vVqVPH5nViYqJ2796twMBAjRgxwq5/cnKyJOnAgQM27VFRUfr666+1fft2Xbp0STdu3LBuO3XqlPXPBw8e1LVr19S4cWN5eHjYHMPJyUlPPPGEDh06dNeahw8frsjIyPSdIAAgVwgLy+kK7i06OjqnSwAAZBOCZzYLCAiweX3p0iUZhqG4uLi7hrurV69a/zx69Gj1799fhQoVUrNmzVSsWDF5enpKksaOHaukpCRr3/j4eEm3RknTU48jAwcOVL9+/ayvExISVLx48XvuBwAAAAASwTPbpc5im8rb21uSVLNmTW3duvWe+6ekpOijjz5SYGCgdu7cqUKFClm3GYahkSNH2vRPnWQordluz549e8/3dHd3l7u7+z37AQAAAIAjPOOZw7y8vFS+fHnt379fly9fvmf/8+fPKz4+XvXq1bMJnZK0detW/fPPPzZtZcuWlYeHh7Zu3apr167ZbLt586Y2bNhw3+cAZIfvv/9e1apVU758+axLE6X67LPPVL58eXl6espisWjq1KmKjY2VxWJR165dM/2eQUFBCgoKuu/aAQAAHnaMeOYCffr00WuvvaZevXpp6tSpyps3r832o0ePymKxKCgoSP7+/vL09NT27duVmJioPHnySLp1y27v3r3tju3m5qa2bdtq+vTpGj16tM2stt9+++09n+80U25+/igrHjtKTEzU2LFjNW/ePB06dEgpKSny8/NTqVKl1KBBA/Xs2VOlS5fWL7/8oubNmys0NFRLly696zH/85//KCoqSvPnz9fzzz+viIgI6y3a7733nsPnhCWpX79++uyzzyTdemZ3wIAB93+C2WjDhg3q3LmzHn30Ub3xxhvy9PRUcHCwJGnmzJnq16+fatSoof/+979yc3NTtWrVcrTe+xETE6OQkBCFh4fbhGsAAIAHGcEzF3jllVe0adMmTZs2TevXr1fTpk0VGBios2fP6sCBA9q8ebNmzpypoKAgOTk56fXXX9fo0aNVtWpVhYWFKSEhQUuWLFHJkiUVGBhod/wRI0ZoxYoV+uCDD7Ru3TpVr15d+/fv1+LFi9WsWTMtW7YsB8763+3KlStq0KCBfv/9d5UpU0adOnWSr6+vTpw4ob1792rEiBEqXbq0SpcuraeeekolSpTQ8uXLdfLkSRUrVszhMc+fP6/o6GgVKlRIYXekdhcXF02fPl1Dhw6Vs7Ozzbbk5GR9//33cnFxUUpKimnnbKbFixdLurVkUL169RxuW7RokQoXLmxtT05O1v79+23WtM2oFStWZHpfAAAA/B9utc0FUm8NnDNnjipWrKiFCxdqzJgxWr58uTw8PDRq1Cg1bdrU2n/48OEaOnSoLBaLJkyYoOXLl6t9+/ZatmyZXF1d7Y5fpEgRbdiwQe3atdOmTZv0+eef68KFC1q+fLkef/zx7DzVh8bYsWP1+++/q0ePHjp06JC++uorjRgxQjNmzNDOnTv1559/qkGDBpJuzS7ctWtX3bx5U9OmTUvzmN99952Sk5P10ksv2X3PTz/9tE6fPq0lS5bY7RcdHa2//vpLzzzzTNaeZDZKnan59mB5r22urq4qV66cihQpkun3Tf3HAQAAANwfgmc2CQ4OlmEYd711rm3btlq+fLkuXryo69ev6+TJk1q1apX69esnPz8/az9XV1e9//77OnTokK5du6Zjx45p9OjRypcvn2JjYxUbG2t37BIlSmj27Nm6dOmSrl69qjVr1ujJJ59URESEDMOw3raIrLFx40ZJ0ptvvmk3oZQklSpVSuXKlbO+7tatm/UfINIyZcoUSbJbB1aSnn/+efn6+mry5Ml22yZPnqxChQrp2Wefzehp6MqVK/rwww9VpUoV5c2bVz4+PqpevboGDx5sXeon1YYNG9SiRQsVKFBAHh4eKleunCIiIpSYmOjw2EePHlXPnj1VokQJubu7q0iRIuratauOHTtm7RMTEyOLxWI991KlSslisVg/K4vFolWrVkmStT31mcy7PeOZ3vNK6xlPwzA0efJk1a9fX97e3sqTJ49q1arl8POPiIiQxWJRTEyM5s6dqxo1asjT01NFihRRnz59bJ7LjoiIUEhIiCQpMjLSek4Wi8X6ex0fH68hQ4aoQoUKypcvn3x8fFSuXDl169ZNJ06ccPhZAwAA5DRutQVMUKBAAUnSH3/8ka7nDYOCgtSkSRP9+uuv1n8UuN3WrVu1e/duPf7446pQoYLd/h4eHmrfvr0mTZqkv/76yzrx1KlTp7R06VL16dPH4Wj43Zw/f16NGjXSvn37VK1aNb366qu6efOmDhw4oE8++URvv/22fH19JUnz589X+/bt5ebmpnbt2snf31+//vqrIiMjtWzZMq1atcpmZuTNmzcrNDRUV69eVVhYmMqUKaPY2FjNmDFDS5Ys0caNG/XII48oKChI4eHhWrBggXbt2qW+ffta37NatWoKDw/X1KlTdezYMYWHh0uSdXtWnJcjhmGoU6dOmjlzph577DF17NhRbm5uWr58uXr06KF9+/Zp1KhRdvt98cUXWrJkiVq2bKng4GAtXbpU48aN04ULFzRjxgxJt/6BKjY2VtOmTVOjRo1s/kHI19dXhmEoNDRUmzdvVv369dW8eXM5OTkpNjZWP/74o7p06cJSRwAAIFcieAImeOGFFzRjxgz16NFDW7duVbNmzVS9enXlz58/zX169OihX3/9VVOmTLELnqkjaT169Ehz/+7du+urr77S999/r//+97+SpGnTpunGjRvq3r17upbrud3rr7+uffv26f3339fQoUNttp09e1b58uWTdGv0sGfPnnJ2dtbGjRtVpUoVSbYB7dNPP9UHH3wg6dazl+3bt9fNmze1detWVa1a1XrcdevWKTg4WH379lV0dLSCgoIUERGh2NhY7dq1S2+99ZbNCGS1atUUExOjY8eOpXsinvSeV1q+/fZbzZw5Uz169NBXX30lF5dbf41ev35dbdq00ejRo9WhQwfVrFnTZr/ly5dr27ZtKlu2rCRp6NChqlatmmbNmqVPP/1UgYGB1qA5bdo0BQcH253T7t27tXnzZrVu3VpRUVE225KSkuxGoQEAAHILbrUFTNCyZUuNHDlSN2/e1CeffKImTZqoQIECKlOmjN58800dPnzYbp/WrVsrf/78+uGHH/T3339b269du6ZZs2Ypb968ateuXZrvWbt2bVWuXNnmds+pU6eqdu3aqlSpUobqP3v2rObNm6fSpUs7DHQBAQHWwLVgwQJdvnxZ3bt3t4ZO6datryNGjJCLi4vNLcQLFy5UbGys3n33XZvQKUkNGjRQy5YttXjxYiUkJGSo5qw+r7SMHz9eefPm1fjx4236urm5WYPsrFmz7Pbr27evNXRKkqenpzp06CDDMLRt27YMnYenp6ddm7u7+z1DMwAAQE5hxBMwyTvvvKNXX31VS5cu1YYNG7R161Zt3rxZX3zxhSZNmqQ5c+boueees/Z3d3fXiy++qPHjx2vu3LnWZzmjoqJ0+fJldevW7Z7Bolu3burXr5+2bNmia9eu6dChQ/ryyy8zXPvWrVtlGIZCQkLueYvujh07JMnhc8LFixdX6dKldfDgQV25ckVeXl7atGmTJOnAgQMOw9+ZM2d08+ZNHTp0SLVq1cpw7XeTkfNyJDExUbt371ZgYKDDpWtSRxwPHDhgt61GjRp2bakzGKdnDV9JKl++vCpXrqyZM2fqxIkTatWqlRo2bKgaNWrYzWYMAACQmxA8ARN5eXnphRde0AsvvCDp1sQw77//viZMmKAePXooLi5Obm5u1v49evTQ+PHjNWXKFGvwTM9ttqk6deqk9957T5MnT9a1a9esz35mVGoQKlq06D37po5MBgQEONxeuHBhHTx4UAkJCfLy8tLFixclyfpcY1quXr2agYrTJyPn5cilS5dkGIbi4uKs66c64qh2R8u6pI6Y3rhxI13v7+LiopUrVyoiIkJRUVF6++23JUl+fn7q3bu3Bg0aRAAFAAC5ErfaAtnIx8dH48ePV8mSJXX+/Hnt3r3bZnu1atVUo0YNrVu3TocOHdLx48e1cuVKlS1bVvXr17/n8VNnr501a5Z++OEH62y3GZW6T1xc3D37ent7S7p1G6sjqe2p/VL/Nzo6WoZhpPnTqFGjDNd9Lxk5L0dSa69Zs+Zda0+dadcMfn5+Gj9+vOLi4rRv3z6NHz9eBQsWVHh4uEaOHGna+wIAANwPgieQzSwWi/LkyZPm9tSRzalTp2rKlCkyDCNdo52punfvrvj4eF29etXh0ivpUatWLTk5OWnVqlX3nLCmevXqkm4tfXKnuLg4HTlyRI888oi8vLwkSXXr1pX0f0vOZKeMnJcjXl5eKl++vPbv35/u22MzKnXE8l6joBaLReXLl9cbb7yh5cuXS5J+/vlnU2oCAAC4XwRPwARff/21tmzZ4nBbVFSUDhw4IF9fX4eT/nTs2FEeHh6aPn26pk6dKhcXF7300kvpfu+nn35aCxYs0IIFC9S4ceNM1R8QEKD//Oc/OnLkiMNbSs+dO6eUlBRJtyZS8vHx0ZQpU7R3715rH8MwNHDgQCUnJ9uspdmyZUuVKFFCY8aM0Zo1a+yOnZycrHXr1mWq7nvJyHmlpU+fPkpMTFSvXr0c3lJ79OhRh2vpplfqUjwnT550eOx9+/bZtaeOKjuadAgAACA34BlPwARLlizRq6++qjJlyqh+/foKDAzU33//rZ07d2rt2rVycnLShAkTbNa2TOXr66v//Oc/1mcgW7Vqlebzk444OzurZcuW930OEyZM0J49ezR06FAtXrxYjRs3lmEYOnTokJYtW6azZ8/K19dX3t7emjhxojp06KC6deuqXbt2KlSokFasWKGtW7eqTp06euedd6zHdXd317x58/T000+rUaNGatKkiTWAHz9+XGvXrlXBggUdTtCTFdJ7Xml55ZVXtGnTJk2bNk3r169X06ZNFRgYqLNnz+rAgQPavHmzZs6cabPsS0aUK1dOgYGBmj17tvLkyaNixYrJYrHotdde065du9S6dWvrTMWFCxdWXFycFixYIGdnZ+sznwAAALkNwRMwwSeffKL69etr+fLlWrNmjU6fPi3p1qQ2Xbp0Ue/eve3Webxdjx49rMEzs7fL3i8/Pz9t2rRJo0aN0g8//KDx48fLw8NDpUqV0oABA5Q3b15r3xdeeEGFCxfW8OHDFRUVpcTERAUFBWnw4MF677335OHhYXPs2rVra9euXfr000+1ePFirVu3Tu7u7ipatKhatWqlDh065IrzcsRisWjq1Kl65plnNHHiRC1cuFB///23/P399eijj2rUqFFq2rRpputzdnZWVFSU3nvvPX333Xe6cuWKJKl9+/aqVauWBgwYoJiYGC1atEiXL19W4cKF1axZM73zzjuqU6dOpt8XAADATBbDMIycLgIPloSEBPn4+Cg+Pt462cqdrl27pqNHj6pUqVJ2oQMAzMTfP7eEheV0BfcWHZ3TFQAA7ld6soHEM54AAAAAAJMRPAEAAAAApiJ4AgAAAABMxeRCAAD8C/H8JAAgN2HEEwAAAABgKoInAAAAAMBUBE8AAAAAgKkInjAVy8QCyG78vQMAQO5D8IQpnJ2dJUnJyck5XAmAh03q3zupfw8BAICcR/CEKVxdXeXu7q74+HhGHwBkG8MwFB8fL3d3d7m6uuZ0OQAA4P9jORWYxs/PT3FxcTp58qR8fHzk6uoqi8WS02UB+BcyDEPJycmKj4/X33//raJFi+Z0SQAA4DYET5jG29tbknT+/HnFxcXlcDUAHgbu7u4qWrSo9e8fAACQOxA8YSpvb295e3srOTlZN27cyOlyAPyLOTs7c3stAAC5FMET2cLV1ZX/IAQAAAAeUkwuBAAAAAAwFcETAAAAAGAqgicAAAAAwFQETwAAAACAqQieAAAAAABTETwBAAAAAKYieAIAAAAATMU6ngCA3CksLKcrgNmio3O6AgBANmHEEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGAqgicAAAAAwFQETwAAAACAqQieAAAAAABTETwBAAAAAKYieAIAAAAATEXwzEIWi0XBwcGmv8/BgwfVsmVLBQQEyGKxKCgoKFPHmTp1qiwWi6ZOnZql9QEAAADA7VxyugBkzI0bN9S6dWv9+eef6ty5s4oWLSpfX9+cLgsAAAAA0kTwfMAcPXpU+/fv1yuvvKKvvvoqp8sBAAAAgHviVtsHzKlTpyRJhQsXzuFKAAAAACB9clXwvHjxopydndWqVSub9i1btshischisejkyZM22+rWrSsvLy+lpKRIklJSUvTZZ5+patWq8vT0lI+Pj0JCQrRo0SK797v9GccVK1aoQYMGyps3rwoWLKguXbrowoULDuv89ttvValSJXl4eKh48eJ69913de3atTTP68qVKwoPD1fFihXl6ekpX19fNW/eXOvWrbPrGxwcLIvFoqSkJA0ZMkRlypSRq6urIiIiFBQUpEaNGkmSIiMjrZ9J6jOaQUFBaT7vmXpcAAAAAMhuuepW2wIFCqhKlSpavXq1bt68KSenW7k4JibG2mfVqlXq3LmzpFuBbvv27Xrqqafk4uIiwzDUrl07RUVF6bHHHtMbb7yhq1evau7cuXr22Wf1+eefq0+fPnbvGx0drYULFyosLEyvvfaa1qxZo+nTp+vIkSN24fCjjz7SkCFDFBAQoF69esnV1VVz5szR/v37HZ7TxYsX9eSTT2rv3r1q2LChQkNDFR8fr59++kkhISH64Ycf7IK2JD3//PPatWuXQkNDVaBAAT3yyCN66623tHPnTk2bNk2NGjWyTmRUrVq1jH/YAAAAAJBNclXwlKSQkBB99tln2rlzp2rUqCHpVtisXLmyzp49axM8165dq5SUFIWEhEiSvv/+e0VFRalRo0ZatmyZ3NzcJEmDBg1SzZo11b9/f4WFhalUqVI27/nzzz8rJiZG9evXl3RrAp+mTZsqJiZGmzZtUr169SRJf/zxhz788EMVLVpU27dvl7+/vyQpIiJCderUcXg+vXv31t69ezV58mR169bN2j5s2DDVrl1bL7/8spo3by4PDw+b/U6dOqXff/9dBQoUsGmPiYnRtGnTFBwcrIiIiAx/vpmRlJSkpKQk6+uEhIRseV8AAAAA/w65LngGBwfrs88+08qVK1WjRg2lpKRo3bp16t69u06fPq2VK1da+65atcq6jyTrLacjR460hk5JKlasmP773/9q4MCBmjFjhj744AOb9+zYsaM1dEqSs7OzunTpopiYGG3ZssUaPGfOnKmUlBT169fPGjolydvbWx988IE1EKc6f/685syZoyZNmtiETkkKCAjQO++8oz59+ujXX3/Vs88+a7M9MjLSLnTmlOHDhysyMjKnywDwsImOzukKFBaW0xX8u+X8NwwAyC65Lng++eSTcnJy0qpVq9S/f39t3bpVV65cUUhIiE6fPq25c+fq6NGjKlWqlFatWiVvb2/ryOiOHTvk6enpcPQxNZzu3LnTblvq/rcrVqyYJOny5cvWtl27dkmSGjZsaNffUduWLVt048YNXbt2zeHo5OHDhyVJBw4csAueaY2g5oSBAweqX79+1tcJCQkqXrx4DlYEAAAA4EGS64Knr6+vqlevbr2NdtWqVXJyctKTTz6pc+fOSbo10lmgQAHt2LFDzzzzjJydnSXdPRClzgIbHx9vt83Hx8euzcXl1kdz48YNa1vqvrePdqYKCAiwa7t48aIkaf369Vq/fn2a53z16tV0HS+nuLu7y93dPafLAAAAAPCAynXBU7r1nOe2bdu0bds2xcTEqFq1asqfP7/y58+vwMBArVq1SoUKFdLNmzetz3dKt255PXv2rMNjprZ7e3tnuq7UgHru3DmVLFnS4fFvl/peb7/9tkaNGpWh98rMDLROTk66fv26w22OAjcAAAAAZIdctZxKqtTbYn/55RetX79ejRs3tm4LCQnRqlWr7J7vlKTq1avrn3/+0W+//WZ3zNWrV0u6vxlgq1atKunWpEZ3ctRWu3ZtWSwWbdy4MdPvmRH58+fXuXPnrEvLpLp69ar1tl4AAAAAyG65Mng2bNhQzs7OGj9+vK5evWozqhkSEqK4uDh9//338vX1tQmSXbp0kXTrmcTk5GRre1xcnMaMGSMXFxe9+OKLma6rY8eOcnZ21pgxY6y3/Uq3bvH9+OOP7foXLlxYbdu21YYNG/Tpp5/KMAy7Pps3b1ZiYmKma7pdrVq1lJycrBkzZljbDMPQwIEDHd7OCwAAAADZIVfeauvt7a2aNWvqt99+k7Ozs83EPakh9K+//lLLli2ta31KUufOnRUVFaWffvpJVapU0bPPPmtdx/PChQsaPXq0HnnkkUzXVaZMGQ0ZMkTh4eGqUqWK2rZtKxcXF82fP1+VK1fWwYMH7faZMGGCDh48qHfffVffffedHn/8cfn4+OjEiRPatm2bDh8+rNOnTytPnjyZrivVm2++qSlTpqhnz55avny5ChUqpLVr1+ry5cuqWrWqdXIkAAAAAMhOuXLEU/q/gFmrVi15eXlZ2x955BHr85W332Yr3Xouct68eRo1apRcXV01btw4ff/996pUqZJ++uknm5lZM2vIkCGaOHGiChYsqK+//lo//PCD2rZtqx9++MFh/wIFCmjDhg3WJV5mzJih8ePHa/PmzapYsaKmT58uPz+/+65LkipXrqylS5eqZs2amjdvnr777jtVqFBB69evl6+vb5a8BwAAAABklMVwdP8ncBcJCQny8fFRfHz8fU3WBAC5Het4misXLNUKALhP6c0GuXbEEwAAAADw70DwBAAAAACYiuAJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBULjldAAAAuVV0dE5XAADAvwMjngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGAqgicAAAAAwFQETwAAAACAqQieAAAAAABTETwBAAAAAKYieAIAAAAATEXwBAAAAACYiuAJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApnLJ6QIAAMBDKiwspysAgAdTdHROV5BhjHgCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGAqgicAAAAAwFQETwAAAACAqQieAAAAAABTETwBAAAAAKYieAIAAAAATEXwBAAAAACYiuAJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4HmHmJgYWSwWRUREaOPGjQoNDZWvr68sFoskyTAMTZ48WfXr15e3t7fy5MmjWrVqafLkyXbHOnXqlMLDw1WvXj35+/vL3d1dQUFBev3113Xu3Dm7/vHx8RoyZIgqVKigfPnyycfHR+XKlVO3bt104sQJm76JiYmKiIhQuXLl5OHhoQIFCqhFixbasGGD3XEjIiJksVgUExOjuXPnqkaNGvL09FSRIkXUp08f/fPPP1n06QEAAACAPYJnGjZs2KBGjRpJkl5++WW1a9dOhmGoU6dO6tGjh86fP6+OHTuqZ8+eunr1qnr06KH+/fvbHGPNmjUaPXq0AgIC1KFDB/Xu3VulS5fWl19+qccff1zx8fHWvoZhKDQ0VB999JEKFCigl19+Wb169VKlSpX0448/6siRI9a+SUlJatKkiSIjI5U3b1699dZbatWqlWJiYtSoUSNFRUU5PKcvvvhC3bt3V/ny5fXaa68pf/78GjdunHr27GnCJwgAAAAAt7jkdAG51fLlyzVp0iR1797d2jZx4kTNnDlTPXr00FdffSUXl1sf3/Xr19WmTRuNHj1aHTp0UM2aNSVJjRs31pkzZ5QvXz6bY0+fPl1dunTR+PHjNWjQIEnSnj17tHnzZrVu3douOCYlJSk5Odn6euTIkdq0aZNefPFFfffdd9bR2Lfeekt16tRRz5499dRTT8nLy8vunLZt26ayZctKkoYOHapq1app1qxZ+vTTTxUYGOjws0hKSlJSUpL1dUJCQvo/SAAAAAAPPYJnGqpXr24TOiVp/Pjxyps3r8aPH28NnZLk5uamoUOHKjo6WrNmzbIGT39/f4fH7ty5s3r37q1ff/3VGjxTeXp62vV3d3eXu7u79fXUqVPl6uqqESNGWEOnJFWpUkVdu3bV119/rZ9++kmdOnWyOU7fvn2toTP1vTp06KDIyEht27YtzeA5fPhwRUZGOtwGAMgeYWE5XUHWi46OzukSAADZhOCZhjp16ti8TkxM1O7duxUYGKgRI0bY9U8dkTxw4IBNe1RUlL7++mtt375dly5d0o0bN6zbTp06Zf1z+fLlVblyZc2cOVMnTpxQq1at1LBhQ9WoUUPOzs7WfgkJCfrzzz9Vvnx5FStWzK6O4OBgff3119q5c6dd8KxRo4Zd/9RjXL58Oa2PQgMHDlS/fv1saihevHia/QEAAADgdgTPNAQEBNi8vnTpkgzDUFxc3F1H/65evWr98+jRo9W/f38VKlRIzZo1U7FixawjmmPHjrW5fdXFxUUrV65URESEoqKi9Pbbb0uS/Pz81Lt3bw0aNEjOzs7W21zvrC9V4cKFJcnm+dFUPj4+dm2pI7e3B+I73TniCgAAAAAZQfBMw+23sEqSt7e3JKlmzZraunXrPfdPSUnRRx99pMDAQO3cuVOFChWybjMMQyNHjrTbx8/PT+PHj9e4ceN04MABrVy5UuPGjVN4eLhcXV01cOBAax1nz551+L6p7an9AAAAACCnMattOnl5eal8+fLav3//XW9LTXX+/HnFx8erXr16NqFTkrZu3XrXJUwsFovKly+vN954Q8uXL5ck/fzzz5JuBcpHHnlEf/zxh+Li4uz2Xb16tSSpWrVq6TwzAAAAADAXwTMD+vTpo8TERPXq1cvmltpUR48eVWxsrKRbEwt5enpq+/btSkxMtPa5dOmSevfu7XDfffv22bWnjmDePulQly5dlJycrIEDB8owDGv7nj17NGXKFPn4+KhVq1aZPU0AAAAAyFLcapsBr7zyijZt2qRp06Zp/fr1atq0qQIDA3X27FkdOHBAmzdv1syZMxUUFCQnJye9/vrrGj16tKpWraqwsDAlJCRoyZIlKlmypN0Msrt27VLr1q1Vu3ZtVapUSYULF1ZcXJwWLFggZ2dn6zOfkvTuu+9q0aJF+u6777R//341adJEf/31l+bMmaPk5GRNnz7dbikVAAAAAMgpBM8MsFgsmjp1qp555hlNnDhRCxcu1N9//y1/f389+uijGjVqlJo2bWrtP3z4cBUoUEBTp07VhAkTFBAQoPbt2ysyMlKVKlWyOXatWrU0YMAAxcTEaNGiRbp8+bIKFy6sZs2a6Z133rGZZdfDw0MrV67UJ598ojlz5uizzz5Tnjx59OSTT+r9999XgwYNsu0zAQAAAIB7sRi336sJpENCQoJ8fHwUHx/PJEYAkE3+net45nQFAID7ld5swDOeAAAAAABTETwBAAAAAKYieAIAAAAATEXwBAAAAACYiuAJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMJVLThcAAADuLTo6pysAACDzGPEEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmCpTwfPvv//W8ePHlZKSYtM+Z84cvfjii+rVq5d27tyZFfUBAAAAAB5wLpnZ6b333tO0adN09uxZubjcOsSXX36pN998U4ZhSJJmz56trVu3qmzZsllXLQAAAADggZOpEc+1a9eqadOmyps3r7Vt+PDhKlq0qNasWaO5c+fqxo0b+vTTT7OsUAAAAADAgylTI55xcXFq2rSp9fXu3bt18uRJjRw5Ug0aNJAkzZs3T6tXr86aKgEAAAAAD6xMjXj+888/cnNzs75et26dLBaLmjVrZm175JFHFBcXd/8VAgAAAAAeaJkKnsWKFdPvv/9ufb1o0SLlz59flStXtrZduHBB+fLlu/8KAQAAAAAPtEzdavv000/riy++0DvvvCMPDw8tXbpUnTt3lsVisfY5cOCASpQokWWFAgAAAAAeTBYjdRraDDhz5oyeeOIJxcbGSpIKFy6sTZs2WYPm8ePH9cgjj6hPnz4aM2ZMlhaMnJeQkCAfHx/Fx8fL29s7p8sBAAAAkEPSmw0yNeJZuHBh7d27VytWrJAkPfnkkzZvcuXKFY0ePVqhoaGZOTwAAAAA4F8kUyOeeLjluhHPsLCcrgAAkBnR0TldAQDgPqU3G2RqciEAAAAAANIrU7faStK+ffs0fvx4bdmyRZcvX9aNGzfs+lgsFh05cuS+CgQAAAAAPNgyFTxXr16t5s2bKykpSS4uLgoICJCLi/2huIsXAAAAAJCp4DlgwAClpKTo22+/VZcuXeTs7JzVdQEAAAAA/iUyFTx37dql9u3bq3v37lldDwAAAADgXyZTkwt5eXnJ398/q2sBAAAAAPwLZSp4tmjRQmvXrs3qWgAAAAAA/0KZCp4jR45UfHy8+vTpo8TExKyuCQAAAADwL5KpZzzbtm2rvHnz6osvvtDUqVP16KOPysfHx66fxWLRihUr7rtIAAAAAMCDK1PBMyYmxvrnv//+Wzt27HDYz2KxZKooAAAAAMC/R6aC582bN7O6DgAAAADAv1SmnvEEAAAAACC9MjXiebu///5bhw4d0tWrV9WwYcOsqAkAAAAA8C+S6RHP2NhYtWzZUvnz51ft2rUVEhJi3bZ+/XpVqFDB5lnQB8n169f1wQcfqHTp0nJzc5PFYlFMTIwsFosiIiJyujwAAAAAeKBkasTz+PHjqlevni5cuKCWLVvqzJkz2rhxo3V73bp1df78ec2aNUvBwcFZVWu2GTVqlIYOHarg4GB16NBBLi4uio2NzemyAAAAAOCBlKngGR4erkuXLmn16tV64oknFBkZaRM8XVxc1LBhQ61fvz7LCs1OixcvVr58+bRs2TK5urpK0gM7egsAAAAAOS1Tt9r+8ssvat26tZ544ok0+5QoUUJxcXGZLiwnnTp1SgULFrSGTgAAAABA5mUqeF68eFFBQUH37JeUlJSZw+eYiIgIWSwWHT16VMeOHZPFYpHFYrnr7cKrVq1S9+7dVbZsWeXLl0/58uVTrVq19M033zjsn3q8uLg4dezYUX5+fvLy8lKLFi30559/SpIOHjyo1q1bq0CBAvLy8tILL7ygc+fOOTze77//rvbt26tIkSJyc3NTyZIl1bt3b124cMFhrU8//bQCAwPl7u6uwMBABQcH69tvv834hwUAAAAA6ZSpW20DAgL0xx9/3LXPnj17VKJEiUwVlVNSA+bYsWMlSW+99ZYk3TVkf/LJJ/rjjz9Ur149tW7dWpcvX9bSpUv1yiuv6ODBgxo9erTdPpcuXVKDBg1UuHBhdenSRYcOHdLChQt14MAB/fzzz2rYsKFq1Kih7t27a9u2bZo3b54uX76s5cuX2xzn559/Vtu2beXs7KznnntOxYsX1759+zR+/Hj98ssv2rx5s/Lnzy9JWrRokcLCwuTr66uWLVuqSJEi+uuvv7Rz507NmDFDPXv2vO/PDwAAAAAcyVTwfOqpp/Tdd99pz549qlSpkt32tWvXasWKFdbg9qAIDg5WcHCwpk6dKkk2M9im9Yznl19+qVKlStm0paSk6JlnntHnn3+uvn372gXw33//Xf/97381ZswYa9trr72mr776Sg0aNFBERIT69u0rSTIMQ88++6wWL16sHTt2qHr16pKkCxcuqHPnzipUqJDWr19v8x6zZs1Sx44dNWTIEI0bN06SNHnyZBmGoZiYGFWpUsWmHkejowAAAACQVTJ1q+0HH3wgT09PNWjQQMOGDbOOfi5ZskSDBw9W8+bN5efnp3feeSdLi82N7gyd0q3JlV599VXduHFDq1atstueL18+ffTRRzZtHTt2lCQVLFhQffr0sbZbLBa1b99ekrRr1y5r+/Tp05WQkKDhw4fbBdsOHTqoRo0amj17tt17e3p62rUVLFjwbqeopKQkJSQk2PwAAAAAQHplasQzKChIv/zyi9q3b68PPvhAFovFOjJnGIZKlCihefPmqUiRIlldb65z5coVjRo1SgsWLNCRI0d09epVm+2nTp2y2+fRRx9V3rx5bdpSP6sqVarIYrE43Hb7ZE2bNm2y/q+j256vXbum8+fP6/z58/Lz81Pbtm0VFRWlunXrqkOHDmrcuLEaNmwof3//e57j8OHDFRkZec9+OSY6OqcrAJALhYXldAW4F/72BoCHR6aCp3Rrrc7Dhw8rOjpamzdv1sWLF+Xt7a26deuqZcuWcnNzy8o6c6Xr168rODhY27dvV/Xq1dW5c2cVLFjQuu7ntGnTHE6w5O3tbdfm4uJyz23JycnWtosXL0qSvvjii7vWePXqVfn5+aldu3ZydXXV2LFj9fXXX2vChAnWiY7GjBmjatWqpXmMgQMHql+/ftbXCQkJKl68+F3fFwAAAABSZSp4btmyRbVr15aLi4tat26t1q1bO+w3btw49e7d+74KzM1++uknbd++XT179tTEiRNtts2ePVvTpk0z7b1TA+ru3bsdPmfryPPPP6/nn39eCQkJ2rBhg6KiojRp0iSFhobq4MGD8vX1dbifu7u73N3ds6p0AAAAAA+ZTD3j2aJFCx0+fPiufcaNG/fATS6UUUeOHJEkPffcc3bb1q5da+p7161bV5K0cePGDO/r7e2t5s2b65tvvlHXrl117tw5bd68OatLBAAAAABJmQyeLi4uatasmU6fPu1w+xdffKG+ffuqTp0691VcbleyZElJ0rp162zaV69ebTcCmtW6desmLy8vDRo0SHv37rXbnpiYaH0OVJJWrFiha9eu2fVLXR/U0aRDAAAAAJAVMnWr7dKlS9WoUSOFhoZqzZo1Nrdofvnll+rdu7fq1KmjZcuWZVWduVJYWJiCgoI0cuRI69IyBw8e1MKFC9WqVSvNnz/ftPcuVKiQZs2apRdeeEFVq1ZV8+bNVa5cOV27dk3Hjh3T6tWr9cQTT2jp0qWSpLffflvHjx9XcHCwgoKCZLFYtG7dOv3222964oknVL9+fdNqBQAAAPBwy1TwrFKlin7++WeFhoYqLCxMy5cvl4eHh7788ku98cYbql27tpYtWyYvL6+srjdXyZcvn1auXKl33nlHa9asUUxMjCpWrKgZM2YoICDA1OAp3brleceOHfr000/166+/avny5cqbN6+KFSumbt26qVOnTta+AwcOVFRUlLZt26ZffvlFrq6uKlWqlEaOHKnXX39dzs7OptYKAAAA4OFlMQzDyOzOP/30k9q0aaOnn35aoaGh6tOnj2rUqKFff/1VPj4+WVkncpGEhAT5+PgoPj7e4Sy8AJAbsJxK7sdqWADw4EtvNsj0ciqS1LJlS3311Vfq1auXFi1apOrVqxM6AQAAAAA20hU816xZk+a2Rx99VM8995zWrVunwYMHa9euXTbbn3zyyfurEAAAAADwQEtX8AwODpbFYrlrH8Mw9Pzzz9u137hxI3OVAQAAAAD+FdIVPIcMGXLP4AkAAAAAgCPpCp4REREmlwEAAAAA+LdyyukCAAAAAAD/bvc1q+3Vq1f1008/aefOndbpc6tVq6ZWrVopb968WVUjAAAAAOABlunguWDBAvXs2VOXLl3S7UuBWiwW+fr6auLEiQ4nGwIAAAAAPFwyFTw3btyotm3bytnZWS+//LKCg4NVuHBhnT17VjExMZo6darat2+v1atX6/HHH8/qmgEAAAAAD5BMBc+hQ4fK3d1dGzduVKVKlWy2tW3bVq+//roef/xxDRs2TNHR0VlSKAAAAADgwZSpyYU2btyodu3a2YXOVJUqVVLbtm21YcOG+yoOAAAAAPDgy1TwTExMlL+//137+Pv7KzExMVNFAQAAAAD+PTIVPIOCgrR8+fK79lmxYoWCgoIyc3gAAAAAwL9IpoJnu3bttG3bNnXp0kWnTp2y2Xb69Gl17dpV27ZtU7t27bKkSAAAAADAg8ti3L4Wyl04OzsrIiJCgwcP1j///KPGjRtr8+bNcnNzU5kyZRQQEKCzZ8/qjz/+0PXr11WnTh2tWrVKnp6eZp8DsllCQoJ8fHysa7cCAAAAeDilNxuke8TTMAzrep2enp5avXq1IiMjVbRoUe3bt0+rVq3Svn37VKxYMUVGRmr16tWETgAAAABA5pZTkSQ3NzcNHjxYgwcP1pUrV5SQkCBvb295eXllZX0AAAAAgAdcpoPn7by8vAicAAAAAACHMjS5kMViMasOAAAAAMC/VLonF3JycpKvr698fX3Tf3CLRUeOHMlsbcilmFwIAAAAgJT+bJChW20vX76sy5cv329tAAAAAICHSIZutY2IiNDNmzcz9AMAAAAAeLhlKHgCAAAAAJBRBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmSvestkwUBAAAAADIDEY8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGCqdM9qCwAAkKXCwnK6gtwrOjqnKwCALMWIJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgmUkRERGyWCyKiYkx7T2Cg4NlsVhMO/7UqVNlsVg0depU094DAAAAAAieAAAAAABTETwz6c0339T+/ftVp06dnC4FAAAAAHI1l5wu4EHl5+cnPz+/nC4DAAAAAHK9XD/iGRMTI4vFooiICG3cuFGhoaHy9fW1PvtoGIYmT56s+vXry9vbW3ny5FGtWrU0efJku2Pd/lzmlClTVLlyZXl6eqpUqVL63//+Zz3e559/rnLlysnDw0OPPfaYvvvuu7seK1VsbKwsFou6du2qP//8U23atFH+/PmVN29eNW3aVLt27XJ4juvWrVOjRo2UN29eFSxYUO3atdOJEyfS/Ewycs6SdPHiRb366qsKCAhQnjx5VLt2bf34449pHh8AAAAAstIDM+K5YcMGDRs2TCEhIXr55Zd1/PhxGYahTp06aebMmXrsscfUsWNHubm5afny5erRo4f27dunUaNG2R1r7NixiomJUcuWLdW4cWPNnz9fffv2VZ48ebRr1y798MMPevbZZ9W4cWPNnj1bL730kkqVKqUGDRqkq9bY2FjVrVtXFSpUUPfu3XXkyBH99NNPCgkJ0f79+xUQEGDtu2LFCj399NNycnJSu3btFBgYqBUrVqh+/frKnz+/3bEzes6JiYkKDg7W7t279fjjj6tRo0Y6ceKE2rVrp2bNmmXimwAAAACAjHlggufy5cs1adIkde/e3do2ceJEzZw5Uz169NBXX30lF5dbp3P9+nW1adNGo0ePVocOHVSzZk2bY61du1bbt2/XI488Iknq37+/ypQpo/79+ysgIEC7d+9WoUKFJEldu3ZV3bp19emnn6Y7eK5evVojRozQe++9Z20bPHiwPv74Y02ZMkUDBgyQJN28eVMvv/yyUlJStGbNGuvxbw+Xd/r2228zdM4jR47U7t271atXL33zzTfW47z00ksKDQ1N1/kkJSUpKSnJ+johISFd+wEAAACA9AAFz+rVq9uETkkaP3688ubNq/Hjx1sDmCS5ublp6NChio6O1qxZs+yCZ58+fayhU5KKFy+uBg0aaOXKlfrf//5nDZ2SVKdOHT3yyCNp3ibrSKlSpfTOO+/YtPXo0UMff/yxtmzZYm1bt26d/vzzT4WFhdmEWovFomHDhmnOnDm6cePGfZ3z9OnT5ebmpg8//NDmOM2aNVOTJk20YsWKe57P8OHDFRkZme7zB4DcJCwspytAWqKjo3O6BABANnlggueds8cmJiZq9+7dCgwM1IgRI+z6JycnS5IOHDhgt6169ep2bUWKFJEkVatWzeG2zZs3p7vWqlWrysnJ9vHZYsWKSZIuX75sbUsNsw0bNrQ7RsmSJVW8eHHFxsZa2zJ6zleuXNHRo0dVoUIFFS5c2K5/w4YN0xU8Bw4cqH79+llfJyQkqHjx4vfcDwAAAACkByh43v5cpCRdunRJhmEoLi7urqNxV69etWvz9va2a0sdPUxrW0pKSrpr9fHxSfP4t49gxsfHS5L8/f0dHicgIMAmeGb0nNNz/PRwd3eXu7t7uvoCAAAAwJ0emOCZOottqtSAWLNmTW3dujUnSrpvqQH13LlzDrefPXvW5nVGzzm1f3qPDwAAAABmyPXLqaTFy8tL5cuX1/79+21uX32QVK1aVdKtyY7udOzYMbslVTJ6zt7e3ipVqpT++OMPnTlzxm67o/cFAAAAgKz2wAZP6dYkQYmJierVq5fDW2qPHj1qc6tqbtOgQQOVKlVKCxcu1Lp166zthmHo/ffft5tYSMr4OXfu3FnXr1/XkCFDbPotW7YsXc93AgAAAMD9emButXXklVde0aZNmzRt2jStX79eTZs2VWBgoM6ePasDBw5o8+bNmjlzpoKCgnK6VIecnJz0zTff6JlnnlHTpk2t63iuXLlSp0+fVpUqVfT777/b7JPRc3733XcVFRWliRMnau/evXryySd14sQJzZ07Vy1atNCiRYty4MwBAAAAPEwe6OBpsVg0depUPfPMM5o4caIWLlyov//+W/7+/nr00Uc1atQoNW3aNKfLvKumTZtqxYoV+uCDD/TDDz/I09NTTZo00Q8//KCXXnrJrn9Gzzlv3rxavXq1Bg4cqB9//FHbt29XxYoVNWfOHMXHxxM8AQAAAJjOYhiGkdNF4MGSkJAgHx8fxcfHO5wFGAByE9bxzL1YxhMAHnzpzQYP9DOeAAAAAIDcj+AJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKlccroAAADMFB2d0xUAAABGPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGAqgicAAAAAwFQETwAAAACAqQieAAAAAABTETwBAAAAAKYieAIAAAAATOWS0wUAAICHVFhYTlcAPFyio3O6AjzEGPEEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGAqgicAAAAAwFQETwAAAACAqQieAAAAAABTETwfYlOnTpXFYtHUqVNzuhQAAAAA/2IETwAAAACAqQieAAAAAABTETwBAAAAAKYieOYia9euVevWrRUQECB3d3cVL15czz//vNatWydJOnXqlMLDw1WvXj35+/vL3d1dQUFBev3113Xu3Dm743Xt2lUWi0V//vmnPvvsM1WsWFHu7u7q2rVrNp8ZAAAAgIeZS04XgFu++OIL9e7dW56enmrdurVKlCihuLg4rVu3TvPmzVODBg20Zs0ajR49Wk2aNFHdunXl6uqqHTt26Msvv9Qvv/yi7du3y8fHx+7YvXv31qZNm9SiRQs9++yzCggIyIEzBAAAAPCwInjmArt371bfvn1VpEgRrV+/XkFBQdZthmHo9OnTkqTGjRvrzJkzypcvn83+06dPV5cuXTR+/HgNGjTI7vi///67duzYoRIlSmSqvqSkJCUlJVlfJyQkZOo4AAAAAB5OBM9c4KuvvtKNGzf08ccf24ROSbJYLAoMDJQk+fv7O9y/c+fO6t27t3799VeHwfOdd97JdOiUpOHDhysyMjLT+wNAdgkLy+kKkBHR0dE5XQIAIJvwjGcu8Ntvv0mSmjVrds++UVFRCg0NVaFCheTi4iKLxSInJyclJCTo1KlTDvepU6fOfdU3cOBAxcfHW39OnDhxX8cDAAAA8HBhxDMXuHz5siwWi4oUKXLXfqNHj1b//v1VqFAhNWvWTMWKFZOnp6ckaezYsTa3w97ufp/pdHd3l7u7+30dAwAAAMDDi+CZC/j6+lqf5SxatKjDPikpKfroo48UGBionTt3qlChQtZthmFo5MiRaR7fYrFkec0AAAAAkF7capsLpN4Ku2zZsjT7nD9/XvHx8apXr55N6JSkrVu36p9//jG1RgAAAADILIJnLvDqq6/K2dlZH3zwgY4dO2azLXUk1N/fX56entq+fbsSExOt2y9duqTevXtnd8kAAAAAkG4Ez1ygcuXKGjt2rE6fPq2KFSuqU6dOGjRokHr06KHHHntMn3zyiZycnPT6668rNjZWVatWVb9+/dSzZ09VqlRJTk5O1plvAQAAACC34RnPXOLNN99UpUqVNHr0aC1ZskR///23/P39VbduXbVt21bSrWVNChQooKlTp2rChAkKCAhQ+/btFRkZqUqVKuXwGQAAAACAYxbDMIycLgIPloSEBPn4+Cg+Pl7e3t45XQ4AWLGO54OFZTwB4MGX3mzArbYAAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwlUtOFwAAQFaJjs7pCgAAgCOMeAIAAAAATEXwBAAAAACYiuAJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmMolpwsAAAAPqbCwnK7gwRUdndMVAECGMOIJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGAqgmcusGbNGlksFvXo0cPh9pMnT8rZ2VlNmjSRJG3btk1vvvmmKlWqJB8fH3l6eqpy5coaMWKEkpOT7fYPCgpSUFCQrl69qn79+qlo0aJyd3dXlSpVNG/ePFPPDQAAAAAInrlAw4YNFRQUpPnz5+vatWt222fMmKGbN2+qc+fOkqSJEyfqxx9/VOXKlfXKK6+oR48eMgxDAwcOVPv27R2+R3Jyspo1a6YlS5bo+eefV6dOnXTkyBG1bdtWy5YtM/X8AAAAADzcXHK6AEgWi0Uvvviihg4dqujoaL3wwgs222fMmCFPT0/95z//kSQNHDhQX3zxhZydna19DMNQz549NXnyZK1fv17169e3OcapU6dUu3ZtrVq1Sm5ubpKkjh07qmnTphozZoyaNWuWZn1JSUlKSkqyvk5ISLjvcwYAAADw8LAYhmHkdBGQDh48qHLlyum5557TTz/9ZG3ftWuXqlWrpvbt22vWrFl3Pcb27dtVs2ZNRUREKDw83NoeFBSkY8eO6c8//1SpUqVs9gkKCtKVK1d04cKFNI8bERGhyMhIu/b4+Hh5e3un9xQBINcJC8vpCh5u0dE5XQEA4H4lJCTIx8fnntmAW21zibJly6pWrVpasmSJLl68aG3/7rvvJMl6m60kXb9+XWPGjFGdOnXk7e0tJycnWSwW1axZU9Kt0c07+fr62oVOSSpWrJguX75819oGDhyo+Ph468+JEycyc4oAAAAAHlLcapuLdO7cWVu3btXcuXP16quv6ubNm5o1a5b8/f1tboVt06aNoqOj9dhjj6ldu3by9/eXq6urLl++rM8//9zmtthUPj4+Dt/TxcVFN2/evGtd7u7ucnd3v7+TAwAAAPDQYsQzF2nfvr1cXFz0/fffS5JWrlypU6dOqUOHDnJxufVvBFu2bFF0dLRCQ0O1b98+TZw4UUOHDlVERESaEwsBAAAAQE4ieOYiqSObGzZs0NGjR60BtFOnTtY+R44ckSS1aNHCZnIhSVq7dm32FQsAAAAA6UTwzGU6d+4swzD07bffKioqSuXKlVOtWrWs20uWLClJWrdunc1+e/fu1fDhw7O1VgAAAABID57xzGVatmwpb29vffrpp0pOTraZVEiS6tSpozp16mju3Lk6ffq06tWrp+PHj+vnn39WixYtNG/evByqHAAAAAAcY8Qzl0ldrzM5Odm6vuftnJ2dtXDhQnXv3l1HjhzRuHHjtG/fPo0aNUojR47MoaoBAAAAIG2s44kMS+9aPQCQ27GOZ85iHU8AePCxjicAAAAAIFcgeAIAAAAATEXwBAAAAACYiuAJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKpecLgAAgJwSHZ3TFQAA8HBgxBMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKkIngAAAAAAUxE8AQAAAACmIngCAAAAAExF8AQAAAAAmIrgCQAAAAAwFcETAAAAAGAqgicAAAAAwFQETwAAAACAqVxyugA8eAzDkCQlJCTkcCUAAAAAclJqJkjNCGkheCLDrly5IkkqXrx4DlcCAAAAIDe4cuWKfHx80txuMe4VTYE73Lx5U6dOnZKXl5csFktOlwMTJSQkqHjx4jpx4oS8vb1zuhzkclwvyAiuF6QX1woygusl+xmGoStXrigwMFBOTmk/ycmIJzLMyclJxYoVy+kykI28vb35yxvpxvWCjOB6QXpxrSAjuF6y191GOlMxuRAAAAAAwFQETwAAAACAqQieANLk7u6u8PBwubu753QpeABwvSAjuF6QXlwryAiul9yLyYUAAAAAAKZixBMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAHYuHTpkjp37iwfHx/5+Pioc+fOunz5cpr9k5OT9d5776ly5crKmzevAgMD9dJLL+nUqVPZVzRyTEavF0mKiopSaGio/Pz8ZLFYtHPnzmypFdlvwoQJKlWqlDw8PFSzZk2tXbv2rv1Xr16tmjVrysPDQ4888oi++uqrbKoUOS0j18rp06fVsWNHlS1bVk5OTnrrrbeyr1DkChm5XqKiovTUU0+pUKFC8vb21uOPP65ffvklG6tFKoInABsdO3bUzp07tXTpUi1dulQ7d+5U586d0+yfmJio7du3a/Dgwdq+fbuioqJ06NAhPffcc9lYNXJKRq8XSbp69arq16+vESNGZFOVyAlz5szRW2+9pUGDBmnHjh1q2LChnn76aR0/ftxh/6NHj+qZZ55Rw4YNtWPHDr3//vvq06eP5s+fn82VI7tl9FpJSkpSoUKFNGjQIFWtWjWbq0VOy+j1smbNGj311FNavHixtm3bppCQEIWFhWnHjh3ZXDlkAMD/t2/fPkOSsWnTJmvbxo0bDUnGgQMH0n2c3377zZBkHDt2zIwykUvc7/Vy9OhRQ5KxY8cOE6tETqlTp47x6quv2rSVK1fOGDBggMP+7777rlGuXDmbtldeecWoV6+eaTUid8jotXK7Ro0aGX379jWpMuRG93O9pKpQoYIRGRmZ1aXhHhjxBGC1ceNG+fj4qG7duta2evXqycfHRxs2bEj3ceLj42WxWOTr62tClcgtsup6wb/P9evXtW3bNjVr1symvVmzZmleGxs3brTrHxoaqq1btyo5Odm0WpGzMnOt4OGVFdfLzZs3deXKFRUoUMCMEnEXBE8AVmfOnJG/v79du7+/v86cOZOuY1y7dk0DBgxQx44d5e3tndUlIhfJiusF/07nz5/XjRs3FBAQYNMeEBCQ5rVx5swZh/1TUlJ0/vx502pFzsrMtYKHV1ZcL6NHj9bVq1fVtm1bM0rEXRA8gYdARESELBbLXX+2bt0qSbJYLHb7G4bhsP1OycnJat++vW7evKkJEyZk+Xkge2TX9YJ/vzuvg3tdG476O2rHv09GrxU83DJ7vcyaNUsRERGaM2eOw384hblccroAAOZ788031b59+7v2CQoK0u+//66zZ8/abfvrr7/s/nXxTsnJyWrbtq2OHj2qlStXMtr5AMuO6wX/bn5+fnJ2drYbgTh37lya10bhwoUd9ndxcVHBggVNqxU5KzPXCh5e93O9zJkzRz169NAPP/ygpk2bmlkm0kDwBB4Cfn5+8vPzu2e/xx9/XPHx8frtt99Up04dSdLmzZsVHx+vJ554Is39UkPn4cOHtWrVKv4j8QFn9vWCfz83NzfVrFlTy5cvV+vWra3ty5cvV8uWLR3u8/jjjys6OtqmbdmyZapVq5ZcXV1NrRc5JzPXCh5emb1eZs2ape7du2vWrFlq0aJFdpQKR3JyZiMAuU/z5s2NKlWqGBs3bjQ2btxoVK5c2Xj22Wdt+pQtW9aIiooyDMMwkpOTjeeee84oVqyYsXPnTuP06dPWn6SkpJw4BWSjjF4vhmEYFy5cMHbs2GEsWrTIkGTMnj3b2LFjh3H69OnsLh8mmj17tuHq6mpMmjTJ2Ldvn/HWW28ZefPmNWJjYw3DMIwBAwYYnTt3tvb/888/jTx58hj//e9/jX379hmTJk0yXF1djXnz5uXUKSCbZPRaMQzD2LFjh7Fjxw6jZs2aRseOHY0dO3YYe/fuzYnykc0yer3MnDnTcHFxMb744gub/0a5fPlyTp3CQ4vgCcDGhQsXjBdffNHw8vIyvLy8jBdffNG4dOmSTR9JxpQpUwzD+L8lMRz9rFq1KtvrR/bK6PViGIYxZcoUh9dLeHh4ttYO833xxRdGyZIlDTc3N6NGjRrG6tWrrdu6dOliNGrUyKZ/TEyMUb16dcPNzc0ICgoyvvzyy2yuGDklo9eKo79DSpYsmb1FI8dk5Hpp1KiRw+ulS5cu2V/4Q85iGP//yX0AAAAAAEzArLYAAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AwAMhKChIQUFBNm0RERGyWCyKiYnJkZpiY2NlsVjUtWvXHHn/B5XFYlFwcHBOl3Hfvv/+e1WrVk358uWTxWJRRESEddtnn32m8uXLy9PTUxaLRVOnTs2S68XR7wEAPAgIngCAe3rppZdksVhUuHBhpaSkOOwzdepU639gOxIcHCyLxWJilebgP/Rzh5SUFE2ZMkXPPPOMChcuLDc3N/n4+Kh27dr64IMPdOzYsWytZ8OGDercubMSExP1xhtvKDw83BqmZ86cqX79+ilPnjz673//q/DwcFWrVi1b68tKMTExdsEaADLKJacLAADkbgkJCZo/f74sFovOnj2rRYsWqWXLltlex4oVK7L9Pe+laNGi2r9/v3x8fHK6lH+1Y8eOqWXLltq1a5cCAgL01FNPqXjx4rp69aq2b9+uESNGaNSoUdqzZ4/KlCmTLTUtXrxYkjR9+nTVq1fP4bZFixapcOHC1vbk5OT7vl5y4+8BAKQHwRMAcFezZs1SYmKi+vfvr9GjR2vSpEk5EjxLly6d7e95L66uripXrlxOl/GvduXKFYWGhurgwYN655139OGHH8rDw8Omzx9//KF+/frp77//zra6Tp06JUk2wfJe27LiesmNvwcAkB7cagsAuKtJkybJzc1NAwcOVP369bV48WKdPn3apk/Xrl3VrVs3SVK3bt1ksVisP9KtZ/pWr15t/XPqT+qzbrc/+3bgwAE9//zz8vPzk8ViUWxsrKR73/I6ceJEVaxYUR4eHipRooQGDhyoa9eu2fS52y2Ddz5/l/r62LFjOnbsmE3dqfvf7Zm948ePq0ePHipatKjc3NxUrFgx9ejRQydOnLDrm3obckpKij766COVKlVK7u7ueuyxxzRhwgS7/teuXdPo0aNVtWpV+fj4KF++fCpdurQ6dOig3bt3p/kZpbp+/brGjRun0NBQFS9eXO7u7vL399fzzz+vHTt22PW//TbqFStWqEGDBsqbN68KFiyoLl266MKFCw7f59tvv1WlSpXk4eGh4sWL691337X7Tu5l1KhROnjwoDp16qSRI0fahU5JKlOmjH7++WdVqFDBpn3Dhg1q0aKFChQoIA8PD5UrV04RERFKTEx0+F5Hjx5Vz549VaJECbm7u6tIkSLq2rWrzW28qdfQlClTJEmlSpWyXhepn9OqVask/d+1nnrd3u16uXLlij788ENVqVJFefPmlY+Pj6pXr67BgwcrOTnZ2i+t3wPDMDR58mTVr19f3t7eypMnj2rVqqXJkyfb9b392ei5c+eqRo0a8vT0VJEiRdSnTx/9888/Nn1DQkIkSZGRkTa/B6m/m/Hx8RoyZIgqVKigfPnyycfHR+XKlVO3bt0cXu8AHk6MeAIA0rR7925t2bJFrVu3VoECBfTSSy9p3bp1mjZtmgYMGGDt16pVK12+fFk//fSTWrZsafc8W3h4uKZOnapjx44pPDzc2n5nvz/++EP16tVTxYoV1aVLF128eFFubm73rHP06NGKiYlRu3bt9Oyzz2rx4sUaMWKEduzYoSVLlmTq2VJfX1+Fh4dr7NixkqS33nrLuu1eE+McPnxYDRo00Llz5xQWFqaKFStq7969mjx5shYuXKj169c7vCW0Q4cO2rx5s55++mk5Oztr7ty5euONN+Tq6qpevXpZ+3Xp0kVz585VlSpV1K1bN7m7u+v48eNatWqVQkNDVbly5bvWd/HiRb311ltq2LChnnnmGeXPn19//vmnfv75Zy1ZskRr1qxR7dq17faLjo7WwoULFRYWptdee01r1qzR9OnTdeTIEa1bt86m70cffaQhQ4YoICBAvXr1kqurq+bMmaP9+/fftbY7pQanIUOG3LPv7dfK/Pnz1b59e7m5ualdu3by9/fXr7/+qsjISC1btkyrVq2Su7u7tf/mzZsVGhqqq1evKiwsTGXKlFFsbKxmzJihJUuWaOPGjXrkkUcUFBSk8PBwLViwQLt27VLfvn3l6+sr6db17OhaT92elvPnz6tRo0bat2+fqlWrpldffVU3b97UgQMH9Mknn+jtt9++6zEMw1CnTp00c+ZMPfbYY+rYsaPc3Ny0fPly9ejRQ/v27dOoUaPs9vviiy+0ZMkStWzZUsHBwVq6dKnGjRunCxcuaMaMGZJuXeuxsbGaNm2aGjVqZHPt+/r6yjAMhYaGavPmzapfv76aN28uJycnxcbG6scff1SXLl1UvHjxe3xzAB4KBgAAaejbt68hyYiKijIMwzAuX75seHh4GI8++qhd3ylTphiSjClTpjg8VqNGjYy0/m/n6NGjhiRDkjF48GCHfUqWLGmULFnSpi08PNyQZHh4eBh79uyxticnJxtPPfWUIcmYPn26tX3VqlWGJCM8PDzNGrp06XLP973XPo0bNzYkGV9//bVN+9dff21IMpo0aWLTnvrZ1K1b14iPj7e2HzhwwHBxcTHKli1rbbt8+bJhsViMWrVqGSkpKTbHSUlJMS5duuSw1ttdu3bNOHnypF37nj17jHz58hlNmza1aU/9bl1cXIx169bZvF9wcLAhydi4caO1/fDhw4aLi4tRtGhR4+zZs9b2+Ph4o2zZsoYko1GjRvesMzY21pBkFCtW7J59b5eQkGD4+voa7u7uxq5du6ztN2/eNDp27GhIMj766CNr+/Xr142goCDDy8vL2Llzp82x1q5dazg7OxvPPvusTXuXLl0MScbRo0ft3j+taz2t6+WFF14wJBnvv/++3T5nzpwxkpOTra8dXY/ffPONIcno0aOHTd+kpCQjLCzMkGRs3brV2p76e+Pj42McOHDA2p6YmGg89thjhsViMeLi4qztd/u9+f333w1JRuvWre22Xbt2zbhy5YpdO4CHE7faAgAcun79ur7//nvlz59fLVq0kCT5+PioZcuWOnz4sNasWZPl71m4cGF98MEHGd6vc+fOqlixovW1i4uLhg0bJkmaNm1altWXHidOnNDKlStVoUIFm1FKSerVq5fKly+vFStWOLwFcfjw4fL29ra+Llu2rOrXr6+DBw/qypUrkm7dvmkYhtzd3eXs7Gyzv7Oz8z1H1yTJ3d1dRYsWtWuvWLGiQkJCtGbNGpvbO1N17NhR9evXt3m/Ll26SJK2bNlibZ85c6ZSUlLUr18/+fv7W9u9vb0z9P2eOXNGklSsWLF07yNJCxYs0OXLl9W9e3dVqVLF2m6xWDRixAi5uLjYzL68cOFCxcbG6t1331XVqlVtjtWgQQO1bNlSixcvVkJCQobqSI+zZ89q3rx5Kl26tMNbwAMCAuTicvcb1MaPH6+8efNq/PjxNn3d3Nw0dOhQSbee1b5T3759VbZsWetrT09PdejQQYZhaNu2bRk6D09PT7s2d3d35cuXL0PHAfDvxa22AACHFixYoAsXLujVV1+1uYXxpZde0pw5czR58mQ9+eSTWfqeVatWTdettXdq2LChXVutWrXk6empnTt3ZkFl6Zf6jGSjRo3sbvG1WCx68skntX//fu3atcvuFsQaNWrYHS81dF2+fFleXl7y9vZW8+bNtXTpUtWoUUNt2rRRw4YNVbdu3Qx9djt37tTIkSO1bt06nTlzxi5onj9/XkWKFMlQfal27dolyfH34qgtq6V+B45uiS5evLhKly5tDfNeXl7atGmTJOnAgQMOw9+ZM2d08+ZNHTp0SLVq1crSWrdu3SrDMBQSEiJXV9cM75+YmKjdu3crMDBQI0aMsNue+r0eOHDAblt6v8+7KV++vCpXrqyZM2fqxIkTatWqlRo2bKgaNWrY/cMIgIcbwRMA4FDqs3WdO3e2aQ8NDVXhwoX1ww8/6H//+5/NCN39CggIyNR+t4+q3dkeFxd3PyVlWOqoWFrnkjrTaXx8vN02R8tspI5g3bhxw9o2b948DRs2TLNmzdKgQYMkSV5eXurevbuGDRumPHny3LXGDRs2qHHjxpKkZs2a6dFHH1W+fPlksViszy4mJSVlur7Uc3P0vWTkO079rDL6HabnOzh48KASEhLk5eWlixcvSpL1uca0XL16NUN1pEdqwHM0Ap0ely5dkmEYiouLU2RkZJr9HNWe3u/zblxcXLRy5UpFREQoKipKb7/9tiTJz89PvXv31qBBgwigACQxqy0AwIETJ05o+fLlkqT69evbzGTp4uKiM2fOKDExUbNnz87S983MJECSdO7cuTTbb/+PayenW/+3l5KSYtfXURDMjNQgfvbsWYfbU9vvJ7DnzZtXQ4cO1Z9//qk///xTkyZNUrly5fT555/rv//97z33Hzp0qJKSkrRixQr9/PPPGj16tCIjIxUREeFweZCMSv3MHX0vaX0ujpQsWVJFixbViRMndPjw4XTvl9HvIPV/o6OjZRhGmj+NGjVKdw3plXprdGb/gSS19po1a9619tSZds3g5+en8ePHKy4uTvv27dP48eNVsGBBhYeHa+TIkaa9L4AHC8ETAGBnypQpunnzpho0aKAePXrY/aSOgk6aNMm6T+qoRlojJffafj/Wrl1r17Z161b9888/NjPn5s+fX5Lj/8h3tIyIdKvujNSc+n5r1qyRYRg22wzDsNZ654y+mVWqVCl1795dq1evVr58+fTzzz/fc58jR46oQIECNs9rSrdu29y+fft915T6nKSj78VR29306NFDkvTxxx/fs+/169clSdWrV5d0a+mTO8XFxenIkSN65JFH5OXlJUmqW7euJGnjxo0Zqi0r1KpVS05OTlq1apXD52rvxcvLS+XLl9f+/fvTfXtsRqX3d9disah8+fJ64403rP9wlZ7rEcDDgeAJALBhGIamTJkii8Wi6dOn69tvv7X7mT59uqpXr67ffvtNe/bskSQVKFBAknTy5EmHx73X9vvx3Xffae/evdbXKSkpev/99yXJOvmNdGuyntRwlnp7pXRrBCytYFOgQAGdP38+3etPlihRQiEhIdblU243efJk7d27V40bN870EhN//fWXfvvtN7v2S5cuKSkpyeEkL3cqWbKkLl26ZPOZ3bhxQ/3799dff/2Vqbpu17FjRzk7O2vMmDE2o54JCQnpCpC369+/v8qWLavp06fr/fffd3gL8NGjR9WqVSvt27dPktSyZUv5+PhoypQpNudoGIYGDhyo5ORkm7U0W7ZsqRIlSmjMmDEOJ81KTk62Wy4mqwQEBOg///mPjhw54vBW2XPnzjkcob9dnz59lJiYqF69ejm8pfbo0aPWNTcz426/u0ePHrV+7rdLHVVOz/UI4OHAM54AABsrVqxQbGysQkJCVKpUqTT7devWTTt27NCkSZP02Wef6fHHH5enp6fGjh2rhIQEFSpUSJKs6302btxY8+bN0wsvvKBnnnlGHh4eqly5snXG3PvRtGlT1atXT+3bt1eBAgW0ePFi7dmzR6GhoerUqZO1n5ubm958802NGDFCNWrUUMuWLXXlyhVFR0erUaNGOnLkiN2xGzdurK1btyosLEwNGzaUm5ubGjRooAYNGqRZz5dffqkGDRqoV69eio6OVoUKFbRv3z79/PPPKlSokL788stMn2tcXJzq1q2rihUrqkaNGipatKguXLign376ScnJyXr33XfveYzevXtr2bJlatCggdq2bSsPDw/FxMQoLi5OwcHBDkcKM6JMmTIaMmSIwsPDVaVKFbVt21YuLi6aP3++KleurIMHD6b7WF5eXvrll1/UsmVLDR8+XFOmTFGzZs1UrFgxJSYmaseOHVq/fr1cXFysa1V6e3tr4sSJ6tChg+rWrat27dqpUKFCWrFihbZu3ao6deronXfesb6Hu7u75s2bp6efflqNGjVSkyZNVKlSJUnS8ePHtXbtWhUsWNDhBD1ZYcKECdqzZ4+GDh2qxYsXq3HjxjIMQ4cOHdKyZct09uzZu85W/Morr2jTpk2aNm2a1q9fr6ZNmyowMFBnz57VgQMHtHnzZs2cOVNBQUGZqq9cuXIKDAzU7NmzlSdPHhUrVkwWi0Wvvfaadu3apdatW6t27dqqVKmSChcurLi4OC1YsEDOzs7WZz4BgHU8AQA22rdvb0gyvvvuu7v2O3/+vOHm5mb4+fkZSUlJhmEYxqJFi4zatWsbnp6e1nU5UyUnJxvvvvuuUaJECcPFxcVmPcO01je83d3W8Vy1apXx9ddfGxUqVDDc3d2NYsWKGQMGDDASExPtjpOSkmIMGTLEKF68uOHm5mY89thjxueff278+eefDmu4cuWK0atXL6NIkSKGk5OTzXqGd6s7NjbW6Natm1GkSBHDxcXFKFKkiNGtWzcjNjbWru/d1ji9c73IS5cuGREREcaTTz5pFClSxHBzczMCAwON5s2bG7/88kuan9+d5s2bZ9SoUcPIkyeP4efnZ7Rt29Y4cuSIw/Up77ZG693WeJw4caJRoUIFw83NzShWrJjRv39/IzExMd3reN7u+vXrxuTJk43mzZsbAQEBhqurq+Hl5WXUqFHDGDhwoHH8+HG7fdasWWM8/fTThq+vr/W7Hjx4sPH33387fI+TJ08affv2NR599FHD3d3d8Pb2NsqXL2/07NnTWLFihU3frFzH0zBurXE6ePBgo1y5coa7u7vh4+NjVKtWzRgyZIhx/fp1a7+7rSs7Z84co2nTpkb+/PkNV1dXo2jRokZwcLAxevRo46+//rL2u/335k5pfdebNm0yGjVqZHh5eVl/t48ePWqcOHHCGDBggFGvXj3D39/fcHNzM0qUKGG0adPG2Lx5s8M6ATycLIZxxwMoAAAAAABkIZ7xBAAAAACYiuAJAAAAADAVwRMAAAAAYCqCJwAAAADAVARPAAAAAICpCJ4AAAAAAFMRPAEAAAAApiJ4AgAAAABMRfAEAAAAAJiK4AkAAAAAMBXBEwAAAABgKoInAAAAAMBUBE8AAAAAgKn+H/z+6f9ghrCfAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(678), plot(678, target='pred')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(4604)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot(4013)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.10.9"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}