user_class.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3d26bd18",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "from sklearn import metrics\n",
    "from scipy.spatial.distance import cdist\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.cluster import AffinityPropagation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "8b50eac6",
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.read_csv(\"train.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "a2108872",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-41-90c5fda747c9>:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  prep_df['cat'] = prep_df['item_id'].apply(lambda x: int(str(x)[:3]))\n"
     ]
    }
   ],
   "source": [
    "prep_df = train[['buyer_id', 'item_id']] # может еще взять count?\n",
    "prep_df['cat'] = prep_df['item_id'].apply(lambda x: int(str(x)[:3]))\n",
    "del prep_df['item_id']\n",
    "user_items_count = prep_df['buyer_id'].value_counts()\n",
    "user_cat_count = prep_df.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "3fa24311",
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in user_cat_count.index:\n",
    "    user_cat_count.loc[i] = user_cat_count[i] / user_items_count[i[0]]\n",
    "    \n",
    "test_df = pd.DataFrame(columns=['buyer_id'] + list(user_cat_count.index.get_level_values('cat').unique()))\n",
    "test_df['buyer_id'] = list(user_cat_count.index.get_level_values('buyer_id').unique())\n",
    "test_df.fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "3b5a550b",
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in user_cat_count.index:\n",
    "    test_df.loc[test_df['buyer_id'] == i[0], i[1]] = user_cat_count[i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "1e819b81",
   "metadata": {},
   "outputs": [],
   "source": [
    "del test_df[218]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "ecc5e964",
   "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>buyer_id</th>\n",
       "      <th>202</th>\n",
       "      <th>203</th>\n",
       "      <th>204</th>\n",
       "      <th>205</th>\n",
       "      <th>211</th>\n",
       "      <th>207</th>\n",
       "      <th>210</th>\n",
       "      <th>212</th>\n",
       "      <th>213</th>\n",
       "      <th>216</th>\n",
       "      <th>215</th>\n",
       "      <th>209</th>\n",
       "      <th>217</th>\n",
       "      <th>214</th>\n",
       "      <th>206</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>95474888</td>\n",
       "      <td>0.542969</td>\n",
       "      <td>0.326172</td>\n",
       "      <td>0.066406</td>\n",
       "      <td>0.025391</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.023438</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.015625</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>94740780</td>\n",
       "      <td>0.328704</td>\n",
       "      <td>0.422840</td>\n",
       "      <td>0.151235</td>\n",
       "      <td>0.035494</td>\n",
       "      <td>0.007716</td>\n",
       "      <td>0.035494</td>\n",
       "      <td>0.009259</td>\n",
       "      <td>0.004630</td>\n",
       "      <td>0.001543</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001543</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001543</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>95363331</td>\n",
       "      <td>0.450586</td>\n",
       "      <td>0.410385</td>\n",
       "      <td>0.087102</td>\n",
       "      <td>0.035176</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.013400</td>\n",
       "      <td>0.001675</td>\n",
       "      <td>0.001675</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>94807631</td>\n",
       "      <td>0.313821</td>\n",
       "      <td>0.437398</td>\n",
       "      <td>0.170732</td>\n",
       "      <td>0.024390</td>\n",
       "      <td>0.006504</td>\n",
       "      <td>0.022764</td>\n",
       "      <td>0.011382</td>\n",
       "      <td>0.006504</td>\n",
       "      <td>0.003252</td>\n",
       "      <td>0.001626</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001626</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>95355948</td>\n",
       "      <td>0.340102</td>\n",
       "      <td>0.455161</td>\n",
       "      <td>0.152284</td>\n",
       "      <td>0.023689</td>\n",
       "      <td>0.003384</td>\n",
       "      <td>0.013536</td>\n",
       "      <td>0.005076</td>\n",
       "      <td>0.003384</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.003384</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47737</th>\n",
       "      <td>94800228</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47738</th>\n",
       "      <td>95806500</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.300000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47739</th>\n",
       "      <td>95412139</td>\n",
       "      <td>0.230769</td>\n",
       "      <td>0.230769</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>0.230769</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47740</th>\n",
       "      <td>95067302</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47741</th>\n",
       "      <td>95779916</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.181818</td>\n",
       "      <td>0.272727</td>\n",
       "      <td>0.090909</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>47742 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       buyer_id       202       203       204       205       211       207  \\\n",
       "0      95474888  0.542969  0.326172  0.066406  0.025391  0.000000  0.023438   \n",
       "1      94740780  0.328704  0.422840  0.151235  0.035494  0.007716  0.035494   \n",
       "2      95363331  0.450586  0.410385  0.087102  0.035176  0.000000  0.013400   \n",
       "3      94807631  0.313821  0.437398  0.170732  0.024390  0.006504  0.022764   \n",
       "4      95355948  0.340102  0.455161  0.152284  0.023689  0.003384  0.013536   \n",
       "...         ...       ...       ...       ...       ...       ...       ...   \n",
       "47737  94800228  0.300000  0.300000  0.300000  0.100000  0.000000  0.000000   \n",
       "47738  95806500  0.300000  0.300000  0.000000  0.200000  0.000000  0.100000   \n",
       "47739  95412139  0.230769  0.230769  0.076923  0.230769  0.076923  0.076923   \n",
       "47740  95067302  0.272727  0.272727  0.272727  0.181818  0.000000  0.000000   \n",
       "47741  95779916  0.181818  0.272727  0.181818  0.272727  0.090909  0.000000   \n",
       "\n",
       "            210       212       213       216  215       209  217       214  \\\n",
       "0      0.000000  0.015625  0.000000  0.000000  0.0  0.000000  0.0  0.000000   \n",
       "1      0.009259  0.004630  0.001543  0.000000  0.0  0.001543  0.0  0.001543   \n",
       "2      0.001675  0.001675  0.000000  0.000000  0.0  0.000000  0.0  0.000000   \n",
       "3      0.011382  0.006504  0.003252  0.001626  0.0  0.001626  0.0  0.000000   \n",
       "4      0.005076  0.003384  0.000000  0.003384  0.0  0.000000  0.0  0.000000   \n",
       "...         ...       ...       ...       ...  ...       ...  ...       ...   \n",
       "47737  0.000000  0.000000  0.000000  0.000000  0.0  0.000000  0.0  0.000000   \n",
       "47738  0.100000  0.000000  0.000000  0.000000  0.0  0.000000  0.0  0.000000   \n",
       "47739  0.076923  0.000000  0.000000  0.000000  0.0  0.000000  0.0  0.000000   \n",
       "47740  0.000000  0.000000  0.000000  0.000000  0.0  0.000000  0.0  0.000000   \n",
       "47741  0.000000  0.000000  0.000000  0.000000  0.0  0.000000  0.0  0.000000   \n",
       "\n",
       "       206  \n",
       "0      0.0  \n",
       "1      0.0  \n",
       "2      0.0  \n",
       "3      0.0  \n",
       "4      0.0  \n",
       "...    ...  \n",
       "47737  0.0  \n",
       "47738  0.0  \n",
       "47739  0.0  \n",
       "47740  0.0  \n",
       "47741  0.0  \n",
       "\n",
       "[47742 rows x 16 columns]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "9c00f8e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_df = test_df.copy()\n",
    "X_train_df.pop('buyer_id')\n",
    "X_train = X_train_df.values\n",
    "model = KMeans(n_clusters=7, random_state=7)\n",
    "model.fit(X_train)\n",
    "all_predictions = model.predict(X_train)\n",
    "pred_df = pd.DataFrame(columns=['buyer_id', 'group_id'])\n",
    "pred_df['buyer_id'] = test_df['buyer_id']\n",
    "pred_df['group_id'] = all_predictions\n",
    "pred_df.to_csv('user_class_df_train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b3cc7343",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "476f2218",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f3b60b35",
   "metadata": {},
   "outputs": [],
   "source": [
    "## kmeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "6e2d185a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# X_train_df = test_df.copy()\n",
    "# X_train_df.pop('buyer_id')\n",
    "# X_train = X_train_df.values\n",
    "# distortions = []\n",
    "# inertias = []\n",
    "# mapping1 = {}\n",
    "# mapping2 = {}\n",
    "# K = range(1, 25)\n",
    "\n",
    "# for k in K:\n",
    "#     model = KMeans(n_clusters=k, random_state=7)\n",
    "#     model.fit(X_train)\n",
    "#     all_predictions = model.predict(X_train)\n",
    "#     distortions.append(sum(np.min(cdist(X_train, model.cluster_centers_,\n",
    "#                                             'euclidean'), axis=1)) / X_train.shape[0])\n",
    "#     inertias.append(model.inertia_)\n",
    "#     mapping1[k] = sum(np.min(cdist(X_train, model.cluster_centers_,\n",
    "#                                    'euclidean'), axis=1)) / X_train.shape[0]\n",
    "#     mapping2[k] = model.inertia_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a065ab5b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plt.plot(K, distortions, 'bx-')\n",
    "# plt.xlabel('Values of K')\n",
    "# plt.ylabel('Distortion')\n",
    "# plt.title('The Elbow Method using Distortion')\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "5207fb15",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZAAAAEWCAYAAABIVsEJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nO3debze453/8dc7kcWWhQTZkxI5QtuUSJVSKUop0XYQ1RHaadqMsVQVmY6hi46iG4qhFOWHYympsQ4prQlJrJEQ2SwnSZOQxC4k+fz+uL53z31O7rPd59znPsv7+Xjcj/t7X9/tur+O+5NrV0RgZmbWVF3KnQEzM2ufHEDMzKwoDiBmZlYUBxAzMyuKA4iZmRXFAcTMzIriAGIFSTpP0o2tcJ/hkkLSZtnnv0j6l1LftzW05HeRdJ2kn7XEtfKueZykB1vymuUi6UpJ55Q7H52NA0gnJendvNdGSR/kfT6uhe91naSPat3zuZa8R7HyAtjTtdL7ZXl+pZHXaZWA25Ii4qaI+FIpri3pFUkHlujaJ0j6W35aRHwvIn5aivtZ3RxAOqmI2Cr3Al4DDs9Lu6kEt7ww/54R8ekS3KM5tpS0W97nbwBLypUZKyxXUrW2wQHE6tNd0g2S3pE0V9LY3A5JAyXdIWmVpCWSTmnB++4oaaaktyTdLWmbvPsekeVlbVZFtEuWfqKkP+cdt1BSZd7n1yWNqeeefwQm5X0+Hrgh/4C6vrOkQ4B/B44pULoaJunx7Bk+KKlfQ98l2/cZSU9n590K9Kwr47VLPwWqBU+QtDi71pJcCbP2v+Szc74naYGkNZJ+J0nZvq6Sfinpjewa/5Z/j/rk7iPp4uy6SyR9OW9/b0nXSFouaamkn0nqmnfu45J+LWk1cCtwJfC57FmvzY77RxWfpL6S7sn+O63Jtgc3lE9rOgcQq88RwC1AH2AacBmApC7An4HngEHAAcBpkg5uofseD3wLGAisBy7J7rszcDNwGtAfuBf4s6TuwKPAvpK6SBoAdAP2yc77BLAV8Hw997wRmJj9UO4CbA08mdtZ33eOiPuBnwO3FihdfQM4EdgO6A6c0dB3yb7PXaSgtg1wG/D1Jj3B6nxvSXp+X46IrYG9gWfrOeUrwJ7Ap4Gjgdx/0+8AXwbGALsDRzYxK58F5gP9gAuBa3LBCbie9N95J+AzwJeAf6l17mLSM/wm8D1gRvas+xS4VxfgD8AwYCjwAdnfrrUsBxCrz98i4t6I2ED6Mcv9MO4J9I+In0TERxGxGLgamFjPtc7I/qWde11fz7F/jIgXIuI94Bzg6OxfpMcA/xMRD0XEx8DFwObA3lke3iH9wH0BeABYKqki+/zXiNhYzz2rSD9wB5JKIjfU2l/Mdwb4Q0S8HBEfAJVZ/qjvuwB7kQLgbyLi44i4HZjVwH3qsxHYTdLmEbE8IubWc+wFEbE2Il4Dpufl92jgtxFRFRFrgAuamIdXI+Lq7G/pemAAsL2k7UmB6bSIeC8iVgK/puZzXRYRl0bE+uw51isi3oyIOyLi/Yh4Bzif9DdgLcz1iVafv+dtvw/0zKoshgEDc9UHma7AX+u51sUR8R+NvO/reduvkn5M+5FKJK/mdkTERkmvk0oEkEoh+5P+JfsosJb0w/G57HNDbgBOIP2I7weMzNtXzHeGTZ/hVtl2fd9lA7A0as50+ipFiIj3JB1DKvlcI+lx4AcR8VIR+c3/75K/3Rj/uG5EvJ8VPrYilbC6AcurCyR0ac69JG1BCkKHAH2z5K0ldc0CmLUQl0CsGK8DSyKiT95r64g4tIWuPyRveyjwMfAGsIz0Qw5AVgUyBFiaJeUCyL7Z9qOkAPIFGhdA7gAOAxZHRO0f7Ia+c1Onta7vuywHBuVV8UB6DnV5D9gi7/MO+Tsj4oGIOIj0r/6XSCWnploO5LcjDKnrwCZ6HVgH9Mt7rr0iYte8Y2o/24ae9Q+AUcBnI6IX6R8DAKr7FCuGA4gVYybwtqSzJG2etRvsJmnPFrr+NyWNzv4l+RPg9uxfjpXAYZIOkNSN9EOxDvi/7LxHgfHA5hFRRSodHAJsCzzT0E2zKrMvUrP+Paeh77wCGJ61lTRGfd9lBqlN4BRJm0n6GjCunms9C+wnaaik3sDU3A5J22eN9Vtm13+XVMJpqkrgVEmDJPUBziriGpuIiOXAg8AvJfXK2rB2lFRfldMKYHDWVlTI1qR2j7VKHTDObYm82qYcQKzJsh/zw0n140tIpYPfA73rOe1M1RwH8kY9x/4RuI5U7dETOCW773xSI+ql2T0PJ3U//ijb/zLpB/Kv2ee3SY2vjze26iIiZkfEoiK+823Z+5uqNaakjvvU+V2y7/M1UnXaGlJ7yZ31XOshUu+k54GngHvydnchBadlwGpSaexfG8pfAVeTfuifJwXje0lBriWqhI4ndTCYR/q+t5NKS3V5BJgL/L2Ov6PfkNqT3gCeAO5vgTxaAfKCUmbWVFk33CsjYliDB1uH5RKImTUoq7Y7NKtSG0SqFvpTufNl5eUSiJk1KGuPehSoILUv/A9walZNaJ2UA4iZmRXFVVhmZlaUDjuQsF+/fjF8+PByZ8PMrF156qmn3oiI/o05tsMGkOHDhzN79uxyZ8PMrF2R1OhZD1yFZWZmRXEAMTOzojiAmJlZUUoWQCQNkTRd0otKi+acmqVvI+khpUVrHpLUN++cqUoLAc3PX1tC0h6S5mT7Lqk1yZyZmZVBKUsg60nTRu9CWt/gJEmjgbOBhyNiJPBw9pls30RgV9IEeJfnViUDrgAmk6bXHpntNzOzMipZAMkWrnk6234HeJG01sEE0oIyZO+5lc0mALdExLqIWAIsBMZlq8v1iogZ2foIN9D01dAadOGFMH16zbTp01O6mZltqlXaQCQNJy1V+SSwfTaFc24q5+2ywwZRc+GYqixtULZdO73QfSZLmi1p9qpVq5qUxz33hKOPrg4i06enz3u21ATlZmYdTMkDiKStSAv1nNbAvDmF2jWinvRNEyOuioixETG2f/9GjYP5h/HjobISjjgCvv71FDwqK1O6mZltqqQBJFso5w7gpojIrWewIquWIntfmaVXUXOVs8GkNQyqqLkSWi69xY0fD1ttBXfeCVOmOHiYmdWnlL2wBFwDvBgRv8rbNQ2YlG1PAu7OS58oqYekEaTG8plZNdc7kvbKrnl83jktavp0WLMGeveGK67YtE3EzMyqlbIEsg/wz8AXJT2bvQ4FLgAOkrQAOCj7TETMJS2bOY+0gthJeavITSGt/rYQWATc19KZzbV5HHssvP02/PGPNdtEzMysppLNhRURf6PuRewPqOOc84HzC6TPBnZrudxtatas1OaxahVcdx0MHJg+z5rlqiwzs0I67GSKTXXmmen9uefS+0svpRKIg4eZWWGeyqSWkSNBgvnzy50TM7O2zQGkli22gGHDUgnEzMzq5gBSQEWFA4iZWUMcQAoYNSpVYXm5eDOzujmAFFBRAe+9B0uXljsnZmZtlwNIARUV6d3VWGZmdXMAKWDUqPTuAGJmVjcHkAJ22AF69XIAMTOrjwNIAVKqxvJYEDOzujmA1MFdec3M6ucAUodRo6CqCt59t9w5MTNrmxxA6pDrifXyy+XNh5lZW+UAUgd35TUzq58DSB123BG6dHEAMTOriwNIHXr0gE98wgHEzKwupVzS9lpJKyW9kJc2RtIT2eqEsyWNy9s3VdJCSfMlHZyXvoekOdm+S7JlbVuFu/KamdWtlCWQ64BDaqVdCPw4IsYA/5l9RtJoYCKwa3bO5ZK6ZudcAUwmrZE+ssA1S6aiIjWib9jQ8LFmZp1NyQJIRDwGrK6dDPTKtnsDy7LtCcAtEbEuIpaQ1j4fJ2kA0CsiZkREADcAR5Yqz7WNGgUffgivvdZadzQzaz9ae0nb04AHJF1MCl57Z+mDgCfyjqvK0j7OtmunFyRpMqm0wtChQ5ud2fyeWCNGNPtyZmYdSms3ok8Bvh8RQ4DvA9dk6YXaNaKe9IIi4qqIGBsRY/v379/szOYCiNtBzMw21doBZBJwZ7Z9G5BrRK8ChuQdN5hUvVWVbddObxX9+sG227onlplZIa0dQJYBX8i2vwgsyLanARMl9ZA0gtRYPjMilgPvSNor6311PHB3a2Z41CgHEDOzQkrWBiLpZmB/oJ+kKuBc4DvAbyVtBnxI1l4REXMlVQLzgPXASRGR6/s0hdSja3PgvuzVaioq4N57W/OOZmbtQ8kCSEQcW8euPeo4/nzg/ALps4HdWjBrTVJRAddeC2vXQp8+5cqFmVnb45HoDcitTuiGdDOzmhxAGuBJFc3MCnMAacCIEdCtm0sgZma1OYA0oFs32Gknl0DMzGpzAGkEd+U1M9uUA0gjVFTAwoWwfn25c2Jm1nY4gDRCRQV8/DEsWVLunJiZtR0OII3gnlhmZptyAGmE3FgQBxAzs2oOII3Qpw9sv7278pqZ5XMAaaSKCpdAzMzyOYA0krvympnV5ADSSBUV8Oab8MYb5c6JmVnb4ADSSF6d0MysJgeQRnJXXjOzmhxAGmnoUOjRwwHEzCynZAFE0rWSVkp6oVb6yZLmS5or6cK89KmSFmb7Ds5L30PSnGzfJdnStq2ua1fYeWcHEDOznFKWQK4DDslPkDQemAB8KiJ2BS7O0kcDE4Fds3Mul9Q1O+0K0tK3I7NXjWu2pooKt4GYmeWULIBExGPA6lrJU4ALImJddszKLH0CcEtErIuIJcBCYJykAUCviJgREQHcABxZqjw3pKICFi+GdevKlQMzs7ajtdtAdgb2lfSkpEcl7ZmlDwJezzuuKksblG3XTi9I0mRJsyXNXrVqVQtnPY0F2bABFi1q8UubmbU7rR1ANgP6AnsBPwQqszaNQu0aUU96QRFxVUSMjYix/fv3b4n81uCuvGZm1Vo7gFQBd0YyE9gI9MvSh+QdNxhYlqUPLpBeFp5U0cysWmsHkLuALwJI2hnoDrwBTAMmSuohaQSpsXxmRCwH3pG0V1ZSOR64u5Xz/A9bbQWDBjmAmJlBqlIqCUk3A/sD/SRVAecC1wLXZl17PwImZY3jcyVVAvOA9cBJEbEhu9QUUo+uzYH7slfZeFJFM7OkZAEkIo6tY9c36zj+fOD8Aumzgd1aMGvNUlEBN94IEVCeESlmZm2DR6I3UUUFvPUWrFhR7pyYmZWXA0gTuSHdzCxxAGkiT6poZpY4gDTRoEGw5ZYeC2Jm5gDSRF26eHVCMzNwACmKA4iZmQNIUSoq4NVX4YMPyp0TM7PycQApQkVFGgeyYEG5c2JmVj4OIEVwV14zMweQoowcmUahO4CYWWfmAFKELbaAYcPcldfMOjcHkCJ5UkUz6+wcQIo0alQqgUSdy1uZmXVsDiBFqqiA996DpUvLnRMzs/JwACmS58Qys86uZAFE0rWSVmaLR9Xed4akkNQvL22qpIWS5ks6OC99D0lzsn2XZCsTlp278ppZZ1fKEsh1wCG1EyUNAQ4CXstLGw1MBHbNzrlcUtds9xXAZNIytyMLXbMcdtgBevVyADGzzqtkASQiHgNWF9j1a+BMIL/5eQJwS0Ssi4glwEJgnKQBQK+ImJEtfXsDcGSp8twUUqrGcldeM+usWrUNRNIRwNKIeK7WrkHA63mfq7K0Qdl27fS6rj9Z0mxJs1etWtVCua6bu/KaWWfWagFE0hbAj4D/LLS7QFrUk15QRFwVEWMjYmz//v2Ly2gTjBoFVVXwzjslv5WZWZvTmiWQHYERwHOSXgEGA09L2oFUshiSd+xgYFmWPrhAepuQ64n18svlzYeZWTm0WgCJiDkRsV1EDI+I4aTgsHtE/B2YBkyU1EPSCFJj+cyIWA68I2mvrPfV8cDdrZXnhuQCiNtBzKwzKmU33puBGcAoSVWSvl3XsRExF6gE5gH3AydFxIZs9xTg96SG9UXAfaXKc1PtuCN07ep2EDPrnDYr1YUj4tgG9g+v9fl84PwCx80GdmvRzLWQHj1gxAgHEDPrnDwSvZncldfMOisHkGaqqEiN6Bs2NHysmVlH4gDSTKNGwYcfwmuvNXysmVlH4gDSTJ5U0cw6KweQZnJXXjPrrBxAmuHCC2HOHNh22+oSyPTpKd3MrKNzAGmGPfeEo49OM/O+9FIKHkcfndLNzDo6B5BmGD8eKith0SKYMQOOOip9Hj++3DkzMyu9Rg8klHQYab2Onrm0iPhJKTLVnowfDxMnwnXXwdixDh5m1nk0qgQi6UrgGOBk0gy5RwHDSpivdmP6dLjnHhg8GB56KL3MzDqDxlZh7R0RxwNrIuLHwOeoOXtup5Rr86ishMsvh40b4etfT+lmZh1dY6uwPsje35c0EHiTNDV7pzZrVnWbx8aNMHp0GlQ4c6arssys42tsALlHUh/gIuBp0qJOvy9ZrtqJM8+s3u7SBX74QzjxRBgzpnx5MjNrLUpLjTfhBKkH0DMi3ipNllrG2LFjY/bs2a16z48+SrPzjhoFjzzSqrc2M2sRkp6KiLGNObbeEoikL0bEI5K+VmAfEXFnsZnsiLp3h+9/P5VEZs3yeBAz69gaakT/QvZ+eIHXV0qYr3Zr8mTo3duj0c2s46s3gETEudnmTyLixPwX8NP6zpV0raSVkl7IS7tI0kuSnpf0p6xdJbdvqqSFkuZLOjgvfQ9Jc7J9l2RL27ZZvXrBlClwxx2wcGG5c2NmVjqN7cZ7R4G02xs45zrgkFppDwG7RcSngJeBqQCSRgMTSQMVDwEul9Q1O+cKYDJpnfSRBa7Z5px6KnTrBhdfXO6cmJmVTr0BRFKFpK8DvSV9Le91Ankj0guJiMeA1bXSHoyI9dnHJ4DB2fYE4JaIWBcRS0jrn4+TNADoFREzIrX23wAc2cTv2Op22AEmTUqj01esKHduzMxKo6ESyChSW0cfarZ/7A58p5n3/hZwX7Y9CHg9b19VljYo266dXpCkyZJmS5q9atWqZmavec44I/XKuvTSsmbDzKxk6u2FFRF3S7oHOCsift5SN5X0I2A9cFMuqdDt60kvKCKuAq6C1I23mdlslp13hq9+FX73OzjrLNh663Lmxsys5TXYBhIRG4CDWuqGkiaRSjXHRfUglCpqTo0yGFiWpQ8ukN4unHkmrF0Lv+/0Qy7NrCNqbCP6/0m6TNK+knbPvZp6M0mHAGcBR0TE+3m7pgETJfWQNILUWD4zIpYD70jaK+t9dTxwd1PvWy6f/Sx84Qvwq1+l6iwzs46ksVOZ7J2950/fHsAX6zpB0s3A/kA/SVXAuaReVz2Ah7LeuE9ExPciYq6kSmAeqWrrpKzkAzCF1KNrc1KbyX20I2eeCYcdBrfcAscfX+7cmJm1nCZPZdJelGMqk0Ii4FOfSu/PP5/mzDIza6uaMpVJY9cD2V7SNZLuyz6PlvTt5mSys5BSKWTuXLivXZWdzMzq19h/D18HPAAMzD6/DJxWigx1RBMnwpAh8ItflDsnZmYtp7EBpF9EVAIbAbLBgBvqP8VyunWD00+Hv/41rZ1uZtYRNDaAvCdpW7IxGJL2Atr0dO5tzb/8C/TtCxddVO6cmJm1jMYGkNNJXW13lPQ4aUqRk0uWqw5oq63gpJPgrrtg/vxy58bMrPkaFUAi4mnS1O57A98Fdo2I50uZsY7o5JOhRw9PsmhmHUNTOpWOAz5NmgfrWEke1dBE222Xlry94QZYvrzcuTEza57GduP9I3Ax8Hlgz+zVqH7CVtPpp8P69fDb35Y7J2ZmzdPYkehjgdHRUUcdtqI774R994UrroCpU9PqhdOnpyVwzzyz3LkzM2u8xlZhvQDsUMqMdBZ77gnPPQdvvw1XXZWCx9FHe/10M2t/GlsC6QfMkzQTWJdLjIgjSpKrDmz8+FQKOfhg+MlPoGdPqKxM6WZm7UljA8h5pcxEZzN+PBxzDNx4Yyp5OHiYWXvUqAASEY+WOiOdyfTpcP/9MGpU2r711hRQzMzak3oDiKR3KLwCoICIiF4lyVUHlmvzqKyEESPSyoWTJqUuvi6JmFl7Um8jekRsHRG9Cry2dvAozqxZ1W0ew4fDv/87rFuX1gsxM2tPvB5ImX3wAYwenaY6eeYZ2KyxrVJmZiXQ4uuBFJmJayWtlPRCXto2kh6StCB775u3b6qkhZLmSzo4L30PSXOyfZdkS9t2GJtvnpa8feEFuPzycufGzKzxSrk+3nXAIbXSzgYejoiRwMPZZySNBiYCu2bnXC6pa3bOFcBk0jrpIwtcs9078kg46CD4z/+ElSvLnRszs8YpWQCJiMeA1bWSJwDXZ9vXA0fmpd8SEesiYgmwEBgnaQDQKyJmZKPgb8g7p8OQ4JJL4L330uh0M7P2oLVX6N4+IpYDZO/bZemDgNfzjqvK0gZl27XTC5I0WdJsSbNXrVrVohkvtYoKOO00uPZamDmz3LkxM2tYaweQuhRq14h60guKiKsiYmxEjO3fv3+LZa61nHMO7LAD/Nu/wcaN5c6NmVn9WjuArMiqpcjeczX+VcCQvOMGA8uy9MEF0jukXr3SioWzZsEf/lDu3JiZ1a+1A8g0YFK2PQm4Oy99oqQekkaQGstnZtVc70jaK+t9dXzeOR3SccfBPvuktpC1a8udGzOzupWyG+/NwAxglKQqSd8GLgAOkrQAOCj7TETMBSqBecD9wEkRsSG71BTg96SG9UXAfaXKc1sgwaWXwhtvwLnnljs3ZmZ180DCNupf/zVN9/7MM/DJT5Y7N2bWWbSJgYTWPD/9aVps6uSToYPGeDNr5xxA2qhtt4Xzz4dHH01zZ5mZtTUOIG3Yd74Dn/kM/OAH8O675c6NmVlNDiBtWNeucNllsHQp/Pzn5c6NmVlNDiBt3N57w/HHwy9/CQsWlDs3ZmbVHEDagV/8Anr0SFOdmJm1FQ4g7cAOO8B++8G998I991SnT58OF15YvnyZWefmANJOnHpqahP57nfhww+rl8bdc89y58zMOisHkHbioIPgggtg2bK0guFRR1UvjWtmVg4OIO3IGWfA4YfDkiWwYQP061fuHJlZZ+YA0o5Mnw4zZsCJJ8Lbb8O4cXD77eXOlZl1Vg4g7USuzaOyMi06deutsH59qsr60Y9SicTMrDU5gLQTs2bVbPP4p39KPbLGjUuDDI84wtO/m1nr8my8HcB//3daxXDECLjrrtTIbmZWDM/G28l897upiuvtt+Gzn01BxMys1MoSQCR9X9JcSS9IullST0nbSHpI0oLsvW/e8VMlLZQ0X9LB5chzW/f5z8Ps2bDLLvDVr8J553lddTMrrVYPIJIGAacAYyNiN6ArMBE4G3g4IkYCD2efkTQ6278rcAhwuaSurZ3v9mDwYHjsMTjhBPjxj2G33WqOXAePXjezllOuKqzNgM0lbQZsASwDJgDXZ/uvB47MticAt0TEuohYQlradlwr57fd6Nkz9dK69FKYPx+OPBKuz56qR6+bWUtq9QASEUuBi4HXgOXAWxHxILB9RCzPjlkObJedMgh4Pe8SVVma1UFKjeoPPwxbbpnGjRxzTHU3YI9eN7OWUI4qrL6kUsUIYCCwpaRv1ndKgbSCXcckTZY0W9LsVatWNT+z7dz++8Pzz6fJGCsrYdiw1MhuZtYSylGFdSCwJCJWRcTHwJ3A3sAKSQMAsveV2fFVwJC88weTqrw2ERFXRcTYiBjbv3//kn2B9mTxYvj4Y9hrL3jqqdTI/txz5c6VmXUE5QggrwF7SdpCkoADgBeBacCk7JhJwN3Z9jRgoqQekkYAI4GZrZzndil/9PqMGanxvKoKxo6F3/zGvbTMrHnK0QbyJHA78DQwJ8vDVcAFwEGSFgAHZZ+JiLlAJTAPuB84KSI8cUcj1B69/sMfprmzRo6E738fDjsMVqwobx7NrP3ySPROKAKuvBJOPx223hquuw4OPbTcuTKztsAj0a1eEkyZktpEBgxIJZFTTkkLVZmZNZYDSCc2ejQ8+WRaa/3SS9P4kBdeKHeuzKy9cADp5Hr2hF//Gu67D1atgjFjUmkkv2bTo9fNrBAHEAPgkEPSmJGxY1NpZO+9YeVKj143s7ptVu4MWNux3Xapu+8pp8Bll8HQodC1K9x0k0evm9mmXAKxGqRUAvne92DdOnj/fTjuOPjBD2BZweGbZtZZOYDYJqZPT+NFzjkH+vaFffaB3/42LVg1ZQq88kq5c2hmbYEDiNWQP3r9Jz+BO+6AZ55JM/qecEKa6XenndL2/Pnlzq2ZlZMDiNVQe/T6+PHp89KlaencxYvh5JNT2i67pGBz2mkp8ORzzy2zjs8j0a0oK1em+bQuuwzeeQe6d4df/QpOOqlmKcaN72btS1NGojuAWLOsWZOCyEUXpUAybBisXp2qvg46qNy5M7Om8lQm1mr69k2N7cuWwYEHwquvpkDyz/8MU6fCokXlzqGZlYoDiLWIWbPg2WfhRz+CXr1gxx1TG8hOO6WSyG23wUcflTuXZtaSHECs2fLbPH72M7jrLnj5Zbj55tST6+WX0/7Bg+Gss2DhwnTehRe68d2sPXMAsWarq+fWK6+k6q3Fi+Hee9N4kl/+Mq1HcuCB8O67KbDkgoinTTFrX9yIbq1q2TL4wx/g6qtTe0nv3qlq69hjYdo099wyK7c234guqY+k2yW9JOlFSZ+TtI2khyQtyN775h0/VdJCSfMlHVyOPFvLGDgwtZMsWpRmAP7iF9M6JNdeC++9l4LLrbem3l1m1raVqwrrt8D9EVEBfJq0JvrZwMMRMRJ4OPuMpNHARGBX4BDgckldy5JrazFdu6YZgE8+GbbZBr761bRG+913w8SJ0L8/7LcfXHABzJlTPb28203M2o5WDyCSegH7AdcARMRHEbEWmABcnx12PXBktj0BuCUi1kXEEmAhMK51c22lkGvzuO02uPPOVCLp3j1N5nj22amNZOpU+NSn0viS730PPv4YjjrK7SZmbUE5pnP/BLAK+IOkTwNPAacC20fEcoCIWC5pu+z4QcATeedXZWmbkDQZmAwwdOjQ0uTeWkxdje+zZqXeXD/7WWozue8++J//SdPKv/sudOsGBx8MX/gCzJ6dApDbTcxaX6s3oksaSwoI+0TEk5J+C7wNnBwRffKOWxMRfTvLzdcAAA4SSURBVCX9DpgRETdm6dcA90bEHfXdx43oHc9HH8Ff/5qCyQ03wJtvpvSttko9vPbbLwWVsWOhR4/q8y68MJVQ8oPM9OkpUJ15Zut+B7O2rq03olcBVRHxZPb5dmB3YIWkAQDZ+8q844fknT8Y8MoUnVD37nDAAXD44WndktNOg623TmlLl6bG+c9/Hvr0SY3z552XAsUnP+nuwmal0OoBJCL+DrwuaVSWdAAwD5gGTMrSJgF3Z9vTgImSekgaAYwEZrZilq0NyR+0+Otfp0b3xx+HSy5Ja7r/6U+preStt+CnP02BZMIE2H57OOww+MY3UhuKuwubNV9ZxoFIGgP8HugOLAZOJAWzSmAo8BpwVESszo7/EfAtYD1wWkTc19A9XIXVMTWlOuqtt+Bvf4PHHoNHH4WZM6t7c40aBXvtBZ/7XHrfdVfYbLPi7mPWkXg2XhxArKbp01PJ49BD00zBY8bAggWp1AKw5ZYwblwKJnvtBevXw3e/W11S8RT11lk0JYCUoxeWWavK7y48fjyceGL6fOutMHw4PPEEzJiR3i+6KAUPgAED4Mtfhv33hyefTL3AHDzMqrkEYh1eU6qj3n8fnn66Oqg8+GDqOgyp4X7kSNh9d/jMZ6pf/fo1/T5mbZWrsHAAsebLlVyOOy5NsfK1r8HatWmN+FdfrT5uyJAUVPr0SY34V1+dqsv+8hdXe1n74yoss2aq3eYxYULNz2++mdY/eeaZVGJ55hmYPz810h9zDJxwQqoKmzAhBZunn4bRo6Fnz03v5ZKLtVcugZgVUMyP+rvvwvPPp+7D99+fJo5cswY++CDt79o19f761KdqvhYsSEHHDfbWFrgKCwcQK4/cj/+UKXDFFWlRraFD4bnnUnDJvV55pfqcvn1TNdjLL6dxK48/DtddB0ceWdddXGqx0nEVllkZ1C45jB9f/fmoo9Ir56230izD+UFlw4a08Bak2Yn790/VXqNHwy67VG/vsEMKHvn3yr+3WWtxCcSshTSnVJALAN/4Rip9HHdcmvtr3rz0euut6mN7906BpE+fNEjy8MPhgQdS8DjwwNLkzzoPV2HhAGLtR+2SS+3PEfD3v1cHkxdfrN7ODYSENJJ+xx1h551Td+Odd65+DRy4aa8wt7VYIa7CMmtH6pvWfvz4NP5kwID0OuCA6vNyo+sPPxxuvx2OOALWrUttKQ89lFZ6zNlyyxRUdtsNvvKVVFJ59FG46qo0g3FdXGqx+rgEYtYONVRq2bgRqqpSMFmwIL3nXosWVc8JBmmW4+HD4ROfgBEj0iu3vXQpfOtbLrV0Ji6BmHVwDZVaunRJvb+GDq3ZLpILAMccAzfemKZ16d4dFi+GJUvSlC2116Pfckv40pdgp53SmJZvfSsNqHzqqbRS5LbbplJSPpdcOgeXQMw6iYZKLTlvvZWCSS6oLF6cGukXLUrtLLm5wnI23zwFqmHDqoPW+++nbsy/+x18/etpWpiGSi4OOm2DG9FxADGrrdgf6NpjW66+OgWJ115Lr1dfrbm9YsWm15DSWJeddkptOQMHVrfr5F4LFqQR/E2pLnPQaXkOIDiAmLWExpZa8n34YWp/ee01+M1v4M9/Tj/yQ4fC8uVpnfvly1ODf209e6buy4MGpUB08MFpwsrttqt+bb99eu/TJ3UEaGr+HHTq5zYQM2sRDbW1FNKzZyppvP56qro655xUcvnFL6rPiUjtKMuXb/p68EGYOzcNpHziCbjnnpqN/jmbbZaO6d07BZpRo1I127HHptLMmjVpf//+acbkbbZJbUPFDsJ04NlU2UogkroCs4GlEfEVSdsAtwLDgVeAoyNiTXbsVODbwAbglIh4oKHruwRiVj7FlFzyz8tVl1VWwr77pskrV66sfq1YUfPz00+nHmPdusHHHxe+dpcuqcG/X7903EsvpW7NL72UOgbss091wMkFnW7dmved2mPQaRdVWJJOB8YCvbIAciGwOiIukHQ20DcizpI0GrgZGAcMBP4X2DkiNtR3fQcQs/Ip5oezpYLOjTemJYpXrYI33qj5nr89b17ark/fvtUBZbvtUnB65JEUbGbMgLPOSuNo+vZNJZy+fVOnguZ8p3IHnTYfQCQNBq4HzgdOzwLIfGD/iFguaQDwl4gYlZU+iIj/ys59ADgvImbUdw8HELP2pZxB56qrUhXYqlWpRJMLNrlXftrKlfV/j549UyDJBZUNG1IJacyYNP/Zd76Tlk3eZpuar169Uimp3EGnPQSQ24H/ArYGzsgCyNqI6JN3zJqI6CvpMuCJiLgxS78GuC8ibi9w3cnAZIChQ4fu8Wr+qj9m1uG0ZtDJP/fEE+Gaa+DnP0/Tx6xendpccu/526tXp+7Qb79d/7W7dElBZ9tt09T/ixalaWgWLkxdoceMSR0Hcq++fau3n3kmtf20xIDPNh1AJH0FODQi/lXS/jQcQH4HzKgVQO6NiDvqu49LIGZWSHO7Mze3tHPllWkyzNWr63/NmZM6FeR6pm3cWP/3yh237bbp2NtuK262gLbeC2sf4AhJhwI9gV6SbgRWSBqQV4WVKyhWAUPyzh8MLGvVHJtZh1EoSOSm369PMT3S6pvivzFBJ9eD7dZbU9Bbu7bu15o1qX3muefgP/6jlaaaiYiyvYD9gXuy7YuAs7Pts4ELs+1dgeeAHsAIYDHQtaFr77HHHmFmVk6/+EXEI4/UTHvkkZRel0ceiejXr/q82p8bOu+ccxp3fF2A2dHI3/C2NA7kAqBS0reB14CjACJirqRKYB6wHjgpGuiBZWbWFhRT2mnNkk5zeSS6mVk716l6YbUGBxAzs6ZrSgDpUurMmJlZx+QAYmZmRXEAMTOzojiAmJlZURxAzMysKB22F5akVcCrQD+ggTk3OwU/h8TPIfFzSPwcquWexbCI6N+YEzpsAMmRNLuxXdI6Mj+HxM8h8XNI/ByqFfMsXIVlZmZFcQAxM7OidIYAclW5M9BG+Dkkfg6Jn0Pi51Ctyc+iw7eBmJlZaXSGEoiZmZWAA4iZmRWlwwYQSYdImi9poaSzy52fcpH0iqQ5kp6V1KmmJ5Z0raSVkl7IS9tG0kOSFmTvfcuZx9ZQx3M4T9LS7O/i2WyF0A5N0hBJ0yW9KGmupFOz9E71N1HPc2jy30SHbAOR1BV4GTiItCTuLODYiJhX1oyVgaRXgLER0ekGS0naD3gXuCEidsvSLgRWR8QF2T8s+kbEWeXMZ6nV8RzOA96NiIvLmbfWlC2VPSAinpa0NfAUcCRwAp3ob6Ke53A0Tfyb6KglkHHAwohYHBEfAbcAE8qcJ2tlEfEYsLpW8gTg+mz7etL/OB1aHc+h04mI5RHxdLb9DvAiMIhO9jdRz3Noso4aQAYBr+d9rqLIB9QBBPCgpKckTS53ZtqA7SNiOaT/kYDtypyfcvo3Sc9nVVwdutqmNknDgc8AT9KJ/yZqPQdo4t9ERw0gKpDW8erqGmefiNgd+DJwUladYXYFsCMwBlgO/LK82Wk9krYC7gBOi4i3y52fcinwHJr8N9FRA0gVMCTv82BgWZnyUlYRsSx7Xwn8iVS915mtyOqAc3XBK8ucn7KIiBURsSEiNgJX00n+LiR1I/1o3hQRd2bJne5votBzKOZvoqMGkFnASEkjJHUHJgLTypynVidpy6yRDElbAl8CXqj/rA5vGjAp254E3F3GvJRN7gcz81U6wd+FJAHXAC9GxK/ydnWqv4m6nkMxfxMdshcWQNYF7TdAV+DaiDi/zFlqdZI+QSp1AGwG/L/O9Bwk3QzsT5qmegVwLnAXUAkMBV4DjoqIDt3AXMdz2J9UVRHAK8B3c+0AHZWkzwN/BeYAG7PkfyfV/3eav4l6nsOxNPFvosMGEDMzK62OWoVlZmYl5gBiZmZFcQAxM7OiOICYmVlRHEDMzKwoDiDW6Un6i6SDa6WdJunyBs57t7Q5q/O+N2fTTXy/Vvp5ks7ItntmM8ueW448WuewWbkzYNYG3EwabPpAXtpE4IflyU7dJO0A7B0Rw+o5pjtplPFTEfHjVsucdTougZjB7cBXJPWAf0wwNxD4m6StJD0s6elsXZVNZnWWtL+ke/I+XybphGx7D0mPZpNZPpA3ZcYpkuZlJYlbClyzp6Q/ZPd8RtL4bNeDwHbZeg37Fvgum5Fmn14QEZ12HRxrHS6BWKcXEW9KmgkcQprGYiJwa0SEpA+Br0bE25L6AU9ImhaNGIGbzTd0KTAhIlZJOgY4H/gWcDYwIiLWSepT4PSTsrx9UlIFaUblnYEjgHsiYkwdtz0T+N+IOK0pz8CsGC6BmCW5aiyy95uzbQE/l/Q88L+kZQG2b+Q1RwG7AQ9Jehb4D9LEngDPAzdJ+iawvsC5nwf+CBARLwGvAjs34p5/Az6XBRuzknIAMUvuAg6QtDuweW7BHeA4oD+wR/av/hVAz1rnrqfm/0u5/QLmRsSY7PXJiPhStu8w4HfAHsBTkmrXBhRakqAxHgNOA+6TNLDIa5g1igOIGRAR7wJ/Aa6luvQB0BtYGREfZ+0QhRqvXwVGS+ohqTdwQJY+H+gv6XOQqrQk7SqpCzAkIqaTqpz6AFvVuuZjpOBFVpoYml2vMd/lDuAi4P46qsfMWoTbQMyq3QzcSXVVFsBNwJ8lzQaeBV6qfVJEvC6pklQttQB4Jkv/SNI/AZdkgWUz0gzRLwM3ZmkCfh0Ra2td9nLgSklzSCWcE7L2kkZ9kYi4MuuxNU3SlyLiw8Y9ArPG82y8ZmZWFFdhmZlZURxAzMysKA4gZmZWFAcQMzMrigOImZkVxQHEzMyK4gBiZmZF+f8OBu6h8j47DQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plt.plot(K, inertias, 'bx-')\n",
    "# plt.xlabel('Values of K')\n",
    "\n",
    "# plt.ylabel('Inertia')\n",
    "# plt.title('The Elbow Method using Inertia')\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "687c45b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "фикс сида + 7 кластеров для трейна + всего датасета"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "1c7726ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "47742"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(train.buyer_id.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "010d9e66",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "47742"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(pred_df.buyer_id.unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "c64d07e8",
   "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>buyer_id</th>\n",
       "      <th>pav_order_id</th>\n",
       "      <th>created</th>\n",
       "      <th>item_id</th>\n",
       "      <th>count</th>\n",
       "      <th>price_sold</th>\n",
       "      <th>flag_weight_goods</th>\n",
       "      <th>weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [buyer_id, pav_order_id, created, item_id, count, price_sold, flag_weight_goods, weight]\n",
       "Index: []"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train[train.buyer_id == 4620121727]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "27e569a8",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}