useful functions for rpg Data usage

rpg_utils.ipynb

@@ -0,0 +1,340 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 69,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "from fuzzywuzzy import process\n",
+    "from typing import List\n",
+    "\n",
+    "\n",
+    "def get_popular_agriculture(\n",
+    "    df: pd.DataFrame,\n",
+    "    region: str,\n",
+    "    surface_parc: str = None,\n",
+    "    groups_agri: List[str] = None,\n",
+    "):  # surface en hectare\n",
+    "    unique_regions = df[\"REGION\"].unique().tolist()\n",
+    "    region, _ = process.extractOne(region, unique_regions)\n",
+    "    df = df[df[\"REGION\"] == region]\n",
+    "    if groups_agri != None:\n",
+    "        unique_agri_group = df[\"LIBELLE_GROUPE_CULTURE\"].unique().tolist()\n",
+    "        groups_agri_new = []\n",
+    "        for group_culture in groups_agri:\n",
+    "            groups_agri_new.append(process.extractOne(group_culture, unique_agri_group)[0])\n",
+    "        df = df[df[\"LIBELLE_GROUPE_CULTURE\"].isin(groups_agri_new)]\n",
+    "    if surface_parc != None:\n",
+    "        popular_cultures = df[\"LIBELLE_CULTURE\"].value_counts().head(10).index\n",
+    "        df = df[df[\"LIBELLE_CULTURE\"].isin(popular_cultures)]\n",
+    "        df[\"proximity\"] = abs(df[\"SURF_PARC\"] - surface_parc)\n",
+    "        sorted_df = df.sort_values(by=\"proximity\").drop(columns=\"proximity\")\n",
+    "        return popular_cultures, sorted_df[[\"LIBELLE_CULTURE\",\"LIBELLE_GROUPE_CULTURE\",\"SURF_PARC\"]]\n",
+    "    popular_cultures = df[\"LIBELLE_CULTURE\"].value_counts().head(10).index\n",
+    "    df = df[df[\"LIBELLE_CULTURE\"].isin(popular_cultures)]\n",
+    "    return popular_cultures, df[[\"LIBELLE_CULTURE\",\"LIBELLE_GROUPE_CULTURE\",\"SURF_PARC\"]]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 72,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv(\"./data_rpg/data_prepared_rpg.csv\")\n",
+    "\n",
+    "pop, sorted_df = get_popular_agriculture(\n",
+    "    df,\n",
+    "    \"Bretagne\",\n",
+    "    132.5\n",
+    "\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 73,
+   "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>LIBELLE_CULTURE</th>\n",
+       "      <th>LIBELLE_GROUPE_CULTURE</th>\n",
+       "      <th>SURF_PARC</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
+       "      <td>Autre plante fourragère annuelle (ni légumineu...</td>\n",
+       "      <td>Fourrage</td>\n",
+       "      <td>118.12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49</th>\n",
+       "      <td>Aïl</td>\n",
+       "      <td>Légumes ou fleurs</td>\n",
+       "      <td>49.67</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>61</th>\n",
+       "      <td>Plantes médicinales et à parfum non pérennes (...</td>\n",
+       "      <td>Autres cultures industrielles</td>\n",
+       "      <td>29.14</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>78</th>\n",
+       "      <td>Plante aromatique pérenne non arbustive ou arb...</td>\n",
+       "      <td>Autres cultures industrielles</td>\n",
+       "      <td>20.14</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>Autre culture pérenne et jachère dans les bana...</td>\n",
+       "      <td>Divers</td>\n",
+       "      <td>5.31</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>37</th>\n",
+       "      <td>Agrume</td>\n",
+       "      <td>Vergers</td>\n",
+       "      <td>3.18</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Plantes aromatiques herbacées non pérennes (&lt; ...</td>\n",
+       "      <td>Autres cultures industrielles</td>\n",
+       "      <td>277.71</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>117</th>\n",
+       "      <td>Avoine de printemps</td>\n",
+       "      <td>Autres céréales</td>\n",
+       "      <td>877.02</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>91</th>\n",
+       "      <td>Artichaut</td>\n",
+       "      <td>Légumes ou fleurs</td>\n",
+       "      <td>2953.38</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>104</th>\n",
+       "      <td>Avoine d’hiver</td>\n",
+       "      <td>Autres céréales</td>\n",
+       "      <td>5176.17</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                       LIBELLE_CULTURE  \\\n",
+       "26   Autre plante fourragère annuelle (ni légumineu...   \n",
+       "49                                                 Aïl   \n",
+       "61   Plantes médicinales et à parfum non pérennes (...   \n",
+       "78   Plante aromatique pérenne non arbustive ou arb...   \n",
+       "14   Autre culture pérenne et jachère dans les bana...   \n",
+       "37                                              Agrume   \n",
+       "2    Plantes aromatiques herbacées non pérennes (< ...   \n",
+       "117                                Avoine de printemps   \n",
+       "91                                           Artichaut   \n",
+       "104                                     Avoine d’hiver   \n",
+       "\n",
+       "            LIBELLE_GROUPE_CULTURE  SURF_PARC  \n",
+       "26                        Fourrage     118.12  \n",
+       "49               Légumes ou fleurs      49.67  \n",
+       "61   Autres cultures industrielles      29.14  \n",
+       "78   Autres cultures industrielles      20.14  \n",
+       "14                          Divers       5.31  \n",
+       "37                         Vergers       3.18  \n",
+       "2    Autres cultures industrielles     277.71  \n",
+       "117                Autres céréales     877.02  \n",
+       "91               Légumes ou fleurs    2953.38  \n",
+       "104                Autres céréales    5176.17  "
+      ]
+     },
+     "execution_count": 73,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sorted_df"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 71,
+   "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>LIBELLE_CULTURE</th>\n",
+       "      <th>LIBELLE_GROUPE_CULTURE</th>\n",
+       "      <th>SURF_PARC</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>Plantes aromatiques herbacées non pérennes (&lt; ...</td>\n",
+       "      <td>Autres cultures industrielles</td>\n",
+       "      <td>277.71</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>14</th>\n",
+       "      <td>Autre culture pérenne et jachère dans les bana...</td>\n",
+       "      <td>Divers</td>\n",
+       "      <td>5.31</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
+       "      <td>Autre plante fourragère annuelle (ni légumineu...</td>\n",
+       "      <td>Fourrage</td>\n",
+       "      <td>118.12</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>37</th>\n",
+       "      <td>Agrume</td>\n",
+       "      <td>Vergers</td>\n",
+       "      <td>3.18</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>49</th>\n",
+       "      <td>Aïl</td>\n",
+       "      <td>Légumes ou fleurs</td>\n",
+       "      <td>49.67</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>61</th>\n",
+       "      <td>Plantes médicinales et à parfum non pérennes (...</td>\n",
+       "      <td>Autres cultures industrielles</td>\n",
+       "      <td>29.14</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>78</th>\n",
+       "      <td>Plante aromatique pérenne non arbustive ou arb...</td>\n",
+       "      <td>Autres cultures industrielles</td>\n",
+       "      <td>20.14</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>91</th>\n",
+       "      <td>Artichaut</td>\n",
+       "      <td>Légumes ou fleurs</td>\n",
+       "      <td>2953.38</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>104</th>\n",
+       "      <td>Avoine d’hiver</td>\n",
+       "      <td>Autres céréales</td>\n",
+       "      <td>5176.17</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>117</th>\n",
+       "      <td>Avoine de printemps</td>\n",
+       "      <td>Autres céréales</td>\n",
+       "      <td>877.02</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                                       LIBELLE_CULTURE  \\\n",
+       "2    Plantes aromatiques herbacées non pérennes (< ...   \n",
+       "14   Autre culture pérenne et jachère dans les bana...   \n",
+       "26   Autre plante fourragère annuelle (ni légumineu...   \n",
+       "37                                              Agrume   \n",
+       "49                                                 Aïl   \n",
+       "61   Plantes médicinales et à parfum non pérennes (...   \n",
+       "78   Plante aromatique pérenne non arbustive ou arb...   \n",
+       "91                                           Artichaut   \n",
+       "104                                     Avoine d’hiver   \n",
+       "117                                Avoine de printemps   \n",
+       "\n",
+       "            LIBELLE_GROUPE_CULTURE  SURF_PARC  \n",
+       "2    Autres cultures industrielles     277.71  \n",
+       "14                          Divers       5.31  \n",
+       "26                        Fourrage     118.12  \n",
+       "37                         Vergers       3.18  \n",
+       "49               Légumes ou fleurs      49.67  \n",
+       "61   Autres cultures industrielles      29.14  \n",
+       "78   Autres cultures industrielles      20.14  \n",
+       "91               Légumes ou fleurs    2953.38  \n",
+       "104                Autres céréales    5176.17  \n",
+       "117                Autres céréales     877.02  "
+      ]
+     },
+     "execution_count": 71,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sorted_df"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "hackathon",
+   "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": "3.13.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}