From 14cca655fb5998ae0635d709ca7758c5f75e0880 Mon Sep 17 00:00:00 2001
From: Antoine Tavant <antoinetavant@hotmail.fr>
Date: Fri, 20 Sep 2024 15:06:03 +0200
Subject: [PATCH] Improve Tempo predictions

---
 .github/workflows/page_deploy.yml           |    7 +-
 .gitignore                                  |    1 +
 notebooks/datascience/tempo_predictor.ipynb | 1916 +++++++++++++++----
 src/energy_forecast/meteo.py                |   73 +
 src/energy_forecast/tempo_rte.py            |   33 +-
 5 files changed, 1699 insertions(+), 331 deletions(-)

diff --git a/.github/workflows/page_deploy.yml b/.github/workflows/page_deploy.yml
index b4fb0de..e1ea6f8 100644
--- a/.github/workflows/page_deploy.yml
+++ b/.github/workflows/page_deploy.yml
@@ -1,6 +1,9 @@
 name: documentation
 
-on: [push, pull_request, workflow_dispatch]
+on: 
+  push:
+    branches:
+      - main
 
 # Sets permissions of the GITHUB_TOKEN to allow deployment to GitHub Pages
 permissions:
@@ -37,4 +40,4 @@ jobs:
           path: './doc/_build'
       - name: Deploy to GitHub Pages
         id: deployment
-        uses: actions/deploy-pages@v4
\ No newline at end of file
+        uses: actions/deploy-pages@v4
diff --git a/.gitignore b/.gitignore
index 5b0805e..c32beb1 100644
--- a/.gitignore
+++ b/.gitignore
@@ -168,3 +168,4 @@ data/silver/*.nc
 data/silver/weather_forecasts/*.nc
 data/geo/*.nc
 .vscode/settings.json
+data/bronze/observations/*.csv.gz
diff --git a/notebooks/datascience/tempo_predictor.ipynb b/notebooks/datascience/tempo_predictor.ipynb
index d098320..20d3f71 100644
--- a/notebooks/datascience/tempo_predictor.ipynb
+++ b/notebooks/datascience/tempo_predictor.ipynb
@@ -7,12 +7,18 @@
     "# Calcul des jours Tempos\n",
     "\n",
     "Voir : [la doc](https://www.services-rte.com/files/live/sites/services-rte/files/pdf/20160106_Methode_de_choix_des_jours_Tempo.pdf\n",
-    ")"
+    ")\n",
+    "\n",
+    "## Data sources\n",
+    "\n",
+    "- the ENR energy production is the one available on the RTE API \"Production Forecast\"\n",
+    "- The Total consumption is the one available on the RTE API \"Consommation\" also available on Eco2mix (the Excel file)\n",
+    "- the temperature is the observed temperature available on the meteo.data.gouv.fr website, each department has a station that provides the temperature on the Day. The Mean temperature is the average of the temperature of the 95 departments.\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [
     {
@@ -31,13 +37,15 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [],
    "source": [
     "import pandas as pd\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
     "from energy_forecast import ROOT_DIR\n",
-    "from energy_forecast.tempo_rte import TempoPredictor"
+    "from energy_forecast.tempo_rte import TempoPredictor\n"
    ]
   },
   {
@@ -49,7 +57,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 51,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [
     {
@@ -161,7 +169,7 @@
        "[80740 rows x 2 columns]"
       ]
      },
-     "execution_count": 51,
+     "execution_count": 37,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -175,14 +183,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 52,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/tmp/ipykernel_1212/3459037049.py:1: DtypeWarning: Columns (5,18,19,20,21,22,26,27,30,31,33,36,37,38,39) have mixed types. Specify dtype option on import or set low_memory=False.\n",
+      "/tmp/ipykernel_4968/3459037049.py:1: DtypeWarning: Columns (5,18,19,20,21,22,26,27,30,31,33,36,37,38,39) have mixed types. Specify dtype option on import or set low_memory=False.\n",
       "  rte_all_data = pd.read_csv(ROOT_DIR / 'data' / 'silver' / 'rte_production.csv')[[\n"
      ]
     }
@@ -201,7 +209,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 53,
+   "execution_count": 39,
    "metadata": {},
    "outputs": [
     {
@@ -319,7 +327,7 @@
        "[90014 rows x 3 columns]"
       ]
      },
-     "execution_count": 53,
+     "execution_count": 39,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -330,7 +338,137 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 71,
+   "execution_count": 6,
+   "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>Prévision J-1</th>\n",
+       "      <th>Solaire</th>\n",
+       "      <th>Eolien</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>2014-01-01 00:00:00+01:00</th>\n",
+       "      <td>1274250.0</td>\n",
+       "      <td>4776.00</td>\n",
+       "      <td>114472.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-01-02 00:00:00+01:00</th>\n",
+       "      <td>1436150.0</td>\n",
+       "      <td>4871.50</td>\n",
+       "      <td>110732.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-01-03 00:00:00+01:00</th>\n",
+       "      <td>1449150.0</td>\n",
+       "      <td>3257.50</td>\n",
+       "      <td>132188.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-01-04 00:00:00+01:00</th>\n",
+       "      <td>1384600.0</td>\n",
+       "      <td>2870.50</td>\n",
+       "      <td>99399.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-01-05 00:00:00+01:00</th>\n",
+       "      <td>1402900.0</td>\n",
+       "      <td>5765.50</td>\n",
+       "      <td>82688.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-04 00:00:00+02:00</th>\n",
+       "      <td>1173500.0</td>\n",
+       "      <td>76644.50</td>\n",
+       "      <td>290434.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-05 00:00:00+02:00</th>\n",
+       "      <td>1125362.5</td>\n",
+       "      <td>72310.50</td>\n",
+       "      <td>230215.75</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-06 00:00:00+02:00</th>\n",
+       "      <td>967137.5</td>\n",
+       "      <td>61951.00</td>\n",
+       "      <td>224055.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-07 00:00:00+02:00</th>\n",
+       "      <td>928362.5</td>\n",
+       "      <td>46914.00</td>\n",
+       "      <td>146855.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-08 00:00:00+02:00</th>\n",
+       "      <td>708787.5</td>\n",
+       "      <td>35448.75</td>\n",
+       "      <td>70008.00</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>3751 rows × 3 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "                           Prévision J-1   Solaire     Eolien\n",
+       "2014-01-01 00:00:00+01:00      1274250.0   4776.00  114472.50\n",
+       "2014-01-02 00:00:00+01:00      1436150.0   4871.50  110732.50\n",
+       "2014-01-03 00:00:00+01:00      1449150.0   3257.50  132188.50\n",
+       "2014-01-04 00:00:00+01:00      1384600.0   2870.50   99399.00\n",
+       "2014-01-05 00:00:00+01:00      1402900.0   5765.50   82688.00\n",
+       "...                                  ...       ...        ...\n",
+       "2024-04-04 00:00:00+02:00      1173500.0  76644.50  290434.00\n",
+       "2024-04-05 00:00:00+02:00      1125362.5  72310.50  230215.75\n",
+       "2024-04-06 00:00:00+02:00       967137.5  61951.00  224055.00\n",
+       "2024-04-07 00:00:00+02:00       928362.5  46914.00  146855.00\n",
+       "2024-04-08 00:00:00+02:00       708787.5  35448.75   70008.00\n",
+       "\n",
+       "[3751 rows x 3 columns]"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "daily_production = rte_all_data.resample('1d').sum()\n",
+    "daily_production\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
    "metadata": {},
    "outputs": [
     {
@@ -429,7 +567,7 @@
        "[3585 rows x 1 columns]"
       ]
      },
-     "execution_count": 71,
+     "execution_count": 40,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -449,14 +587,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 72,
+   "execution_count": 41,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/tmp/ipykernel_1212/2617323631.py:4: PerformanceWarning: Non-vectorized DateOffset being applied to Series or DatetimeIndex.\n",
+      "/tmp/ipykernel_4968/2617323631.py:4: PerformanceWarning: Non-vectorized DateOffset being applied to Series or DatetimeIndex.\n",
       "  daily_consumption.index = daily_consumption.index + pd.DateOffset(hour=0)\n"
      ]
     },
@@ -477,7 +615,7 @@
        "Name: Prévision J-1, Length: 3752, dtype: float64"
       ]
      },
-     "execution_count": 72,
+     "execution_count": 41,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -492,14 +630,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 73,
+   "execution_count": 42,
    "metadata": {},
    "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/tmp/ipykernel_1212/452278787.py:3: PerformanceWarning: Non-vectorized DateOffset being applied to Series or DatetimeIndex.\n",
+      "/tmp/ipykernel_4968/994817580.py:3: PerformanceWarning: Non-vectorized DateOffset being applied to Series or DatetimeIndex.\n",
       "  daily_production.index = daily_production.index + pd.DateOffset(hour=0)\n"
      ]
     },
@@ -524,8 +662,8 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>SOLAR_FORECAST_D1</th>\n",
-       "      <th>EOLIEN_FORECAST_D1</th>\n",
+       "      <th>Solaire</th>\n",
+       "      <th>Eolien</th>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>start_date</th>\n",
@@ -595,24 +733,24 @@
        "</div>"
       ],
       "text/plain": [
-       "                           SOLAR_FORECAST_D1  EOLIEN_FORECAST_D1\n",
-       "start_date                                                      \n",
-       "2014-12-15 00:00:00+01:00               0.00             5085.00\n",
-       "2014-12-16 00:00:00+01:00            4870.89            39909.00\n",
-       "2014-12-17 00:00:00+01:00            4609.62           103617.00\n",
-       "2014-12-18 00:00:00+01:00            5258.76           122720.00\n",
-       "2014-12-19 00:00:00+01:00            6289.70           104494.00\n",
-       "...                                      ...                 ...\n",
-       "2024-08-12 00:00:00+02:00          116114.62            67128.44\n",
-       "2024-08-13 00:00:00+02:00           96650.51            36417.80\n",
-       "2024-08-14 00:00:00+02:00           74620.37            43453.81\n",
-       "2024-08-15 00:00:00+02:00          101109.47            61722.23\n",
-       "2024-08-16 00:00:00+02:00           95051.52            37254.94\n",
+       "                             Solaire     Eolien\n",
+       "start_date                                     \n",
+       "2014-12-15 00:00:00+01:00       0.00    5085.00\n",
+       "2014-12-16 00:00:00+01:00    4870.89   39909.00\n",
+       "2014-12-17 00:00:00+01:00    4609.62  103617.00\n",
+       "2014-12-18 00:00:00+01:00    5258.76  122720.00\n",
+       "2014-12-19 00:00:00+01:00    6289.70  104494.00\n",
+       "...                              ...        ...\n",
+       "2024-08-12 00:00:00+02:00  116114.62   67128.44\n",
+       "2024-08-13 00:00:00+02:00   96650.51   36417.80\n",
+       "2024-08-14 00:00:00+02:00   74620.37   43453.81\n",
+       "2024-08-15 00:00:00+02:00  101109.47   61722.23\n",
+       "2024-08-16 00:00:00+02:00   95051.52   37254.94\n",
        "\n",
        "[3533 rows x 2 columns]"
       ]
      },
-     "execution_count": 73,
+     "execution_count": 42,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -620,24 +758,246 @@
    "source": [
     "origin = production_forcasted.index[0]+ pd.DateOffset(hour=6, minute=0)\n",
     "daily_production = production_forcasted.resample(\"1D\", origin=origin).sum()\n",
-    "daily_production.index = daily_production.index + pd.DateOffset(hour=0) \n",
+    "daily_production.index = daily_production.index + pd.DateOffset(hour=0)\n",
+    "daily_production.rename(columns={\"SOLAR_FORECAST_D1\": \"Solaire\", \"EOLIEN_FORECAST_D1\": \"Eolien\"}, inplace=True)\n",
     "daily_production"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 85,
+   "execution_count": 44,
    "metadata": {},
    "outputs": [],
    "source": [
-    "daily_consumption_naive_tz = daily_consumption.tz_localize(None)\n",
-    "daily_production_naive_tz = daily_production.tz_localize(None)\n",
-    "tempos_naive_tz = tempos.tz_localize(None)\n"
+    "daily_production[\"Prévision J-1\"] = daily_consumption\n",
+    "daily_production[\"Production_nette\"] = daily_production[\"Prévision J-1\"] - (\n",
+    "    daily_production[\"Solaire\"] + daily_production[\"Eolien\"]\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "quantils = daily_production[\"Production_nette\"].rolling(365, center=False).aggregate({\"q40\": lambda x: x.quantile(0.4),\n",
+    "                                                                           \"q80\": lambda x: x.quantile(0.8)}).bfill()\n",
+    "daily_production[\"Production_nette_q40\"] = quantils[\"q40\"]\n",
+    "daily_production[\"Production_nette_q80\"] = quantils[\"q80\"]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# historical weather"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 46,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from energy_forecast.meteo import aggregates_observations, download_observations_all_departments\n",
+    "\n",
+    "all_dep_filenames_2022 = download_observations_all_departments()\n",
+    "all_dep_mean_temperature_2022 = aggregates_observations(all_dep_filenames_2022)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "AAAAMMJJ\n",
+       "2023-01-01    12.163991\n",
+       "2023-01-02     9.506840\n",
+       "2023-01-03     7.085597\n",
+       "2023-01-04     8.494031\n",
+       "2023-01-05     9.781422\n",
+       "                ...    \n",
+       "2024-09-12    10.936362\n",
+       "2024-09-13    10.629364\n",
+       "2024-09-14    11.353159\n",
+       "2024-09-15    12.184974\n",
+       "2024-09-16    13.959155\n",
+       "Length: 625, dtype: float64"
+      ]
+     },
+     "execution_count": 47,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "all_dep_mean_temperature_2022"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "all_dep_filenames_1950 = download_observations_all_departments(cache_duration=\"1200h\",\n",
+    "                                                               file_type=\"previous-1950-2022_RR-T-Vent\",\n",
+    "                                                               verbose=True,\n",
+    "                                                               )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_01_previous-1950-2022_RR-T-Vent.csv.gz (1/95)\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_02_previous-1950-2022_RR-T-Vent.csv.gz (2/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_03_previous-1950-2022_RR-T-Vent.csv.gz (3/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_04_previous-1950-2022_RR-T-Vent.csv.gz (4/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_05_previous-1950-2022_RR-T-Vent.csv.gz (5/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_06_previous-1950-2022_RR-T-Vent.csv.gz (6/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_07_previous-1950-2022_RR-T-Vent.csv.gz (7/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_08_previous-1950-2022_RR-T-Vent.csv.gz (8/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_09_previous-1950-2022_RR-T-Vent.csv.gz (9/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_10_previous-1950-2022_RR-T-Vent.csv.gz (10/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_11_previous-1950-2022_RR-T-Vent.csv.gz (11/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_12_previous-1950-2022_RR-T-Vent.csv.gz (12/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_13_previous-1950-2022_RR-T-Vent.csv.gz (13/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_14_previous-1950-2022_RR-T-Vent.csv.gz (14/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_15_previous-1950-2022_RR-T-Vent.csv.gz (15/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_16_previous-1950-2022_RR-T-Vent.csv.gz (16/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_17_previous-1950-2022_RR-T-Vent.csv.gz (17/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_18_previous-1950-2022_RR-T-Vent.csv.gz (18/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_19_previous-1950-2022_RR-T-Vent.csv.gz (19/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_20_previous-1950-2022_RR-T-Vent.csv.gz (20/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_21_previous-1950-2022_RR-T-Vent.csv.gz (21/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_22_previous-1950-2022_RR-T-Vent.csv.gz (22/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_23_previous-1950-2022_RR-T-Vent.csv.gz (23/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_24_previous-1950-2022_RR-T-Vent.csv.gz (24/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_25_previous-1950-2022_RR-T-Vent.csv.gz (25/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_26_previous-1950-2022_RR-T-Vent.csv.gz (26/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_27_previous-1950-2022_RR-T-Vent.csv.gz (27/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_28_previous-1950-2022_RR-T-Vent.csv.gz (28/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_29_previous-1950-2022_RR-T-Vent.csv.gz (29/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_30_previous-1950-2022_RR-T-Vent.csv.gz (30/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_31_previous-1950-2022_RR-T-Vent.csv.gz (31/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_32_previous-1950-2022_RR-T-Vent.csv.gz (32/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_33_previous-1950-2022_RR-T-Vent.csv.gz (33/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_34_previous-1950-2022_RR-T-Vent.csv.gz (34/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_35_previous-1950-2022_RR-T-Vent.csv.gz (35/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_36_previous-1950-2022_RR-T-Vent.csv.gz (36/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_37_previous-1950-2022_RR-T-Vent.csv.gz (37/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_38_previous-1950-2022_RR-T-Vent.csv.gz (38/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_39_previous-1950-2022_RR-T-Vent.csv.gz (39/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_40_previous-1950-2022_RR-T-Vent.csv.gz (40/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_41_previous-1950-2022_RR-T-Vent.csv.gz (41/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_42_previous-1950-2022_RR-T-Vent.csv.gz (42/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_43_previous-1950-2022_RR-T-Vent.csv.gz (43/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_44_previous-1950-2022_RR-T-Vent.csv.gz (44/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_45_previous-1950-2022_RR-T-Vent.csv.gz (45/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_46_previous-1950-2022_RR-T-Vent.csv.gz (46/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_47_previous-1950-2022_RR-T-Vent.csv.gz (47/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_48_previous-1950-2022_RR-T-Vent.csv.gz (48/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_49_previous-1950-2022_RR-T-Vent.csv.gz (49/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_50_previous-1950-2022_RR-T-Vent.csv.gz (50/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_51_previous-1950-2022_RR-T-Vent.csv.gz (51/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_52_previous-1950-2022_RR-T-Vent.csv.gz (52/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_53_previous-1950-2022_RR-T-Vent.csv.gz (53/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_54_previous-1950-2022_RR-T-Vent.csv.gz (54/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_55_previous-1950-2022_RR-T-Vent.csv.gz (55/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_56_previous-1950-2022_RR-T-Vent.csv.gz (56/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_57_previous-1950-2022_RR-T-Vent.csv.gz (57/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_58_previous-1950-2022_RR-T-Vent.csv.gz (58/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_59_previous-1950-2022_RR-T-Vent.csv.gz (59/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_60_previous-1950-2022_RR-T-Vent.csv.gz (60/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_61_previous-1950-2022_RR-T-Vent.csv.gz (61/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_62_previous-1950-2022_RR-T-Vent.csv.gz (62/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_63_previous-1950-2022_RR-T-Vent.csv.gz (63/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_64_previous-1950-2022_RR-T-Vent.csv.gz (64/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_65_previous-1950-2022_RR-T-Vent.csv.gz (65/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_66_previous-1950-2022_RR-T-Vent.csv.gz (66/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_67_previous-1950-2022_RR-T-Vent.csv.gz (67/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_68_previous-1950-2022_RR-T-Vent.csv.gz (68/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_69_previous-1950-2022_RR-T-Vent.csv.gz (69/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_70_previous-1950-2022_RR-T-Vent.csv.gz (70/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_71_previous-1950-2022_RR-T-Vent.csv.gz (71/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_72_previous-1950-2022_RR-T-Vent.csv.gz (72/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_73_previous-1950-2022_RR-T-Vent.csv.gz (73/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_74_previous-1950-2022_RR-T-Vent.csv.gz (74/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_75_previous-1950-2022_RR-T-Vent.csv.gz (75/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_76_previous-1950-2022_RR-T-Vent.csv.gz (76/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_77_previous-1950-2022_RR-T-Vent.csv.gz (77/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_78_previous-1950-2022_RR-T-Vent.csv.gz (78/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_79_previous-1950-2022_RR-T-Vent.csv.gz (79/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_80_previous-1950-2022_RR-T-Vent.csv.gz (80/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_81_previous-1950-2022_RR-T-Vent.csv.gz (81/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_82_previous-1950-2022_RR-T-Vent.csv.gz (82/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_83_previous-1950-2022_RR-T-Vent.csv.gz (83/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_84_previous-1950-2022_RR-T-Vent.csv.gz (84/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_85_previous-1950-2022_RR-T-Vent.csv.gz (85/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_86_previous-1950-2022_RR-T-Vent.csv.gz (86/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_87_previous-1950-2022_RR-T-Vent.csv.gz (87/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_88_previous-1950-2022_RR-T-Vent.csv.gz (88/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_89_previous-1950-2022_RR-T-Vent.csv.gz (89/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_90_previous-1950-2022_RR-T-Vent.csv.gz (90/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_91_previous-1950-2022_RR-T-Vent.csv.gz (91/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_92_previous-1950-2022_RR-T-Vent.csv.gz (92/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_93_previous-1950-2022_RR-T-Vent.csv.gz (93/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_94_previous-1950-2022_RR-T-Vent.csv.gz (94/95)\n",
+      "INFO:energy_forecast.meteo:Reading /home/antoine/code/energetic-stress-production/data/bronze/observations/Q_95_previous-1950-2022_RR-T-Vent.csv.gz (95/95)\n"
+     ]
+    }
+   ],
+   "source": [
+    "all_dep_mean_temperature_2013 = aggregates_observations(all_dep_filenames_1950, cut_before=\"2013-09-01\", verbose=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from energy_forecast import ROOT_DIR\n",
+    "all_dep_mean_temperature_2013.to_csv(ROOT_DIR / 'data' / 'silver' / 'all_dep_mean_temperature_2013.csv')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Reload"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "all_dep_mean_temperature_2013 = pd.read_csv(ROOT_DIR / 'data' / 'silver' / 'all_dep_mean_temperature_2013.csv', index_col=0, parse_dates=True)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 115,
+   "execution_count": 51,
    "metadata": {},
    "outputs": [
     {
@@ -661,144 +1021,1035 @@
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
        "      <th></th>\n",
-       "      <th>Prévision_J-1</th>\n",
-       "      <th>Solaire</th>\n",
-       "      <th>Eolien</th>\n",
-       "      <th>Type_de_jour_TEMPO</th>\n",
+       "      <th>0</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>AAAAMMJJ</th>\n",
+       "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>2015-09-01</th>\n",
-       "      <td>1167400.0</td>\n",
-       "      <td>21617.70</td>\n",
-       "      <td>30097.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2013-09-01</th>\n",
+       "      <td>16.312595</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2015-09-02</th>\n",
-       "      <td>1137500.0</td>\n",
-       "      <td>24844.99</td>\n",
-       "      <td>18895.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2013-09-02</th>\n",
+       "      <td>16.497312</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2015-09-03</th>\n",
-       "      <td>1127150.0</td>\n",
-       "      <td>21967.80</td>\n",
-       "      <td>24162.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2013-09-03</th>\n",
+       "      <td>18.602833</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2015-09-04</th>\n",
-       "      <td>1117350.0</td>\n",
-       "      <td>25466.61</td>\n",
-       "      <td>24709.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2013-09-04</th>\n",
+       "      <td>20.945904</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2015-09-05</th>\n",
-       "      <td>980050.0</td>\n",
-       "      <td>27009.82</td>\n",
-       "      <td>32334.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2013-09-05</th>\n",
+       "      <td>21.346506</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>...</th>\n",
        "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2024-04-04</th>\n",
-       "      <td>1163350.0</td>\n",
-       "      <td>62687.02</td>\n",
-       "      <td>279577.48</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2022-12-27</th>\n",
+       "      <td>6.259121</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2024-04-05</th>\n",
-       "      <td>1110187.5</td>\n",
-       "      <td>71763.38</td>\n",
-       "      <td>175006.34</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2022-12-28</th>\n",
+       "      <td>8.497527</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2024-04-06</th>\n",
-       "      <td>950700.0</td>\n",
-       "      <td>0.00</td>\n",
-       "      <td>0.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2022-12-29</th>\n",
+       "      <td>8.885538</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2024-04-07</th>\n",
-       "      <td>930150.0</td>\n",
-       "      <td>0.00</td>\n",
-       "      <td>0.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2022-12-30</th>\n",
+       "      <td>10.010722</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2024-04-08</th>\n",
-       "      <td>468475.0</td>\n",
-       "      <td>0.00</td>\n",
-       "      <td>0.00</td>\n",
-       "      <td>BLEU</td>\n",
+       "      <th>2022-12-31</th>\n",
+       "      <td>13.010294</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "<p>3143 rows × 4 columns</p>\n",
+       "<p>3409 rows × 1 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "            Prévision_J-1   Solaire     Eolien Type_de_jour_TEMPO\n",
-       "2015-09-01      1167400.0  21617.70   30097.00               BLEU\n",
-       "2015-09-02      1137500.0  24844.99   18895.00               BLEU\n",
-       "2015-09-03      1127150.0  21967.80   24162.00               BLEU\n",
-       "2015-09-04      1117350.0  25466.61   24709.00               BLEU\n",
-       "2015-09-05       980050.0  27009.82   32334.00               BLEU\n",
-       "...                   ...       ...        ...                ...\n",
-       "2024-04-04      1163350.0  62687.02  279577.48               BLEU\n",
-       "2024-04-05      1110187.5  71763.38  175006.34               BLEU\n",
-       "2024-04-06       950700.0      0.00       0.00               BLEU\n",
-       "2024-04-07       930150.0      0.00       0.00               BLEU\n",
-       "2024-04-08       468475.0      0.00       0.00               BLEU\n",
+       "                    0\n",
+       "AAAAMMJJ             \n",
+       "2013-09-01  16.312595\n",
+       "2013-09-02  16.497312\n",
+       "2013-09-03  18.602833\n",
+       "2013-09-04  20.945904\n",
+       "2013-09-05  21.346506\n",
+       "...               ...\n",
+       "2022-12-27   6.259121\n",
+       "2022-12-28   8.497527\n",
+       "2022-12-29   8.885538\n",
+       "2022-12-30  10.010722\n",
+       "2022-12-31  13.010294\n",
        "\n",
-       "[3143 rows x 4 columns]"
+       "[3409 rows x 1 columns]"
       ]
      },
-     "execution_count": 115,
+     "execution_count": 51,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "\n",
-    "data = pd.concat([daily_consumption_naive_tz, daily_production_naive_tz, tempos_naive_tz],\n",
-    "                 axis=1).sort_index().dropna(axis=0, how=\"any\")\n",
-    "\n",
-    "data = data[~data.index.duplicated()]\n",
-    "year = data.index[0].year\n",
-    "first_september = data.index[0] + pd.DateOffset(month=9, day=1, year=year)\n",
-    "if first_september < data.index[0]:\n",
-    "    first_september += pd.DateOffset(years=1)\n",
-    "\n",
-    "last_august = data.index[-1] - pd.DateOffset(month=8, day=31)\n",
-    "data.rename(columns={\n",
-    "    \"Prévision J-1\": \"Prévision_J-1\",\n",
-    "    \"SOLAR_FORECAST_D1\":\"Solaire\",\n",
+    "all_dep_mean_temperature_2013"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 58,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "all_dep_mean_temperature = pd.concat([all_dep_mean_temperature_2013.loc[:, \"0\"],\n",
+    "                                      all_dep_mean_temperature_2022], axis=0)\n",
+    "all_dep_mean_temperature.sort_index(inplace=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<Axes: xlabel='AAAAMMJJ'>"
+      ]
+     },
+     "execution_count": 62,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "all_dep_mean_temperature.plot()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "mean_temperature_q30 = all_dep_mean_temperature.rolling(365, center=False).aggregate(lambda x: x.quantile(0.3)).bfill()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<Axes: xlabel='AAAAMMJJ'>"
+      ]
+     },
+     "execution_count": 64,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGwCAYAAAB7MGXBAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAABwfUlEQVR4nO2deXwU9f3/X3tlk0ASroQznMrpAXiCrUpFUGm/ar2KWg/Urwf9WrWVr/zaatVavq3aaqutbQW8tdBatWJFBMEDPFBAEDkU5EzCmWxCks0e8/tj85mdmZ29ktk5dl/PxyOP7DG7+9nZmc/nNe/TJUmSBEIIIYQQi3BbPQBCCCGEFDYUI4QQQgixFIoRQgghhFgKxQghhBBCLIVihBBCCCGWQjFCCCGEEEuhGCGEEEKIpXitHkAmRKNR7N27F2VlZXC5XFYPhxBCCCEZIEkSGhsb0a9fP7jdye0fjhAje/fuRXV1tdXDIIQQQkgH2LVrFwYMGJD0eUeIkbKyMgCxL1NeXm7xaAghhBCSCYFAANXV1fI6ngxHiBHhmikvL6cYIYQQQhxGuhALBrASQgghxFKyFiONjY247bbbMGjQIJSUlGDixIn45JNPUr5m+fLlGD9+PPx+P4466ig89dRTHR0vIYQQQvKMrMXI9ddfjyVLluDZZ5/F+vXrMWXKFEyePBl79uzR3X779u2YNm0aJk2ahLVr1+K2227D9ddfj8WLF3d68IQQQghxPi5JkqRMN25paUFZWRleffVVTJs2TX78hBNOwLnnnotf/epXCa/53//9XyxatAgbNmyQH/vBD36A+vp6vPnmmxl9biAQQEVFBRoaGhgzQgghhDiETNfvrCwj4XAYkUgExcXFqsdLSkrw/vvv675m1apVmDx5suqxqVOnYtWqVUk/JxgMIhAIqP4IIYQQkp9kJUbKysowYcIE3H///di7dy8ikQiee+45rFq1CjU1Nbqvqa2tRe/evVWP9e7dG4FAAC0tLbqvmTNnDioqKuQ/1hghhBBC8pesY0aeffZZSJKE/v37w+/34w9/+AOmT5+esrJatsyePRsNDQ3y365duwx7b0IIIYTYi6zrjAwbNgwrVqzAkSNHEAgE0LdvX1x22WUYOnSo7vZ9+vRBXV2d6rG6ujqUl5ejpKRE9zV+vx9+vz/boRFCCCHEgXTYnNGlSxf07dsXhw8fxuLFi3H++efrbjdhwgQsXbpU9diSJUswYcKEjn40IYQQQvKIrMXI4sWL8eabb2L79u1YsmQJJk2ahJEjR+Laa68FEHOxXHXVVfL2N910E7Zt24ZZs2Zh06ZN+NOf/oQFCxbg9ttvN+5bEEIIIcSxZC1GGhoaMHPmTIwcORJXXXUVvvWtb2Hx4sXw+XwAgJqaGuzcuVPefsiQIVi0aBGWLFmC448/Hg8//DCefPJJTJ061bhvQQghhBDHklWdEatgnRFCCCHEeeSkzgghhBBC8puWtgjMtlNQjBBCCCEEAFDT0ILx9y/B7X9fa+rnUowQQgghBtEUDOPtjXXYVOvMyuEvfbwLLaEIXlm719TPzbrOCCGEEEL0ue/fX2DB6t0AgPdmTUJ1j1KLR5QdPo/Lks+lZYQQQggxiJqGVvn2rkPNFo6kY4Sj1uS0UIwQQgghBhFoDcu39yqEiVMIR+JiJBSJmva5FCOEEEKIAXy1rxHrdtXL93+6cB3G378E9c1t1g0qS5SWkWCYYoQQQghxFAs/3Z3w2KEjbfjz8q8tGE3HUKb0BkMR0z6XYoQQQggxgGSlOdYorCV2p03hmmmlZYQQQgjJD6zJT+kYStcMLSOEEEJInuB2OUeOtCnFCC0jhBBCSH7gdtBKSzFCCCGE5CEuBzlqguG4a6aVbprCZNv+Jvzz0904Egyn35gQQogjcJCXxjLLCMvB24jpf/sQdYEgvt7fhFnnjLR6OIQQQgqInQeb8c7m/fJ9BrAWKHWBIADg890NFo+EEEKIUTglgHXVtgOq+0ztLXAcctwSQgjJALdD5vTWkFp80DJS4LioRgghJG+wYk4/0BRES1t2YqJFIz4YM1LgUIoQQkj+YPac/vg7X+HBxZtRUeLDW7efjt7lxQnbvP75Xvz13W2IRCVcfMIAXHvaENRqGvsxm6bAoWGEEELyB7Pn9BXtQagNLSGs2XlYd5sn39uOz3c34Iu9AfypvXfO5tpGAHG3krJpXq6hGLEh1CKEEJI/mO2mCUXTp+cqrR6t7e6cUHtfmm6lRbH7DGAtbBgzQggh+YPZAayRqLLzrr6gCOrUExFipLTIE7tPy0hhQylCCCH5Qy4rsIYiUVw972P87F/rFY/FRcTbX9YlvGbtrnpsP3BEvt8WieKsh5djXXtZCVmMRGgZKWhoGCGEkPwhl3P657vrsWLLfjz/0U65lHtE4abRs21c//QnCY99vT8uTkqKYrktdNMUIJKkPGSoRgghJF/IZdGzsMIKItwzysBT9doS40BTW8r3LPV5Et4n11CM2ATlb07LCCGE5A+5nNOVMYZCPOgJlGR4dAJauvhjYqSNbprCQ3nAUIsQQojz+Ou723Qfz2VSgvKtI+0iRLmeRNIYN6I6lpPSdjdNmGKk8FAeELSMEEKIvfiyJoBnP9yBaAdcF3oLfi6QLSOKmBG98Z44qLt8W29o8QBW89w0rMBqE1RihLYRQgixFec++h4AoMTnwcUnDMjqteFINGeZKcrU3UiGbppeXf0AgPvPH4PnP9qJTe3FzgTCMmKmm4ZixCYwZoQQQuzPF3sbshYji7+ow9E/+0+ORhRHWETCKjdNohgRwqjIq+8cETEjdNMUIKqYEYoRQgixJSZ5XDqEWEeU64mem6ZNIUb0LCcldNMULhLdNIQQkjds/tU5cLtciESlnHa/veSJldhS1wRAP2YklWXE53HrPi9Se80sekYxYhNU6pRahBBCHI3fG1vQfR6guH1xzwUlivcOhqJoCoZVMSO6lpFwXIzoPS9iRihGChCVOrWxGZAQQgoZvSJiVqJMGz7vD+8lPK9vGYk9VpTMMuKnm6ZgUR4Pi9bX4K5DzajuUWrdgAghhCTw9KodCdeLpUVezDhtsBXDka0cydgXCCY8pnTTRHVeXkI3TeGiDSL6+Ssb8PSMky0aDSGEkGQ8s2pHwmM+T9xCcXRVV9PG8u3hvbCxJiDfH1rZBW/c+m3sCwRx+oPvoLE1nPAaZQDrtacNxq8WfSk/V1nml11M6YSOkVCM2ARtUZz6lpBFIyGEEJKKihIfrp4wCACwattBfPLNYdWi/7tLx5o2ltvOGo6/rIhXfi3yuFHs88iulpZQBJIkye6clrYItrU3xfN5XJhx2hCcOLgHBvcsxac7DmNMvwrsqW8GADS3RUz7HkzttQlaU5nd/JKEEEJiVJb5cceUEbhjyghMGNoTQOyCUrg3upX6TBtLSZEHRyksMd52C40qsFVh4fh0x2H5dv/uJXC7XRhb3Q3dSotw1qje6FNRjBJfzE7REqIYKTi0lhGzygcTQgjJDq+iuZy7/XYkKsnBoG6d5nO5RPlpHndsWVdm8CgtHE3BmAWnT3kxqsqKdd9PlINvoWWk8NBGNOsFFRFCCLEeZeVST7v7IypJcuyf12QxUhtolW/72j/b43bJ41RaOFpCMTFyVIq4FiFGmtvCplnpKUZsgvYHp2WEEELsSVLLSLsYcZtcRvv0oyvjY1ME0upZOISVRFRZ1aO4/bmohJwWbFNCMWITtBlU1CKEEGJPvJ740imEh7IfjMdky4jy87zu+NhE3Eir0jLSLkZKU4iRUoWLxyxXDbNpbAJjRggh6WgNRXD906ux63Az5nz/WEwc1svqIRUkRQoxIm4qC4R5TLaMKK0hyttCjLRkKUa8HjeKPG60RaLYeai5U9aRRoULKRUUIzZBW2eEUoQQomXtrnq8/9UBAMC/19VQjFiEcsEXlpGQYsF2m+xz8CaxjIgg1iUb63DS4B4AgOZ2YZKuRH1JkQdtLVGc//gHnRpbNNic0XZ009gErSGElhFCiJYjwXgtiwBrEVmGT2UZaRcjCl+712Q1onQbKYVJQ/sxouxVk4llBAD+6/h+8Hlchvxl9B0y/rYkp2izaahFCCFalOb2BooRy/DpWEa+VFRBtdQyohjbJScOwCNvb0VrWBnAGhO0ohleMu6/4Bjcf8ExnR5bIBBAxcPpt6NlxCYUWswIi7oRkj3KYML3vzqAV9fusXA0hYvSMiIKnO1tiMVGdCnymG8ZUXyecmyirHswFLfayNk0Oewk3BFoGbEJ2jbO+SxGnljxNf727jYsvGkChlaa18OBEKezt14dDPjjl9ZiYI9SjBvY3aIRFSbKxX/qmD647/wxONjUBgA4ZWgP07NplNYQ5Wf72+uMBNstI42tIbz+eQ2A9G4as6EYsQlRbcxIHhc9+7//bAIA3P/6Rsy/ls0ACckU0TNkTL9yfLE35ha4//WNePmW06wcVt6jvVgs8sYX/GKfB1dNGGzyiNQo3TTKgmwiSFVkw9zz6hfyc12L7bX8001jExKyafLYMkII6RhiUblwXH/cePpQAPEOrCR3aPex2W6YdKjEiMpNIywjsfHvqW+Rtz9jeCXshL32aIEiKcoIC7QBrYQQElK0fv92e9VNZaYEyQ0hjRhRxmXYAa+OAAEAv69djLQHPovCbI9dPg5lxeY188sEe9lpCpCPtx/CdU99gkZFyh4A1AWCCIYjcgASIYSIwlpet1vO6KBlJPe0hbVixNyYkHR4krhp5ADW9vGHo/Hjx27Yb0QFxqqvD6qEyKUnDpBv7zrUYsWQCCE2RVyh+zwu+WqYlpHcoxV8TZqLR6tRumYONAXl21o3Tbj9e3htJqaALMVIJBLBL37xCwwZMgQlJSUYNmwY7r///pTxDcuXL4fL5Ur4q62t7fTg84FIe6TqpScOwLq7p+C3Fx+PihJf+3OcZAghcYTw8HnilpEwLSM5JxRWz8XKei92YNLIePyH8iJWFiPCTaM4fuxGVm6a3/zmN/jzn/+Mp59+GmPGjMHq1atx7bXXoqKiArfeemvK127evBnl5eXy/aqqqo6NOM8QsSGlRV5UtOerF/vcaGgBwvmcUkMIyZo22TLilk3tIV605Jy2SFx8eNwuXHHKIAtHk8hRVWXybWVZiGJNo7zNdY0AzG/klwlZiZGVK1fi/PPPx7Rp0wAAgwcPxosvvoiPP/447WurqqrQrVu3Dg0ynxEXNXpdF2kZIYQoUZrZaRkxD+HmqCzzY+lPzkC5zYI/lSgdFSKAdW9Dq8q11LNLkdnDSktWtpqJEydi6dKl2LJlCwBg3bp1eP/993Huueemfe3YsWPRt29fnH322fjgg9SNd4LBIAKBgOovXxEqVilG4r0OKEYIIXHEnFDkccumds4TuUe53+0sRADg5CE95NvKBIiXP9st3x7Sq4upY8qErCwjd911FwKBAEaOHAmPx4NIJIIHHngAV1xxRdLX9O3bF0888QROPPFEBINBPPnkkzjzzDPx0UcfYfz48bqvmTNnDu69997svolDEdYPt0tpGXGpniOEECAewOr1uOQgRG3aKTEekU2jTJu1G8t/eibe2bwP008eKD+mzKzZeTBWMK9bqU+VCmwXshIjCxYswPPPP48XXngBY8aMwdq1a3HbbbehX79+uPrqq3VfM2LECIwYMUK+P3HiRHz99df4/e9/j2effVb3NbNnz8Ydd9wh3w8EAqiurs5mqI4hIqdaJVpG9GJGln5Zh28ONuO/ju+HyjK/OYMkhNiCkCJmRFhGwrxoyTlCjBTZWIwM7tUF1/YaonrMo7jI3XEoJkZ62NBFA2QpRu68807cdddd+MEPfgAAOPbYY7Fjxw7MmTMnqRjR4+STT8b777+f9Hm/3w+/vzAWWtkyoidGNObXmoYWXPf0agDA9gNN+NUFx5o0SkKIHRDCw+dxqSyo0aikmkOIsShFoJMQmZkAsPtwLMvGjvEiQJYxI83NzXBriqV4PB5Es8z6WLt2Lfr27ZvVa/IVkU2jVLBCjGib5Slbhh9udkb78Lc31uH//rMJOw4esXoohDieUFhhGVFcpYeYeZdT5GJhNqzPkYqSIg++fXQvAMDm2ljsZf9uJVYOKSlZWUa+973v4YEHHsDAgQMxZswYrFmzBr/73e8wY8YMeZvZs2djz549eOaZZwAAjzzyCIYMGYIxY8agtbUVTz75JJYtW4a33nrL2G/iUEQDJqXgFvEjWjGinG+0jZvsSDgSxfXPxCw5uw414/Er9GOECCGZ0aasM6K4MAxHJPhZTztniHpQyotGpzCidxne23pAbsYq0n3tRlaH7x//+Ef84he/wC233IJ9+/ahX79+uPHGG3H33XfL29TU1GDnzp3y/ba2NvzkJz/Bnj17UFpaiuOOOw5vv/02Jk2aZNy3cDB6bhpxW3uxoxQnWqFiR5RR/vUtbQnP2/8bEGIvRByZTxHACrAKa64RMcJOdIVpa4q4bCqoshIjZWVleOSRR/DII48k3eapp55S3Z81axZmzZrVkbEVBHoBrOJmqmZ5TgigV5ZQdtv0BCDESSjdNMo5I5Wb5tCRNvzohc/QGorgtxcfpyqQRTJDzMVeB4oRrYCy61dwVjROHiIOcuViLUyBWleM8ywjDlBMhDiIkBy74IbLFS98lupce2/rfqz8+iA+21mP/6xnG46OILtp7LqSp0DrWrLrhSHFiMVE5JgRHTeNRm8o7ztBjChNx3rjtecpQYh9UTbKA+LVmlO5aYKhuFBpDdurp4pTkN00Nl3IU6G1jNhVUFGMWIxeBdZkbhplQ0InFERTXq05YbyE2JlIVJJLfYvg1UwKnwUVzymFCcmcqI473SloLSN21VMUIxYjrmj0ysFruyEr13MHGEYoRggxEOX5JNJ6Myl89oelW+Xbq3ccxoY9DfjJgnXYtr8pRyPNP8I6iQZOQVsaxa7WHSaDWUxUp86IOFi0C7jzLCPOGi8hdkYpRsQVuvj/6NtbMbpfOc4e3RvDe8cDVOub27C/MSjfd7uAWf/4HBtrAqgLtOK5608xafTORq8elFOgm4ZkhG5qbxIxorybKtPGLignz8921mNXezliQkj2KMW91iKyaH0NHly8GT9duE71mpaQOkbks5312FgTK371/lcHcjncvCLSPpd5HFb0DKCbhmSImF/0KrBq9YbSMqJ14dgRrR/7//1rvUUjIcT5hNvPJ7dL0dk7rD7HDh1R1/NpCzNGxAj05mmnoLWE2NVNQzFiMVG9bJokAawqy4gD3B7a1uYHmxILnxFCMqMtg/4o2nmB6fXGoDdPO4VxA7up7tv1K1CMWIzeBJNRzIj9tYh8JSfgxEhIxwkrSsHLaBYWbSBrsN0yUlXmRxnrxXeYsIPFyAmDeqga5tnVusOj02L0WlMLMaJ1xSjvOcFN06YRH9r7hOQbR4JhPLh4M/Y3BdNvnCE9uxThJ1NGJNQYARJN7omWkdj9Iq879pxxwyoo9BINnERFiU9utJoX5eCJ8ehNMEJ9JwawOis7RZm2HIlKCf5tQvKNdzbvw1MrvzH8fY/pX4Fj+lUAiFVfFWjXFa01Ur7Y8bgR9th/zrAreokGTqKsOL7U50WjPGI8QoyoLCMZVWDN+dA6jfhupT4PGoNhueMoIflKc1sse+Xoqq644pSBnX6/Bat3Y2NNAK2hSHyuUIoRzfbJYkaKvO6UtUhIasIOLnoGAP97zkj8ffUudC3y4sJx/a0eji4UIxajvHIRiONdW0Jd1ZvGAROL6KNR6o+JEcaMkHxHiIFBPbvgmtOGdPr9Vu84jI01AUSiktyxV9mtV2tyD2nmhTZFY70ir/3nDLvi5ABWADh9eCVOH15p9TBSQjFiMUqfrkD49n616EvMe3+7/Hirws1h9zojn+08jFfW7AEAdCnyAggmiBF7fwNCsice6GjM+yldtm1hnQBWDZGohLpAKy79yyrsONiM0X3LAdAy0ln0GpoSY6EYsZhgODGb5pPth+TbextadV9n50Z5kiTh+39aKd/v1dWPbQeOIBSJamqlWDE6QnJHvIeJMWpE7uAtxS0jPh0rqiASlfDzVzZgx8FYgUFR4GzP4Rb0rig2ZEyFiLB4eR1Y9MwpUIxYTEgntVd5BfPaj05Tbb9mZz3uee0LW7tpahQC6uoJg3D+uP74/p9WIhSRVEWZVmzZj9ZQxLYBVYRki9EpoG7ZMqIf7K7X+/qbA0cSHjvnmD74Ym+DIWMqROQAVlpGcgbFiMUIn65f4aZRWj2OG9BNtb2Y7GysRVAXiIuRe88/Bk3BsHz/gj99oNr2z8u/xu1nDzdtbITkkoiI6zBIjHjkmkNR2aXrS5FNA+in0Hf1e1O6d0hqIga730gi3LUWIkmSbtGzVGm7yQqi2YnnP9oJIFZoCYhNhKPafde7DrWothVmZELyAaMtI6IXitIyohQ6ep+iVwK+C8VIp4iLEe7DXME9ayHBcFQ+yLsq8sBT6QxxpbSnviX5RhYjJsh+3Urkx346Rd/60app5EWIk4lEjI0tkC0jkiTX7VEGu+taRnTESFe/R+PeieGE4ol2wMlde50CxYiFBFpFRbxYLQ6B6CXQX7GYC3ze+MlQ32zPXi/ixD3nmD7yY17NVVlJ+/dlui/JJ8IGxxYIC0s0Greiqi0jiZ9z8EjivFBapG8ZCbIQYUZE6abJOdy1FrIvEKvN3KurX1XZ71cXHINJIyrxyA/GJrzm6Kqy+Osb7VnbOaJTIEh7VSbuh1kIjeQRIt7LqJgRt45lRCkqRvWNzQddilIHgXfxexMuCAAgGKIYyYQw3TQ5hwGsFiKqNWobWI3pV4H5156s+xqP24V+FcXY29Bq24lEL/K8SDMRClMzLSMknzB60RKnTSQq6Wbe/WzaaHQt3orLTqzG4+98hVXbDuq+TyyANVEgBcMRAL7EFxAVtIzkHooRC9GrqJgJ/nYXR2vYnvEW8tWh4ntpr8pEHYZ1uxtwoCmIXl395g2QkBxhdD0KIWrUYiT+3kdVdcUfp48DEBMWycRIF78HPh2BRDdNZrDoWe6hzrOQeCO57H4GYWWwq2VEfC/liVtZphYb3zq6l3z79XV7zRkYITlG2RzSCNSWESF09OeLkhT1err6vap4M8FHigKLJDlO703jBGgZsRC92IpMkEtE2zQSXs9v3r9bCV684VS8/WUdenYtwozThuDdLfuxrzGoKnNPiJPJVZ2RqKTvplFyXHU3HDegArUNrQnxZEVet25V2Mb2IHqSGqf3pnECFCMWEu6gSVcZYW9H5IwCzYk7YVhPTBjWU74/aUQV/r56l61rphCSDUZn07gVvWnCuhVY43T1e/Haj74l35/zny/xlxXb5PHovY4xW5kRSTKnEeOgm8ZCOnoVJU4Iuza+ytTio5xoCbEzzW1hnP27Fbjn1Q0ptzM6m8ajKHLYppNNkwrlGNxul+7rQsxmy4iOWrFJ5lCMWEiog/5lr80X8UiGJk2vzUUVIYJFn9dg674mPL1qR8rt5JgRowJYPYmWkUwtqcpYNLdLP9ZEr0AaSYQBrLmHYsRCxKKdbZlmpR/ZjmQqRuzubiJEEFV1m05+vBp9Ba2swCpcKto0+WQox+BxuVBEN02HyXROIx2HYsRCOtrHwq2IsLcj8omb5irCQ8sIcQjKK+JRd7+ZdDvj64zEPvc/62tlq4xeIGqq1wKAy+XStYxQjGQGxUjuoRixkLBOeedMkC0KdrWMSJkFe3lt/j0IEZQVxwuDtaZIqTfaMjKsqisAoEXRwylTN41SQHkYM9IpKEZyD8WIhcRz17P7GezeuTfbAFaWhCd25/F3vlLdT+aqEYUMjcq6mDSiCivuPBMv/fep8mOZujWVhkm3S52FI1w9tIxkRqbWXtJxmNprIfIB3sHUXruLEVpGSL6w/cAR1f3mtgi6+BOnz1xkXQzq2UXVATtTt6ZyCNpsmmKfG22RKMVICt7cUIM/LP0KkaiEHYdivz8tI7mDlhELCXXUTZMvlhGXiBnhhEiANzfU4rxH38OanYetHkoC2kW7oUW/WFiuzPlKIZHpea/s6Ot2uVTnY7HcNduec4gdeO7DndhYE8DmukbZNTewZ6nFo8pfaBmxkEhH3TQ2r8CaqUkznqKc8yERB/CjFz5DOCrhl//eiFdnnmb1cFRorRENLSG8snYPuhR5cfXEwQnb5bIeRaaeAuV2HpdLJZD8vtic08aTLyli39x61tE4dUgP9KkoxtDKrhaPKn+hGLGQjk5ccmqvXS0jGQawxouecUIk8fNh274mi0eiJhKVZIHdrdSH+uYQNtUG8Ns3NwMAvj++vxzgakago6sDcQsulzqgVVha7DqH2AGxb0b3LcPEo3ql2Zp0FrppLCSUZREjgbIQkh0Jy5UiWfSMOB+li6a03b3R1BqWH1Nm13Q0KD0bMp0ttNk0yosDEcBq1znEDrDQmblQjFhIOMvyzoJ4ISTDh2QI4TQNvQQsekb0sNvRoBTLItYiGFYKkPjtuGUkd+PpiJvG7XKp3KbiAojB48lhOq+5UIxYSKiDvWnsvoiLHhrprg5Z9IzokarCqRWEFMLDryNGQuH4eI0ueqZHplfqyq3cLrVA8tEykhaKEXOhGLEQYRnRq4yYCreiRLQdEVeKRV6m9pLssdvRIC4a3C6gyBs7V1ViRGEZiZoQwJrpWytjS1wul+q+HDNit51tIyhGzIUBrBaSriV4Mjw2Kwe/L9CKL2sb0b3Uh2P7V6C+OZb2mM4ywqJnxAmEFBcNQmQEw/GKqMrjt6MtHrIh0wBW7RA8KjHCC4F0iH3DQmfmQDFiIaEOBrvZqehZOBLFOY++h0NH2gAAt5w5TH5OpA8mg5YR4gSEm6bI45bPvaAiaFUZ4BrpoOs1GzJeGzUbKgWSmHPsMIfYlUyLNxJjoBixkGxbggvsVA5+674mWYgAwOuf18i3+5QXp3xtvOiZ9d+D2Ae7adP1exoAxKwJQmTsbwrKz7eGIvjvZ1bjq/1N2HWoGUCOLSMZ5tNoh6BcVBkzkh6xa+imMQeKEQsJZZgCq8VOjfJqA62q+/sbY5P0OWP6pDUne22eokysQbJZ1IjoSxMMR3GkLeae+VpRC+WLvQG8tbFOvu9xu1Ddw/hKndU9SrDrUAvOHt07o+21okW5ptJNkx4R+0YxYg4UIxYSLwfv3EZ5EU28h+guWlrkSftaD03FRAe7+uhv+PZQrN1VD0C9QIljvlfXIvzpihPQv3uJqpeMUSy5/Qwcbm5D34rM3lu7Gz0MYM0KEZds1+Mx32A2jYVkWhxMi1xG3QZXNclcLMWZiBG6aYgOdrsSFcfnqUN7YmSfMgBAo6roWUyMlBR5cPKQHuifAyECxGqcZCpEgMTiaEo3Da2S6WE2jblQjFjIut31ALJP7bVTnZFkk5moVJkKO30PYh+yPR9yjTK2y9+e2tvYGm+UJyqw2u0KWluPRLmo+tzCMsJzLxmswGou9jrrC4g99S2oaYjFWxSnyTrRIqfE2mART9Zxt2txeg8gi54RPex2JSqn9rpdcp0RPcuI7bIutAGsypgRLy0j6YjSMmIqjBmxiH2KwM9JI6qyeq2dGuUJV9PIPmUY3LMLDje3oazYhwvG9k/7Wqb2Ej1ymRbbEcSC7fO44ffGLH5KAS0KoNlt3AluGmU5eDdjRtJhRs0YEodixCLEJDCwRym6lRZl9Vq3jWJGxETdr1sJnvjhCVm9VnyPEIueEQXZprpniyRJ+PkrG7CnvgW/v3QsundJff4J65/X49KtnSMKoNnNnD95VCzr5qTB3QGoF1Vh4bHDBY1doWXEXChGLEKu7teBA11ulKfvITGVeJfS7L+HCNwN2+GLEEtRHgNJPH+GURtoxfMf7QQALNu0DxedMCDl9keC7QGqPo8cM6JEFECz26LVvUsRNt1/jtyhV20Zsc8FjV2JsAKrqVCMWIRQ3R05zkV8nx2uaiLRjhVuAyBP7G0UIwWP8hhQBofmglZF9dQDiuJl+ttG5NTdHl2KZDeNEmEZsZsYAeJdhgGNGPHQMpIO4YLOYc9DooC72SLEHNAR066d3DSd6VJa5GnvgBqiGMmWNzfU4jsPLcfP/rXe6qEYglIg5PqwVpZvD6URws1t8R40XYq8qsVd8PaX+wDYz02jReWm8dhnDrEjH28/JAtkO4rMfCSrFSQSieAXv/gFhgwZgpKSEgwbNgz3339/2pbfy5cvx/jx4+H3+3HUUUfhqaee6syY8wJJThvL/rWyidUGVzWRTrhpimgZ6TAvfrwT2w4cwfMf7UQgx5YEM1A2nsv18dCm7LibJl5JWEWKvG643S6UpKifU2SzlGQtyuHFi55ZP4fYkZVfH5Bv9+rqt3AkhUNWbprf/OY3+POf/4ynn34aY8aMwerVq3HttdeioqICt956q+5rtm/fjmnTpuGmm27C888/j6VLl+L6669H3759MXXqVEO+hBPpjGVETCR2WMQ7E3Eui5Gw9d/DaSgX74NNbSgv9lk4ms6zubZRvh0MRyFJUsbdabNFed4kS00XyAXN2i0iqSoLZ1Loz0r03TRWjcbeCBF67WmD5fmW5JasxMjKlStx/vnnY9q0aQCAwYMH48UXX8THH3+c9DVPPPEEhgwZgocffhgAMGrUKLz//vv4/e9/X9BiRJhHOzLh2mkRlwtCUYyYitJ9kA/774X2gFJBMBzVdYkYQSisFCNpLCPt+1nUAipJMaZineBWO6EUI0WswJqS1vbfvaufYZVmkdXZM3HiRCxduhRbtmwBAKxbtw7vv/8+zj333KSvWbVqFSZPnqx6bOrUqVi1alXS1wSDQQQCAdVfvhHPpsn+tcIcbIdFqFOWEYWFJ52rj6g5EowX3VJaSZyK1l0QzOGxrXTNhNO4acS+FcIolUDqYvOFy6cQS3JfKJ53urQoSvwTc8jq7LnrrrsQCAQwcuRIeDweRCIRPPDAA7jiiiuSvqa2tha9e6u7TPbu3RuBQAAtLS0oKUnstTBnzhzce++92QzNcUidKDUsLAortuxHNCpZWvnRiJgRICZI9DIViD5Ky0guF24zaGgJJXyHYCgClOTG9aQMWk1nGWhpi20rLCLlJcmnzKOquhowutzRu8wPr9uFcFRC327FANr3M0lAnF+pLGHEWLK6Ll+wYAGef/55vPDCC/jss8/w9NNP46GHHsLTTz9t6KBmz56NhoYG+W/Xrl2Gvr8dEL7ajrhphvcuk2/vPtxi1JA6hFxnpAMmHmUZ/H2B1CmWRE1LKD/cNP/3n004/t638N7WA6rH533wTc4+MxjOPJtGxIz42xelqrLipNvOOG2IAaPLHV6PG2/8+NtYeNMEjB3QDQAQaA07+vjJFdpYIZJ7srKM3Hnnnbjrrrvwgx/8AABw7LHHYseOHZgzZw6uvvpq3df06dMHdXV1qsfq6upQXl6uaxUBAL/fD78/vyOYo53IphnVt1y+bXUQa2csI0pLSENLCNWGjSr/UV7ROzkj4okVX+s+vutQc84+c9uBJvl2OstIa1gsSunFthNM+uJCRunaawlFVFZKQjeNFWR1BDY3N8OtqSfh8XgQTRGSPWHCBCxdulT12JIlSzBhwoRsPjrv6Ew2DQD06horYZ3uyi7XCJ97R3PxRbt1NsvLDqX+yKcgxO8e1xcA0NwWTrNlxzl8pE2+nSq19/2tB/DKmr0AUseKOJEij1suuJgPMUdGEw9czq/f3c5kZRn53ve+hwceeAADBw7EmDFjsGbNGvzud7/DjBkz5G1mz56NPXv24JlnngEA3HTTTXjssccwa9YszJgxA8uWLcOCBQuwaNEiY7+Jw4h2stSwaHSVLgAv18gVWDsoRrxyVD9NxdmgtIY42DCSQPf2Pk3KmBijCbTEhU6y4y4UieKGZ1bLV8jds+wfZXdcLhf8XjdaQ1EWHdShpX2fpErlJsaSlRj54x//iF/84he45ZZbsG/fPvTr1w833ngj7r77bnmbmpoa7NwZT9MbMmQIFi1ahNtvvx2PPvooBgwYgCeffLKg03qB+GLS0VIKogV4yOJFvDMVWGOvE/1p8mhFNQGlGMkny0j30ljQai7FiPKcCSXZd4eOtKElFIHbBVz3rSG4/JRButuV+b1oDIZtX/BMj2KfJyZGaBlJoKYhFotXWmTvDKl8Iqs9XVZWhkceeQSPPPJI0m30qqueeeaZWLNmTbZjy2s666bxtS/+IYuDz4SI6GinVTtVk3USyt3l5JgRLRWyZSSMd7fsx1f7mvDDCYMMLTwVVqX26p8/+xtjAdU9u/rxs2mjk77XfReMwQOLvsTsc0cZNj6zEL2hWmkZUbG/MYj65lhVYyGOSe6h7LMIObW3g3OsWPytjrXoTJ0RIO5uSnaFSvRR1mXJp10nYoi+3n8EV82LFVOs7lGKs0f3TvWyrFBWXU0mgh9duhUA0Ls8MZB+5qRhePydr3HB2H64cNwAXDgudddfuyICyGkZUbP9wBH59uCeXSwcSWFBMWIR0U7UGQHsUxKeMSPWkA+WEb1Cd6cP74XqHiXYdSiesh5oMbb3jjJoNVkA66H2INc+5YmpvLdNHo7ThvXC+EHdDR2X2YjUesaMqBHibGSfMktrOBUaFCMWIdbejooRUdfD6lgLIYb8HUwNZMxIx4iqLCPO3HfaYd985jCUFnnx9h1n4EBTG6576hNsqm1M2z8mW8IZFD0TdSauPDUxVsTncWPiUb0MHZMViIqxz364A+GohNOHVxryvgebgnjho5040haB1+3CBeP6274gnBLhtmImjblQjFhEpBN1RoB4b4lkPm+zEFdVHa1TwJiR7JEkKS9Se7Uiytd+LPi9HvTvVoLBPbtgU20j2gwWqkrXZrLUeLnYWR5XBe7XrQRrdtbjPxtqsWLLfmz45VRDLAFPr9qBP7S7uQBgw94GPHXtyZ1+X7MQv31xBrVliHFQjFhEZ8rBA/FYizW76jF1TB/LzImimmVHJ23ZMuLQBdUKtBYFhxpGkG7YopeK0YI7k3Lw8avj/F2Q7pwyAgO6leAv725Dc1sEB5qCKDWgv05NfczF1qe8GLWBVsdVVy4EIWpHKEYsQsyBHW2TLiwRf313GypKfJg56SijhpYVopR0Ry0jIvbFyKv71d8cwtZ9TTh7dG/06pp/lXy1zc2c6qZJN25hKTG6sJ/SJRiKStiwpwEAMLpvOZZu2ofqHiWyyM5nU/3gXl1w17kj8bf3tiEqASf/emn6F2XB2OpuePOLWmysCaA1FHHMvoz/9vkrRO0I97ZFdKYcPABMPzlePH3jXuu6Gotgr07HjBgkRuqb23DxE6sw++X1+NXrGw15T7uhXcQXfV4jV4x0Euk0lBCqqaqkdgRl5lZzMIzv/vF9fPeP72PJl3W44ZnV+O4f3pcbyDllAe0oLpcLZ40yLlNJUOb34qqJ8XibLXWNhn9GrmgtkN/ebtAyYhFiPuxoSuw5x/TF7y87Hrf/fR0aDM42yIbOumm8bmNjXw40xUt9v7J2LyQAv7t0bIf3sx3RLuJLN+3DY+9sxZ1TR1ozoA6SVox4c2UZib9fY2u8Guv63TELSTgqoam9HH0hXB3/7aoT5QXYKLxuF7weN04Y1B2f7jiMHQebcVx7cz67I4LynVjIzslQjFhEZ2NGAEVJeAvTYjvrphEiYcPeBizZWJdm6/QIk7vg1bV7cc3EwRg30NlpmEr03Bt1DvPLA+ndNHINmhy6aSJJspLEzUKJG8iVFaCiJFY0zEmWu4hcyJFixEwoRixCxEh0QovA57E+LTZuGenYiVvUPtk/9+FOPPfhzjRbd4x/rdmTZ2IkfvsnZw/Hw0u2IOrAAOB0YqTIm5v0daV4b1NUMNbG4gCFYRnJJeJiw+q2Fdkg3HgdrZ1EOgbFiMm0haO48smP8PE3hwB0zjLikS0jNhAjHZy0rzxlIOoCrapFoTOs3VUv3x7dtxwbawJ55aIB1Iu4uHrTW0jtTtpsmnaxbXRhP+X5oqw+qifoigvEMpIr7HDBlC3CjdfRFhekY1CMmMzX+5tkIQJ0PIAVUJaEt+6qQ0zmHfWvnjK0JxbcOMGw8Qy+K94NevKoKmysCThqIswESfFz+zzOrdMipTlszXDTBBUi+NARdexVsc/NCpydJFe/YS6J0DJiCbRBmoz2pOzMZCea5dnCTWPDyHO5Sq0DF+pUqCwj7cePE9N7rXLTKM9B5RA++OqAaru7znFWQLAdsUsPrWwIMWbEEmgZMRltmmLn3DSx126qtSZtTpKkeACrDU9cr02q1BqNrpvGQZO9wA5uGiVKkbLu7imoYMfWThO/YHLOOdjZflukY9hvBclztCdlZ473Xl2L5Nt6TcdyjdLE3dGYkVySr6Xm4wXz4oLUQXO9TObZNLmzjChpaU9vvfTEARQiBuFIy4jsprHfnJbPcG+bjPak7IxlpH/3Evl2swWpc8punyV2dNOIxcxBE2EmKNPCPa78ddPkqhx8MrePECOdOSeJGp9NGnpmAwNYrYFixETe3bIfb3+prqXR0XLwQEwACMtKUzCcemOD+HDbQfzmzU04Egzj4JFYbQuP2yVPOnbCl7dumth/tyueGu5EMZLOT5OzcvBJAr4l2eLERcgovA5M7Q0zgNUSGDNiEl/sbcBV8z5OeLwzx7vL5UJXvxeB1jCagmEYX9Q5kf9+ZjUCrWF0K/HJTbXs6gZx562bRtSocSncNM77jumGLASukV17JUlK6/bhGmQcXkdaRhjAagXc2yYhOleWFav1X2drYHRtFwRHTLKMBNrLZ3+xN4CG5ljp9eMGVJjy2dnill0YFg/EYJR9jTx5nE2TCzdNJqKNbhrjcKJ1kqm91kAxYhLC1HxUVVfcOXWE/HhnJz5vjpqJpcPtigewjqvuZupnZ4qT4ylSISzebpdLPn6caRlRj7mqvFh1PxdumkwCKbkGGYcT47ZCjBmxBIoRkxCToM/tVgmQzl6EiYnT7GwaCfb3see7m8atcNM4yCUvoxzz9JMH4qLxA1TP56JrbybCxq7HsxMRzQ6NqrBsBowZsQbGjJiEmAR9XheUrsjOWkbE681ebiVJGbtg8odniNjPeWcZUbhpZMuIA7+jGHNZsRdzvn9swvPCTRNoCeGznYcN+cym1vTuTLppjKNbSaz8QH1zW5ot7UOYqb2WQDFiEnJVP41lpNPiW2RTmHz1L0GZ1WHPydudr24asd/dzg5gTVdcShTS23bgCL7/p5WmjYsXxMbRxR9L+T8SdE7XXqb2WgPFiEnIlhGPRox0cuazzjIiQUL8Ct2OOCWeIhqVsLmuEeGIhJIiD4ZVdknpKlDVGXGw9UdcgSYL4h5b3Q0Th/XErsPNhn/2d0ZU4elVO3SfYz8a4xDWBSdZ7mgZsQaKEZMIy2LEpZp8O2tVEK+2YjGye8yIU+Ip7nt9I55a+Y18/5ffG41rThuSdHu5Aiucbf0RKZTJxEhJkQcv3HBqjj47mlSM2PRwdiSyWLb5BYESMVfnW7dvu0PpZxJtitx15THe2eNdFjMmn+uxAFZ7x4w4JZ7iq31NAOKT39b2+8lQRvvHrT85HGCOEALKiivQVAuNXd2OTsQp56ASOdmAbhpToRgxCaVlxG2kZUSuwNmpt8keSXmFbs+T1ik1OERF0KOrugJI71YSjeOKvG6F9cfe31EPMelbYQ13uVxJLwR4QWwcTjw+WfTMGri3TUKOGdEEsDa0hDr1vi4TzfTK9GEJkuymsevkLcZl94lQuJH83sxSWZWdkp145SmIWOybT2YdoWXEOOT0egcdn2F27bUEihGTEAuMzxtvbgbEgvQ6g3gnM0515XzihNRep0yEYnx+byzzIFnvFMHGvQEAQJHXY+srzw17GvD//rUeb26o0X0+kiaANdckEx12jYFyIh4HuhFlse/l8mgm3NsmIfv53W6Vm6azE7HbxGwK7WcoszrsiFMmQuGu8PvcqvvJeP6jWOBlOBKVjx87tmi///WNeOGjnbjpuc90n7e67HZyy4jJA8lj7CyWkyHEiJ9ixFS4t01CLBZFXnUAa2dN1GYGsEY1lhFx105Xkt86qhcA4PThlfJEaHZ12mwRE7U/w14sojLpxScMkGtxGN3Z1ggOHokXutL7DeSYEYuOH0+Sz7WruHYicbFsv+MzGUFZjHgsHklhwdRekxBq2+tWp/Z21jJiZmqvpFA8EqS4mybnn5w5D1x4DF74aCeuOGUQdtfH6lPYvc6IbBlpn/wyDWAdW91NLrdtRzGiPLQjUSmhiJQQYVYVl0pWT4SWEeOIB5FbPJAsoJvGGihGTEJcGWiLnnXWRB0PYO3U22SENmYkHsBqn9l7UM8umH3eKABATUMLgNzEjLS0RQwTgMISIiwjwXAUoUhUtoBoibcWcMuWkaANe38oj4tQRIL2QjNd0bNck+xz7WTpczpOKTyohL1prMFRYuRQUxAhd9DqYXQIUQ7Zp6gNAXT+qlD20pgcM6IsB2/XuTtX/urfLdmCPyzdauh7AvErsfe2HsDRP/sP7j9/DH44YbD8vCRJuPffG7HrUExkFXnc8mvs2IhMuai3RaIogVqNiHLwydwluYbZNLnHie0K5N5PFCOm4igxcvqDy+H2l1o9jE7h9bhVjfI6HcBqoWUEdi8Hn6Nsmne37Df0/QCgT3kxxvSvAD7ZJT/2i1e/UImRXYda5EqtXrcLvcuLZRFqRzeNEr3xiYcss4wkjRkxeSB5jMfE0gNGYfeSBfmKo8SI06ko8WHC0J6oKvejT3kxPG4XjulX0an3jJ8vZmfTSHJ9DLuatWWhZvA6La7y/vLDE3DG8EpD3tPncWPJxtqU27QpFvRFt34blWV+HGiKWQqjUsxyYqffQmmR0hMjcj0Hi2JGaBnJPSI+35GWER4HpuIoMbLh3qkoLy+3ehiG8OH/O8uQ9zHVMqK8LcUDWu16zubqqkwsrKVFHhT7jIu4T5dZJb5Hjy5FGNGnDID66j4qAXaqYK20SIXCib9BxOJsmmQiyK7HsxNxShVkJXavn5SvOEqMkETECbNofY3c48RIvB4Xzh/bH/27lUDSXNzKrextetbm6qosV109PWmUhN7irfRrR6KSrZp7Kfd7m65lxK51RuyzD52OV07tdZIYif3ncWAuFCMOp4s/9hMu+rwGi6Bf6bKzbNwbwGOXj9cJYLVfaq+SXF2VKfsMGYkvjbiJVyyNP6ZcUO129amsLaHnponK38eaFMpkIshGes7xOC2bRpkIQDFiLhQjDuenU0ZgQPeSnFx57DrUjPe2HsD+xlhcgnZBiVqcmpkOYbkwOtMklKNGWun2oxAbSteM8rbdJvyIosdOWKffTlhHXJlJMhHELArjcFoFVuUweRiYC8WIwxndrxz3nX9MTt572aY6vLf1AFpCsbRkpak9EpUsN7Ono6w4dngHWsOGBnfmqpFWOkuLbD5Wdn1WrKd268ETTuOmsbpRXrLfz05BwE7HaY0cldZFHgfmwhJzJCklvthifiQYBqDuJhuVpLjbwKattitKfPJtUZvDCMRVfrKiZB0lnWVEL2ZEFcBqs6vPSJpsGqsb5SULYLWptnYk4rdtDUXlecTORFVuGgsHUoDYcxUhtqC0KJYp0tIWQVMwjIfe2iw/F5XsbxlRZro0GTgRioXV6EU0nbiR3TRJ2gnYzU0TtrsYYQBrzunV1S/f3lgTsHAkmSGp3DQ8DsyEYoQkpYs/tpg3hyJ4be1eLPo8HiAbjSoWExuftP0qigEY26hLfG+jA1gzt4zEH3O5XHJG1Rd77TXZp7OM2LUcvE21tSMp8rpR3aMEgP3Esh5RBrBaBsUISUpJUcxN0xyMoC7QqnouFIlavphkglfuamvcRBiSG7wZe/qkszAlCxgW8+frn+81dDydRSkA9wUS2ziIid8qy1qyWBXGChhLabu71xliJH6bh4G5UIyQpJS2uznaIlHsa8+oOaZ/rOhcSyiC7QdidU2sqqCZCWJsYQPKpR9oCmL34WY5O8dn8CKaroBasvoHPzx1EABjBVdnkSRJNZ6dh5oTtsmVuytTkn2u0RavQsdJ/WloGbEOZtOQpJQUxRfHFz/eCQDoW1GCDXsCKpeAnS0jonZHZ1Ofl22qw4ynVqseM9oyUt2jFCN6l2FzXaPu85EkZaqP7t0VABAMRwwdT2cIhqOqxUdvIWoNxcSIkVVssyGZ6LAquydfERcEThAjysKONp7W8hKedSQpfq8bE4b2lO93KfLo9mI5VbGN3RATYWcbyX2xJzEeo3upT2fLzvHk1ScmfS6Zm6bYG1vMgyH7NMvbtv+I6n59cyhhm9b2lPFinzXTEC0j5iDEsxOqsNIyYh20jJCkuFwuvHDDKWhuiy0aPo8bOw+pF5kbzxiqipi3G8J6oVd0KxtCmon0+m8NyUlsQaqMGjmAVbOIFnljrwkaXNytoyz4ZBdm/fNz1WOHm9sStpPFiNcay0gyCwgtI8bild009jg+UxFR1RmxcCAFCM86khKXy4Uufi+6+L0o8rrloFaBVQtJpoi4js5m02hjTox20QiUV+vaC/eIXIFV/bhR1h+j+HxPvXy7T3ksm0lPKIlCaEJMmU0yy4idY6CcSDxmxOKBZICySR4Dmc2FYoRkRZFmEbZqIckUsbDc+tJazHzhM/lqPFu0C32uTPnKzBKtLUfSqTMSe40xcTFGIXbV7ZOH43/PHQFAv29OKEfF4zIlmeiwajz5ihxE7gDLiJQkSJzkHp51JCu0F5N2Dl4F4gtLWziKRZ/X4LMdhzv0PtpMlVwtWMrOvdr1Wyzy2is2n4EZQ0YQF02pG6XlquFgpiRrTGj3Y9ppOKlZnl7/J2IOFCMkK7RXDHaft7U1LDpaiVV7VWeFKT+SZKLMRS2VzqCMbUmV1hmSxYg105DPywBWM/DKrlJ7HJ+pEEOkFjGfrGaBwYMHt1d8VP/NnDlTd/unnnoqYdvi4mJDBk6sIVGM2Pus1cZ2HDySGEiZCa2aTBV/jmJluipickqL1J+RLJvGqLgYo1CKJiGc7OimSfa5DGA1FtEd2W69k/SIylWO7T2v5SNZnXWffPIJampq5L8lS5YAAC655JKkrykvL1e9ZseOHZ0bMbEUl+aIsXuQl+jcK5j98vqs63EcOtKGf3y6W/2+/twkorndLrw3axIAoLktIrs8gOTZNEZlDBmFUjS5M7CMWBUwqo1/EjCA1VjEbnaCZSQeM2LtOAqRrGbUykp1jYn/+7//w7Bhw3DGGWckfY3L5UKfPn06NjpiO7TnqN3n7UqdtON9gSCqe5Rm/B7/XpdYZr1PRe4sfJVl8TFvP3AEQytjRc2i6bJpbGMZif13KywjejpJiKdkoiDXdEkiKBnAaizC0uSkmBFaRsynw5d3bW1teO6553DHHXekvDpuamrCoEGDEI1GMX78ePz617/GmDFjUr53MBhEMBjvZREI2KsBWCGT4Kax+SWEXyfbR89lkIqW9gwcj9uFedechMNH2vCto3oZMj49lBVJA63xGJdkE6VcZdaWlhH1Y0raZMuINYv/ZSdVY39jEFv3NeLDbYfkx+3ahdqpOLEcPLWI+XR4FnjllVdQX1+Pa665Juk2I0aMwLx58/Dqq6/iueeeQzQaxcSJE7F79+6krwGAOXPmoKKiQv6rrq7u6DCJwWgXQru7afRSj7OdFMUEddH4/jhjeCUuGNc/5yLsqKqYNWTB6l3yY6InjvY7xeuM2GOyV3YXTpVJEbI4m6Z3eTHuv+AY3D55uOpxWkaMxesoMRL7b/eLrHykw2fd3Llzce6556Jfv35Jt5kwYQKuuuoqjB07FmeccQZefvllVFZW4i9/+UvK9549ezYaGhrkv127dqXcnpiHVnvY/ZzVEyPZWkbE5q4EJ1XuEGP8fHe9/JgoHKbt5eKzWR0HuYeOIptmY00AX+xtUG1ntZtGoD1GmNprLGJht4sbMRUS3TSW0aFZYMeOHXj77bdx/fXXZ/U6n8+HcePG4auvvkq5nd/vR3l5ueqP2INEMWLvk1Zvocv2Ak2eoExcM3/x3dEAgP2NQbz82W68/NluuVmh1vUkfPJ6/V+sQHbTuFyqrKNLn1ilKjoXsthNI9BmRjGA1Vi6tsfmNLV2LK3eTKIMYLWMDs0C8+fPR1VVFaZNm5bV6yKRCNavX4++fft25GOJDXBanRG/TkfY7N004pZ5X7Z3WSxAti4QxB0L1uGOBevwdXvzuRJNyq/y/uZa/Y6/ZqIs8z5uYDdcfspAAMCRtggCLXHBFE6Sqmw2fk2jvmTF0EjH6NGlCIB+fyK7EY8ZsfnElodkfdZFo1HMnz8fV199NbxedfzrVVddhdmzZ8v377vvPrz11lvYtm0bPvvsM1x55ZXYsWNH1hYVYh8cFzOic9WdrRixIt1vZJ8yzDhtCE4fXonTNZ2SjxtQobrfuzye2fNljfXB3kFFbIvP48avLzxWToV+b+sBeTvxO1hdZEx7jNAyYiyiK7OdukonIx4kbvFACpCss2nefvtt7Ny5EzNmzEh4bufOnXArrioOHz6MG264AbW1tejevTtOOOEErFy5EqNHj+7cqIllaM9R27tpDIgZsSLC3u124e7vxc+Tu/75OV76JBY7dXRVWcL2U0b3xlsb61DT0GraGJMhAm2V7o/G9sq3/9lQi4tOGAAgHuNi9TGUYBlhAKuhiOPALl2lU2FFfBiJkbUYmTJliqoQk5Lly5er7v/+97/H73//+w4NjNgT7bph93nbiGwasbWVi2Y6C5Rwdby5oQY3nznMjCElJaiT9XPrd47CH5Z9pbriFL+D1ZYIbUAwU3uNRcQ4OUGMRGziOixEclNGkuQtsbL+zuluaUSdESG+rfymStGnt89FEbbu7f55K4m0WzyUi3rfbiUA1J2IZTFi8cRfXuxT3edCZCzC8qStfLxhTwNueGY1GlpCqCrz44UbTkW/9uPEKuRWBjwGTMfm17XEjigXQ9vHjOhaRrJ7DzsEtXlU+zzx+TH9YnEkdijlIJu6FeN0yc/FBxgPYLV+GuqvWATtfkw7jWRumhVb9qOmoRXNbRF8c7AZn3xzSO/lpkLLiHVYPwsQx6E8Te1+zupZRjoewGrdl1UWYdIbh3g6mQvVTPQqxYrbyuFFFCnAJH9J5qbRVgy2Q9E+ZcE+Yi4UIyRr9BYZu1LkSUztzT6ANfbfyq+azjLiksWISQNKgZ54EzeV+14WI8xeyWtky0hI7abRFukLZ2uyzAHJOmOT3EMxQrJGuRja/ZwdVtUl4bHsA1itv1ryqCwjic+LhT9boZUL9LKPXPL44o/ZJWaE5BYRM9KmsYxoLSEhG/gYI6zAahkUIyRr9BYZu1JalBijHeloOXibuGn0QmldOm4Qq4hnH8Ufk91IYhtJkmNGOPHnN8ndNPazjDBmxDooRkjWOMlNo4deB9lUSBbUGdHicaW2jIiH7GUZSXTTiH2p/AnsYBk555g+AIATB3W3eCT5RzyAVeum0caMWC9GosymsQym9pKsUQoQu9cZ0aOj5eCtLISktIzoWWjkAFHTRpQcEQqQKoBV+RvYIWbkZ+eNwoXj+svdkolxiIy2A01t+HDbQZw6tCeARPFhjwDW2H8nXmQ5HQcuJcRqnOSm0aOjXXstjRlRWhl0nrdTNk2quixi3yvFiB0sI263C8f0r0gogEY6T1WZX749/4Pt8m1tNo32vhXo1cgh5kAxQrJGndrrvJO2o3VGrPyu6YqexbNVTBpQCvQq1moDbJWZFE48hkjmdPF7cefUEQCAVkV/Gq2bRptdYwWyZYRixHQoRkjWuNNkdtidbANY99S3ALA2ZkTtpkl8Ph7Aar0a0cum0bpplOsOr0Lzn37dYhWClVZJIT7EcWILN42IGaFANh2KEZI1hRTAunzzPizZWAfA3hVY4wGs5ownFXp1WbR1UJRXwQwWzH/0Us+FW6ak3TVmh2wa1hmxDooRkjWqJFMHnrPZBLBurWuSb08eVZWL4WSEx0EBrJKOWyue2quOGXG7nBl3RLJDHAvKc08EsMpixAZKWj4uKUZMh2KEZI3L4ZaRbNw0YoK8+IQBOG5AtxyNKD16C7vq+fYz2Q5uGv3y+eqiZ2G54BmnoEJAiGmle04cAyJo2A6pvXE3jcUDKUA4E5CsURezct5Zm42bJmoTH7Jyn+ulGLuQaAa3injAb/wxbbaPXH2Vs35BIFtGpETLSHF7hVZ7ZNPQTWMVFCMka9JdpduNJ686EVecMhCTRlQCyNIyErFH/5R05eDt1JtGL2YkHjMQux/mpF9QiJ9Z6aaRY0aK2i0jtsim4XFpFRQjJGuUi0yvrv7kG9qEyaN744ELj5XNwdlYRuxSd8CtNo0kPq/T+8UqpAwqsIpgRav3KzEH2U2jk00TD2C1/uBlBVbrYAVWkjXKq+/yEp91A8kSsaBnE8Bql8ZZ6qJnyeuM2DVmRBtgG7eM8HqoEHDriBGRylssB7Cmt4wEwxFVULnR7DkcS+O3+nwvRChGSNYoJxQn+fw9st8689fEAy2t/Z4+RdUzvbFo63hYiV7MiNBP2gqsPgcdP6TjiHMv0BLGyq8PYPzA7gmWkY17A2nf54dPfoyPvzmUu4G2Y/X5XohQjJCsUYoRn4OubOMR/Zmv2PsCQdVrreKMEZU4c0QlhvTqgu5dihKet1ejvNh/pQVHK5ZE8KLV+5WYg7ho2XmoGZf/7SOce0wf2S3TxR9bhr452Ix1u+pxfHW3pO+zua4RANCra1HOMrH8Pjf+a2y/nLw3SQ7FCMkaVZMzBy0mehH96Vj6ZazgmdVLfK+ufjx17clJn3fZqc4I9Cqwxv6LQ+dIMNbBtYS9YAqC8QO7Y/Ko3thcF8CuQy3YeahZFqTfGVmFf63ZAwDYcag5pRgRsUb/vHkiBvXskvNxE/NwzmUtsQ12a3KWKWKon+44nPFrerRbIUb2KcvFkAzDrXGDWIk4PFQl7KEuV3/wSMziVFlm/wBo0nmKfR48efWJmHPhcQBic4hwgfatKMa3j+4FIH0V1pCcEs6lK9/gL0qyJtAalm87qVKhv72ewZKNddhx8EhGr2ls/65j+lXkbFxG4LJRzIiUss5I7L8w0fu4qBQUwpIajkryMeD1uOXjIF1GjRArPgfNOyQzOBOQguHykwfJt0Xzu3Q0BmNipGuxvT2a2qJiVqIXM5IsgNVJbj7SeUTsSMwyEk/vFsdBqlojkagkH1u0jOQf/EVJhyly2IQwul85jhsQs3C0hiJptw+GI2gLxybHrn57ixGXJibDSvQtI8lSeylGCom4ZSSqsIy45KyqVJYRZbl4J2Xxkcxw1mpCSCcp9sYCJltD6WsaNCncUfYXI3YqBx/7r9fDSLaM2KTMPjEXkX0XiUiyuPC63XJmTKr+NMpGek7K4iOZwV+UFBTF7aWnW9rSW0aa2l00XYo8tr+Ct0udEaWbSKkztOXqIyK1l1e4BYUqZkRRa0ZYOlJ17g3TMpLXUIyQDiPZIpE0O0rag1hbMnDTyOmnRfa2igDxOiNWx4wo1xK9HkZyozzh+7e5yCPGoo4ZibvqhKUjVTZNSOHC4XGTf1CMkIJC1LXIJGZEBNgVOeAqTBuTYRVKMaReL9S9c0TPH7ppCgulZSSqECNCpIRSxIwoA15dPG7yDooRUlAUZyFGZJ+2AwJ14wGs9rGMuPQsIxDZNLH7dnd/EWMRFo1IVFLFDcmpvSmyaZQBryT/sP8sS4iBCDGSiZsmFIn7tO2OXbJpokljRkQp/th92TJCMVJQiN+7LRKV44c8bpcsUjLJpmHwan7CX5UUFCVyAGv6bBonFeaySwCrklQxI0ztLUxE1oxImQeEm0Zk06Ry09Ayks/Yf5YltsXu6a56lGRjGRE+agdMfi7NYm8V0SQxI9qYFhEvwEDEwkJPfLrdijojKdw0TnKbkuzhr0qy5qdThsPjdmH2uaOsHkrWFLdn0wQzESPheB0Eu2OXANZk2TTx59WWESe1EyCdR098elwuRZ2RVKm97ZZKHjN5ifMubYnl/Og7R+P6bw+V4y+cRDaWEbFgOqHSrF0a5SX7fK0bSQQv0jJSWOjVlVFm06RK7ZWzaRxwPpLs4a9KOoQThQiQbQCrc9w0cuqsxRGsUhLLiDbANhKhZaQQ0bWMqNw0qQJYnRNQTrKHYoQUFCVZVGBVdhW1O/HUWWtRiiG9mBExQlpGChO9mBG1myZ9aq8TAspJ9vBXJQWF3JsmnEE2TdQ57crFYt/YGs6ohkquUHZdVS484qbo1ivEIIueFRba+CuXSxPAmiq110EB5SR7KEZIQSEsI60ZWEbaHHQlJr4XAPx73V7LxhFWmNJdKjeNOsD2pU92iSfMHB6xGI/bpfrJhRj1ZlP0zAEB5SR7+KuSgiKbmJGwg2JGepcXy7cPHWmzbByRqP6CIQfYtj8/tFcXAEBlmd+8wRFboHTNiZgh8VjqbJp2S6UDzkeSPRQjpKDIqjeNgywjAHDxCQMAxOMxrCBZ0K9wI4mQEpF1M7pvmXmDI7ZA6b4TlhFxjkVSBbAmEbokP+CvSgqK4iy69oYUjbmcgBinlRk18bbw6qlFLEBChIghsuFZ4aEUE+KYjTfKS+WmcY6lkmQPxQgpKOSYkYyKnrUvrF5nnCbC5J1iPs85YjHRZk24NAGsQpToFUYj+Y1H100jYkYyKHrmEEslyQ7+qqSgEG6aUERKWWAJcFY2DRA3eVvppklWJVMsQGJo4r9Ddi0xEKWlURwXvgyKnjnNUkmyg2KEFBTKYm3p0ntDDqozAsQn9kiKjIRcE29mpg1g1bppaBkpVNQp3+psmozKwTvkfCTZwV+VFBR+hcslnavGaT5qjw3cNMn2meymaRchwl1DLVJ4qC0jsf/CkpZZozweNPkIxQgpKFwulzwZpiqwBMQnP59Dove1QaJWIAewavaZR9GbRpIkOYCVlpHCQ9mfRsSKyHVGUllGmE2T1/BXJQVHJpH7QDyV0ClmYbGw/+PT3ZaNIV1qLxAXJNrHSWGgyqbxaLJpUhY9Y52RfIZde0nB4XO70YpoWjHiNDdNt1IfgFjRsyPBMLr4zT+9k/XzUYqOiCQpYkbMGxuxBx69AFZ3csvITxeuw+INtWgMhgE453wk2eGMSz5CDMSbQYdQQF3a3AlcccpA+faR9onbbJJlICkt61GFm4Z1RgoPZcyIT3bT6FdgjUYl/OPT3bIQib2ey1Y+wl+VFBzxyP3M3DROmfzKin2KCrPWRLHGM5CSu2miUdAyUsDoWkY8+gGsem4bp1wckOxwxixLiIEUZRAsBwCh9tRfpxQ9A+IVZv+1Zo8lny9bRpJUYAViQkRiAGvBonThyTEjSdw0bTrp905JtSfZwV+VFBzeJFdhWoLhWOqv30FiREzeB48ELfl82TKSpAIrINw0DGAtVPSKniULKtcTI04pQkiyI6tZdvDgwXC5XAl/M2fOTPqahQsXYuTIkSguLsaxxx6LN954o9ODJqQzZNIhFACC7ROhk8TIzO8cBQBobktf7j4XZBLAqnTTUIsUHkormbCgif/aOK42HVdqoNWaeCiSW7KaZT/55BPU1NTIf0uWLAEAXHLJJbrbr1y5EtOnT8d1112HNWvW4IILLsAFF1yADRs2dH7khHQQX4ZumrgY8aTczk6UtseMtFglRqL66Zcel9pNI4xSbl7lFhxKy0jX9owvr1ywT5LTvoF4fyglegKFOJ+sxEhlZSX69Okj/73++usYNmwYzjjjDN3tH330UZxzzjm48847MWrUKNx///0YP348HnvsMUMGT0hHyKSmARA3ETvJMlLULpysmrDjbhr1PtO6aUIO6/tDjENpGRHp50pLmtJi2RZJFNUemtPykg7Psm1tbXjuuecwY8aMpOl5q1atwuTJk1WPTZ06FatWrUr53sFgEIFAQPVHiFEkC5bTIldgdVDAnPIK0wqSl4N3yYIkHI0HsDIYsfBQp/aqs2kAdSxXUCdmhPo1P+nwTPDKK6+gvr4e11xzTdJtamtr0bt3b9VjvXv3Rm1tbcr3njNnDioqKuS/6urqjg6TkAR8mVZgdWDFx0yry+aKeMnuxH0mrmiVQYkeriwFh0dhNRNuOqUlTWkZ0YvrmnhUrxyOjlhFh8XI3Llzce6556Jfv35GjgcAMHv2bDQ0NMh/u3btMvwzSOEiYkDSixFnde0FMuvxkUvi5eAT95kIYlW6kJwk9IgxqLJpXDqWEcXxoZdNM2V074THiPPpUL3oHTt24O2338bLL7+ccrs+ffqgrq5O9VhdXR369OmT8nV+vx9+v78jQyMkLSIGJJimMJhYWIucJEYsd9OIRnmJIsPtBhChZaTQUTbKE5YRl8sFj9uFSFRSZdToiRFW7c1POjTLzp8/H1VVVZg2bVrK7SZMmIClS5eqHluyZAkmTJjQkY8lxBD87YXBdh9uTrmdsIz4vM6Z/OS05TTBublCfG4qy8h7W/fLjzmlui0xDnWdkcTHlRZLvQBWkp9kPRNEo1HMnz8fV199NbxetWHlqquuwuzZs+X7P/7xj/Hmm2/i4YcfxqZNm/DLX/4Sq1evxo9+9KPOj5yQDiIWxec+2plyO9nl4KAFU8SMWGEZaQtHZWuTXjMzsd9//cYm+TFaRgoP5W+urD+jl3LfppPaS/KTrGfZt99+Gzt37sSMGTMSntu5cydqamrk+xMnTsQLL7yAv/71rzj++OPxj3/8A6+88gqOOeaYzo2akE5w8pAeAOI1DpIRdqSbRvTdMXcS37CnAePuewtPrfwGQLwBmhLqDgKoLSNKMaJXGZk1RQqHrGNGpkyZoipKo2T58uUJj11yySVJi6IRYgVCjDS3pa7kmKzpm52RJ3STJ/HPdh7GEUWhNV3LCNUIgTqbRmkl0RPSejEjJD9xziUfIQZR6otp8FQl0yVJkq/KnFVnJDZWs9002kVDb5+xWBUB1Jkz6tLwQkgrU3spRgoF58yyhBiEV2fS06JczB3lpsmwuqzRaHuK6NUZ0Vay/csPT8jpmIg9UQoQpT7VO3ZpGSkcnDPLEmIQ4qq9LRJN6nJUmoqd5KbxZVhd1mhCmkVDL5vG71P3+Blb3S2XQyI2Ra/OCKB/7FKMFA4UI6TgUFo6krkzlFdnTnLTiKtOraUi12jN6XrFzLSWESf1/CHGkTRmRCfeiQGshQNnA1JwKC0dybJOlFf6TqoSWiQXdDO3PkOr5gpWz7Wl7TPipG7IxDiSZtOIAFadomf9u5XA73Xj59NGmTRKYjYUI6TgUFo6kl15xbvPuhxV8bGsOBace6QtktQFlQsaW0OacfgStrnj7OHy7dIijyycSGHhcacLYE20jEwd0wcb7p2K67891KRRErPpUDl4QpxMsj4YSpzYsReI106JRCW0hqIoKTLH+tDYqk6T7lqcOLVMO7YvAq0h1DW04tShPVnwrEBRFz2LPy7ijPRSe4u8bsediyQ7KEZIweFyueB1uxCOSsndNJHklUTtjDehFbs5YkSbJt3Vn/i5brcLV5wyyJTxEPuiEiOqOiOJRc/k/lC0ouU9/IVJQeKTr8JSu2mclNYLqLMTzKw1ciSotoyU+HidQ/RJmk2jWw5eVEF21kUByR5nzbSEGIRc0yCJGGltDwAt9jkryFJ51WmmGGnRBMx20bGMEAIkjxnROyeVbhqS3/DyhRQkRTr+aSUi88Np6aculwtuFxCV1GJk3a567ErTpbijjOxTjs93N6geKy3i1EL0SZdNo0xLb3NgfyjSMThjkIJEXGm1JkmBFY9rC3U5Aa/bjbZIVJ7Ut+1vwvmPf2DqGGgZIcnwKISFMoBVN5um/aLA57CLApI9FCOkIKko8aGmoRW/XbwJz19/asLzTrWMAO2m70jcMlLT0AoAKPF5cHx1haGf9dH2Q1BmEJ8zpg+OHVBBywhJijepm0Ynm4aWkYKBMwYpSCpKYnUwPvjqIHYdakZ1j1LV88GwiBlx3iQoJnghRoQPfmhlF7z03xMM/axj71mMxvbg1eMHVOAJ9pshaUiWTePTyaZhzEjhwF+YFCR/vjK+aK7f05DwfGtIWEac524Qk72o/SFEiV7zus6iXCT0+tEQosWrk84LKANYE7v2OtFCSbKDvzApSHp0KcIZwysBAPXNoYTnN9cGADjTMtLQEvs+tYGYe0bEjuSiyJhSjLCIGckE5XGizFYTYvbBxZvx2zc3AVDEjFDo5j38hUnB0r005qppbgsnPLe3PraQO3ESHN67KwAg2h7MIeo25MJyodw/TurhQ6zDq2iUp7R4+BQi5U/Lv0ZTMCzHbtFNk//wFyYFi3DB6GXUiMeOH9DNzCEZQreSIgBx94zwwefCTaOs+KrsxkpIMpSWEaUbVCuWH317CzbVNgJgAGshwF+YFCzCBaPtJgsASzftAwD0rig2dUxGoC0elVPLiEKA5ELskPyjvCSeN1FRGm+oqG298Lf3tsu3e5X5cz8wYikUI6RgETVEtGJEWeege2li91m7o82myWUAq3IBoRghmXB0VZnidlf59ndGVKGyzI8Sndo+Q3t1MWVsxDqY2ksKFrkxl6YKq7IC5NjqbmYOyRC0PT7CORUjymwaihGSnsoyP3578XHweVwoK46L/VOG9sQnP5sMABh81yL58UcuGwuXi8dWvkMxQgoWr2xBUFtGlEW8nJghEu9+qokZyYFY8KnSNGloJZlx6YnVKZ8/fkAF1rW3GHDiOUiyh7MHKVg8Or0wACCiUCNuB16RCdEhRIiwkOQiwJRuGpILlAKEWVqFAcUIKVjEQqrtbht1uhhxa9007bUaciAWfHTTkBygdP8xS6sw4K9MChYhNLSWESmq3MbMERmDV1NWO5dFz9R9RjidEGNIVqWV5C+cPUjB4nXnqWXEoxZZETm1N8cBrFw0iEEwMLrwoBghBYuwFLy39QD+9x+fy6WnlWLEgVokHgvTLkLiHYiN77Oj9OezfwgxiiJVZV8eV4UAf2VSsPRtL2h2oCmIv6/ehdU7DgGIB7C6XHBkSqFPYxlpCYkOxMaLEaVrpqTIeU0FiT354YRBGDewG847tg/GDexm9XCICVCMkIJlypg+eGbGyajuUQIACLZ36hWGEY8DhQgQt/j8YelWNLeFMff9WCXLXDT9O9gUlG/nQuyQwuSM4ZX41y2n4U9XnODIztkkeyhGSMHicbtw+vBK9OoaKzUtyqcLN40T40UAYFhlvKrlPz/bI98e3rtMb/NOsfLrg/JtvcqZhBCSCRQjpODxaeqNiHhWh2oRXHHKQHnse+tbAMR88Ocd29fwzzppcHf5ds+uRYa/PyGkMKAYIQWPtrFcNOpsy4jL5cIx/SoAAEeCYQDAyL7GW0UA4I/Tx6O0yIPTh1di6pg+OfkMQkj+w3LwpODxanq5xN00lg2p04i4kea2WPBqrlqw96koxsb7zsnJexNCCgdaRkjBoy0SJtw0TrWMAPHv1CLECNNuCSE2hjMUKXjEwh2S63LEFnB/DrJPzCJuGYm5aShGCCF2hjMUKXh8spsmZhkRrg0n180QcTDiu7BwFCHEznCGIgWPtnx6qxAjDk5VFcXIRMEzWkYIIXaGMxQpeLya1F6xgDtZjAjX01f7mgAAflpGCCE2hjMUKXjk8ukaN42TK4r2aS91L75L327FVg6HEEJSwtReUvDE64zELCOffBPrUePkmJHZ547EacN6IRSJotjnxhnDq6weEiGEJIVihBQ8RZ6Y6BDdbbv4Y6fFvkAw6WvsTlmxD9OOM77iKiGE5AK6aUjBIxrItbbHioTaRcnpwystGxMhhBQSFCOk4BGBqk+t/AabaxvlsvBFHucWPSOEECdBMUIKHmWg6rQ/vIe2djHC2hyEEGIOnG1JwVOsCFQNRyW0hWOBrD7W5iCEEFPgbEsKnmKN6AjRMkIIIabC2ZYUPNoUXsaMEEKIuVCMkIKn2JtEjNBNQwghpsDZlhQ82kqrbe3Fz+imIYQQc+BsSwqekiL1afDulv0AKEYIIcQsONuSgsfv1S/73rNrkckjIYSQwoRihBQ8yXrQnDCou8kjIYSQwoRihBQ8ybrzFtFNQwghpsDZlhQ8FSU+3cddLqb2EkKIGWQtRvbs2YMrr7wSPXv2RElJCY499lisXr066fbLly+Hy+VK+Kutre3UwAkxiq5+L24+c5jqsQcuPMai0RBCSOHhzWbjw4cP47TTTsOkSZPwn//8B5WVldi6dSu6d0/vW9+8eTPKy8vl+1VVVdmPlpAccfrRlfjz8q/l++OqGS9CCCFmkZUY+c1vfoPq6mrMnz9ffmzIkCEZvbaqqgrdunXLaNtgMIhgMCjfDwQC2QyTkKzxaqqtetx00RBCiFlk5aZ57bXXcOKJJ+KSSy5BVVUVxo0bh7/97W8ZvXbs2LHo27cvzj77bHzwwQcpt50zZw4qKirkv+rq6myGSUjWeN0UI4QQYhVZiZFt27bhz3/+M44++mgsXrwYN998M2699VY8/fTTSV/Tt29fPPHEE/jnP/+Jf/7zn6iursaZZ56Jzz77LOlrZs+ejYaGBvlv165d2QyTkKzRFjjTihNCCCG5wyVJkpTpxkVFRTjxxBOxcuVK+bFbb70Vn3zyCVatWpXxh55xxhkYOHAgnn322Yy2DwQCqKioQENDgyruhBCj+LImgHMffU++/96sSajuUWrhiAghxPlkun5nZRnp27cvRo8erXps1KhR2LlzZ1aDO/nkk/HVV19l9RpCcolPEzOijSEhhBCSO7ISI6eddho2b96semzLli0YNGhQVh+6du1a9O3bN6vXEJJLPG635j7FCCGEmEVW2TS33347Jk6ciF//+te49NJL8fHHH+Ovf/0r/vrXv8rbzJ49G3v27MEzzzwDAHjkkUcwZMgQjBkzBq2trXjyySexbNkyvPXWW8Z+E0I6gTZGxOtmPUBCCDGLrMTISSedhH/961+YPXs27rvvPgwZMgSPPPIIrrjiCnmbmpoaldumra0NP/nJT7Bnzx6UlpbiuOOOw9tvv41JkyYZ9y0I6SRM7SWEEOvIKoDVKhjASnLN/sYgTnrgbfn+F/dORRd/VlqdEEKIhpwEsBKSr2gDWGkZIYQQ86AYIQSJ4oN1RgghxDwoRghBYtEzWkYIIcQ8KEYIQaL4cLkoRgghxCwoRggB3TKEEGIlFCOEgJYQQgixEooRQgghhFgKxQghhBBCLIVihBBCCCGWQjFCCCGEEEuhGCGEEEKIpVCMEEIIIcRSKEYIIYQQYikUI4S0w1IjhBBiDRQjhLTzraN6WT0EQggpSChGCCGEEGIpFCOEtFPi81g9BEIIKUi8Vg+AELvwi++OxtZ9TbjuW0OsHgohhBQUFCOEtFPdoxTv/PRMq4dBCCEFB900hBBCCLEUihFCCCGEWArFCCGEEEIshWKEEEIIIZZCMUIIIYQQS6EYIYQQQoilUIwQQgghxFIoRgghhBBiKRQjhBBCCLEUihFCCCGEWArFCCGEEEIshWKEEEIIIZZCMUIIIYQQS6EYIYQQQoileK0eQCZIkgQACAQCFo+EEEIIIZki1m2xjifDEWLk4MGDAIDq6mqLR0IIIYSQbDl48CAqKiqSPu8IMdKjRw8AwM6dO1N+GT0CgQCqq6uxa9culJeX52J4OeGkk07CJ598YvUwMob7OfdwH5sD97M5cD+bg9X7uaGhAQMHDpTX8WQ4Qoy43bHQloqKig7vzPLyckcd8B6Px1HjFXA/5x7uY3PgfjYH7mdzsHo/i3U86fMmjYNkycyZM60eQkHA/Zx7uI/NgfvZHLifc4NLShdVYgMCgQAqKirQ0NCQtbLrzGtJ5nA/5x7uY3PgfjYH7mdzsHo/Z/r5jrCM+P1+3HPPPfD7/aa+lmQO93Pu4T42B+5nc+B+Nger93Omn+8IywghhBBC8hdHWEYIIYQQkr9QjBBCCCHEUihGCCGEEGIpFCOEEEIIsRTbi5E5c+bgpJNOQllZGaqqqnDBBRdg8+bNqm1aW1sxc+ZM9OzZE127dsVFF12Euro61Ta33norTjjhBPj9fowdOzblZ3711VcoKytDt27dDP429sXM/SxJEh566CEMHz4cfr8f/fv3xwMPPJCrr2YrzNzPixcvxqmnnoqysjJUVlbioosuwjfffJOjb2YvjNjP69atw/Tp01FdXY2SkhKMGjUKjz76aMJnLV++HOPHj4ff78dRRx2Fp556KtdfzzaYtZ9ffvllnH322aisrER5eTkmTJiAxYsXm/IdrcbMY1nwwQcfwOv1pl0rjcT2YmTFihWYOXMmPvzwQyxZsgShUAhTpkzBkSNH5G1uv/12/Pvf/8bChQuxYsUK7N27F9///vcT3mvGjBm47LLLUn5eKBTC9OnT8e1vf9vw72JnzNzPP/7xj/Hkk0/ioYcewqZNm/Daa6/h5JNPzsn3shtm7eft27fj/PPPx3e+8x2sXbsWixcvxoEDB3TfJx8xYj9/+umnqKqqwnPPPYcvvvgCP/vZzzB79mw89thj8jbbt2/HtGnTMGnSJKxduxa33XYbrr/++oJZKM3az++++y7OPvtsvPHGG/j0008xadIkfO9738OaNWtM/b5WYNY+FtTX1+Oqq67CWWedZcr3k5Ecxr59+yQA0ooVKyRJkqT6+nrJ5/NJCxculLf58ssvJQDSqlWrEl5/zz33SMcff3zS9581a5Z05ZVXSvPnz5cqKiqMHr5jyNV+3rhxo+T1eqVNmzblbOxOIlf7eeHChZLX65UikYj82GuvvSa5XC6pra3N+C9iczq7nwW33HKLNGnSJPn+rFmzpDFjxqi2ueyyy6SpU6ca/A2cQa72sx6jR4+W7r33XmMG7iByvY8vu+wy6ec//3natdJobG8Z0dLQ0AAg3jzv008/RSgUwuTJk+VtRo4ciYEDB2LVqlVZvfeyZcuwcOFCPP7448YN2KHkaj//+9//xtChQ/H6669jyJAhGDx4MK6//nocOnTI2C/gEHK1n0844QS43W7Mnz8fkUgEDQ0NePbZZzF58mT4fD5jv4QDMGo/NzQ0qBp+rVq1SvUeADB16tSs5558IVf7WUs0GkVjY2Pa5mv5SC738fz587Ft2zbcc889ORh5ahzRKE8QjUZx22234bTTTsMxxxwDAKitrUVRUVFCfEfv3r1RW1ub8XsfPHgQ11xzDZ577rmCL02cy/28bds27NixAwsXLsQzzzyDSCSC22+/HRdffDGWLVtm5NewPbncz0OGDMFbb72FSy+9FDfeeCMikQgmTJiAN954w8iv4AiM2s8rV67E3//+dyxatEh+rLa2Fr179054j0AggJaWFpSUlBj7ZWxMLvezloceeghNTU249NJLDRu/E8jlPt66dSvuuusuvPfee/B6zZcGjhIjM2fOxIYNG/D+++8b/t433HADLr/8cpx++umGv7fTyOV+jkajCAaDeOaZZzB8+HAAwNy5c3HCCSdg8+bNGDFihOGfaVdyuZ9ra2txww034Oqrr8b06dPR2NiIu+++GxdffDGWLFkCl8tl+GfaFSP284YNG3D++efjnnvuwZQpUwwcXf5g1n5+4YUXcO+99+LVV19FVVVVhz/LieRqH0ciEVx++eW499575XnZbBzjpvnRj36E119/He+88w4GDBggP96nTx+0tbWhvr5etX1dXR369OmT8fsvW7YMDz30ELxeL7xeL6677jo0NDTA6/Vi3rx5Rn0N25Pr/dy3b194vV7VAT9q1CgAwM6dOzs3eAeR6/38+OOPo6KiAr/97W8xbtw4nH766XjuueewdOlSfPTRR0Z9DdtjxH7euHEjzjrrLPz3f/83fv7zn6ue69OnT0KmU11dHcrLywvKKpLr/Sx46aWXcP3112PBggUJ7rF8J5f7uLGxEatXr8aPfvQjeQ287777sG7dOni9XnOs1qZFp3SQaDQqzZw5U+rXr5+0ZcuWhOdF8M4//vEP+bFNmzZ1KLBy/fr18t+vfvUrqaysTFq/fr106NAhQ7+THTFrPy9evFgCIH311VfyY2vXrpUASJs3bzbmy9gYs/bzHXfcIZ188smqx/bu3SsBkD744IPOfxGbY9R+3rBhg1RVVSXdeeedup8za9Ys6ZhjjlE9Nn369IIJYDVrP0uSJL3wwgtScXGx9Morrxj7JWyOGfs4Eomo1r/169dLN998szRixAhp/fr1UlNTU26+nALbi5Gbb75ZqqiokJYvXy7V1NTIf83NzfI2N910kzRw4EBp2bJl0urVq6UJEyZIEyZMUL3P1q1bpTVr1kg33nijNHz4cGnNmjXSmjVrpGAwqPu5hZZNY9Z+jkQi0vjx46XTTz9d+uyzz6TVq1dLp5xyinT22Web+n2twqz9vHTpUsnlckn33nuvtGXLFunTTz+Vpk6dKg0aNEj1WfmKEft5/fr1UmVlpXTllVeq3mPfvn3yNtu2bZNKS0ulO++8U/ryyy+lxx9/XPJ4PNKbb75p6ve1CrP28/PPPy95vV7p8ccfV21TX19v6ve1ArP2sRazs2lsL0YA6P7Nnz9f3qalpUW65ZZbpO7du0ulpaXShRdeKNXU1Kje54wzztB9n+3bt+t+bqGJETP38549e6Tvf//7UteuXaXevXtL11xzjXTw4EGTvqm1mLmfX3zxRWncuHFSly5dpMrKSum//uu/pC+//NKkb2otRuzne+65R/c9Bg0apPqsd955Rxo7dqxUVFQkDR06VPUZ+Y5Z+znZ8X711Veb92UtwsxjWYnZYsQlSZKUvXOHEEIIIcQYHBPASgghhJD8hGKEEEIIIZZCMUIIIYQQS6EYIYQQQoilUIwQQgghxFIoRgghhBBiKRQjhBBCCLEUihFCCCGEWArFCCGEEEIshWKEkAJn1apV8Hg8mDZtWtJtXnzxRXg8HsycOTPle82ZMwcejwcPPvhgyu1uvPFGeDweLFy4MOG5X/7yl3C5XDjnnHMSnnvwwQfhcrlw5plndmr7sWPHJr1PCDEfihFCCpy5c+fif/7nf/Duu+9i7969SbeZNWsWXnzxRbS2tiZ9r3nz5mHWrFmYN29e0m2am5vx0ksvpdyub9++eOedd7B79+6E9x84cGCntyeE2AuKEUIKmKamJvz973/HzTffjGnTpuGpp55K2Gb79u1YuXIl7rrrLgwfPhwvv/yy7nutWLECLS0tuO+++xAIBLBy5Urd7RYuXIjRo0fjrrvuwrvvvotdu3YlbFNVVYUpU6bg6aeflh9buXIlDhw4oGvByXZ7Qoi9oBghpIBZsGABRo4ciREjRuDKK6/EvHnzoO2dOX/+fEybNg0VFRW48sorMXfuXN33mjt3LqZPnw6fz4fp06en3O7KK69ERUUFzj33XF0BBAAzZsxQPTdv3jxcccUVKCoqMmR7Qoh9oBghpIARwgAAzjnnHDQ0NGDFihXy89FoFE899ZS8zQ9+8AO8//772L59u+p9AoEA/vGPf8jbXXnllViwYAGamppU223duhUffvghLrvsMnm7+fPnJwggAPjud7+LQCCAd999F0eOHMGCBQswY8aMpN8l2+0JIfaBYoSQAmXz5s34+OOPMX36dACA1+vFZZddprJoLFmyBEeOHMF5550HAOjVqxfOPvvshFiPF198EcOGDcPxxx8PABg7diwGDRqEv//976rt5s2bh6lTp6JXr14AgPPOOw8NDQ1YtmxZwvh8Pp8sVhYuXIjhw4fjuOOOS/p9st2eEGIfvFYPgBBiDXPnzkU4HEa/fv3kxyRJgt/vx2OPPYaKigrMnTsXhw4dQklJibxNNBrF559/jnvvvRdut1t+ry+++AJer1e13bx583DdddcBACKRCJ5++mnU1taqtotEIpg3bx7OOuushDHOmDEDp5xyCjZs2JCRlSPb7Qkh9oBihJACJBwO45lnnsHDDz+MKVOmqJ674IIL8OKLL+KSSy7Bq6++ipdeegljxoyRn49EIvjWt76Ft956C+eccw7Wr1+P1atXY/ny5ejRo4e83aFDh3DmmWdi06ZNGDlyJN544w00NjZizZo18Hg88nYbNmzAtddei/r6enTr1k01ljFjxmDMmDH4/PPPcfnll6f9XtluTwixBxQjhBQgr7/+Og4fPozrrrsOFRUVqucuuugizJ07F62trejZsycuvfRSuFwu1TbnnXce5s6di3POOQdz587FySefjNNPPz3hc0466STMnTsXDz74IObOnYtp06bJrhzB6NGjcfvtt+P555/XrWOybNkyhEKhBKGSjGy3J4RYD2NGCClA5s6di8mTJycIESAmRlavXo077rgDF154YYIQEdu89tpr2LdvH5577jlcdNFFup9z0UUX4ZlnnkFdXR0WLVqku53b7caFF16YNPumS5cuWQmLdNtHo9EEd5LyPiHEfFySXhg7IYTkKTfddBN2796N119/Xfc+IcR8aBkhhBQEjY2NePfdd/Hyyy9j8uTJCfcJIdZBMUIIKQjuvvtuXHzxxbjwwgtx0003JdwnhFgH3TSEEEIIsRRaRgghhBBiKRQjhBBCCLEUihFCCCGEWArFCCGEEEIshWKEEEIIIZZCMUIIIYQQS6EYIYQQQoilUIwQQgghxFL+P7mzN73TT79jAAAAAElFTkSuQmCC",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "mean_temperature_q30.plot()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 65,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 72,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "daily_consumption_naive_tz = daily_consumption.tz_localize(None)\n",
+    "daily_production_naive_tz = daily_production.tz_localize(None)\n",
+    "tempos_naive_tz = tempos.tz_localize(None)\n",
+    "daily_production_naive_tz[\"Mean_temp_q30\"] = mean_temperature_q30\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>Solaire</th>\n",
+       "      <th>Eolien</th>\n",
+       "      <th>Prévision J-1</th>\n",
+       "      <th>nette</th>\n",
+       "      <th>Production_nette</th>\n",
+       "      <th>Production_nette_q40</th>\n",
+       "      <th>Production_nette_q80</th>\n",
+       "      <th>Mean_temp_q30</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>start_date</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>2014-12-15</th>\n",
+       "      <td>0.00</td>\n",
+       "      <td>5085.00</td>\n",
+       "      <td>1643500.0</td>\n",
+       "      <td>1638415.00</td>\n",
+       "      <td>1638415.00</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>8.537848</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-12-16</th>\n",
+       "      <td>4870.89</td>\n",
+       "      <td>39909.00</td>\n",
+       "      <td>1643650.0</td>\n",
+       "      <td>1598870.11</td>\n",
+       "      <td>1598870.11</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>8.537848</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-12-17</th>\n",
+       "      <td>4609.62</td>\n",
+       "      <td>103617.00</td>\n",
+       "      <td>1584100.0</td>\n",
+       "      <td>1475873.38</td>\n",
+       "      <td>1475873.38</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>8.537848</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-12-18</th>\n",
+       "      <td>5258.76</td>\n",
+       "      <td>122720.00</td>\n",
+       "      <td>1494350.0</td>\n",
+       "      <td>1366371.24</td>\n",
+       "      <td>1366371.24</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>8.666227</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2014-12-19</th>\n",
+       "      <td>6289.70</td>\n",
+       "      <td>104494.00</td>\n",
+       "      <td>1449000.0</td>\n",
+       "      <td>1338216.30</td>\n",
+       "      <td>1338216.30</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>8.829757</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",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-08-12</th>\n",
+       "      <td>116114.62</td>\n",
+       "      <td>67128.44</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>8.601025</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-08-13</th>\n",
+       "      <td>96650.51</td>\n",
+       "      <td>36417.80</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>8.601025</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-08-14</th>\n",
+       "      <td>74620.37</td>\n",
+       "      <td>43453.81</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>8.601025</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-08-15</th>\n",
+       "      <td>101109.47</td>\n",
+       "      <td>61722.23</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>8.601025</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-08-16</th>\n",
+       "      <td>95051.52</td>\n",
+       "      <td>37254.94</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>8.601025</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>3533 rows × 8 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "              Solaire     Eolien  Prévision J-1       nette  Production_nette  \\\n",
+       "start_date                                                                      \n",
+       "2014-12-15       0.00    5085.00      1643500.0  1638415.00        1638415.00   \n",
+       "2014-12-16    4870.89   39909.00      1643650.0  1598870.11        1598870.11   \n",
+       "2014-12-17    4609.62  103617.00      1584100.0  1475873.38        1475873.38   \n",
+       "2014-12-18    5258.76  122720.00      1494350.0  1366371.24        1366371.24   \n",
+       "2014-12-19    6289.70  104494.00      1449000.0  1338216.30        1338216.30   \n",
+       "...               ...        ...            ...         ...               ...   \n",
+       "2024-08-12  116114.62   67128.44            NaN         NaN               NaN   \n",
+       "2024-08-13   96650.51   36417.80            NaN         NaN               NaN   \n",
+       "2024-08-14   74620.37   43453.81            NaN         NaN               NaN   \n",
+       "2024-08-15  101109.47   61722.23            NaN         NaN               NaN   \n",
+       "2024-08-16   95051.52   37254.94            NaN         NaN               NaN   \n",
+       "\n",
+       "            Production_nette_q40  Production_nette_q80  Mean_temp_q30  \n",
+       "start_date                                                             \n",
+       "2014-12-15           1103302.956           1476113.766       8.537848  \n",
+       "2014-12-16           1103302.956           1476113.766       8.537848  \n",
+       "2014-12-17           1103302.956           1476113.766       8.537848  \n",
+       "2014-12-18           1103302.956           1476113.766       8.666227  \n",
+       "2014-12-19           1103302.956           1476113.766       8.829757  \n",
+       "...                          ...                   ...            ...  \n",
+       "2024-08-12                   NaN                   NaN       8.601025  \n",
+       "2024-08-13                   NaN                   NaN       8.601025  \n",
+       "2024-08-14                   NaN                   NaN       8.601025  \n",
+       "2024-08-15                   NaN                   NaN       8.601025  \n",
+       "2024-08-16                   NaN                   NaN       8.601025  \n",
+       "\n",
+       "[3533 rows x 8 columns]"
+      ]
+     },
+     "execution_count": 73,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "daily_production_naive_tz"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 74,
+   "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>Prévision_J-1</th>\n",
+       "      <th>Solaire</th>\n",
+       "      <th>Eolien</th>\n",
+       "      <th>Prévision_J-1</th>\n",
+       "      <th>nette</th>\n",
+       "      <th>Production_nette</th>\n",
+       "      <th>Production_nette_q40</th>\n",
+       "      <th>Production_nette_q80</th>\n",
+       "      <th>Mean_temp_q30</th>\n",
+       "      <th>Type_de_jour_TEMPO</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>2015-09-01</th>\n",
+       "      <td>1167400.0</td>\n",
+       "      <td>21617.70</td>\n",
+       "      <td>30097.00</td>\n",
+       "      <td>1167400.0</td>\n",
+       "      <td>1115685.30</td>\n",
+       "      <td>1115685.30</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>7.644259</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2015-09-02</th>\n",
+       "      <td>1137500.0</td>\n",
+       "      <td>24844.99</td>\n",
+       "      <td>18895.00</td>\n",
+       "      <td>1137500.0</td>\n",
+       "      <td>1093760.01</td>\n",
+       "      <td>1093760.01</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>7.644259</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2015-09-03</th>\n",
+       "      <td>1127150.0</td>\n",
+       "      <td>21967.80</td>\n",
+       "      <td>24162.00</td>\n",
+       "      <td>1127150.0</td>\n",
+       "      <td>1081020.20</td>\n",
+       "      <td>1081020.20</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>7.644259</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2015-09-04</th>\n",
+       "      <td>1117350.0</td>\n",
+       "      <td>25466.61</td>\n",
+       "      <td>24709.00</td>\n",
+       "      <td>1117350.0</td>\n",
+       "      <td>1067174.39</td>\n",
+       "      <td>1067174.39</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>7.644259</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2015-09-05</th>\n",
+       "      <td>980050.0</td>\n",
+       "      <td>27009.82</td>\n",
+       "      <td>32334.00</td>\n",
+       "      <td>980050.0</td>\n",
+       "      <td>920706.18</td>\n",
+       "      <td>920706.18</td>\n",
+       "      <td>1103302.956</td>\n",
+       "      <td>1476113.766</td>\n",
+       "      <td>7.644259</td>\n",
+       "      <td>BLEU</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",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-04</th>\n",
+       "      <td>1163350.0</td>\n",
+       "      <td>62687.02</td>\n",
+       "      <td>279577.48</td>\n",
+       "      <td>1163350.0</td>\n",
+       "      <td>821085.50</td>\n",
+       "      <td>821085.50</td>\n",
+       "      <td>871976.352</td>\n",
+       "      <td>1132033.542</td>\n",
+       "      <td>8.796117</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-05</th>\n",
+       "      <td>1110187.5</td>\n",
+       "      <td>71763.38</td>\n",
+       "      <td>175006.34</td>\n",
+       "      <td>1110187.5</td>\n",
+       "      <td>863417.78</td>\n",
+       "      <td>863417.78</td>\n",
+       "      <td>871826.910</td>\n",
+       "      <td>1132033.542</td>\n",
+       "      <td>8.800151</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-06</th>\n",
+       "      <td>950700.0</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>950700.0</td>\n",
+       "      <td>950700.00</td>\n",
+       "      <td>950700.00</td>\n",
+       "      <td>871826.910</td>\n",
+       "      <td>1132033.542</td>\n",
+       "      <td>8.826117</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-07</th>\n",
+       "      <td>930150.0</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>930150.0</td>\n",
+       "      <td>930150.00</td>\n",
+       "      <td>930150.00</td>\n",
+       "      <td>871826.910</td>\n",
+       "      <td>1132033.542</td>\n",
+       "      <td>8.826117</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2024-04-08</th>\n",
+       "      <td>468475.0</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>468475.0</td>\n",
+       "      <td>468475.00</td>\n",
+       "      <td>468475.00</td>\n",
+       "      <td>871034.898</td>\n",
+       "      <td>1132033.542</td>\n",
+       "      <td>8.826117</td>\n",
+       "      <td>BLEU</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>3143 rows × 10 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "            Prévision_J-1   Solaire     Eolien  Prévision_J-1       nette  \\\n",
+       "2015-09-01      1167400.0  21617.70   30097.00      1167400.0  1115685.30   \n",
+       "2015-09-02      1137500.0  24844.99   18895.00      1137500.0  1093760.01   \n",
+       "2015-09-03      1127150.0  21967.80   24162.00      1127150.0  1081020.20   \n",
+       "2015-09-04      1117350.0  25466.61   24709.00      1117350.0  1067174.39   \n",
+       "2015-09-05       980050.0  27009.82   32334.00       980050.0   920706.18   \n",
+       "...                   ...       ...        ...            ...         ...   \n",
+       "2024-04-04      1163350.0  62687.02  279577.48      1163350.0   821085.50   \n",
+       "2024-04-05      1110187.5  71763.38  175006.34      1110187.5   863417.78   \n",
+       "2024-04-06       950700.0      0.00       0.00       950700.0   950700.00   \n",
+       "2024-04-07       930150.0      0.00       0.00       930150.0   930150.00   \n",
+       "2024-04-08       468475.0      0.00       0.00       468475.0   468475.00   \n",
+       "\n",
+       "            Production_nette  Production_nette_q40  Production_nette_q80  \\\n",
+       "2015-09-01        1115685.30           1103302.956           1476113.766   \n",
+       "2015-09-02        1093760.01           1103302.956           1476113.766   \n",
+       "2015-09-03        1081020.20           1103302.956           1476113.766   \n",
+       "2015-09-04        1067174.39           1103302.956           1476113.766   \n",
+       "2015-09-05         920706.18           1103302.956           1476113.766   \n",
+       "...                      ...                   ...                   ...   \n",
+       "2024-04-04         821085.50            871976.352           1132033.542   \n",
+       "2024-04-05         863417.78            871826.910           1132033.542   \n",
+       "2024-04-06         950700.00            871826.910           1132033.542   \n",
+       "2024-04-07         930150.00            871826.910           1132033.542   \n",
+       "2024-04-08         468475.00            871034.898           1132033.542   \n",
+       "\n",
+       "            Mean_temp_q30 Type_de_jour_TEMPO  \n",
+       "2015-09-01       7.644259               BLEU  \n",
+       "2015-09-02       7.644259               BLEU  \n",
+       "2015-09-03       7.644259               BLEU  \n",
+       "2015-09-04       7.644259               BLEU  \n",
+       "2015-09-05       7.644259               BLEU  \n",
+       "...                   ...                ...  \n",
+       "2024-04-04       8.796117               BLEU  \n",
+       "2024-04-05       8.800151               BLEU  \n",
+       "2024-04-06       8.826117               BLEU  \n",
+       "2024-04-07       8.826117               BLEU  \n",
+       "2024-04-08       8.826117               BLEU  \n",
+       "\n",
+       "[3143 rows x 10 columns]"
+      ]
+     },
+     "execution_count": 74,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "data = pd.concat([daily_consumption_naive_tz, daily_production_naive_tz, tempos_naive_tz],\n",
+    "                 axis=1).sort_index().dropna(axis=0, how=\"any\")\n",
+    "\n",
+    "data = data[~data.index.duplicated()]\n",
+    "year = data.index[0].year\n",
+    "first_september = data.index[0] + pd.DateOffset(month=9, day=1, year=year)\n",
+    "if first_september < data.index[0]:\n",
+    "    first_september += pd.DateOffset(years=1)\n",
+    "\n",
+    "last_august = data.index[-1] - pd.DateOffset(month=8, day=31)\n",
+    "data.rename(columns={\n",
+    "    \"Prévision J-1\": \"Prévision_J-1\",\n",
+    "    \"SOLAR_FORECAST_D1\":\"Solaire\",\n",
     "    \"EOLIEN_FORECAST_D1\":\"Eolien\",\n",
     "    \"tempo_type\":\"Type_de_jour_TEMPO\",\n",
     "}, inplace=True)\n",
     "\n",
-    "data[first_september: last_august]\n"
+    "data[first_september: last_august]\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 76,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2014\n"
+     ]
+    },
+    {
+     "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>col_0</th>\n",
+       "      <th>prediction_blanc</th>\n",
+       "      <th>prediction_bleu</th>\n",
+       "      <th>prediction_rouge</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Type_de_jour_TEMPO</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>BLANC</th>\n",
+       "      <td>40</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>BLEU</th>\n",
+       "      <td>4</td>\n",
+       "      <td>297</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>ROUGE</th>\n",
+       "      <td>1</td>\n",
+       "      <td>0</td>\n",
+       "      <td>21</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
+       "Type_de_jour_TEMPO                                                     \n",
+       "BLANC                             40                0                 3\n",
+       "BLEU                               4              297                 0\n",
+       "ROUGE                              1                0                21"
+      ]
+     },
+     "execution_count": 76,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
+    "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
+    "predictor.confusion_matrix(data_pred=predictions)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 77,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2016/2016\n"
+     ]
+    },
+    {
+     "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>col_0</th>\n",
+       "      <th>prediction_blanc</th>\n",
+       "      <th>prediction_bleu</th>\n",
+       "      <th>prediction_rouge</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Type_de_jour_TEMPO</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>BLANC</th>\n",
+       "      <td>17</td>\n",
+       "      <td>1</td>\n",
+       "      <td>25</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>BLEU</th>\n",
+       "      <td>10</td>\n",
+       "      <td>287</td>\n",
+       "      <td>3</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>ROUGE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>4</td>\n",
+       "      <td>18</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
+       "Type_de_jour_TEMPO                                                     \n",
+       "BLANC                             17                1                25\n",
+       "BLEU                              10              287                 3\n",
+       "ROUGE                              0                4                18"
+      ]
+     },
+     "execution_count": 77,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "first_september = second_septembre\n",
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
+    "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
+    "\n",
+    "predictor.confusion_matrix(data_pred=predictions)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 78,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2017/2018\n"
+     ]
+    },
+    {
+     "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>col_0</th>\n",
+       "      <th>prediction_blanc</th>\n",
+       "      <th>prediction_bleu</th>\n",
+       "      <th>prediction_rouge</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Type_de_jour_TEMPO</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>BLANC</th>\n",
+       "      <td>24</td>\n",
+       "      <td>2</td>\n",
+       "      <td>17</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>BLEU</th>\n",
+       "      <td>16</td>\n",
+       "      <td>282</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>ROUGE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>22</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
+       "Type_de_jour_TEMPO                                                     \n",
+       "BLANC                             24                2                17\n",
+       "BLEU                              16              282                 2\n",
+       "ROUGE                              0                0                22"
+      ]
+     },
+     "execution_count": 78,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "first_september = second_septembre\n",
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
+    "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
+    "\n",
+    "predictor.confusion_matrix(data_pred=predictions)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 79,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2018/2019\n"
+     ]
+    },
+    {
+     "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>col_0</th>\n",
+       "      <th>prediction_blanc</th>\n",
+       "      <th>prediction_bleu</th>\n",
+       "      <th>prediction_rouge</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>Type_de_jour_TEMPO</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>BLANC</th>\n",
+       "      <td>30</td>\n",
+       "      <td>0</td>\n",
+       "      <td>13</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>BLEU</th>\n",
+       "      <td>8</td>\n",
+       "      <td>292</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>ROUGE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>20</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
+       "Type_de_jour_TEMPO                                                     \n",
+       "BLANC                             30                0                13\n",
+       "BLEU                               8              292                 0\n",
+       "ROUGE                              0                2                20"
+      ]
+     },
+     "execution_count": 79,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "first_september = second_septembre\n",
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
+    "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
+    "\n",
+    "predictor.confusion_matrix(data_pred=predictions)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 113,
+   "execution_count": 80,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2019/2020\n"
+     ]
+    },
     {
      "data": {
       "text/html": [
@@ -834,21 +2085,21 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>BLANC</th>\n",
-       "      <td>41</td>\n",
-       "      <td>0</td>\n",
-       "      <td>2</td>\n",
+       "      <td>28</td>\n",
+       "      <td>9</td>\n",
+       "      <td>10</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>BLEU</th>\n",
-       "      <td>4</td>\n",
-       "      <td>297</td>\n",
-       "      <td>0</td>\n",
+       "      <td>8</td>\n",
+       "      <td>292</td>\n",
+       "      <td>1</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ROUGE</th>\n",
        "      <td>1</td>\n",
        "      <td>0</td>\n",
-       "      <td>21</td>\n",
+       "      <td>17</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -857,38 +2108,39 @@
       "text/plain": [
        "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
        "Type_de_jour_TEMPO                                                     \n",
-       "BLANC                             41                0                 2\n",
-       "BLEU                               4              297                 0\n",
-       "ROUGE                              1                0                21"
+       "BLANC                             28                9                10\n",
+       "BLEU                               8              292                 1\n",
+       "ROUGE                              1                0                17"
       ]
      },
-     "execution_count": 113,
+     "execution_count": 80,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
+    "first_september = second_septembre\n",
     "second_septembre = first_september + pd.DateOffset(years=1)\n",
     "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
     "predictor = TempoPredictor(data_first_year)\n",
     "predictions = predictor.predict()\n",
-    "predictor.confusion_matrix(data_pred=predictions)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Now using Forecasted production\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
     "\n",
-    "The RTE classification is done the day before, based on the forecasted production. We will use the same data to classify the days."
+    "predictor.confusion_matrix(data_pred=predictions)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 54,
+   "execution_count": 81,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2020/2021\n"
+     ]
+    },
     {
      "data": {
       "text/html": [
@@ -909,82 +2161,75 @@
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>FORECAST_D1_SOLAR</th>\n",
-       "      <th>FORECAST_D1_EOLIEN</th>\n",
+       "      <th>col_0</th>\n",
+       "      <th>prediction_blanc</th>\n",
+       "      <th>prediction_bleu</th>\n",
+       "      <th>prediction_rouge</th>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>start_date</th>\n",
+       "      <th>Type_de_jour_TEMPO</th>\n",
+       "      <th></th>\n",
        "      <th></th>\n",
        "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>2022-02-01</th>\n",
-       "      <td>0.00</td>\n",
-       "      <td>53922.45</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2022-02-02</th>\n",
-       "      <td>25611.40</td>\n",
-       "      <td>101260.82</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2022-02-03</th>\n",
-       "      <td>26034.16</td>\n",
-       "      <td>73235.63</td>\n",
+       "      <th>BLANC</th>\n",
+       "      <td>37</td>\n",
+       "      <td>1</td>\n",
+       "      <td>5</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2022-02-04</th>\n",
-       "      <td>19710.59</td>\n",
-       "      <td>159568.08</td>\n",
+       "      <th>BLEU</th>\n",
+       "      <td>7</td>\n",
+       "      <td>293</td>\n",
+       "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2022-02-05</th>\n",
-       "      <td>30702.25</td>\n",
-       "      <td>175855.14</td>\n",
+       "      <th>ROUGE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>22</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "text/plain": [
-       "            FORECAST_D1_SOLAR  FORECAST_D1_EOLIEN\n",
-       "start_date                                       \n",
-       "2022-02-01               0.00            53922.45\n",
-       "2022-02-02           25611.40           101260.82\n",
-       "2022-02-03           26034.16            73235.63\n",
-       "2022-02-04           19710.59           159568.08\n",
-       "2022-02-05           30702.25           175855.14"
+       "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
+       "Type_de_jour_TEMPO                                                     \n",
+       "BLANC                             37                1                 5\n",
+       "BLEU                               7              293                 0\n",
+       "ROUGE                              0                0                22"
       ]
      },
-     "execution_count": 54,
+     "execution_count": 81,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "data_forecast_file = ROOT_DIR / \"data/silver/forecasted_production.csv\"\n",
-    "data_forecast = pd.read_csv(data_forecast_file, index_col=0)\n",
-    "data_forecast.index = pd.to_datetime(data_forecast.index, utc=True).tz_localize(None)  # type: ignore\n",
-    "# resample to daily from 6am to 6am\n",
-    "daily_forecast = data_forecast.resample(\"D\", origin=\"06:00:00\").sum()\n",
+    "first_september = second_septembre\n",
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
+    "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
     "\n",
-    "daily_forecast.index = daily_forecast.index - pd.Timedelta(hours=6)  # type: ignore\n",
-    "daily_forecast.head()"
+    "predictor.confusion_matrix(data_pred=predictions)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 56,
+   "execution_count": 82,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "2022-09-01 00:00:00 2023-08-31 00:00:00\n"
+      "Analyse de l'année 2021/2022\n"
      ]
     },
     {
@@ -1007,137 +2252,168 @@
        "<table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
        "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>Prévision_J-1</th>\n",
+       "      <th>col_0</th>\n",
+       "      <th>prediction_blanc</th>\n",
+       "      <th>prediction_bleu</th>\n",
+       "      <th>prediction_rouge</th>\n",
+       "    </tr>\n",
+       "    <tr>\n",
        "      <th>Type_de_jour_TEMPO</th>\n",
-       "      <th>Solaire</th>\n",
-       "      <th>Eolien</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>2022-09-01</th>\n",
-       "      <td>1106650.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>68375.04</td>\n",
-       "      <td>42995.35</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2022-09-02</th>\n",
-       "      <td>1081925.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>50872.25</td>\n",
-       "      <td>43115.24</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2022-09-03</th>\n",
-       "      <td>945125.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>56447.35</td>\n",
-       "      <td>46623.69</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2022-09-04</th>\n",
-       "      <td>903875.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>71938.19</td>\n",
-       "      <td>66545.15</td>\n",
+       "      <th>BLANC</th>\n",
+       "      <td>32</td>\n",
+       "      <td>1</td>\n",
+       "      <td>9</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2022-09-05</th>\n",
-       "      <td>1075950.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>68783.20</td>\n",
-       "      <td>66041.44</td>\n",
+       "      <th>BLEU</th>\n",
+       "      <td>84</td>\n",
+       "      <td>217</td>\n",
+       "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
-       "      <td>...</td>\n",
+       "      <th>ROUGE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>22</td>\n",
        "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2023-08-27</th>\n",
-       "      <td>878250.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>69547.21</td>\n",
-       "      <td>99322.30</td>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
+       "Type_de_jour_TEMPO                                                     \n",
+       "BLANC                             32                1                 9\n",
+       "BLEU                              84              217                 0\n",
+       "ROUGE                              0                0                22"
+      ]
+     },
+     "execution_count": 82,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "first_september = second_septembre\n",
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
+    "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
+    "\n",
+    "predictor.confusion_matrix(data_pred=predictions)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 83,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2022/2023\n"
+     ]
+    },
+    {
+     "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>col_0</th>\n",
+       "      <th>prediction_blanc</th>\n",
+       "      <th>prediction_bleu</th>\n",
+       "      <th>prediction_rouge</th>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2023-08-28</th>\n",
-       "      <td>987675.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>70295.54</td>\n",
-       "      <td>84276.31</td>\n",
+       "      <th>Type_de_jour_TEMPO</th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
+       "      <th></th>\n",
        "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
        "    <tr>\n",
-       "      <th>2023-08-29</th>\n",
-       "      <td>1013650.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>80138.51</td>\n",
-       "      <td>58030.63</td>\n",
+       "      <th>BLANC</th>\n",
+       "      <td>31</td>\n",
+       "      <td>2</td>\n",
+       "      <td>10</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2023-08-30</th>\n",
-       "      <td>1031375.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>86203.59</td>\n",
-       "      <td>93396.01</td>\n",
+       "      <th>BLEU</th>\n",
+       "      <td>5</td>\n",
+       "      <td>295</td>\n",
+       "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>2023-08-31</th>\n",
-       "      <td>1025125.0</td>\n",
-       "      <td>BLEU</td>\n",
-       "      <td>71614.36</td>\n",
-       "      <td>110535.58</td>\n",
+       "      <th>ROUGE</th>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>22</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
-       "<p>365 rows × 4 columns</p>\n",
        "</div>"
       ],
       "text/plain": [
-       "            Prévision_J-1 Type_de_jour_TEMPO   Solaire     Eolien\n",
-       "2022-09-01      1106650.0               BLEU  68375.04   42995.35\n",
-       "2022-09-02      1081925.0               BLEU  50872.25   43115.24\n",
-       "2022-09-03       945125.0               BLEU  56447.35   46623.69\n",
-       "2022-09-04       903875.0               BLEU  71938.19   66545.15\n",
-       "2022-09-05      1075950.0               BLEU  68783.20   66041.44\n",
-       "...                   ...                ...       ...        ...\n",
-       "2023-08-27       878250.0               BLEU  69547.21   99322.30\n",
-       "2023-08-28       987675.0               BLEU  70295.54   84276.31\n",
-       "2023-08-29      1013650.0               BLEU  80138.51   58030.63\n",
-       "2023-08-30      1031375.0               BLEU  86203.59   93396.01\n",
-       "2023-08-31      1025125.0               BLEU  71614.36  110535.58\n",
-       "\n",
-       "[365 rows x 4 columns]"
+       "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
+       "Type_de_jour_TEMPO                                                     \n",
+       "BLANC                             31                2                10\n",
+       "BLEU                               5              295                 0\n",
+       "ROUGE                              0                0                22"
       ]
      },
-     "execution_count": 56,
+     "execution_count": 83,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "used_cols = [\"Prévision_J-1\", \"Type_de_jour_TEMPO\"]\n",
-    "all_data = pd.concat([data_agg[used_cols], daily_forecast], axis=1).dropna()\n",
-    "all_data.rename(\n",
-    "    columns={\"FORECAST_D1_SOLAR\": \"Solaire\", \"FORECAST_D1_EOLIEN\": \"Eolien\"},\n",
-    "    inplace=True,\n",
-    ")\n",
-    "period_start = all_data.index[0] + pd.DateOffset(day=1, month=9)\n",
-    "period_end = period_start + pd.DateOffset(years=1, days=-1)\n",
-    "print(period_start, period_end)\n",
-    "all_data = all_data.loc[period_start:period_end]\n",
-    "all_data"
+    "first_september = second_septembre\n",
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
+    "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
+    "\n",
+    "predictor.confusion_matrix(data_pred=predictions)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 104,
+   "execution_count": 84,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Analyse de l'année 2023/2024\n"
+     ]
+    },
     {
      "data": {
       "text/html": [
@@ -1173,20 +2449,20 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>BLANC</th>\n",
-       "      <td>34</td>\n",
-       "      <td>1</td>\n",
-       "      <td>8</td>\n",
+       "      <td>29</td>\n",
+       "      <td>2</td>\n",
+       "      <td>4</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>BLEU</th>\n",
-       "      <td>11</td>\n",
-       "      <td>289</td>\n",
+       "      <td>7</td>\n",
+       "      <td>157</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>ROUGE</th>\n",
-       "      <td>1</td>\n",
        "      <td>0</td>\n",
+       "      <td>1</td>\n",
        "      <td>21</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
@@ -1196,32 +2472,30 @@
       "text/plain": [
        "col_0               prediction_blanc  prediction_bleu  prediction_rouge\n",
        "Type_de_jour_TEMPO                                                     \n",
-       "BLANC                             34                1                 8\n",
-       "BLEU                              11              289                 0\n",
-       "ROUGE                              1                0                21"
+       "BLANC                             29                2                 4\n",
+       "BLEU                               7              157                 0\n",
+       "ROUGE                              0                1                21"
       ]
      },
-     "execution_count": 104,
+     "execution_count": 84,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
-    "predictor = TempoPredictor(all_data)\n",
+    "first_september = second_septembre\n",
+    "second_septembre = first_september + pd.DateOffset(years=1)\n",
+    "data_first_year = data[first_september: second_septembre - pd.DateOffset(days=1) ].copy()\n",
+    "predictor = TempoPredictor(data_first_year)\n",
     "predictions = predictor.predict()\n",
+    "print(f\"Analyse de l'année {first_september.year}/{second_septembre.year}\")\n",
+    "\n",
     "predictor.confusion_matrix(data_pred=predictions)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 105,
+   "execution_count": 86,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1231,27 +2505,27 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 106,
+   "execution_count": 87,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "2022-09-01    0\n",
-       "2022-09-02    0\n",
-       "2022-09-03    0\n",
-       "2022-09-04    0\n",
-       "2022-09-05    0\n",
+       "2023-09-01    0\n",
+       "2023-09-02    0\n",
+       "2023-09-03    0\n",
+       "2023-09-04    0\n",
+       "2023-09-05    0\n",
        "             ..\n",
-       "2023-08-27    0\n",
-       "2023-08-28    0\n",
-       "2023-08-29    0\n",
-       "2023-08-30    0\n",
-       "2023-08-31    0\n",
-       "Freq: D, Name: ROUGE, Length: 365, dtype: int64"
+       "2024-04-04    0\n",
+       "2024-04-05    0\n",
+       "2024-04-06    0\n",
+       "2024-04-07    0\n",
+       "2024-04-08    0\n",
+       "Freq: D, Name: ROUGE, Length: 221, dtype: int64"
       ]
      },
-     "execution_count": 106,
+     "execution_count": 87,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1262,7 +2536,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 107,
+   "execution_count": 88,
    "metadata": {},
    "outputs": [
     {
@@ -1271,13 +2545,13 @@
        "Text(0.5, 1.0, 'Production normée et seuils de déclenchement des couleurs Tempo')"
       ]
      },
-     "execution_count": 107,
+     "execution_count": 88,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 1500x500 with 1 Axes>"
       ]
@@ -1316,7 +2590,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 109,
+   "execution_count": 90,
    "metadata": {},
    "outputs": [
     {
@@ -1325,13 +2599,13 @@
        "Text(0.5, 1.0, 'Production normée et seuils de déclenchement des couleurs Tempo')"
       ]
      },
-     "execution_count": 109,
+     "execution_count": 90,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 1500x500 with 1 Axes>"
       ]
@@ -1400,7 +2674,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
+   "version": "3.12.3"
   }
  },
  "nbformat": 4,
diff --git a/src/energy_forecast/meteo.py b/src/energy_forecast/meteo.py
index ad54fae..0965df0 100644
--- a/src/energy_forecast/meteo.py
+++ b/src/energy_forecast/meteo.py
@@ -570,6 +570,79 @@ def instant_flux_from_cumul(df_unstacked):
         df_instant_flux.index -= pd.Timedelta("1h")
     return df_instant_flux
 
+
+def download_observations(url, filename):
+    """Download the observations from the url and save it in the filename.
+
+    Parameters
+    ----------
+    url : str
+        the url to download the data.
+    filename : str
+        the filename to save the data.
+    """
+    response = requests.get(url)
+    response.raise_for_status()
+    with open(filename, "wb") as f:
+        f.write(response.content)
+
+def download_observations_all_departments(cache_duration="12h",
+                                          file_type="latest-2023-2024_RR-T-Vent",
+                                          verbose=False):
+    """Download the temperature for each department of France."""
+    url_template = "https://object.files.data.gouv.fr/meteofrance/data/synchro_ftp/BASE/QUOT/Q_{DEP_ID:0>2}_{file_type}.csv.gz"
+    list_files = []
+    download_root = ROOT_DIR / "data/bronze/observations"
+    download_root.mkdir(parents=True, exist_ok=True)
+    now = pd.Timestamp("now")
+    for dep_id in range(1, 96):
+        url = url_template.format(DEP_ID=dep_id, file_type=file_type)
+        filename = download_root / "Q_{DEP_ID:0>2}_{file_type}.csv.gz".format(DEP_ID=dep_id, file_type=file_type)
+        try:
+            if filename.exists():
+                modification_time = pd.Timestamp(filename.stat().st_mtime, unit="s")
+                if now - modification_time < pd.Timedelta(cache_duration):
+                    list_files.append(filename)
+                    continue
+            if verbose:
+                logger.info(f"Downloading {url} to {filename}")
+            download_observations(url, filename)
+            list_files.append(filename)
+        except requests.exceptions.HTTPError:
+            logger.warning(f"Could not download {url}")
+    return list_files
+
+def aggregates_observations(list_files, cut_before="2022-01-01", verbose=False):
+    """Aggregate the observations for each department of France.
+
+    Parameters
+    ----------
+    list_files : list[str]
+        the list of the files to aggregate.
+
+    Returns
+    -------
+    pd.DataFrame
+        the DataFrame containing the observations for each department.
+    """
+    all_deps = []
+    for i, file_name in enumerate(list_files):
+        if verbose:
+            logger.info(f"Reading {file_name} ({i+1}/{len(list_files)})")
+        tem_df = pd.read_csv(file_name, sep=";", usecols=["AAAAMMJJ", "TM"], compression="gzip", date_format="%Y%m%d", parse_dates=["AAAAMMJJ"])
+        
+        tem_df = (tem_df.set_index("AAAAMMJJ")
+                  .dropna(subset=["TM"])
+                  .sort_index()
+        )
+        tem_df = (tem_df
+                  .groupby(tem_df.index).mean()
+                  
+        )[cut_before:]
+        all_deps.append(tem_df)
+        
+    return pd.concat(all_deps, axis=1).mean(axis=1)
+        
 if __name__ == "__main__":
     logger.info("Fetching data for today")
     warm_cache(logger)
diff --git a/src/energy_forecast/tempo_rte.py b/src/energy_forecast/tempo_rte.py
index e632bb7..ce0ed36 100644
--- a/src/energy_forecast/tempo_rte.py
+++ b/src/energy_forecast/tempo_rte.py
@@ -1,4 +1,5 @@
 
+import logging
 import numpy as np
 import pandas as pd
 import requests
@@ -6,6 +7,7 @@
 
 from energy_forecast.rte_api_core import RTEAPROAuth2
 
+logger = logging.getLogger(__name__)
 
 class TempoSignalAPI(RTEAPROAuth2):
     """Class to interact with the Tempo Signal API from RTE.
@@ -70,6 +72,9 @@ class TempoPredictor:
     production_nette_f = "Production_nette"
     known_jour_tempo_f = "Type_de_jour_TEMPO"
     production_normed_f = "Production_norm"
+    production_nette_q40_f = "Production_nette_q40"
+    production_nette_q80_f = "Production_nette_q80"
+    mean_temp_q30_f = "Mean_temp_q30"
 
     def __init__(self, data: pd.DataFrame):
         """Initialize the class.
@@ -83,20 +88,32 @@ def __init__(self, data: pd.DataFrame):
             - Solaire : the solar production (forecasted)
             - Eolien : the wind production (forecasted)
             - Type_de_jour_TEMPO : the known tempo signal (used for the training and stock calculation)
+            Optional columns:
+            - Production_nette : the net production (forecasted)
+            - Production_nette_q40 : the 40th percentile of the net production
+            - Production_nette_q80 : the 80th percentile of the net production
+            - Mean_temp_q30 : the 30th percentile of the temperature
 
         Note
         ----
         The data should start from the 1st of September, and last one year.
         """
         self.data: DataFrame = data
-        self.data[self.production_nette_f] = self.data[self.prevision_consumtion_f] - (self.data[self.production_eolien_f] + self.data[self.production_solar_f])
-
-        self.q80 = data[self.production_nette_f].quantile(0.8)  # TODO : should instead be computed on rolling window of 1 year, using data from previous season
-        self.q40 = data[self.production_nette_f].quantile(0.4)  # TODO : should instead be computed on rolling window of 1 year, using data from previous season
-        self.qtemp30 = 9  # TODO : should instead be computed on rolling window of 1 year, using data from previous season of the 30th percentile of the temperature
+        if self.production_nette_f not in self.data.columns:
+            self.data[self.production_nette_f] = self.data[self.prevision_consumtion_f] - (self.data[self.production_eolien_f] + self.data[self.production_solar_f])
+        if self.production_nette_q40_f not in self.data.columns:
+            logger.warning("The 40th percentile of the net production is not in the data. It will be computed.")
+            self.data[self.production_nette_q40_f] = self.data[self.production_nette_f].quantile(0.4)
+        if self.production_nette_q80_f not in self.data.columns:
+            self.data[self.production_nette_q80_f] = self.data[self.production_nette_f].quantile(0.8)
+        if self.mean_temp_q30_f not in self.data.columns:
+            logger.warning("The 30th percentile of the temperature is not in the data. It will assumed 9°C.")
+            self.data[self.mean_temp_q30_f] = 9
         self.gamma = -0.1176
         self.kappa = 8.3042
-        self.data[self.production_normed_f] = (self.data[self.production_nette_f] - self.q40) / ( (self.q80 - self.q40) * np.exp(self.gamma * (self.kappa - self.qtemp30)))
+        self.data[self.production_normed_f] = (self.data[self.production_nette_f] - self.data[self.production_nette_q40_f]) / (
+            (self.data[self.production_nette_q80_f] - self.data[self.production_nette_q40_f]) * np.exp(self.gamma * (self.kappa - self.data[self.mean_temp_q30_f]))
+            )
         self.categories: DataFrame = pd.get_dummies(data[self.known_jour_tempo_f]).astype(int)
         self.start_BLANC = 43
         self.start_ROUGE = 22
@@ -115,7 +132,7 @@ def __init__(self, data: pd.DataFrame):
 
         self.data["seuil_rouge"] = (
             self.A_rouge
-            + self.B_rouge * ( data["jour_tempo"] -1 )
+            + self.B_rouge * ( data["jour_tempo"] -1 )  # Not sur if start from 0 or 1
             + self.C_rouge * data["stock_rouge"]
         )
         self.data.loc[ self.data["stock_rouge"] == 0  ,"seuil_rouge"] = 2
@@ -183,4 +200,4 @@ def confusion_matrix(self, data_true: pd.Series | None=None, data_pred: pd.Serie
             data_pred = data_pred.idxmax(axis=1)
 
         confusion_matrix = pd.crosstab(data_true, data_pred)
-        return confusion_matrix
\ No newline at end of file
+        return confusion_matrix