From 15adab6a44cf366371f28ea4695a2997ba47e4a9 Mon Sep 17 00:00:00 2001
From: 504e2a253167de0c9e327e795bb40e5b
 <504e2a253167de0c9e327e795bb40e5b@app-learninglab.inria.fr>
Date: Tue, 10 Dec 2024 14:03:53 +0000
Subject: [PATCH] Pas fini

---
 module3/exo3/exercice_fr.ipynb | 746 +++++++++++++++++----------------
 1 file changed, 386 insertions(+), 360 deletions(-)

diff --git a/module3/exo3/exercice_fr.ipynb b/module3/exo3/exercice_fr.ipynb
index 3900d1d..7da01cf 100644
--- a/module3/exo3/exercice_fr.ipynb
+++ b/module3/exo3/exercice_fr.ipynb
@@ -9,16 +9,24 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
     "%matplotlib inline\n",
     "import matplotlib.pyplot as plt\n",
     "import pandas as pd\n",
-    "# import isoweek\n",
     "import os\n",
-    "from urllib.request import urlretrieve"
+    "from urllib.request import urlretrieve\n",
+    "import datetime\n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Importation et formatage des données"
    ]
   },
   {
@@ -30,14 +38,14 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "File downloaded and saved as data_keeling.csv\n"
+      "File data_keeling.csv found at /home/jovyan/work/module3/exo3/data_keeling.csv\n"
      ]
     }
    ],
@@ -65,7 +73,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
@@ -471,7 +479,7 @@
        "[3403 rows x 2 columns]"
       ]
      },
-     "execution_count": 18,
+     "execution_count": 3,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -491,7 +499,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
@@ -530,7 +538,7 @@
        "Index: []"
       ]
      },
-     "execution_count": 19,
+     "execution_count": 4,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -548,363 +556,85 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 5,
    "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>Concentration en CO2 (ppm)</th>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>Date</th>\n",
-       "      <th></th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>1958-03-29</th>\n",
-       "      <td>316.19</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-04-05</th>\n",
-       "      <td>317.31</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-04-12</th>\n",
-       "      <td>317.69</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-04-19</th>\n",
-       "      <td>317.58</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-04-26</th>\n",
-       "      <td>316.48</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-05-03</th>\n",
-       "      <td>316.95</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-05-17</th>\n",
-       "      <td>317.56</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-05-24</th>\n",
-       "      <td>317.99</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-07-05</th>\n",
-       "      <td>315.85</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-07-12</th>\n",
-       "      <td>315.85</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-07-19</th>\n",
-       "      <td>315.46</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-07-26</th>\n",
-       "      <td>315.59</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-08-02</th>\n",
-       "      <td>315.64</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-08-09</th>\n",
-       "      <td>315.10</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-08-16</th>\n",
-       "      <td>315.09</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-08-30</th>\n",
-       "      <td>314.14</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-09-06</th>\n",
-       "      <td>313.54</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-11-08</th>\n",
-       "      <td>313.05</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-11-15</th>\n",
-       "      <td>313.26</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-11-22</th>\n",
-       "      <td>313.57</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-11-29</th>\n",
-       "      <td>314.01</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-12-06</th>\n",
-       "      <td>314.56</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-12-13</th>\n",
-       "      <td>314.41</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-12-20</th>\n",
-       "      <td>314.77</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1958-12-27</th>\n",
-       "      <td>315.21</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1959-01-03</th>\n",
-       "      <td>315.24</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1959-01-10</th>\n",
-       "      <td>315.50</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1959-01-17</th>\n",
-       "      <td>315.69</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1959-01-24</th>\n",
-       "      <td>315.86</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1959-01-31</th>\n",
-       "      <td>315.42</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>...</th>\n",
-       "      <td>...</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-04-20</th>\n",
-       "      <td>426.91</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-04-27</th>\n",
-       "      <td>427.13</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-05-04</th>\n",
-       "      <td>426.51</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-05-11</th>\n",
-       "      <td>427.20</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-05-18</th>\n",
-       "      <td>426.26</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-05-25</th>\n",
-       "      <td>426.68</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-06-01</th>\n",
-       "      <td>426.78</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-06-08</th>\n",
-       "      <td>427.01</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-06-15</th>\n",
-       "      <td>427.10</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-06-22</th>\n",
-       "      <td>426.54</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-06-29</th>\n",
-       "      <td>425.41</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-07-06</th>\n",
-       "      <td>425.73</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-07-13</th>\n",
-       "      <td>426.10</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-07-20</th>\n",
-       "      <td>424.36</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-07-27</th>\n",
-       "      <td>424.72</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-08-03</th>\n",
-       "      <td>424.42</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-08-10</th>\n",
-       "      <td>422.50</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-08-17</th>\n",
-       "      <td>422.80</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-08-24</th>\n",
-       "      <td>421.45</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-08-31</th>\n",
-       "      <td>421.57</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-09-07</th>\n",
-       "      <td>421.81</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-09-14</th>\n",
-       "      <td>421.39</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-09-21</th>\n",
-       "      <td>421.77</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-09-28</th>\n",
-       "      <td>421.51</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-10-05</th>\n",
-       "      <td>421.86</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-10-12</th>\n",
-       "      <td>422.13</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-10-19</th>\n",
-       "      <td>422.16</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-10-26</th>\n",
-       "      <td>422.36</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-11-02</th>\n",
-       "      <td>423.15</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>2024-11-09</th>\n",
-       "      <td>423.18</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>3403 rows × 1 columns</p>\n",
-       "</div>"
-      ],
       "text/plain": [
-       "            Concentration en CO2 (ppm)\n",
-       "Date                                  \n",
-       "1958-03-29                      316.19\n",
-       "1958-04-05                      317.31\n",
-       "1958-04-12                      317.69\n",
-       "1958-04-19                      317.58\n",
-       "1958-04-26                      316.48\n",
-       "1958-05-03                      316.95\n",
-       "1958-05-17                      317.56\n",
-       "1958-05-24                      317.99\n",
-       "1958-07-05                      315.85\n",
-       "1958-07-12                      315.85\n",
-       "1958-07-19                      315.46\n",
-       "1958-07-26                      315.59\n",
-       "1958-08-02                      315.64\n",
-       "1958-08-09                      315.10\n",
-       "1958-08-16                      315.09\n",
-       "1958-08-30                      314.14\n",
-       "1958-09-06                      313.54\n",
-       "1958-11-08                      313.05\n",
-       "1958-11-15                      313.26\n",
-       "1958-11-22                      313.57\n",
-       "1958-11-29                      314.01\n",
-       "1958-12-06                      314.56\n",
-       "1958-12-13                      314.41\n",
-       "1958-12-20                      314.77\n",
-       "1958-12-27                      315.21\n",
-       "1959-01-03                      315.24\n",
-       "1959-01-10                      315.50\n",
-       "1959-01-17                      315.69\n",
-       "1959-01-24                      315.86\n",
-       "1959-01-31                      315.42\n",
-       "...                                ...\n",
-       "2024-04-20                      426.91\n",
-       "2024-04-27                      427.13\n",
-       "2024-05-04                      426.51\n",
-       "2024-05-11                      427.20\n",
-       "2024-05-18                      426.26\n",
-       "2024-05-25                      426.68\n",
-       "2024-06-01                      426.78\n",
-       "2024-06-08                      427.01\n",
-       "2024-06-15                      427.10\n",
-       "2024-06-22                      426.54\n",
-       "2024-06-29                      425.41\n",
-       "2024-07-06                      425.73\n",
-       "2024-07-13                      426.10\n",
-       "2024-07-20                      424.36\n",
-       "2024-07-27                      424.72\n",
-       "2024-08-03                      424.42\n",
-       "2024-08-10                      422.50\n",
-       "2024-08-17                      422.80\n",
-       "2024-08-24                      421.45\n",
-       "2024-08-31                      421.57\n",
-       "2024-09-07                      421.81\n",
-       "2024-09-14                      421.39\n",
-       "2024-09-21                      421.77\n",
-       "2024-09-28                      421.51\n",
-       "2024-10-05                      421.86\n",
-       "2024-10-12                      422.13\n",
-       "2024-10-19                      422.16\n",
-       "2024-10-26                      422.36\n",
-       "2024-11-02                      423.15\n",
-       "2024-11-09                      423.18\n",
-       "\n",
-       "[3403 rows x 1 columns]"
+       "Date\n",
+       "1958-03-29    316.19\n",
+       "1958-04-05    317.31\n",
+       "1958-04-12    317.69\n",
+       "1958-04-19    317.58\n",
+       "1958-04-26    316.48\n",
+       "1958-05-03    316.95\n",
+       "1958-05-17    317.56\n",
+       "1958-05-24    317.99\n",
+       "1958-07-05    315.85\n",
+       "1958-07-12    315.85\n",
+       "1958-07-19    315.46\n",
+       "1958-07-26    315.59\n",
+       "1958-08-02    315.64\n",
+       "1958-08-09    315.10\n",
+       "1958-08-16    315.09\n",
+       "1958-08-30    314.14\n",
+       "1958-09-06    313.54\n",
+       "1958-11-08    313.05\n",
+       "1958-11-15    313.26\n",
+       "1958-11-22    313.57\n",
+       "1958-11-29    314.01\n",
+       "1958-12-06    314.56\n",
+       "1958-12-13    314.41\n",
+       "1958-12-20    314.77\n",
+       "1958-12-27    315.21\n",
+       "1959-01-03    315.24\n",
+       "1959-01-10    315.50\n",
+       "1959-01-17    315.69\n",
+       "1959-01-24    315.86\n",
+       "1959-01-31    315.42\n",
+       "               ...  \n",
+       "2024-04-20    426.91\n",
+       "2024-04-27    427.13\n",
+       "2024-05-04    426.51\n",
+       "2024-05-11    427.20\n",
+       "2024-05-18    426.26\n",
+       "2024-05-25    426.68\n",
+       "2024-06-01    426.78\n",
+       "2024-06-08    427.01\n",
+       "2024-06-15    427.10\n",
+       "2024-06-22    426.54\n",
+       "2024-06-29    425.41\n",
+       "2024-07-06    425.73\n",
+       "2024-07-13    426.10\n",
+       "2024-07-20    424.36\n",
+       "2024-07-27    424.72\n",
+       "2024-08-03    424.42\n",
+       "2024-08-10    422.50\n",
+       "2024-08-17    422.80\n",
+       "2024-08-24    421.45\n",
+       "2024-08-31    421.57\n",
+       "2024-09-07    421.81\n",
+       "2024-09-14    421.39\n",
+       "2024-09-21    421.77\n",
+       "2024-09-28    421.51\n",
+       "2024-10-05    421.86\n",
+       "2024-10-12    422.13\n",
+       "2024-10-19    422.16\n",
+       "2024-10-26    422.36\n",
+       "2024-11-02    423.15\n",
+       "2024-11-09    423.18\n",
+       "Length: 3403, dtype: float64"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 5,
      "metadata": {},
      "output_type": "execute_result"
     }
    ],
    "source": [
     "raw_data[\"Date\"] = pd.to_datetime(raw_data[\"Date\"])\n",
-    "data = raw_data.set_index(\"Date\").sort_index()\n",
+    "data = pd.Series(data = raw_data[\"Concentration en CO2 (ppm)\"].tolist(), index = raw_data[\"Date\"])\n",
     "data"
    ]
   },
@@ -917,7 +647,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 29,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -971,27 +701,185 @@
    "source": [
     "Quelques données sont manquantes, mais leur nombre est largement inférieur au nombre de données, et sont éparses après les années 1960. Nous pouvons donc considérer l'erreur induite par le manque de données négligeable.\n",
     "\n",
+    "Néanmoins, nous pouvons combler ces données en utilisant des moyennes glissantes pour aider à la visualisation, tout en gardant en mémoire les dates avec des données interpolées, afin de les exclure de l'analyse quantitative. Cela est effectué à l'aide du tableau `interpolated_marks` :"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "1958-03-29    0\n",
+       "1958-04-05    0\n",
+       "1958-04-12    0\n",
+       "1958-04-19    0\n",
+       "1958-04-26    0\n",
+       "1958-05-03    0\n",
+       "1958-05-10    1\n",
+       "1958-05-17    0\n",
+       "1958-05-24    0\n",
+       "1958-05-31    1\n",
+       "1958-06-07    1\n",
+       "1958-06-14    1\n",
+       "1958-06-21    1\n",
+       "1958-06-28    1\n",
+       "1958-07-05    0\n",
+       "1958-07-12    0\n",
+       "1958-07-19    0\n",
+       "1958-07-26    0\n",
+       "1958-08-02    0\n",
+       "1958-08-09    0\n",
+       "1958-08-16    0\n",
+       "1958-08-23    1\n",
+       "1958-08-30    0\n",
+       "1958-09-06    0\n",
+       "1958-09-13    1\n",
+       "1958-09-20    1\n",
+       "1958-09-27    1\n",
+       "1958-10-04    1\n",
+       "1958-10-11    1\n",
+       "1958-10-18    1\n",
+       "             ..\n",
+       "2024-04-20    0\n",
+       "2024-04-27    0\n",
+       "2024-05-04    0\n",
+       "2024-05-11    0\n",
+       "2024-05-18    0\n",
+       "2024-05-25    0\n",
+       "2024-06-01    0\n",
+       "2024-06-08    0\n",
+       "2024-06-15    0\n",
+       "2024-06-22    0\n",
+       "2024-06-29    0\n",
+       "2024-07-06    0\n",
+       "2024-07-13    0\n",
+       "2024-07-20    0\n",
+       "2024-07-27    0\n",
+       "2024-08-03    0\n",
+       "2024-08-10    0\n",
+       "2024-08-17    0\n",
+       "2024-08-24    0\n",
+       "2024-08-31    0\n",
+       "2024-09-07    0\n",
+       "2024-09-14    0\n",
+       "2024-09-21    0\n",
+       "2024-09-28    0\n",
+       "2024-10-05    0\n",
+       "2024-10-12    0\n",
+       "2024-10-19    0\n",
+       "2024-10-26    0\n",
+       "2024-11-02    0\n",
+       "2024-11-09    0\n",
+       "Freq: 7D, Length: 3477, dtype: int64"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "full_index = pd.date_range(start=data.index[0], end=data.index[-1], freq='7D')\n",
+    "full_data = data.reindex(full_index)\n",
+    "while full_data.isna().any():\n",
+    "    rolling_mean = full_data.rolling(window=5, min_periods=3).mean()\n",
+    "    full_data[full_data.isna()] = rolling_mean[full_data.isna()]\n",
+    "interpolated_marks = pd.Series(data=np.where(data.reindex(full_index).isna(), 1, 0), index=full_index)\n",
+    "interpolated_marks"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Interprétation des données"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
     "Commençons par représenter l'évolution de la concentration en CO$_2$ depuis 1958."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.axes._subplots.AxesSubplot at 0x7f82cf75a828>"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "full_data.plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "On observe une oscillation rapide autour d'une évolution plus lente. Nous pouvons donc modéliser l'évolution temporelle comme :\n",
+    "\n",
+    "$$C(t)=f(t)+A\\cos\\Big(\\frac{2\\pi}{T}(t-t_0)\\Big)\\ ,$$\n",
+    "\n",
+    "où $C(t)$ est la concentration en CO$_2$, $t$ est le temps, $f(t)$ est une fonction monotone qui croît lentement et $A$ est l'amplitude des oscillations autour de $f$, $T$ est leur fréquence et $t_0$ est leur origine temporelle."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Caractérisation des oscillations"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Regardons les dernières années afin de mieux caractériser les oscillations rapides."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.axes._subplots.AxesSubplot at 0x7f5e1baf33c8>"
+       "<matplotlib.axes._subplots.AxesSubplot at 0x7f82d59caa90>"
       ]
      },
-     "execution_count": 31,
+     "execution_count": 9,
      "metadata": {},
      "output_type": "execute_result"
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1003,14 +891,152 @@
     }
    ],
    "source": [
-    "data[\"Concentration en CO2 (ppm)\"].plot()"
+    "data[-200:].plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Les oscillations semblent avoir une période d'un an (variations annuelles). Le pic de concentration a lieu au début de l'été, soit vers juin. Afin d'affiner l'analyse, nous pouvons considérer que sur un an, l'évolution de $f$ peut s'apparenter à une droite (sa tangente au milieu de l'an donné).\n",
+    "\n",
+    "Concentrons-nous sur la dernière année : entre automne 2023 et automne 2024. Nous cherchons à isoler les données entre début septembre 2023 et fin septembre 2024, afin d'être certains d'englober toute la variation annuelle."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 14,
    "metadata": {},
    "outputs": [],
+   "source": [
+    "period1 = pd.Period(pd.Timestamp(2023, 9, 1), 'W')\n",
+    "period2 = pd.Period(pd.Timestamp(2024, 9, 30), 'W')\n",
+    "data_last_year = full_data[full_data[((full_data.index >= period1.start_time) & (full_data.index <= period1.end_time))].index[0]:\n",
+    "     full_data[((full_data.index >= period2.start_time) & (full_data.index <= period2.end_time))].index[0]]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Vérifions si nous comprenons bien toute l'année voulue :"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.axes._subplots.AxesSubplot at 0x7f82cf6c3a20>"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "data_last_year.plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Parfait. Maintenant, trouvons les deux données minimales (au début et à la fin de l'oscillation respectivement) afin de trouver la pente de $f$. $f$ étant croissante, le premier minimum est le minimum de l'année isolée. Les valeurs de début d'année 2024 étant supérieures au deuxième minimum, nous pouvons le trouver en contraignant `year == 2024`. On vérifie par la suite que les minima trouvés sont bien ceux attendus par analyse graphique.\n",
+    "\n",
+    "Nous pourrons ensuite enlever cette contribution lente pour avoir une oscillation brute, et mesurer son amplitude.\n",
+    "\n",
+    "__Remarque__ : nous faisons en fait un traitement de signal à la main ; nous aurrions aussi pu utiliser un filtre passe-haut avec une coupure de fréquence choisie soigneusement pour obtenir le même résultat. Ce document étant censé être accessible à tous, il est préférable de ne pas utiliser de méthode trop technique."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "2023-09-23    417.77\n",
+      "Freq: 7D, dtype: float64\n",
+      "2024-09-14    421.39\n",
+      "Freq: 7D, dtype: float64\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.axes._subplots.AxesSubplot at 0x7f82cf5dac88>"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "min_2023 = data_last_year.where(data_last_year == min(data_last_year)).dropna()\n",
+    "data_last_year_2024 = data_last_year[data_last_year.index.year == 2024]\n",
+    "min_2024 = data_last_year_2024.where(data_last_year_2024 == min(data_last_year_2024)).dropna()\n",
+    "print(min_2023)\n",
+    "print(min_2024)\n",
+    "\n",
+    "# Pour définir la droite correspondant à l'évolution de f(t), il est plus simple de repasser sur des indices numériques\n",
+    "data_last_year_filtered = pd.Series(data=data_last_year.values, index=range(len(data_last_year)))\n",
+    "dy = min_2024.values[0] - min_2023.values[0]\n",
+    "min_2023_filtered = data_last_year_filtered.where(data_last_year_filtered == min(data_last_year_filtered)).dropna()\n",
+    "data_last_year_filtered_2024 = data_last_year_filtered[data_last_year.index.year == 2024]\n",
+    "min_2024_filtered = data_last_year_filtered_2024.where(data_last_year_filtered_2024 == min(data_last_year_filtered_2024)).dropna()\n",
+    "dx = min_2024_filtered.index[0] - min_2023_filtered.index[0]\n",
+    "data_f = min_2023.values[0] + (dy / dx) * (range(len(data_last_year)) - min_2023_filtered.index[0])\n",
+    "f = pd.Series(data=data_f, index=data_last_year.index)\n",
+    "data_last_year_filtered.index = data_last_year.index\n",
+    "\n",
+    "data_last_year_filtered -= f\n",
+    "data_last_year_filtered.index = data_last_year.index\n",
+    "amp = data_last_year_filtered.max()\n",
+    "data_last_year_filtered.plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Modélisation de l'évolution en arrière-plan"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
    "source": []
   }
  ],
-- 
2.18.1