Exercice 3 - stats and analysis on weather in Paris

module2/exo3/exercice_fr.ipynb

 {
- "cells": [],
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Analyse de la distribution de la température à Paris le 30 mai 2023\n",
+    "Je n'ai pas beaucoup d'idées de données personnelles que je pourrais utiliser pour cet execice donc je suis allée chercher des données de météo à Paris et vais analyser comment la température a évolué au cours de la journée.\n",
+    "\n",
+    "# Source des données\n",
+    "* Le site web d'où sortent ces données: [https://www.timeanddate.com/weather/france/paris/historic](https://www.timeanddate.com/weather/france/paris/historic).\n",
+    "* La méthode de téléchargement: j'ai enregistré au format html, récupéré le dictionnaire javascript au format json dans le code source, puis importé les données dans python pour les reformater en csv comme ceci:\n",
+    "\n",
+    "```\n",
+    "import json, csv\n",
+    "text = \"\"\"copy-paste here the json dictionnary taken from html source code\"\"\"\n",
+    "my_data = json.loads(text)\n",
+    "csv_columns = ['date','ts','ds','icon','desc','temp','templow','baro','wind','wd','hum']\n",
+    "my_data2 = [{k:d[k] for k in csv_columns} for d in my_data['detail']]\n",
+    "csv_file = \"temps_Paris_30_05_2023.csv\"\n",
+    "try:\n",
+    "    with open(csv_file, 'w') as csvfile:\n",
+    "    writer = csv.DictWriter(csvfile, fieldnames=csv_columns)\n",
+    "    writer.writeheader()\n",
+    "    for data in my_data2:\n",
+    "        writer.writerow(data)\n",
+    "except IOError:\n",
+    "    print(\"I/O error\")\n",
+    "```"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "On peut jeter un coup d'oeil à ce fichier:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%load_ext rpy2.ipython"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "date,ts,ds,icon,desc,temp,templow,baro,wind,wd,hum\n",
+      "1684173600000.0,18:00,\"Monday, 15 May 2023, 18:00 — 00:00\",14,Passing clouds.,11,9,1019,15,0,85\n",
+      "1684195200000.0,00:00,\"Tuesday, 16 May 2023, 00:00 — 06:00\",14,Passing clouds.,9,5,1021,13,350,90\n",
+      "1684216800000.0,06:00,\"Tuesday, 16 May 2023, 06:00 — 12:00\",2,Scattered clouds.,12,5,1022,19,350,79\n",
+      "1684238400000.0,12:00,\"Tuesday, 16 May 2023, 12:00 — 18:00\",2,Passing clouds.,16,12,1023,22,10,51\n",
+      "1684260000000.0,18:00,\"Tuesday, 16 May 2023, 18:00 — 00:00\",2,Passing clouds.,16,11,1024,16,10,57\n",
+      "1684281600000.0,00:00,\"Wednesday, 17 May 2023, 00:00 — 06:00\",13,Clear.,11,7,1025,8,350,84\n",
+      "1684303200000.0,06:00,\"Wednesday, 17 May 2023, 06:00 — 12:00\",2,Passing clouds.,14,7,1027,14,50,77\n",
+      "1684324800000.0,12:00,\"Wednesday, 17 May 2023, 12:00 — 18:00\",2,Passing clouds.,17,14,1026,21,40,51\n",
+      "1684346400000.0,18:00,\"Wednesday, 17 May 2023, 18:00 — 00:00\",2,Passing clouds.,16,12,1026,17,30,56\n"
+     ]
+    }
+   ],
+   "source": [
+    "%%sh\n",
+    "head temps_Paris_30_05_2023.csv"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "            date     ts                                     ds  icon  \\\n",
+      "0   1.684174e+12  18:00     Monday, 15 May 2023, 18:00 — 00:00    14   \n",
+      "1   1.684195e+12  00:00    Tuesday, 16 May 2023, 00:00 — 06:00    14   \n",
+      "2   1.684217e+12  06:00    Tuesday, 16 May 2023, 06:00 — 12:00     2   \n",
+      "3   1.684238e+12  12:00    Tuesday, 16 May 2023, 12:00 — 18:00     2   \n",
+      "4   1.684260e+12  18:00    Tuesday, 16 May 2023, 18:00 — 00:00     2   \n",
+      "5   1.684282e+12  00:00  Wednesday, 17 May 2023, 00:00 — 06:00    13   \n",
+      "6   1.684303e+12  06:00  Wednesday, 17 May 2023, 06:00 — 12:00     2   \n",
+      "7   1.684325e+12  12:00  Wednesday, 17 May 2023, 12:00 — 18:00     2   \n",
+      "8   1.684346e+12  18:00  Wednesday, 17 May 2023, 18:00 — 00:00     2   \n",
+      "9   1.684368e+12  00:00   Thursday, 18 May 2023, 00:00 — 06:00    13   \n",
+      "10  1.684390e+12  06:00   Thursday, 18 May 2023, 06:00 — 12:00     2   \n",
+      "11  1.684411e+12  12:00   Thursday, 18 May 2023, 12:00 — 18:00     2   \n",
+      "12  1.684433e+12  18:00   Thursday, 18 May 2023, 18:00 — 00:00     2   \n",
+      "13  1.684454e+12  00:00     Friday, 19 May 2023, 00:00 — 06:00    13   \n",
+      "14  1.684476e+12  06:00     Friday, 19 May 2023, 06:00 — 12:00     2   \n",
+      "15  1.684498e+12  12:00     Friday, 19 May 2023, 12:00 — 18:00     2   \n",
+      "16  1.684519e+12  18:00     Friday, 19 May 2023, 18:00 — 00:00     2   \n",
+      "17  1.684541e+12  00:00   Saturday, 20 May 2023, 00:00 — 06:00    13   \n",
+      "18  1.684562e+12  06:00   Saturday, 20 May 2023, 06:00 — 12:00     2   \n",
+      "19  1.684584e+12  12:00   Saturday, 20 May 2023, 12:00 — 18:00     2   \n",
+      "20  1.684606e+12  18:00   Saturday, 20 May 2023, 18:00 — 00:00     2   \n",
+      "21  1.684627e+12  00:00     Sunday, 21 May 2023, 00:00 — 06:00    13   \n",
+      "22  1.684649e+12  06:00     Sunday, 21 May 2023, 06:00 — 12:00     1   \n",
+      "23  1.684670e+12  12:00     Sunday, 21 May 2023, 12:00 — 18:00     2   \n",
+      "24  1.684692e+12  18:00     Sunday, 21 May 2023, 18:00 — 00:00     2   \n",
+      "25  1.684714e+12  00:00     Monday, 22 May 2023, 00:00 — 06:00    14   \n",
+      "26  1.684735e+12  06:00     Monday, 22 May 2023, 06:00 — 12:00     2   \n",
+      "27  1.684757e+12  12:00     Monday, 22 May 2023, 12:00 — 18:00     6   \n",
+      "28  1.684778e+12  18:00     Monday, 22 May 2023, 18:00 — 00:00     6   \n",
+      "29  1.684800e+12  00:00    Tuesday, 23 May 2023, 00:00 — 06:00    17   \n",
+      "..           ...    ...                                    ...   ...   \n",
+      "31  1.684843e+12  12:00    Tuesday, 23 May 2023, 12:00 — 18:00     6   \n",
+      "32  1.684865e+12  18:00    Tuesday, 23 May 2023, 18:00 — 00:00    14   \n",
+      "33  1.684886e+12  00:00  Wednesday, 24 May 2023, 00:00 — 06:00    13   \n",
+      "34  1.684908e+12  06:00  Wednesday, 24 May 2023, 06:00 — 12:00     2   \n",
+      "35  1.684930e+12  12:00  Wednesday, 24 May 2023, 12:00 — 18:00     2   \n",
+      "36  1.684951e+12  18:00  Wednesday, 24 May 2023, 18:00 — 00:00     2   \n",
+      "37  1.684973e+12  00:00   Thursday, 25 May 2023, 00:00 — 06:00    13   \n",
+      "38  1.684994e+12  06:00   Thursday, 25 May 2023, 06:00 — 12:00     1   \n",
+      "39  1.685016e+12  12:00   Thursday, 25 May 2023, 12:00 — 18:00     2   \n",
+      "40  1.685038e+12  18:00   Thursday, 25 May 2023, 18:00 — 00:00     2   \n",
+      "41  1.685059e+12  00:00     Friday, 26 May 2023, 00:00 — 06:00    13   \n",
+      "42  1.685081e+12  06:00     Friday, 26 May 2023, 06:00 — 12:00     2   \n",
+      "43  1.685102e+12  12:00     Friday, 26 May 2023, 12:00 — 18:00     2   \n",
+      "44  1.685124e+12  18:00     Friday, 26 May 2023, 18:00 — 00:00     2   \n",
+      "45  1.685146e+12  00:00   Saturday, 27 May 2023, 00:00 — 06:00    13   \n",
+      "46  1.685167e+12  06:00   Saturday, 27 May 2023, 06:00 — 12:00     2   \n",
+      "47  1.685189e+12  12:00   Saturday, 27 May 2023, 12:00 — 18:00     2   \n",
+      "48  1.685210e+12  18:00   Saturday, 27 May 2023, 18:00 — 00:00     1   \n",
+      "49  1.685232e+12  00:00     Sunday, 28 May 2023, 00:00 — 06:00    14   \n",
+      "50  1.685254e+12  06:00     Sunday, 28 May 2023, 06:00 — 12:00     1   \n",
+      "51  1.685275e+12  12:00     Sunday, 28 May 2023, 12:00 — 18:00     2   \n",
+      "52  1.685297e+12  18:00     Sunday, 28 May 2023, 18:00 — 00:00     2   \n",
+      "53  1.685318e+12  00:00     Monday, 29 May 2023, 00:00 — 06:00    13   \n",
+      "54  1.685340e+12  06:00     Monday, 29 May 2023, 06:00 — 12:00     2   \n",
+      "55  1.685362e+12  12:00     Monday, 29 May 2023, 12:00 — 18:00     2   \n",
+      "56  1.685383e+12  18:00     Monday, 29 May 2023, 18:00 — 00:00     2   \n",
+      "57  1.685405e+12  00:00    Tuesday, 30 May 2023, 00:00 — 06:00    14   \n",
+      "58  1.685426e+12  06:00    Tuesday, 30 May 2023, 06:00 — 12:00     2   \n",
+      "59  1.685448e+12  12:00    Tuesday, 30 May 2023, 12:00 — 18:00     2   \n",
+      "60  1.685470e+12  18:00    Tuesday, 30 May 2023, 18:00 — 00:00     2   \n",
+      "\n",
+      "                     desc  temp  templow  baro  wind   wd  hum  \n",
+      "0         Passing clouds.    11        9  1019    15    0   85  \n",
+      "1         Passing clouds.     9        5  1021    13  350   90  \n",
+      "2       Scattered clouds.    12        5  1022    19  350   79  \n",
+      "3         Passing clouds.    16       12  1023    22   10   51  \n",
+      "4         Passing clouds.    16       11  1024    16   10   57  \n",
+      "5                  Clear.    11        7  1025     8  350   84  \n",
+      "6         Passing clouds.    14        7  1027    14   50   77  \n",
+      "7         Passing clouds.    17       14  1026    21   40   51  \n",
+      "8         Passing clouds.    16       12  1026    17   30   56  \n",
+      "9                  Clear.    12        7  1025    13   30   74  \n",
+      "10        Passing clouds.    14        7  1026    13   10   73  \n",
+      "11        Passing clouds.    16       14  1024    19   40   47  \n",
+      "12        Passing clouds.    16       12  1024    16   20   52  \n",
+      "13                 Clear.    12        7  1024    13  350   76  \n",
+      "14        Passing clouds.    16        7  1024    14   10   73  \n",
+      "15        Passing clouds.    19       16  1022    18   30   47  \n",
+      "16        Passing clouds.    19       13  1021    17   20   50  \n",
+      "17                 Clear.    13       10  1021    17   10   75  \n",
+      "18        Passing clouds.    18       10  1019    17  350   72  \n",
+      "19        Passing clouds.    20       18  1018    20   40   49  \n",
+      "20        Passing clouds.    19       15  1017    18   20   56  \n",
+      "21                 Clear.    15       11  1017    13  350   82  \n",
+      "22                 Sunny.    18       11  1017    18   10   78  \n",
+      "23        Passing clouds.    21       18  1016    18   20   57  \n",
+      "24        Passing clouds.    20       15  1016    15    0   69  \n",
+      "25        Passing clouds.    15       11  1017    14  350   83  \n",
+      "26        Passing clouds.    18       11  1018    11  340   81  \n",
+      "27         Broken clouds.    21       18  1017    14  330   64  \n",
+      "28  More clouds than sun.    20       14  1018    16    0   76  \n",
+      "29            Low clouds.    14       12  1020    16   10   89  \n",
+      "..                    ...   ...      ...   ...   ...  ...  ...  \n",
+      "31          Partly sunny.    18       13  1022    20   30   73  \n",
+      "32        Passing clouds.    16       12  1023    21   30   64  \n",
+      "33                 Clear.    12        8  1023    19   20   77  \n",
+      "34        Passing clouds.    15        8  1024    16   30   69  \n",
+      "35        Passing clouds.    19       15  1023    23   50   43  \n",
+      "36        Passing clouds.    18       15  1022    17   20   55  \n",
+      "37                 Clear.    15       11  1023    15   30   70  \n",
+      "38                 Sunny.    17       11  1024    22   40   63  \n",
+      "39        Passing clouds.    21       17  1024    25   50   44  \n",
+      "40        Passing clouds.    20       16  1024    20   30   54  \n",
+      "41                 Clear.    16       11  1025    18   20   72  \n",
+      "42        Passing clouds.    18       10  1026    16   20   75  \n",
+      "43        Passing clouds.    22       18  1025    18   40   53  \n",
+      "44        Passing clouds.    22       16  1024    19   40   52  \n",
+      "45                 Clear.    16       12  1024    18   40   63  \n",
+      "46        Passing clouds.    19       12  1023    15   30   61  \n",
+      "47        Passing clouds.    23       19  1021    17   30   50  \n",
+      "48                 Sunny.    23       19  1019    15   40   53  \n",
+      "49        Passing clouds.    19       13  1019    13   20   68  \n",
+      "50                 Sunny.    22       13  1018    14   30   69  \n",
+      "51        Passing clouds.    25       22  1018    21   40   41  \n",
+      "52        Passing clouds.    25       19  1017    20   20   44  \n",
+      "53                 Clear.    19       13  1018    22   30   70  \n",
+      "54        Passing clouds.    19       13  1020    21   20   74  \n",
+      "55        Passing clouds.    23       19  1020    23   30   53  \n",
+      "56        Passing clouds.    23       15  1020    26   20   57  \n",
+      "57        Passing clouds.    15       11  1022    23   30   67  \n",
+      "58        Passing clouds.    18       11  1023    20   30   64  \n",
+      "59        Passing clouds.    22       18  1022    22   30   44  \n",
+      "60        Passing clouds.    21       20  1021    20   30   38  \n",
+      "\n",
+      "[61 rows x 11 columns]\n"
+     ]
+    }
+   ],
+   "source": [
+    "import pandas\n",
+    "df = pandas.read_csv('temps_Paris_30_05_2023.csv')\n",
+    "print(df)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Analyse des données"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Informations qu'on a"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['date',\n",
+       " 'ts',\n",
+       " 'ds',\n",
+       " 'icon',\n",
+       " 'desc',\n",
+       " 'temp',\n",
+       " 'templow',\n",
+       " 'baro',\n",
+       " 'wind',\n",
+       " 'wd',\n",
+       " 'hum']"
+      ]
+     },
+     "execution_count": 43,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "list(df.columns)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "La plage horaire/date qu'on a à disposition va du"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'Mon 15 May 2023 06:00:00 PM '"
+      ]
+     },
+     "execution_count": 39,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import datetime\n",
+    "datetime.datetime.fromtimestamp(df['date'].min()/1000).strftime('%c')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "au"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'Tue 30 May 2023 06:00:00 PM '"
+      ]
+     },
+     "execution_count": 40,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "datetime.datetime.fromtimestamp(df['date'].max()/1000).strftime('%c')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Température\n",
+    "En moyenne, à Paris, il a fait une température en degrés Celsius de :"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "1684173600000"
+      ]
+     },
+     "execution_count": 60,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df['temp'].mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 75,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "import matplotlib.pyplot as plt\n",
+    "df.plot(x='date',y='temp')\n",
+    "ax = plt.subplot()\n",
+    "xticks = list(range(int(round(df.date.min(),0)),int(round(df.date.max(),0)),int(round((df.date.max()-df.date.min())/6.,0))))\n",
+    "xtick_labels = [datetime.datetime.fromtimestamp(d/1000).strftime('%Y-%m-%d %Hh') for d in xticks]\n",
+    "ax.set_xticks(xticks)\n",
+    "ax.set_xticklabels(xtick_labels, rotation=45)\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 86,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "%matplotlib inline\n",
+    "df.groupby(df['desc']).count().plot(kind='pie', y='date', autopct='%1.0f%%')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
  "metadata": {
   "kernelspec": {
    "display_name": "Python 3",
@@ -16,10 +429,9 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.3"
+   "version": "3.6.4"
   }
  },
  "nbformat": 4,
  "nbformat_minor": 2
 }
-