Revisite logistique

module3/exo3/exercice.ipynb

@@ -9,7 +9,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -17,7 +17,8 @@
     "import matplotlib.pyplot as plt\n",
     "import pandas as pd\n",
     "import isoweek\n",
-    "import csv"
+    "import numpy\n",
+    "from sklearn.linear_model import LogisticRegression"
    ]
   },
   {
@@ -1041,7 +1042,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1069,7 +1070,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
@@ -1121,7 +1122,7 @@
        "1   Dead         230     139"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 7,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1142,7 +1143,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 8,
    "metadata": {},
    "outputs": [
     {
@@ -1201,7 +1202,7 @@
        "2  Mortality (%)          31      23"
       ]
      },
-     "execution_count": 25,
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1241,7 +1242,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": 9,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1328,7 +1329,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 10,
    "metadata": {},
    "outputs": [
     {
@@ -1450,7 +1451,7 @@
        "11    Mortality (65+(%))          85      85"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 10,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -1481,7 +1482,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 30,
+   "execution_count": 11,
    "metadata": {},
    "outputs": [
     {
@@ -1536,41 +1537,43 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 44,
+   "execution_count": 15,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
       "text/plain": [
-       "<Figure size 432x288 with 1 Axes>"
+       "21.0"
       ]
      },
-     "metadata": {
-      "needs_background": "light"
-     },
-     "output_type": "display_data"
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
-    "sorted_data = data.set_index(\"Age\").sort_index()\n",
+    "sorted_data = data.sort_values(\"Age\")\n",
     "\n",
-    "Bool_Death = []\n",
-    "Ages = sorted_data.index\n",
-    "Status = sorted_data.loc[ : , \"Status\"]\n",
+    "Bool_Death_smoker = []\n",
+    "Bool_Death_no_smoker = []\n",
+    "Ages_smoker = []\n",
+    "Ages_no_smoker = []\n",
     "\n",
-    "for it in Status:\n",
-    "    if(it==\"Alive\"):\n",
-    "        Bool_Death.append(1)\n",
-    "    else:\n",
-    "        Bool_Death.append(0)\n",
-    "        \n",
-    "fig = plt.figure()\n",
-    "plt.plot(Ages,Bool_Death,label = \"Death\")\n",
-    "plt.xlabel(\"Ages\")\n",
-    "plt.ylabel(\"Mortalité\")\n",
-    "plt.legend()\n",
-    "plt.show()"
+    "for it in range(len(sorted_data)):\n",
+    "    if(sorted_data[\"Status\"][it]==\"Alive\" and sorted_data[\"Smoker\"][it]==\"Yes\"):\n",
+    "        Bool_Death_smoker.append(1)\n",
+    "        Ages_smoker.append(sorted_data[\"Age\"][it])\n",
+    "    if(sorted_data[\"Status\"][it]==\"Dead\" and sorted_data[\"Smoker\"][it]==\"Yes\"):\n",
+    "        Bool_Death_smoker.append(0)\n",
+    "        Ages_smoker.append(sorted_data[\"Age\"][it])\n",
+    "    if(sorted_data[\"Status\"][it]==\"Alive\" and sorted_data[\"Smoker\"][it]==\"No\"):\n",
+    "        Bool_Death_no_smoker.append(1)\n",
+    "        Ages_no_smoker.append(sorted_data[\"Age\"][it])\n",
+    "    if(sorted_data[\"Status\"][it]==\"Dead\" and sorted_data[\"Smoker\"][it]==\"No\"):\n",
+    "        Bool_Death_no_smoker.append(0)\n",
+    "        Ages_no_smoker.append(sorted_data[\"Age\"][it])\n",
+    "\n",
+    "sorted_data[\"Age\"][0]"
    ]
   },
   {