diff --git a/module3/exo3/exercice.ipynb b/module3/exo3/exercice.ipynb index c634483ebbbeaf6da436a43ccc7bf87cca2c0305..5f2944bc8187b97c085eb7f1b6814f3146802e25 100644 --- a/module3/exo3/exercice.ipynb +++ b/module3/exo3/exercice.ipynb @@ -31,13 +31,16 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "%matplotlib inline\n", "import matplotlib.pyplot as plt\n", - "import pandas as pd\n" + "import pandas as pd\n", + "from scipy import stats\n", + "import numpy as np\n", + "import seaborn as sns\n" ] }, { @@ -49,7 +52,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 2, "metadata": {}, "outputs": [], "source": [ @@ -58,7 +61,7 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 3, "metadata": {}, "outputs": [ { @@ -526,7 +529,7 @@ "[1314 rows x 3 columns]" ] }, - "execution_count": 20, + "execution_count": 3, "metadata": {}, "output_type": "execute_result" } @@ -538,7 +541,7 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -578,7 +581,7 @@ "Index: []" ] }, - "execution_count": 21, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -596,7 +599,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -612,41 +615,1138 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 6, "metadata": {}, "outputs": [], "source": [ - "\n", - "fumeuse = raw_data.loc[raw_data.Smoker == \"Yes\"]\n", - "non_fumeuse = raw_data.loc[raw_data.Smoker == \"No\"]" + "sm_st = pd.crosstab(raw_data['Smoker'],raw_data['Status'], margins = True)" ] }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 7, "metadata": {}, "outputs": [ { "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StatusAliveDeadAllMortalityRate
Smoker
No5022307320.314208
Yes4431395820.238832
All94536913140.280822
\n", + "
" + ], "text/plain": [ - "str" + "Status Alive Dead All MortalityRate\n", + "Smoker \n", + "No 502 230 732 0.314208\n", + "Yes 443 139 582 0.238832\n", + "All 945 369 1314 0.280822" ] }, - "execution_count": 27, + "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "type(fumeuse.Status[1])" + "sm_st['MortalityRate']=sm_st.Dead / sm_st.All\n", + "sm_st" ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, - "outputs": [], - "source": [] + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "fig = plt.figure()\n", + "ax = fig.add_axes([0,0,1,1])\n", + "langs = ['Smoker','Non_Smoker']\n", + "students = [0.23,0.31]\n", + "ax.bar(langs,students)\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Le taux de mortalité est plus élevé chez les femmes non fumeuses que les femmes fumeuses, ce qui semble surprenant. \n", + "On peut regarder la significativité de ce résultat en faisant un test de comparaison des proportions de ces deux populations différentes (fumeuses et non fumeuses).\n" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "False" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "from statsmodels.stats.proportion import proportions_ztest\n", + "count = np.array([ sm_st.Dead.No, sm_st.Dead.Yes])\n", + "nobs = np.array([sm_st.All.No, sm_st.All.Yes])\n", + "stat, pval = proportions_ztest(count, nobs)\n", + "pval > 0.05" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "La p-value est inférieur à 0.05, nous considérons donc que la proportion de fumeuses est significativement différente de celle des non fumeuses." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Taux de mortalité: Non fumeuse VS Fumeuse en fonction de la tranche d'âge" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "On considére maintenant les classes suivantes : 18-34 ans, 34-54 ans, 55-64 ans, plus de 65 ans afin de voir s'il y a des différences de taux de mortalité entre les fumeuses et les non-fumeuses." + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
SmokerStatusAgeAgeGroup
0YesAlive21.018-34
1YesAlive19.318-34
2NoDead57.555-64
3NoAlive47.135-54
4YesAlive81.465-Plus
5NoAlive36.835-54
6NoAlive23.818-34
7YesDead57.555-64
8YesAlive24.818-34
9YesAlive49.535-54
10YesAlive30.018-34
11NoDead66.065-Plus
12YesAlive49.235-54
13NoAlive58.455-64
14NoDead60.655-64
15NoAlive25.118-34
16NoAlive43.535-54
17NoAlive27.118-34
18NoAlive58.355-64
19YesAlive65.765-Plus
20NoDead73.265-Plus
21YesAlive38.335-54
22NoAlive33.418-34
23YesDead62.355-64
24NoAlive18.018-34
25NoAlive56.255-64
26YesAlive59.255-64
27NoAlive25.818-34
28NoDead36.935-54
29NoAlive20.218-34
...............
1284YesDead36.035-54
1285YesAlive48.335-54
1286NoAlive63.155-64
1287NoAlive60.855-64
1288YesDead39.335-54
1289NoAlive36.735-54
1290NoAlive63.855-64
1291NoDead71.365-Plus
1292NoAlive57.755-64
1293NoAlive63.255-64
1294NoAlive46.635-54
1295YesDead82.465-Plus
1296YesAlive38.335-54
1297YesAlive32.718-34
1298NoAlive39.735-54
1299YesDead60.055-64
1300NoDead71.065-Plus
1301NoAlive20.518-34
1302NoAlive44.435-54
1303YesAlive31.218-34
1304YesAlive47.835-54
1305YesAlive60.955-64
1306NoDead61.455-64
1307YesAlive43.035-54
1308NoAlive42.135-54
1309YesAlive35.935-54
1310NoAlive22.318-34
1311YesDead62.155-64
1312NoDead88.665-Plus
1313NoAlive39.135-54
\n", + "

1314 rows × 4 columns

\n", + "
" + ], + "text/plain": [ + " Smoker Status Age AgeGroup\n", + "0 Yes Alive 21.0 18-34\n", + "1 Yes Alive 19.3 18-34\n", + "2 No Dead 57.5 55-64\n", + "3 No Alive 47.1 35-54\n", + "4 Yes Alive 81.4 65-Plus\n", + "5 No Alive 36.8 35-54\n", + "6 No Alive 23.8 18-34\n", + "7 Yes Dead 57.5 55-64\n", + "8 Yes Alive 24.8 18-34\n", + "9 Yes Alive 49.5 35-54\n", + "10 Yes Alive 30.0 18-34\n", + "11 No Dead 66.0 65-Plus\n", + "12 Yes Alive 49.2 35-54\n", + "13 No Alive 58.4 55-64\n", + "14 No Dead 60.6 55-64\n", + "15 No Alive 25.1 18-34\n", + "16 No Alive 43.5 35-54\n", + "17 No Alive 27.1 18-34\n", + "18 No Alive 58.3 55-64\n", + "19 Yes Alive 65.7 65-Plus\n", + "20 No Dead 73.2 65-Plus\n", + "21 Yes Alive 38.3 35-54\n", + "22 No Alive 33.4 18-34\n", + "23 Yes Dead 62.3 55-64\n", + "24 No Alive 18.0 18-34\n", + "25 No Alive 56.2 55-64\n", + "26 Yes Alive 59.2 55-64\n", + "27 No Alive 25.8 18-34\n", + "28 No Dead 36.9 35-54\n", + "29 No Alive 20.2 18-34\n", + "... ... ... ... ...\n", + "1284 Yes Dead 36.0 35-54\n", + "1285 Yes Alive 48.3 35-54\n", + "1286 No Alive 63.1 55-64\n", + "1287 No Alive 60.8 55-64\n", + "1288 Yes Dead 39.3 35-54\n", + "1289 No Alive 36.7 35-54\n", + "1290 No Alive 63.8 55-64\n", + "1291 No Dead 71.3 65-Plus\n", + "1292 No Alive 57.7 55-64\n", + "1293 No Alive 63.2 55-64\n", + "1294 No Alive 46.6 35-54\n", + "1295 Yes Dead 82.4 65-Plus\n", + "1296 Yes Alive 38.3 35-54\n", + "1297 Yes Alive 32.7 18-34\n", + "1298 No Alive 39.7 35-54\n", + "1299 Yes Dead 60.0 55-64\n", + "1300 No Dead 71.0 65-Plus\n", + "1301 No Alive 20.5 18-34\n", + "1302 No Alive 44.4 35-54\n", + "1303 Yes Alive 31.2 18-34\n", + "1304 Yes Alive 47.8 35-54\n", + "1305 Yes Alive 60.9 55-64\n", + "1306 No Dead 61.4 55-64\n", + "1307 Yes Alive 43.0 35-54\n", + "1308 No Alive 42.1 35-54\n", + "1309 Yes Alive 35.9 35-54\n", + "1310 No Alive 22.3 18-34\n", + "1311 Yes Dead 62.1 55-64\n", + "1312 No Dead 88.6 65-Plus\n", + "1313 No Alive 39.1 35-54\n", + "\n", + "[1314 rows x 4 columns]" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "bins= [18,34,54,64,200]\n", + "labels = ['18-34','35-54','55-64','65-Plus']\n", + "raw_data['AgeGroup'] = pd.cut(raw_data['Age'], bins=bins, labels=labels, right=False)\n", + "raw_data" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
AgeGroupSmokerStatuscounts
018-34NoAlive213
118-34NoDead6
218-34YesAlive174
318-34YesDead5
435-54NoAlive180
535-54NoDead19
635-54YesAlive198
735-54YesDead41
855-64NoAlive80
955-64NoDead39
1055-64YesAlive64
1155-64YesDead51
1265-PlusNoAlive29
1365-PlusNoDead166
1465-PlusYesAlive7
1565-PlusYesDead42
\n", + "
" + ], + "text/plain": [ + " AgeGroup Smoker Status counts\n", + "0 18-34 No Alive 213\n", + "1 18-34 No Dead 6\n", + "2 18-34 Yes Alive 174\n", + "3 18-34 Yes Dead 5\n", + "4 35-54 No Alive 180\n", + "5 35-54 No Dead 19\n", + "6 35-54 Yes Alive 198\n", + "7 35-54 Yes Dead 41\n", + "8 55-64 No Alive 80\n", + "9 55-64 No Dead 39\n", + "10 55-64 Yes Alive 64\n", + "11 55-64 Yes Dead 51\n", + "12 65-Plus No Alive 29\n", + "13 65-Plus No Dead 166\n", + "14 65-Plus Yes Alive 7\n", + "15 65-Plus Yes Dead 42" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "sm_st_age = raw_data.groupby(['AgeGroup','Smoker','Status']).size().reset_index(name='counts')\n", + "sm_st_age" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "# width of the bars\n", + "barWidth = 0.2\n", + " \n", + "# Choose the height of the blue bars - Smoker\n", + "bars1 = [5/(5+174), 41/(41+198), 51/(51+64) , 42/(42+7)]\n", + " \n", + "# Choose the height of the cyan bars - Non Smoker\n", + "bars2 = [6/(6+213), 19/(19+180), 39/(39+80) , 166/(166+29)]\n", + " \n", + "\n", + " \n", + "# The x position of bars\n", + "r1 = np.arange(len(bars1))\n", + "r2 = [x + barWidth for x in r1]\n", + " \n", + "# Create blue bars\n", + "plt.bar(r1, bars1, width = barWidth, color = 'blue', edgecolor = 'black', capsize=7, label='Smoker')\n", + " \n", + "# Create cyan bars\n", + "plt.bar(r2, bars2, width = barWidth, color = 'cyan', edgecolor = 'black', capsize=7, label='Non Smoker')\n", + " \n", + "# general layout\n", + "plt.xticks([r + barWidth for r in range(len(bars1))], ['18-34','35-54','55-64','65-Plus'])\n", + "plt.ylabel('Mortality Rate')\n", + "plt.legend()\n", + " \n", + "# Show graphic\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "En considérant les catégories d'âge, il n'y a pas de différences entre le taux de mortalité des fumeuses et les non fumeuses pour les 18-34 ans et pour les 65-plus ans. \n", + "En revanche pour les catégories 35-54 et 55-64, le taux de mortalité des fumeuses est plus élevé. " + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Regression logistique: Mortalité en fonction de l'âge" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Afin d'éviter un biais induit par des regroupements en tranches d'âges arbitraires et non régulières, il est envisageable d'essayer de réaliser une régression logistique. Si on introduit une variable Death valant 1 ou 0 pour indiquer si l'individu est décédé durant la période de 20 ans, on peut étudier le modèle Death ~ Age pour étudier la probabilité de décès en fonction de l'âge selon que l'on considère le groupe des fumeuses ou des non fumeuses. " + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:5: SettingWithCopyWarning: \n", + "A value is trying to be set on a copy of a slice from a DataFrame.\n", + "Try using .loc[row_indexer,col_indexer] = value instead\n", + "\n", + "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n", + " \"\"\"\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[4.63502546]\n", + "[[-0.0730934]]\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/conda/lib/python3.6/site-packages/scipy/stats/stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n", + " return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 13, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "#Fumeuses\n", + "fum = raw_data.loc[raw_data.Smoker == \"Yes\"]\n", + "\n", + "from sklearn.linear_model import LogisticRegression\n", + "fum['Death']= fum.Status.map({'Dead': 0, 'Alive': 1})\n", + "#clf = LogisticRegression(random_state=0).fit(list(raw_data.Death),list(raw_data.Age))\n", + "X_f = np.array(fum.Age).reshape(-1, 1)\n", + "y_f = np.array(fum.Death)\n", + "clf_f = LogisticRegression(random_state=0).fit(X_f,y_f)\n", + "\n", + "\n", + "\n", + "# Check trained model intercept\n", + "print(clf_f.intercept_)\n", + "\n", + "# Check trained model regression coefficients\n", + "print(clf_f.coef_)\n", + "\n", + "# Make predictions\n", + "preds_f = clf_f.predict(X = X_f)\n", + "\n", + "clf_f.score(X = X_f ,\n", + " y = y_f)\n", + "\n", + "sns.regplot(x='Age', y='Death', data=fum, logistic=True)" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:4: SettingWithCopyWarning: \n", + "A value is trying to be set on a copy of a slice from a DataFrame.\n", + "Try using .loc[row_indexer,col_indexer] = value instead\n", + "\n", + "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n", + " after removing the cwd from sys.path.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[5.70698877]\n", + "[[-0.08999613]]\n" + ] + }, + { + "data": { + "text/plain": [ + "" + ] + }, + "execution_count": 14, + "metadata": {}, + "output_type": "execute_result" + }, + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "#Non Fumeuses\n", + "nfum = raw_data.loc[raw_data.Smoker == \"No\"]\n", + "\n", + "nfum['Death']= nfum.Status.map({'Dead': 0, 'Alive': 1})\n", + "X_nf = np.array(nfum.Age).reshape(-1, 1)\n", + "y_nf = np.array(nfum.Death)\n", + "clf_nf = LogisticRegression(random_state=0).fit(X_nf,y_nf)\n", + "\n", + "\n", + "\n", + "# Check trained model intercept\n", + "print(clf_nf.intercept_)\n", + "\n", + "# Check trained model regression coefficients \n", + "print(clf_nf.coef_)\n", + "\n", + "\n", + "# Make predictions\n", + "preds_nf = clf_f.predict(X = X_nf)\n", + "\n", + "clf_nf.score(X = X_nf,\n", + " y = y_nf)\n", + "\n", + "\n", + "sns.regplot(x='Age', y='Death', data=nfum, logistic=True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Le coefficient de régression du modèle avec les fumeuses est supérieur à celui avec les non fumeuses; la mortalité des jeunes commencerait plus tôt chez les fumeuses. \n", + "Ces régressions ne permettent pas de conclure sur la nocivité du tabagisme.\n", + "\n" + ] } ], "metadata": {