From 2a256dd79b033e93f8e6f749e0ea1d497d158300 Mon Sep 17 00:00:00 2001 From: 7eba932125d7468e05c00632ef18215f <7eba932125d7468e05c00632ef18215f@app-learninglab.inria.fr> Date: Fri, 11 Jun 2021 15:53:03 +0000 Subject: [PATCH] =?UTF-8?q?Ajout=20Extrapol=20lin=C3=A9aire?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- module3/exo3/exercice.ipynb | 257 ++++++++++++++++++++---------------- 1 file changed, 144 insertions(+), 113 deletions(-) diff --git a/module3/exo3/exercice.ipynb b/module3/exo3/exercice.ipynb index c4cc194..07d0fa4 100644 --- a/module3/exo3/exercice.ipynb +++ b/module3/exo3/exercice.ipynb @@ -84,7 +84,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -206,7 +206,7 @@ "4 314.91 315.70 314.44 " ] }, - "execution_count": 3, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" } @@ -225,7 +225,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 5, "metadata": { "scrolled": true }, @@ -349,7 +349,7 @@ "6 315.07 317.51 314.70 " ] }, - "execution_count": 4, + "execution_count": 5, "metadata": {}, "output_type": "execute_result" } @@ -374,7 +374,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 6, "metadata": {}, "outputs": [ { @@ -511,7 +511,7 @@ "1958-05 315.07 317.51 314.70 " ] }, - "execution_count": 5, + "execution_count": 6, "metadata": {}, "output_type": "execute_result" } @@ -534,7 +534,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 7, "metadata": {}, "outputs": [], "source": [ @@ -549,12 +549,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Toutes les périodes sont bien renseignées. Quand il n'y a pas de données pour la période, la valeur -99.99 est entrée. Nous enlevons pour le moment ces valeurs. Mais avant cela, il faut convertir les valeurs de CO2 en données numériques:" + "Toutes les périodes sont bien renseignées. Quand il n'y a pas de données pour la période, la valeur -99.99 est entrée. Nous enlevons pour le moment ces valeurs. La colonne `'index'` est créée avant pour tenir compte de l'espacement irrégulier des périodes. Il faut convertir les valeurs de CO2 en données numériques:" ] }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 17, "metadata": {}, "outputs": [ { @@ -588,6 +588,7 @@ " CO2_4\n", " C02_5\n", " CO2_6\n", + " index\n", " \n", " \n", " period\n", @@ -601,6 +602,7 @@ " \n", " \n", " \n", + " \n", " \n", " \n", " \n", @@ -616,6 +618,7 @@ " 314.91\n", " 315.70\n", " 314.44\n", + " 3.0\n", " \n", " \n", " 1958-04\n", @@ -629,6 +632,7 @@ " 314.99\n", " 317.45\n", " 315.16\n", + " 4.0\n", " \n", " \n", " 1958-05\n", @@ -642,6 +646,7 @@ " 315.07\n", " 317.51\n", " 314.70\n", + " 5.0\n", " \n", " \n", " 1958-07\n", @@ -655,6 +660,7 @@ " 315.22\n", " 315.86\n", " 315.19\n", + " 7.0\n", " \n", " \n", " 1958-08\n", @@ -668,6 +674,7 @@ " 315.29\n", " 314.93\n", " 316.19\n", + " 8.0\n", " \n", " \n", "\n", @@ -682,16 +689,16 @@ "1958-07 1958 07 21381 1958.5370 315.86 315.19 315.86 \n", "1958-08 1958 08 21412 1958.6219 314.93 316.19 313.99 \n", "\n", - " CO2_4 C02_5 CO2_6 \n", - "period \n", - "1958-03 314.91 315.70 314.44 \n", - "1958-04 314.99 317.45 315.16 \n", - "1958-05 315.07 317.51 314.70 \n", - "1958-07 315.22 315.86 315.19 \n", - "1958-08 315.29 314.93 316.19 " + " CO2_4 C02_5 CO2_6 index \n", + "period \n", + "1958-03 314.91 315.70 314.44 3.0 \n", + "1958-04 314.99 317.45 315.16 4.0 \n", + "1958-05 315.07 317.51 314.70 5.0 \n", + "1958-07 315.22 315.86 315.19 7.0 \n", + "1958-08 315.29 314.93 316.19 8.0 " ] }, - "execution_count": 7, + "execution_count": 17, "metadata": {}, "output_type": "execute_result" } @@ -699,6 +706,7 @@ "source": [ "data_valuesonly = data.copy()\n", "data_valuesonly['CO2'] = pd.to_numeric(data_valuesonly['CO2'])\n", + "data_valuesonly['index'] = np.linspace(1,len(data_valuesonly['CO2']), len(data_valuesonly['CO2']))\n", "\n", "periods_novalue = []\n", "for i in data_valuesonly.index:\n", @@ -710,16 +718,16 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 21, "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "" + "" ] }, - "execution_count": 8, + "execution_count": 21, "metadata": {}, "output_type": "execute_result" }, @@ -750,16 +758,16 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 22, "metadata": {}, "outputs": [ { "data": { "text/plain": [ - "" + "" ] }, - "execution_count": 9, + "execution_count": 22, "metadata": {}, "output_type": "execute_result" }, @@ -785,7 +793,7 @@ "metadata": {}, "source": [ "## La contribution lente \n", - "On veut maintenant extraire la contribution lente et l'extrapoler à 2025. Une première approche est une évolution linéaire à partir de l'année 2000." + "On veut extraire la contribution lente et l'extrapoler à 2025. Une première approche est une évolution linéaire à partir de 1958." ] }, { @@ -822,7 +830,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 23, "metadata": {}, "outputs": [ { @@ -840,16 +848,16 @@ " Model Family: Gaussian Df Model: 1 \n", "\n", "\n", - " Link Function: identity Scale: 19.998 \n", + " Link Function: identity Scale: 20.521 \n", "\n", "\n", - " Method: IRLS Log-Likelihood: -2195.3 \n", + " Method: IRLS Log-Likelihood: -2205.0 \n", "\n", "\n", - " Date: Thu, 10 Jun 2021 Deviance: 15019. \n", + " Date: Fri, 11 Jun 2021 Deviance: 15412. \n", "\n", "\n", - " Time: 13:48:15 Pearson chi2: 1.50e+04 \n", + " Time: 15:22:50 Pearson chi2: 1.54e+04 \n", "\n", "\n", " No. Iterations: 3 Covariance Type: nonrobust\n", @@ -860,10 +868,10 @@ " coef std err z P>|z| [0.025 0.975] \n", "\n", "\n", - " Intercept 306.1259 0.326 938.290 0.000 305.486 306.765\n", + " Intercept 305.3562 0.334 913.159 0.000 304.701 306.012\n", "\n", "\n", - " index 0.1329 0.001 177.232 0.000 0.131 0.134\n", + " index 0.1326 0.001 174.904 0.000 0.131 0.134\n", "\n", "" ], @@ -875,21 +883,21 @@ "Dep. Variable: CO2 No. Observations: 753\n", "Model: GLM Df Residuals: 751\n", "Model Family: Gaussian Df Model: 1\n", - "Link Function: identity Scale: 19.998\n", - "Method: IRLS Log-Likelihood: -2195.3\n", - "Date: Thu, 10 Jun 2021 Deviance: 15019.\n", - "Time: 13:48:15 Pearson chi2: 1.50e+04\n", + "Link Function: identity Scale: 20.521\n", + "Method: IRLS Log-Likelihood: -2205.0\n", + "Date: Fri, 11 Jun 2021 Deviance: 15412.\n", + "Time: 15:22:50 Pearson chi2: 1.54e+04\n", "No. Iterations: 3 Covariance Type: nonrobust\n", "==============================================================================\n", " coef std err z P>|z| [0.025 0.975]\n", "------------------------------------------------------------------------------\n", - "Intercept 306.1259 0.326 938.290 0.000 305.486 306.765\n", - "index 0.1329 0.001 177.232 0.000 0.131 0.134\n", + "Intercept 305.3562 0.334 913.159 0.000 304.701 306.012\n", + "index 0.1326 0.001 174.904 0.000 0.131 0.134\n", "==============================================================================\n", "\"\"\"" ] }, - "execution_count": 11, + "execution_count": 23, "metadata": {}, "output_type": "execute_result" } @@ -898,19 +906,18 @@ "import statsmodels.api as sm\n", "\n", "data_valuesonly[\"Intercept\"]=1\n", - "data_valuesonly['index'] = np.linspace(1,len(data_valuesonly['CO2']), len(data_valuesonly['CO2']))\n", "logmodel=sm.GLM(data_valuesonly['CO2'], data_valuesonly[['Intercept','index']]).fit()\n", "logmodel.summary()" ] }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 19, "metadata": {}, "outputs": [ { "data": { - "image/png": "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\n", + "image/png": "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\n", "text/plain": [ "
" ] @@ -935,97 +942,121 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Avec cette prédiction, la teneur en CO2 dans l'atmosphère en avril 2025 serait de $$" + "Avec cette prédiction, la teneur en CO2 dans l'atmosphère en avril 2025 serait de $412\\ ppm$. On reste donc en dessous des dernières valeurs atteintes. Pour être plus réalistes, on veut maintenant estimer la teneur en CO2 avec une approximation linéaire à partir de l'an 2000 (index 505 pour le mois de Janvier 2000) au vue de la croissance plus rapide sur les dernières années:" ] }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 31, "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", + "
InterceptindexCO2period
7481749.0405.6480732020-12
7491750.0405.7809472021-01
7501751.0405.9138202021-02
7511752.0406.0466932021-03
7521753.0406.1795672021-04
\n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", "
Generalized Linear Model Regression Results
Dep. Variable: CO2 No. Observations: 248
Model: GLM Df Residuals: 246
Model Family: Gaussian Df Model: 1
Link Function: identity Scale: 5.4151
Method: IRLS Log-Likelihood: -560.35
Date: Fri, 11 Jun 2021 Deviance: 1332.1
Time: 15:34:11 Pearson chi2: 1.33e+03
No. Iterations: 3 Covariance Type: nonrobust
\n", - "
" + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "\n", + " \n", + "\n", + "
coef std err z P>|z| [0.025 0.975]
Intercept 272.7958 1.322 206.343 0.000 270.205 275.387
index 0.1866 0.002 90.406 0.000 0.183 0.191
" ], "text/plain": [ - " Intercept index CO2 period\n", - "748 1 749.0 405.648073 2020-12\n", - "749 1 750.0 405.780947 2021-01\n", - "750 1 751.0 405.913820 2021-02\n", - "751 1 752.0 406.046693 2021-03\n", - "752 1 753.0 406.179567 2021-04" + "\n", + "\"\"\"\n", + " Generalized Linear Model Regression Results \n", + "==============================================================================\n", + "Dep. Variable: CO2 No. Observations: 248\n", + "Model: GLM Df Residuals: 246\n", + "Model Family: Gaussian Df Model: 1\n", + "Link Function: identity Scale: 5.4151\n", + "Method: IRLS Log-Likelihood: -560.35\n", + "Date: Fri, 11 Jun 2021 Deviance: 1332.1\n", + "Time: 15:34:11 Pearson chi2: 1.33e+03\n", + "No. Iterations: 3 Covariance Type: nonrobust\n", + "==============================================================================\n", + " coef std err z P>|z| [0.025 0.975]\n", + "------------------------------------------------------------------------------\n", + "Intercept 272.7958 1.322 206.343 0.000 270.205 275.387\n", + "index 0.1866 0.002 90.406 0.000 0.183 0.191\n", + "==============================================================================\n", + "\"\"\"" ] }, - "execution_count": 17, + "execution_count": 31, "metadata": {}, "output_type": "execute_result" } ], "source": [ - "data_pred.tail()" + "logmodel=sm.GLM(data_valuesonly['CO2'][505:], data_valuesonly[['Intercept','index']][505:]).fit()\n", + "logmodel.summary()" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "data_pred = pd.DataFrame({'index': data_valuesonly['index'][505:],\n", + " 'Intercept': 1})\n", + "data_pred['CO2'] = logmodel.predict(data_pred[['Intercept','index']])\n", + "data_pred['period'] = data_valuesonly.index[505:]\n", + "data_pred.plot(x=\"period\",y=\"CO2\",kind='line',color='r')\n", + "plt.scatter(x=data_valuesonly.index,y = data_valuesonly[\"CO2\"])\n", + "plt.grid(True)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Cette fois, la teneur en CO2 en avril 2025 est estimée à $422\\ ppm$. " ] }, { -- 2.18.1