diff --git a/module3/exo3/exercice.ipynb b/module3/exo3/exercice.ipynb index c5b86087438ce770849af2ffdb5ff123dc3692f2..6d558d9413f50ed79776e905a299295cf6d7cb5b 100644 --- a/module3/exo3/exercice.ipynb +++ b/module3/exo3/exercice.ipynb @@ -590,7 +590,7 @@ "hidePrompt": false }, "source": [ - "La concentration est donc bien un nombre on doit pas changer. Les dates sont des chaînes de caractère, on doit donc traiter pour avoir des nombres." + "La concentration est donc bien un nombre on doit pas changer. Les dates sont des chaînes de caractère, on doit donc traiter pour avoir une forme utilisable dans un graphique." ] }, { @@ -649,7 +649,7 @@ { "data": { "text/plain": [ - "" + "" ] }, "execution_count": 7, @@ -695,11 +695,19 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 8, "metadata": { "scrolled": true }, "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/conda/lib/python3.6/site-packages/scipy/signal/_arraytools.py:45: 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", + " b = a[a_slice]\n" + ] + }, { "data": { "image/png": "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\n", @@ -729,12 +737,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Afin de pouvoir caractériser la période, nous ramenons les données autour de 0 en retirant aux données la courbe du fit." + "Afin de pouvoir caractériser la période, nous ramenons les données autour de 0 en retirant aux données la courbe du fit obtenue avec l'utilisation du filtre *savgol*." ] }, { "cell_type": "code", - "execution_count": 27, + "execution_count": 16, "metadata": {}, "outputs": [ { @@ -762,12 +770,12 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Nous rappliquons un filtre à ces nouvelles données pour avoir un signal plus facile à visualiser." + "Nous rappliquons le même filtre à ces nouvelles données pour avoir un signal plus facile à visualiser." ] }, { "cell_type": "code", - "execution_count": 32, + "execution_count": 10, "metadata": {}, "outputs": [ { @@ -799,21 +807,18 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "On peut estimer la période plus facilement." + "Visuellement nous remarquons que la période de l'oscillation est de plus ou moins un an et nous besoin la suite de l'analyse sur cette supposition. On peut donc estimer la période plus facilement. Ne connaissant pas de moyen pour déterminer à l'aide du code et d'une fonction la période des données, nous utilisons les étapes suivantes:\n", + "- Le signal ressemble aà un signal sinusoidal. De ce fait, nous savons que la période correspond au passage des données à la valeur 0 une fois sur deux.\n", + "- Nous affichons donc des barres verticales une fois sur deux, là où les données passent par zéro.\n", + "- Nous faisons une nouvelle inspection visuelle et remarquons que sur un zoom de deux ans le lieu de données égales à 0 passe aux alentour de juillet de l'année pour les deux années consécutives\n", + "- Nous confirmons que la période est proche de un an" ] }, { "cell_type": "code", - "execution_count": 42, + "execution_count": 12, "metadata": {}, "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[22, 23, 26, 26, 26, 26, 26, 25, 46, 1, 26, 1, 1, 1, 1, 1, 1, 1, 1, 1, 47, 26, 24, 24, 24, 52, 27, 26, 26, 26, 26, 26, 51, 28, 27, 51, 26, 26, 51, 52, 52, 27, 25, 27, 26, 51, 53, 27, 26, 24, 24, 25, 26, 26, 26, 26, 25, 27, 26, 26, 26, 26, 25, 27, 79, 26, 25, 27, 25, 27, 26, 26, 26, 27, 24, 28, 26, 25, 26, 27, 25, 27, 25, 27, 25, 27, 25, 25, 25, 50, 24, 25, 51, 52, 52, 104, 27, 25, 51, 26, 26, 26, 25, 27, 78, 27, 24, 54]\n" - ] - }, { "data": { "image/png": "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\n", @@ -840,17 +845,7 @@ " raw_data_period.append(i)\n", " raw_data_date.append(raw_data[\"convert\"][i])\n", " raw_data_value.append(raw_data[\"smooth_b\"][i])\n", - " \n", - " \n", - "#print(raw_data_period) \n", - "#print(len(raw_data_period))\n", "\n", - "for i in range(len(raw_data_period)-1):\n", - " #print(i)#,raw_data_period[i+1]-raw_data_period[i])\n", - " raw_data_period_mean.append(raw_data_period[i+1]-raw_data_period[i])\n", - " \n", - "print(raw_data_period_mean)\n", - "2.*np.mean(raw_data_period_mean)\n", "\n", "fig, ax = plt.subplots()\n", "ax.plot(raw_data[\"convert\"],raw_data[\"smooth_b\"])\n", @@ -863,6 +858,37 @@ "plt.show()" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Puisque l'indice des données correspond au nombr ede semaines édcoulées dpeuis le début des données, une autre méthode consisterait à récolter l'indice des données qui passent par 0. C'est ce qui est fait dans *raw_data_period*. Nous regardons le nombre de semaine moyen entre deux données consécutives passant par 0 et multiplions par 2 pour avoir la période. Cette méthode nous donne une réponse en nombre de semaine proche de la valeur de un an. " + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "57.81481481481482" + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "for i in range(len(raw_data_period)-1):\n", + " raw_data_period_mean.append(raw_data_period[i+1]-raw_data_period[i])\n", + " \n", + "#print(raw_data_period_mean)\n", + "2.*np.mean(raw_data_period_mean)\n" + ] + }, { "cell_type": "code", "execution_count": null,