Skip to content

M
mooc-rr
Commit 2dfad9fe authored by 39781cc7cca0dc30af9d6060ede9947c's avatar 39781cc7cca0dc30af9d6060ede9947c
Browse files
Options

Ajout modélidation d'une oscillation annuelle.

parent 511fa753
Hide whitespace changes
Inline Side-by-side
Showing with 170 additions and 127 deletions
module3/exo3/exercice.ipynb
View file @ 2dfad9fe
...@@ -469,7 +469,7 @@ ...@@ -469,7 +469,7 @@
"hidePrompt": true "hidePrompt": true
}, },
"source": [ "source": [
"Pour cela, nous allons ajuster une courbe quadratique à la base complète en utilisant la librairie $lmfit$. \n", "Pour cela, nous allons ajuster une courbe quadratique à la base complète en utilisant la librairie *lmfit*. \n",
"Cette courbe suivra la tendance générale des données initiales en \"annulant\" les oscillations périodiques.\n", "Cette courbe suivra la tendance générale des données initiales en \"annulant\" les oscillations périodiques.\n",
"\n", "\n",
"Nous utiliserons le coefficient de détermination ($r²$) pour estimer la qualité de la régression." "Nous utiliserons le coefficient de détermination ($r²$) pour estimer la qualité de la régression."
...@@ -488,10 +488,10 @@ ...@@ -488,10 +488,10 @@
"output_type": "stream", "output_type": "stream",
"text": [ "text": [
"Requirement already satisfied: lmfit in /opt/conda/lib/python3.6/site-packages (1.0.0)\n", "Requirement already satisfied: lmfit in /opt/conda/lib/python3.6/site-packages (1.0.0)\n",
"Requirement already satisfied: scipy>=1.2 in /opt/conda/lib/python3.6/site-packages (from lmfit) (1.4.1)\n",
"Requirement already satisfied: numpy>=1.16 in /opt/conda/lib/python3.6/site-packages (from lmfit) (1.18.2)\n", "Requirement already satisfied: numpy>=1.16 in /opt/conda/lib/python3.6/site-packages (from lmfit) (1.18.2)\n",
"Requirement already satisfied: asteval>=0.9.16 in /opt/conda/lib/python3.6/site-packages (from lmfit) (0.9.18)\n", "Requirement already satisfied: asteval>=0.9.16 in /opt/conda/lib/python3.6/site-packages (from lmfit) (0.9.18)\n",
"Requirement already satisfied: uncertainties>=3.0.1 in /opt/conda/lib/python3.6/site-packages (from lmfit) (3.1.2)\n" "Requirement already satisfied: uncertainties>=3.0.1 in /opt/conda/lib/python3.6/site-packages (from lmfit) (3.1.2)\n",
"Requirement already satisfied: scipy>=1.2 in /opt/conda/lib/python3.6/site-packages (from lmfit) (1.4.1)\n"
] ]
} }
], ],
...@@ -510,7 +510,7 @@ ...@@ -510,7 +510,7 @@
"outputs": [], "outputs": [],
"source": [ "source": [
"# Import des librairies\n", "# Import des librairies\n",
"from lmfit.models import QuadraticModel\n", "from lmfit.models import QuadraticModel, Model\n",
"from sklearn.metrics import r2_score" "from sklearn.metrics import r2_score"
] ]
}, },
...@@ -621,7 +621,7 @@ ...@@ -621,7 +621,7 @@
}, },
"source": [ "source": [
"Le coefficient de détermination est très bon. \n", "Le coefficient de détermination est très bon. \n",
"Avec cet unique modèle de prédiction, il serait possible de calculer de bonnes prédictions de CO2 pour les années futures." "Ce modèle de prédiction suffirait à calculer de bonnes prédictions de CO2 pour les années futures."
] ]
}, },
{ {
...@@ -719,7 +719,7 @@ ...@@ -719,7 +719,7 @@
{ {
"data": { "data": {
"text/plain": [ "text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x7f8512415a58>" "<matplotlib.axes._subplots.AxesSubplot at 0x7f5b08ff9d30>"
] ]
}, },
"execution_count": 17, "execution_count": 17,
...@@ -769,81 +769,133 @@ ...@@ -769,81 +769,133 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 18, "execution_count": null,
"metadata": { "metadata": {},
"hideCode": true,
"hidePrompt": true
},
"outputs": [],
"source": [
"# Import des librairies\n",
"import scipy.fftpack"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"hideCode": true,
"hidePrompt": true
},
"outputs": [], "outputs": [],
"source": [ "source": []
"yf = pd.Series(scipy.fftpack.fft(data['untrend_data']))"
]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 20, "execution_count": 18,
"metadata": { "metadata": {},
"hideCode": true,
"hidePrompt": true
},
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>CO2</th>\n",
" <th>trend_fit</th>\n",
" <th>untrend_data</th>\n",
" </tr>\n",
" <tr>\n",
" <th>date</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>1960-01-02</th>\n",
" <td>315.72</td>\n",
" <td>315.96</td>\n",
" <td>315.95</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1960-01-09</th>\n",
" <td>316.40</td>\n",
" <td>315.98</td>\n",
" <td>316.61</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1960-01-16</th>\n",
" <td>316.73</td>\n",
" <td>316.00</td>\n",
" <td>316.92</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1960-01-23</th>\n",
" <td>316.57</td>\n",
" <td>316.01</td>\n",
" <td>316.75</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1960-01-30</th>\n",
" <td>316.68</td>\n",
" <td>316.03</td>\n",
" <td>316.84</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [ "text/plain": [
"dtype('complex128')" " CO2 trend_fit untrend_data\n",
"date \n",
"1960-01-02 315.72 315.96 315.95\n",
"1960-01-09 316.40 315.98 316.61\n",
"1960-01-16 316.73 316.00 316.92\n",
"1960-01-23 316.57 316.01 316.75\n",
"1960-01-30 316.68 316.03 316.84"
] ]
}, },
"execution_count": 20, "execution_count": 18,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
} }
], ],
"source": [ "source": [
"yf.dtype" "data[mask1_1y].head(5)"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 21, "execution_count": 19,
"metadata": { "metadata": {
"hideCode": true, "hideCode": true,
"hidePrompt": true "hidePrompt": true
}, },
"outputs": [],
"source": [
"data_1y = data[mask1_1y].drop(['trend_fit', 'CO2'], axis=1) - 315.95\n",
"time_1y = dates_to_idx(data_1y.index)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [ "outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/opt/conda/lib/python3.6/site-packages/numpy/core/_asarray.py:85: ComplexWarning: Casting complex values to real discards the imaginary part\n",
" return array(a, dtype, copy=False, order=order)\n"
]
},
{ {
"data": { "data": {
"text/plain": [ "text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x7f84f8413f98>" "<matplotlib.axes._subplots.AxesSubplot at 0x7f5b08f40b38>"
] ]
}, },
"execution_count": 21, "execution_count": 20,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
}, },
{ {
"data": { "data": {
"image/png": "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\n", "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXYAAAEVCAYAAAD0Ps6RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3Xd8VFX6+PHPk056D4QkhB4DhkBAqgoWUGSxt9XdVXcX21q+btHddS3rrj931W26Flx1iw1d10ITUIkgoBh6gIQOSYA0ShIgbeb8/pghBkifmUzJ83698mJy751znpm5eThz7rnniDEGpZRSvsPP3QEopZRyLk3sSinlYzSxK6WUj9HErpRSPkYTu1JK+RhN7Eop5WM0sSullI/RxK6UUj5GE7tSSvmYAHdUGh8fb9LT091RtVJKea01a9ZUGGMS2jvOLYk9PT2dvLw8d1StlFJeS0T2duQ47YpRSikfo4ldKaV8jCZ2pZTyMW7pY1dKeYaGhgaKi4upra11dyiqmZCQEFJSUggMDOzS8zWxK9WDFRcXExERQXp6OiLi7nAUYIyhsrKS4uJi+vfv36UynNYVIyL+IrJOROY5q0yllGvV1tYSFxenSd2DiAhxcXEOfYtyZh/7fcBWJ5aneiBjDLUNFipr6qhrtLg7nB5Bk7rncfQzcUpXjIikAJcBvwcecEaZqmd46+t9vL5iN9W1jRyrb+R4vQWL1bZcY1ZKFB//ZFKHyvlwXQlnp0QxMCHcleEq5RWc1cf+F+AXQERrB4jILGAWQFpampOqVZ6mtsHCki2lXJCRSFhw26fXS1/s5KmFBWSnRjMyLZrQoADCgwMIDfZnR1kN/1tbwrbSaoYktXpaAVBy5AT3z1nP9LN788JNOc58OUp5JYe7YkRkBlBmjFnT1nHGmNnGmNHGmNEJCe3eEau8UGlVLTfM/op73l7HjOe+ZFPx0RaPM8bwpyXbeGphAd8Zkcx7d4znj9eM4LGZw/jZtKHcNXkQD12agQjM33ig3XrnbdgPQG5hObUN2n3jy9avX8+CBQtcVn5ubi4zZszo8PHp6elUVFS0ecyTTz7paFid5owW+0RgpohMB0KASBF5wxhzsxPKVl5i3b7D3P6fNdTUNfLgJRn8Z9UernpxBT+dOpRZ5w7Az8/WZ2iM4ckFW3ll+W6uzUnhqauz8Pc7sz8xMSKEsf1jmb/pAPdfNLjNPsePN+wnLMifY/UWVu2sZEpGoqtepk97fO5mtuyvcmqZmcmRPPqdYU4rb/369eTl5TF9+vQz9jU2NhIQ4HkD/Z588kl+9atfdWudDrfYjTG/NMakGGPSgRuAzzWp9yzvrynm+tlfERzox//umsCdkwey8L7zuDgziacWFvC9176mtKoWq9Xwm4/yeWX5br4/vh9/aCWpn3RZVjI7ymrYVlrT6jE7ymrYvL+Kn1wwmLAgfxZvOeiKl6hcZM+ePQwfPrzp92eeeYbHHnuMyZMn8+CDD3LOOecwZMgQli9fTn19PY888ghz5swhOzubOXPm8NhjjzFr1iymTp3K97//fSwWCz//+c8ZM2YMWVlZvPzyy4CtJT558mSuueYaMjIyuOmmmzDGdi3nk08+ISMjg0mTJvG///2vzXgrKyuZOnUqI0eO5Pbbb28qA+CKK64gJyeHYcOGMXv2bAAeeughTpw4QXZ2NjfddFOrxzmdMcZpP8BkYF57x+Xk5Bjl/RoaLeaJuZtNvwfnmRtnrzKHaupO2W+1Ws07q/eajIcXmuzHF5lbX19t+j04zzy5YIuxWq3tll9WVWv6PzTPPLOooNVjnl1caNIfmmcOHj1h7npzjcl5YrFptLRftrLZsmWLW+vfvXu3GTZsWNPvTz/9tHn00UfN+eefbx544AFjjDHz5883F154oTHGmNdff93cfffdTcc/+uijZtSoUeb48ePGGGNefvll88QTTxhjjKmtrTU5OTlm165dZunSpSYyMtIUFRUZi8Vixo0bZ5YvX25OnDhhUlJSzLZt24zVajXXXnutueyyy1qN95577jGPP/64McaYefPmGcCUl5cbY4yprKw0xhhz/PhxM2zYMFNRUWGMMSYsLOyUMlo77nQtfTZAnulALnbqlALGmFxjTMc7qJTXslgNP/p3Hv/4cje3TEjnX7edQ0xY0CnHiAjXj0lj3r2TSI7uxecFZfzfRUN46JKMDg3nSogIZtyAOOZvPHBKy+gkYwxzN+xnXP84kiJDmJqZREVNPeuLDjvtdSr3ueqqqwDIyclhz549rR43c+ZMevXqBcDixYv597//TXZ2NmPHjqWyspLt27cDcM4555CSkoKfnx/Z2dns2bOHgoIC+vfvz+DBtu6+m29uu7Nh2bJlTcdcdtllxMTENO3729/+xogRIxg3bhxFRUVN9Z6uo8c5wvM6pJRXWLfvMLmF5Tx4SQZ3Th7Y5rEDE8L54K6J7CirITM5slP1zMhK5lcfbGLrgeoznptfUsXuimPMOm8AAJOHJhLgJyzeUkpOv9jOvSDlFgEBAVit1qbfm9+UExwcDIC/vz+NjY2tlhEWFtb02BjDc889x7Rp0045Jjc3t6m808vs7Jjxlo7Pzc3l008/ZdWqVYSGhjJ58uQWbzDq6HGO0knAVJfkFpbj7yd895yODV0NCvDrdFIHmDYsCX8/Yf6m/Wfs+3hDCYH+wqXDewMQ1SuQ8QPjWLy5tMUWvvI8SUlJlJWVUVlZSV1dHfPmtX3jekREBNXV1a3unzZtGi+++CINDQ0AbNu2jWPHjrV6fEZGBrt372bnzp0AvP32223Wf9555/Hmm28CsHDhQg4ftn07PHr0KDExMYSGhlJQUMBXX33V9JzAwMCmeNo6zpk0sasuyd1Wxqi0aKJCuzZJUUfFhQczYeCZ3TFWq2HexgOcNziB6NBvu4CmZiaxu+IYO8tbv+CqPEdgYCCPPPIIY8eOZcaMGWRkZLR5/JQpU9iyZUvTxdPT/ehHPyIzM5NRo0YxfPhwbr/99jZb+yEhIcyePZvLLruMSZMm0a9fvzbrf/TRR1m2bBmjRo1i8eLFTffkXHLJJTQ2NpKVlcVvfvMbxo0b1/ScWbNmkZWVxU033dTmcU7VkY54Z//oxVPvVlp1wvR7cJ55/vPt3VLf21/vNf0enGc2FR9p2vbVzgrT78F55sN1xaccu//I8W6Nzdu5++Kpap3HXDxVPcOybbYbMs4f0j03mk0b1psAP2Fes5uVPt6wn16B/lycmXTKsX2iepGVEsWSLaXdEptSnkgTu+q0pYVlJEQEM6wLfeZdERMWxMRB8czftB9jDA0WKws2HeCizCRCg868/j81M4n1RUcordI5xlXXvP7662RnZ5/yc/fdd7s7rA7TUTGqUxotVpZvK2fasN7dOivgZVl9+MV/N7Kp5CiVx+o5fLyBmSOSWzx26rDePLN4G0u2lHLzuLb7TJWtO1ZneDzVrbfeyq233uq2+o2DF/+1xa46ZX3REapqG5k8tHtv25+W2ZtAf2H+xgPMXb+fyJAAzhsS3+KxgxPDSY8L1e6YDggJCaGyslJHEXkQY19oIyQkpMtlaItddcrJYY6TBrecVF0lKjSQSYPimbthP1W1jVx2dh+CA/xbPFZEuDgziX+u3EN1bQMRIa4duePNUlJSKC4upry83N2hqGZOLo3XVZrYVac0DXPs1f3J8rKsZJYWbgBgZnbL3TAnTR3Wm1eW7ya3sJzvtNJlo2zDDbu6/JryXNoV0wM9u7iQa15cSYPF2v7BzZRV15JfUtXt3TAnXZyZRJC/X9NUA20ZlRZDXFgQi7U7RvVA2mLvYYwxvJdXzMGqWv6zai+3Tep4a+2LQtvX9clD3TOfflSvQO67aDAJ4cFtzgoJ4O8nXHRWEgs2HaC+0UpQgLZhVM+hZ3sPU3CwmoNVtUSGBPDnJdsoq+74kMDcbeUkRgST2ad7hjm25O4pg7huTGqHjr04M4nqukZW7Gh7IQSlfI0m9h5maWEZALO/P5raRgt/WFjYoeedHOZ4/pAErxkaN2lwPH2iQnh6UWHTOqpK9QSa2HuYpQVlDEuOZNyAOH507gDeX1vMmr2H2n2eu4Y5OiIk0J+HL8tky4Eq3vp6r7vDUarbaGLvQY4eb2DN3sNcYF867p4LBtEnKoTffLi53Ratu4Y5Omr62b2ZOCiOpxcVUllT5+5wlOoWmth7kGXby7EamlrdoUEBHW7RLi0sIyctxi3DHB0hIjw+cxjH6y388ZOOdTsp5e00sfcgSwvLiA4NJDs1umnb9LN7M2Fg2y3asupaNu+v4nw3jYZx1KDECG6b1J85eUWs26erKynfp4m9h7BaDV8U2i5+Nh8qKCL89vK2W7TuHuboDPdeOJikyGAe+aj9bielvJ3DiV1EQkRktYhsEJHNIvK4MwJTzrXRPnnWlBYufjZv0X6Sf5DN+4+e8rMw/6Dbhzk6Kjw4gF9NP4tNJUeZ802Ru8NRyqWccYNSHXCBMaZGRAKBL0VkoTHGNWs+qS5ZWlCGSOtzqN974WA+Wl/CHW+saXH/DWNSvWaYY2tmjkjmra/38cdFBVw6vPcZi28r5SscTuz2VT1OrkMWaP/R77oeJrewjJGp0a0ms/DgAOb+ZBLrio6csU+Asf3bvoXfG9i6nYYz/W/LeXpxIU9eeba7Q1LKJZwypYCI+ANrgEHA340xX7dwzCxgFtC0TqDqHuXVdWwoPspPLx7S5nGJkSFMG9a7m6Jyj6G9I/jB+HReX7mb2yb2Z1BiuLtDUsrpnHLx1BhjMcZkAynAOSIyvIVjZhtjRhtjRickeO9FOG+0bJvt4ueUDO+5uciVfnhuf4xBpxpQPsupo2KMMUeAXOASZ5arHPO5fSk7b7746Ux9o3vRJyqENXt16KPyTc4YFZMgItH2x72Ai4ACR8tVztFosbJsWzmThyTg186MiD3JqH4xmtiVz3JGi70PsFRENgLfAEuMMfOcUK5ygrX7jlBd29g0jYCyyUmLoeTICQ4e1QWvle9xxqiYjcBIJ8SiXGBpYRkBfsJEL5vjxdVy+sUAsHbfYaaf3cfN0SjlXHrnqY9bWlDG6PQYInXdz1NkJkcSEuin3THKJ2li92H7j5yg4GB1i3eb9nSB/n5kpURrYlc+SRO7D8st1GGObcnpF8Pm/UepbbC4OxSlnEoTuw9bsaOC3pEhDNabcFqUkxZDg8Wwsfiou0NRyqk0sfsoYwxf7apk/MA4r5/jxVVG2S+ganeM8jWa2H3UttIaKo/VM36A98/x4iqxYUEMiA/TxK58jiZ2H7Vqp+12+fEDNbG3ZVS/GNbuO4xtLjulfIMmdh+1alclKTG9SI0NdXcoHi2nXwyHjtWzp/J4q8fsP3KCxZsPdmNUSjlGE7sPsloNX+8+pN0wHZDTgX72n767gVn/WcMbX7W9LqxSnkITuw/aerCKI8cbtBumAwYlhBMREtBqYl+77zCrdlUSHx7MIx/ls7SgrJsjVKrzNLH7oFU7KwHtX+8IPz9hVFoMa1tJ7C8s3Ul0aCAL7ptEZnIkd7+1lvwSHR6pPJsmdh+0amcl/ePD6BPVy92heIWcfjFsK6vm6ImGU7YXHqzm062l3DIhncSIEF77wRiiewVy2z+/oeTICTdFq1T7NLH7mEaLldW7DzFO+9c7bHS/GIyB9actC/hi7g5Cg/y5ZUI6YFth6vVbz+FEvYXbXv+GqtqGFkpTyv00sfuYzfurqK5r1G6YThiRGo2fnHoBdV/lceZuPMBNY9OIDv12ndihvSN46Xs57Cyv4c431lDfaHVHyEq1SRO7j1m1y9a/Pm5ArJsj8R5hwQGc1SeSNXsPNW17edlO/EX40bkDzjh+4qB4nro6ixU7Knli3pbuDFWpDtHE7mNW7axkUGI4iREh7g7Fq+T0i2H9viM0WqyUVdXy3ppirs5JISmy5ffxmpwUbh6Xxtur91FWrYt1KM+iid2L1DVamPPNPqpb6dttsFj5Zo+OX++KnH4xHKu3UFhazatf7qbRYuWO889srTd368T+NFoN7+UVd1OUSnWMJnYv8tTCAh58fxOPfLS5xf0bi49wvN7CBO1f77RRabYblZYWlPHGV3uZkZVMv7iwNp8zMCGcsf1jeeebfVitOiWB8hya2L3E0oIyXl+xh35xoXywroRFLdzifnL8+lhtsXdaSkwvEiOCee7zHRyrt3Dn5IEdet53x6ZRdOgEX+6ocHGESnWcw4ldRFJFZKmIbBWRzSJynzMCU98qq67lZ+9tIKN3BPPvPZfMPpH8+oNNHDpWf8pxq3ZVktE7gtiwoFZKUq0REXL6xVDXaOWCjETO6hPZoeddMrw3sWFBvPX1PhdHqFTHOaPF3gj81BhzFjAOuFtEMp1QrsI278tP391ATV0jz904kvDgAJ65dgRHTzTwyEf5TcfVNVrI23NYhzk6YEy6bSTRXR1srQMEB/hzTU4KS7aWUlalF1GVZ3A4sRtjDhhj1tofVwNbgb6OlqtsXluxm+XbK/jNjEwGJ0UAtoWY771gMPM2HmD+xgMArN93hLpGq144dcB3x6bx/p0TGJ3euaGiN4xJxWI1vLdGL6Iqz+DUPnYRSQdGAl+3sG+WiOSJSF55ebkzq/VZ+SVH+cMnBUzNTOKmsWmn7Ltz8kDO7hvFbz7Kp6KmjpU7K/ET7V93REigf9Nsj50xICGc8QPieHu1XkRVnsFpiV1EwoH3gfuNMVWn7zfGzDbGjDbGjE5ISHBWtT7reH0j976zjtiwIP5wddYZy9sF+Pvx7HUjqKlt5NcfbGLVrkqGJUcR1SvQTRH3bN8dm0bx4RMs266NFuV+TknsIhKILam/aYz5nzPK7OmemLeF3RXH+PN12cS0cjF0SFIED0wdwqLNpazefUj7191o2rDexOlFVOUhnDEqRoBXga3GmD85HpIqOXKCt1cX8cOJ/ZkwKL7NY3987gBGpkUDaP+6GwUF+HFNTgqfFZRRqhdRlZs5o8U+EfgecIGIrLf/THdCuT3WCvuY6GtGp7R7rL+f8NfrR3LLhHRtsbvZjeekYbEa3v2myN2hqB4uwNECjDFfAtLugarDVuyoID48iKH2UTDtSYsL5bGZw1wclWpPenwYEwfF8c43Rdw1ZRD+fvpnodxD7zz1MMYYVuyoZMLA+DMumCrP991z+lFy5ATLtulFVOU+mtg9zLbSGipq6pjUTt+68kwXZyYRHx7EO9/oRVTlPprYPczJOUcmDtbE7o2CAvyYkZXM0sJyXWFJuY0mdg+zckcF6XGh9I3W9Uq91czsZOobrSzKP3OiNqW6gyZ2J6upa+Tg0a4Nd2uwWPlqVyUTtRvGq41MjSY1thcfb9jv7lBUD6WJ3cnuf2c90/+2nGN1jZ1+7oaiIxyrt2j/upcTEb6TlczKnZVU1NS5OxzVA2lid6L8kqN8urWUQ8fqeXt15y+erdhRiQg6Ht0HzMxOxmI1LNh0wN2hqB5IE7sTvZC7g4jgAEamRfPK8l3UNVo69fwVOyoYnhxFdKjOp+7tMnpHMjQpgo/Xa3eM6n6a2J1ke2k1C/MP8oMJ6Txw8RBKq+r439qSDj//WF0ja/cd1v51HzIzO5m8vYcpOXLC3aGoHkYTu5P8fekOegX6c9uk/kwaFE9WShQvfbGTRou1Q89fvecQjVaj/es+5DtZyQDM1YuoqptpYneCPRXH+HjDfm4e14/YsCBEhLsmD2Rv5XEWdHDI24rtFQQF+DE6vfPzgSvPlBYXSnZqtHbHqG6nid0JXszdSYC/Hz86t3/TtqmZvRmYEMYLS3dgTPuLL6zYWcnofjGEBPq7MlTVzWaOSGbLgSp2lNW4OxTVg2hid1Dx4eO8v7aYG8ekkhgR0rTdz0+4c/IgCg5Ws7SwrM0yKmrq2HqgSvvXfdCMrD74CTqmXXUrTewOevmLXYjA7eefuQDy5dnJ9I3uxd+X7myz1b5yZyWAJnYflBgZwrgBcczdsL9D39yUcgZN7A4oraplTl4R1+SkkNzCFACB/n7MOm8Aa/YeZvXuQ62Ws3JHBREhAZzdN8qV4So3mTkimd0Vx8gvOWPFSKVcQhO7A15ZtguL1XDn+YNaPeb6ManEhwfxQu7OFvcbY1i+vYIJA+N0/m4fdenwPgT6Cx9v6PjwV6UcoYm9iypr6njz631cPiKZtLjQVo8LCfTn1on9+WJbOfklR8/Yv+/QcUqOnNBuGB8WFRrI+UMSmLvhAFardsco19PE3kV/X7qT2kYLd005s2/9dN8b34+I4ADuenMtzywqZM3eQ1jsf+Ardmj/ek/wnRHJHKyqZfWe1rvklHIWh5fG64kKD1bzr1V7uGFMGoMS21++LjIkkD9fn83sZbt48YudPL90B1G9AjlvSAL7Dh2nT1QIA+LDXB+4cpuLM5MICfRj4aYDjNNFx5WLOSWxi8hrwAygzBgz3BlleipjDI98lE9ESAC/mDa0w8+7KDOJizKTOHq8geU7yskttP1U1NRxw5hUXQbPx4UGBTBxYDxLC8t5zBj9vJVLOavF/k/geeDfTirPY328YT9f7z7E768cTkxY5yfrigoNZEZWMjOykrFaDdvLaugbo4tq9ASTMxL5rKCMXRXHGJgQ7u5wlA9zSh+7MWYZ4POdhzV1jTy5YCvD+0Zyw5g0h8vz8xOG9o4gPFh7xHqCyUMSAFha0PYNa0o5qtsunorILBHJE5G88nLXr+BusRrue2cdTy7YSkMHJ+Jqz3Ofbae0qo7fXj5chyaqTkuNDWVwYji5ha4//1XP1m2J3Rgz2xgz2hgzOiEhweX1vfblbj5av5/Zy3bxvVe/5tCxeofK21FWzatf7ua60SmMStOJulTXTMlI5OvdldR0YYUtpTrKJ4c77iir4enFhVycmcSfrx/B2n1HmPn8l2w90PKdf8YY1uw9xO/nb2HBpgNnLJBhjOGxj7cQGuTPg5dkdMdLUD5qytBEGiyGFTsq3B2K8mE+17nbaLHy0/c2EBrkz++vHE5iRAgD4sOZ9Z88rnphJX+6bgSXnt0HgLpGC/M2HOCfK/ewqeQoImDMbmJCA7lyZArXjUkho3ckC/MP8uWOCh6fOYy48GA3v0LlzUanxxAeHEBuYRnThvV2dzjKRzlruOPbwGQgXkSKgUeNMa86o+zOmr18FxuKjvDcjSObZlsckRrN3J9M4vY31nDnm2u5a/JAAvyEt1bvo6KmnkGJ4fzuiuFcnp3M2n1HeDeviDe+2strK3Zzdt8oSqtqOatPJDeNdfyCqerZAv39OHdwPEsLyjE67FG5iFMSuzHmRmeU46jCg9X8Zcl2pp/dmxlZfU7ZlxgZwjuzxvHwB/m8kLsTEbhgaCK3TuzPxEFxTX9g5w9J4PwhCRw+Vs+H60uY800Rh47V88JNowjw98meK9XNpgxNZGH+QbYeqCYzOdLd4Sgf5DNdMQ0WKz99bz0RIQE8cfnwFltCwQH+/PGaLKZn9aF/XBjpbdztGRMWxK0T+3PLhHSO11sI0yGJykkmD7UPeyws08SuXMJnmqAv5u4kv6SK310xvM1+cBFhytDENpP66cdrUlfOlBgZwvC+keS2swCLUl3lE4l98/6j/O2z7cwckdx0YVQpTzZlaCJr9h7m6PEGd4eifJBPJPYnF2wlOjSIx2cOc3coSnXI5KGJWA18sV1vVlLO5/WJ/dCxelbtrOSGMaldmrtFKXfITo0mJjSQXJ1eQLmA1yf2T7eWYjXomGDlVfz9hPOHJJC7rVwX31BO5/WJffHmg/SN7sXwvjq6QHmXKRmJHDpWz8YWVtZSyhFendiP1TWybHsFF2cm6Y0eyuucNzgBEfhcu2OUk3l1Yv9iWzn1jVbthlFeKSYsiJGp0acMe6ypa+ST/IM89P5Grn1pJZU1dW6MUHkrrx6gvWjzQWJCAxmTrrMtKu80ZWgizy7ZxnOfbWflzkry9h6iwWIICvCjvtHKun1HuCgzyd1hKi/jtS32+kYrnxeUcdFZSXqrv/JaF55lS9rPLtnG4eP13DapP+/MGseyn08BoOjwcXeGp7yU17bYV+2qpLq2UbthlFfLTI7kv3eMJzm6F8nR3y6RaIyhV6A/RYdOuDE65a28NrEv2nyQ0CB/Jg2Od3coSjlkdHrsGdtEhLTYUPYd0ha76jyv7MOwWg1LtpQyeWgCIYH+7g5HKZdIje1FsXbFqC7wysS+rugw5dV12g2jfFpKTChFh45jjN7ApDrHKxP7os2lBPoLUzIS3R2KUi6TGhvKsXoLh3WiMNVJXpfYjTEs2nyQ8QPjiQwJdHc4SrlMaoztYmqR9rOrTvK6xF5YWs3eyuNMG6Zje5VvS4sLBdALqD6ku66ZeF1iX5RfighcrDdtKB+XGmNL7DqW3TcsLSxj8tO5LNp80OV1OSWxi8glIlIoIjtE5CFnlNmaRZsPMiotpmmhaqV8VVhwALFhQTqW3QdsK63mnrfWMSQpgkmDXD9E2+HELiL+wN+BS4FM4EYRyXS03JYUHTrOlgNV2g2jeozUGB3y6O0OHavnh//6hpBAf/7xg9HdstSmM1rs5wA7jDG7jDH1wDvA5U4o9wwnv8LoMEfVU6ToTUperb7Ryh1vrKG0qo5Xvp9zyt3FruSMxN4XKGr2e7F92ylEZJaI5IlIXnl515YDO1FvYUx6DP3iOrYQtVLeLi02lP1HTmDRxTi8jjGGhz/cxOrdh3j6mixGpnXfZIXOSOwtTYR+xllojJltjBltjBmdkJDQpYruuXAw794+vkvPVcobpcaE0mAxHKyqdXcoqpNe/XI37+YVc88Fg7g8+4y2rks5I7EXA6nNfk8B9juh3BbpghqqJ0mN1bHs3uizraX8fsFWLh3em/+7aEi31++MxP4NMFhE+otIEHAD8LETylWqx2sa8qiJ3Wss21bOvW+vY1hyJM9eNwI/v+5vjDp8edYY0ygiPwEWAf7Aa8aYzQ5HppQiOboXIprYvcXbq/fx8If5DE4M59UfjCE0yD0T6DqlVmPMAmCBM8pSSn0rKMCP5KheFB3WseyezGo1/HFRIS99sZPzhyTw/HdHEuHGKU+8dj52pXqKlJhe2mL3YLUNFn767gbmbzrAd8dJMzQbAAAaUElEQVSm8duZw9y+qpsmdqU8XGpsKMu3d22IsHKtypo6fvzvPNbuO8Kvpmfw43MHeMQAD03sSnm41JhQSqvqqG2w6MIyHsIYw/xNB3hy/lYqj9Xz4k2juPTsPu4Oq4kmdqU83Mkhj8WHTzAoMdzN0aj8kqP8du4WVu85xFl9Innx5hxGpEa7O6xTaGJXysOlxX47y6Mmdvcpq67lmUWFvLemmNjQIJ688myuH5OKvxuGM7ZHE7tSHi7VntiL9QKq27zx1V6eWlhAXaOFH03qzz0XDvbohX40sSvl4RLCgwkK8NMhj24ye9lOnlxQwLmD4/nt5cPpH+/5c1VpYlfKw/n5iQ55dJN/LN/FkwsKmJHVh79cn+32YYwd5R1RKtXDpcbo9L3d7R/Ld/G7+Vu57GzvSuqgiV0pr5AWG6ot9m706pe7+d38rUw/uzd/ucG7kjpoYlfKK6TG9qKqtpGjJxrcHYrPe33Fbp6Yt4VLh/fmrzeMJNDLkjpoYlfKK+gsj93jvbwiHp+7hWnDkvjbjd6Z1EETu1Je4eSQR03srnO8vpGnFhZwTv9YnrtxlNcmddBRMUp5haYWuy5s7TJvfLWXymP1zL4kg6AA703qoC12pbxCVGggkSEBFB3SseyucLy+kdnLdnHu4Hhy+nXf2qSuooldKS+RGhuqLXYXefOrfVTU1HPfhYPdHYpTaGJXykukxuiQR1c4UW/h5WU7mTQontHpse4Oxyk0sSvlJVJjbSspWa3G3aH4lDe/3mtrrV/kG6110MSulNdIjQ2lvtFKeU2du0PxGSfqLbz0xS4mDIxjjI+01sHBxC4i14rIZhGxishoZwWllDqTDnl0vrdW76Oips5n+tZPcrTFng9cBSxzQixKqTbokEfnqm2w8NIXOxk/II6xA+LcHY5TOTSO3RizFfCINf6U8nUpMbaVlHTIo3O89fU+yqvreO7Gke4Oxem6rY9dRGaJSJ6I5JWX68K8SnVWSKA/iRHBOsujE5xsrY/tH8s4H2utQwcSu4h8KiL5Lfxc3pmKjDGzjTGjjTGjExISuh6xUj1Yqs7y6BSvr9hDWXUd9180xN2huES7XTHGmIu6IxClVPvSYkNZvfuQu8PwavklR/nTkkKmZiYxfqDvtdZBhzsq5VVSY3px4OgJahss7g7FKx2vb+Ted9YRGxbEH67Ocnc4LuPocMcrRaQYGA/MF5FFzglLKdWScQPjsBp4Zdkud4filX47dwu7K47x5+uyiQkLcnc4LuNQYjfGfGCMSTHGBBtjkowx05wVmFLqTBMGxnPZ2X14fukO9lVqX3tnLNh0gHe+KeKO8wcyYVC8u8NxKe2KUcrL/GZGJgF+wiMf52OMTi/QEfuPnOCh9zcyIiWKBy72zQumzWliV8rL9I4K4YGpQ8ktLOeT/IPuDsfjWayG++esx2I1XrvUXWf5/itUygf9YHw/MvtE8vjcLdTUNbo7HI9QXdtAdW0DltMmSXsxdwerdx/it5cPJz0+zE3RdS9dQUkpLxTg78fvrhzO1S+u5C9LtvHwjEx3h+RW20qrufSvy5uSekigH2FBAYQFB1By5AQzRyRz1ai+bo6y+2hiV8pLjUqL4YYxaby+cg9XjUohMznS3SG5zX/XFCPALy/NoLbByvH6RmrqGjleb+G8IfH84pKMHjX1iSZ2pbzYg5cMZfHmgzz84Sb+e8cE/Px6TvI6yWI1fLS+hMlDE7n9/IHuDscjaB+7Ul4sOjSIX04/i7X7jvDON0XuDsctVu2spLSqjitH9pyulvZoYlfKy109qi9j+8fy6w838b1Xv2buhv096s7UD9aVEBEcwIVnJbo7FI+hXTFKeTkR4cWbc/jnyj38N6+Ie95eR1SvQK7ITua6MakMS45yd4guc6Lewif5B5iRlUxIoL+7w/EYmtiV8gGxYUE8cPEQ7rtwMCt3VvBuXjFvf1PEv1btZfyAOJ65bgR9o3u5O0ynW7K1lGP1Fq7QbphTaGJXyof4+wnnDk7g3MEJHDlez3/XFPPnJdu49C/L+H9XZXFZVh93h+hUH64roU9UCGP7+856pc6gfexK+ajo0CB+dO4AFtx3Lv0Twrn7rbX84r8bOOYjNzRV1tTxxbZyLs/u2yNHA7VFE7tSPq5fXBj/vWM8d08ZyHtripnx3JdsKj7q7rAcNnfDfixWo6NhWqCJXakeINDfj59Py+CtH43jRL2Fq15cwftrit0dlkM+WL+fs/pEMrR3hLtD8Tia2JXqQcYPjOOT+89lVFoMD3+Y77VT/+4qr2FD0RGuHJns7lA8kiZ2pXqY6NAg/nx9Nv5+wi8/2OiVU/9+uH4/IjBzhHbDtEQTu1I9UHJ0L345PYMVOyqZ42V3rBpj+HBdCRMHxtM7KsTd4XgkTexK9VA3jklj3IBYfj9/KweP1ro7nA5bu+8w+w4d17HrbdDErlQP5ecn/OHqLBqsVh7+cJPXdMl8sK6EkEA/pg1LcncoHsvRxayfFpECEdkoIh+ISLSzAlNKuV6/uDB+NnUon24t4+MN+90dTrtKjpxg7oYDXJzZm4iQQHeH47EcbbEvAYYbY7KAbcAvHQ9JKdWdbp3Yn+zUaB6fu4XKmjp3h9OqbaXVXP3CSqzGcPt5A9wdjkdzKLEbYxYbY07exvYVkOJ4SEqp7uTvJ/zxmiyqaxt49OPN7g6nRWv2HuLal1ZhNYZ3bx/P8L6+O7GZMzizj/02YGFrO0VklojkiUheeXm5E6tVSjlqSFIE91wwmHkbD7C0sMzd4Zzis62l3PSPr4kNC+L9OydwVp+eu1JUR7Wb2EXkUxHJb+Hn8mbH/BpoBN5srRxjzGxjzGhjzOiEhATnRK+Ucpo7Jw8kOjSQTzYddHcoTd7LK2LWf9YwODGC9+4YT2psqLtD8grtzu5ojLmorf0i8gNgBnCh8ZbL6kqpMwT6+zEyNZp1RYfdHQoHjp7gja/28velO5k4KI6Xvzea8GCdjLajHHqnROQS4EHgfGOMd96brJRqkp0aQ+62cqprG7p91ElpVS0LNh1g/sYD5O21/ecyc0QyT1+bRXCALqLRGY7+F/g8EAwssa8A/pUx5g6Ho1JKucXItGiMgY3FR5k4KL5b6vy8oJSXcnfxzd5DGAMZvSP46cVDmJ7Vh4EJ4d0Sg69xKLEbYwY5KxCllPuNSLXdirJu3+FuSezri45wx3/Wkhwdwv0XDuGyrN4MStTZGh2lnVZKqSZRvQIZmBDGun1HXF7X4WP13P3mWhIigvngronEhAW5vM6eQqcUUEqdYmRaDOuLjrh0igGL1XDfnPWUV9fx4s2jNKk7mSZ2pdQpRqZFU3msnqJDJ1xWx3Ofb2fZtnIemzmMrBSdicTZNLErpU6RfbKf3UXDHnMLy/jrZ9u5elQKN56T6pI6ejpN7EqpUwxNiqBXoL9L+tmLDx/n/jnrGZoUwe+uGI59NJ1yMk3sSqlTBPj7kZUSxboi5yb2ukYLd725FovF8NLNOfQK0rHprqKJXSl1huy0aLbur6Ku0eK0Mp/7bAcbi4/y7HUjSI8Pc1q56kya2JVSZxiZGkO9xcrm/VVOKc9iNbybV8TFmUlMHdbbKWWq1mliV0qdYWTayRuVnNMd89WuSsqq67hSl7PrFprYlVJnSIoMITkqhPVO6mf/aH0J4cEBXJCR6JTyVNs0sSulWjQyLYZ1+xwf8ljbYGHhpoNcMrw3IYF6wbQ7aGJXSrVoZFo0xYdPUF7t2HJ5SwvKqK5r5PLsZCdFptqjiV0p1aKTNyq11h1TVdvAnG/2Ud9obbOcj9bvJyEimAkDu2e2SKWJXSnViuF9owjwk1a7Y379QT4Pvr+Jf3y5q9Uyjp5o4POCMr6TlYy/n96M1F00sSulWhQS6E9mcmSLI2MWbDrA3A37iQ0L4vnPd1BaVdtiGYvyD1JvsWo3TDfTxK6UalV2ajQbi49gsX4702NFTR0Pf5jP2X2jePf28TRaDH/4pKDF53+4voT+8WFkpUR1V8gKTexKqTaMTIvmWL2F7WXVABhj+M2H+dTUNvLsdSMYlBjOD8/tz//WlpzRZXPwaC2rdlUyc0SyzgnTzTSxK6VaNTI1BoD19u6YjzfsZ2H+Qf7v4iEMSbKtdHT3lEEkRgTz2NwtWJu17Odt3I8xaDeMG2hiV0q1ql9cKDGhgazbd4Syqloe+Wgz2anR/Pjc/k3HhAcH8OAlGWwoOsIH60qatn+4voSslCgG6Lql3c6hxC4iT4jIRhFZLyKLRUT/a1bKh4gI2anRrCs6zK8+2ERtg4VnrxtBgP+pqePKkX0ZkRrNU58UUFPXyI6yGvJLqrg8W6cQcAdHW+xPG2OyjDHZwDzgESfEpJTyICPTYthWWsOnW8v4+bShDGyhBe7nJzz2nUzKq+v4+9IdfLy+BD+B72T1cUPEyqHFrI0xzad+CwNct0iiUsotTt6oNCY9hlsn9m/1uJFpMVw1qi+vLt9NdGggEwbGkxgZ0l1hqmYc7mMXkd+LSBFwE2202EVklojkiUheeXm5o9UqpbrJ2AGx/HBSf/50XXa7Nxk9eEkGAf5CWXWdXjR1I2lvJXIR+RRoaQLlXxtjPmp23C+BEGPMo+1VOnr0aJOXl9fZWJVSXuDVL3fz96U7yP35ZCJDAt0djk8RkTXGmNHtHtdeYu9Ehf2A+caY4e0dq4ldKd9mtRr8dAoBp+toYnd0VMzgZr/OBFq+/Uwp1aNoUncvhy6eAk+JyFDACuwF7nA8JKWUUo5wdFTM1c4KRCmllHPonadKKeVjNLErpZSP0cSulFI+RhO7Ukr5GE3sSinlY5x2g1KnKhUpxzY8EiAN2NfCYVHA0Ra2xwMVnTjeGdtdXaczyvek96u1evX98t73q7XtnvR+tXa8J71frW3vaJ39jDEJLRx3KmOMW3+A8la2z25le14nj3d4ezfU6XD5nvR+tVavvl/e+3618T56zPvVhc/aY86BztbZ3o8ndMWcuVKuzdxOltPa8c7a3t1ld7Ycfb/0/XLXa3Jl2c54TZ72fjnjfWyTW7piTglAJM90YO6Drh7vDK6u05Xlu+P9cnW93lq2u+r11rLdVa8v5BhPaLHPdvHxzuDqOl1ZvjveL1fX661lu6teby3bXfV6fY5xe4tdKaWUc3lCi10ppZQTaWJXSikf47GJXURqurEui4isb/aT3saxk0VkXifKNiLyn2a/B4hIeWfK6GA9V9rrynBmuS3U0y2vx162y8+B9uoQkVwRcfiilis/HxH5tYhsFpGN9vN3rJPLTxGRj0Rku4jsFJG/ikhQG8ffLyKhDtRnROTZZr//TEQe62p5p5V98m99s4hsEJEHRKTb8mB35TWPTezd7IQxJrvZzx4nln0MGC4ivey/XwyUdKYAEenI9Mo3Al8CN3SybP/OHI8TXk8P1aXPpz0iMh6YAYwyxmQBFwFFTixfgP8BHxpjBgNDgHDg92087X6gy4kdqAOuEpF4B8pozcm/9WHYzt3pQLvLeXobj07sIhIuIp+JyFoR2SQil9u3p4vIVhF5xf4/7+JmicZZdfuLyNMi8o29JXR7s92RIvKBiGwRkZc68D/+QuAy++Mbgbeb1XOOiKwUkXX2f4fat98iIu+JyFxgcTuxhgMTgR9iTxz2bxbLWopTRGpE5Lci8jUwvuPvikOvZ7mIZDc7boWIZLVX0enfkETkeRG5xf54j4g83uz86FJruK06nKGNz6e11zVdRApE5EsR+Vs734b6ABXGmDoAY0yFMWa/iOSIyBciskZEFolIH3vZuSLyF/tnky8i57QT/gVArTHmdXv5FuD/gNtEJExEnrG/9xtF5B4RuRdIBpaKyNLOv1sANGIbJfJ/p+8QkX72nLDR/m+aiETZz4WT53eoiBSJSJsLrhpjyoBZwE/EptW/eRH5hf11bhCRp7r4uk6W5fq81pW7mrrjB6jBthBIpP33eGAHIEA6tg8/277vXeBmB+qyAOvtPx/Yt80CHrY/DgbygP7AZKAWGAD4A0uAa9p5HVnAf4EQex2TgXn2/ZFAgP3xRcD79se3AMVAbAfivxl41f54JTCqrTgBA1znwOfSldfzA+Av9sdDaOVOuxbqairbvu154Bb74z3APfbHdwH/6OLraauOXGC0g+dya5/PGXXa39MioL99+9vNj2uh7HD7Z7ANeAE4Hwi015NgP+Z64LVmr+cV++PzgPx2Yr8X+HML29cB9wHvN/u8Y5t9LvEOvF819vNoD7Zb6n8GPGbfNxf4gf3xbdi+SQB8BExp9npbPBeAmha2HQaSaP1v/lL7+xna/HU68Npcntc8usWO7cU+KSIbgU+Bvtg+AIDdxpj19sdrsL0pXdW8K+ZK+7apwPdFZD3wNRAHnFzjdbUxZpextV7eBia1VbgxZqM9vhuBBaftjgLeE5F84M/AsGb7lhhjDnUg/huBd+yP37H/3lacFmx/kF3SxdfzHjDD3oq6DfhnV+s/zf/s/zp6DrhSa59PSzKAXcaY3fbf327jWIwxNUAOtqRUDswBbgeGA0vs5+/DQEqzp71tf+4ybN8+o9uoQrA1BFrafh7wkjGm0V5eR87VDjHGVAH/xvYfS3Pjgbfsj//Dt+f0HGwJHWzfiuZ0orqTC7S29jd/EfC6Mea4PTZHX6fL85qja5662k1AApBjjGkQkT3YWjRg64c7yQI4tSsG25t/jzFm0SkbRSZz5onekZsBPgaewdZSi2u2/QlgqTHmSrFdtM1ttu9Yu0GKxGH7ujxcRAy21rnBlnBbi7PWnuwd0anXY4w5LiJLgMuB64COXpBs5NQuw5DT9p88Dyx0/Xxur44ua+Pz+biVOju9CrT9s8wFckVkE3A3sNkY01o3W2fO383AKUtgikgkkArsaue5jvoLsBZ4vY1jTtb/MfD/RCQW2390n3ekAhEZgO3cKaP1v/lLcO7rdHle8/QWexRQZn/xU4B+3Vj3IuDOk/10IjJERMLs+84Rkf72Pr3rsV0Ua89rwG+NMZtO2x7Ftxcfb+lCnNcA/zbG9DPGpBtjUoHd2FoyXYmzo7ryev4B/A34phOtnr1ApogEi0gUcGEX43VXHa19PrRSZwEwQL4dmXU9bRCRoSIyuNmmbGArkCC2C6uISKCINP8meL19+yTgqDGmpRkIT/oMCBWR79uf4w88i+0b12LgDrFf3LcnVYBqIKKtuDvCfo68i+3axEkr+fYC9E3Yz2n7N5fVwF+xdV2123ARkQTgJeB5Y+v7aO1vfjG2awqh9u2xrZXZQS7Pax7ZYrefKHXAm8BcEcnD1o9Y0I1h/APb16C1IiLYvuZeYd+3CngKOBtYBnzQXmHGmGJsJ93p/gj8S0QeoIOtjNPcaI+lufeBO7sSZ0d15fUYY9aISBVtt8CAb88BY0yRiLwLbAS2Y+vbdYruqIPWP5/vYktap9RpjDkhIncBn4hIBbZk1ZZw4Dl7d0ojtv7aWdguPv7N/p9GALbW72b7cw6LyEps/di3tVW4McaIyJXACyLyG2yNwQXAr7C1KIcAG0WkAXgF27WC2cBCETlgjJnSTvzteRb4SbPf7wVeE5GfY/ubvLXZvjnYuvwmt1FeL3tXSyC29+s/wJ/s+1r8mzfGfCK2C/95IlLPt6+/U7ozr3nklAIiMgLbBZ72rtirVti7jH5mjJnh7lhOEpFkbF0GGcYYazvHuvwc8NTzTETCjTE19uTyd2C7MebPTio7F9t5keeM8lTHdef55nFdMSJyB7aLOw+7OxblPPav8l8Dv+5AUnf5OeDh59mP7a3Kzdi+tr/s5niUg7r7fPPIFrtSSqmu87gWu1JKKcd4RGIXkddEpMw+9vnkthEissp+Z9Zc+xCrk/uy7Ps22/eH2Lfn2H/fIbY79jo9dEwppZzBiXnt92K7k7bD88x4RGLHNnTqktO2/QN4yBhzNrbRHD+HpivLbwB3GNt8D5OBBvtzXsQ2ImCw/ef0MpVSqrv8E+fktblApy64ekRit98Bd/q45qHYhuiB7Xb4kzdJTAU2GmM22J9baYyxiG0ujEhjzCr7mNR/8+3wRKWU6lbOyGv2x18ZYw50pm6PSOytyAdm2h9fi+1ON7CNmzVim9horYj8wr69L7a5VU4qtm9TSilP0dm81iWenNhvA+4WkTXY7mKrt28PwHZX5U32f68UkQtp+VZsHfKjlPIknc1rXeKRd54CGGMKsH09QUSG8O00scXAF8aYCvu+Bdhmy3uDUyc6SgH2d1vASinVji7ktc+6Uo/HtthFJNH+rx+2Qf0v2XctArLENudyALZpSrfY+6CqRWScfTTM97FN5amUUh6hs3mtq/V4RGIXkbexzWsyVESKReSHwI0isg3bPAr7sc8vYow5jG1uh2+wzbOw1hgz317UndiuOu8AdmJbEEIppbqds/KaiPxRRIqxTcZWLB1YJlDvPFVKKR/jES12pZRSzqOJXSmlfIwmdqWU8jGa2JVSysdoYldKKR+jiV31SCLymIj8rI39V4hIZnfGpJSzaGJXqmVXAJrYlVfSceyqxxCRX2O7I7kI20LFa4Cj2KZ6DsJ2Y9v3gGxgnn3fUb6dge/vQAJwHPix/fZwpTyOJnbVI4hIDrb5scdimyNpLbbbuV83xlTaj/kdUGqMeU5E/gnMM8b8177vM2xzZW8XkbHA/zPGXND9r0Sp9nnsJGBKOdm5wAfGmOMAIvKxfftwe0KPBsKxzdlxChEJByYA7zVblCvY5REr1UWa2FVP0tLX038CVxhjNojILdhWrjmdH3DEGJPtutCUch69eKp6imXY5rjuJSIRwHfs2yOAAyISiG0u7JOq7fswxlQBu0XkWgCxGdF9oSvVOdrHrnqMZhdP92Kb/3oLcAz4hX3bJiDCGHOLiEwEXgHqgGsAK7Y1dfsAgcA7xpjfdvuLUKoDNLErpZSP0a4YpZTyMZrYlVLKx2hiV0opH6OJXSmlfIwmdqWU8jGa2JVSysdoYldKKR/z/wFzKBBPschO3gAAAABJRU5ErkJggg==\n",
"text/plain": [ "text/plain": [
"<Figure size 432x288 with 1 Axes>" "<Figure size 432x288 with 1 Axes>"
] ]
...@@ -855,125 +907,115 @@ ...@@ -855,125 +907,115 @@
} }
], ],
"source": [ "source": [
"yf.plot()" "data_1y.plot()"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 22, "execution_count": 58,
"metadata": { "metadata": {},
"hideCode": true,
"hidePrompt": true
},
"outputs": [], "outputs": [],
"source": [ "source": [
"from lmfit.model import Model, save_model\n",
"\n",
"def mysine(x, amp, freq, shift):\n", "def mysine(x, amp, freq, shift):\n",
" return amp * np.sin(x*freq + shift)" " return amp * np.sin(2*np.pi*x*freq) + shift"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 32, "execution_count": 59,
"metadata": { "metadata": {},
"hideCode": true, "outputs": [],
"hidePrompt": true
},
"outputs": [
{
"ename": "NotImplementedError",
"evalue": "guess() not implemented for Model",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNotImplementedError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-32-5a291762f4b2>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0msinemodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mModel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmysine\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2\u001b[0m \u001b[0mpars\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msinemodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmake_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mamp\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfreq\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.25\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mshift\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mparams\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msinemodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mguess\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'untrend_data'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0mresult\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msinemodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'untrend_data'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/opt/conda/lib/python3.6/site-packages/lmfit/model.py\u001b[0m in \u001b[0;36mguess\u001b[0;34m(self, data, **kws)\u001b[0m\n\u001b[1;32m 737\u001b[0m \u001b[0mcname\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__class__\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 738\u001b[0m \u001b[0mmsg\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'guess() not implemented for %s'\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0mcname\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 739\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mNotImplementedError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmsg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 740\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 741\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_residual\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mparams\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mweights\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mNotImplementedError\u001b[0m: guess() not implemented for Model"
]
}
],
"source": [ "source": [
"sinemodel = Model(mysine)\n", "sinemodel = Model(mysine)\n",
"pars = sinemodel.make_params(amp=1, freq=0.25, shift=0)\n", "params = sinemodel.make_params(amp=7, freq=1, shift = 0)"
"params = sinemodel.guess(data['untrend_data'], x=time)\n",
"\n",
"result = sinemodel.fit(data['untrend_data'], params, x=time)"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 24, "execution_count": 60,
"metadata": { "metadata": {},
"hideCode": true, "outputs": [],
"hidePrompt": true "source": [
}, "result = sinemodel.fit(data_1y['untrend_data'], params, x=time_1y)"
]
},
{
"cell_type": "code",
"execution_count": 61,
"metadata": {},
"outputs": [],
"source": [
"coeffs = result.params.valuesdict()"
]
},
{
"cell_type": "code",
"execution_count": 62,
"metadata": {},
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": [ "text/plain": [
"OrderedDict([('amp', 53.93685622928183),\n", "OrderedDict([('amp', 2.913657292010378),\n",
" ('freq', 0.3543795056268845),\n", " ('freq', 1.0733831551163455),\n",
" ('shift', 0.06850006173238188)])" " ('shift', 0.8693924761163696)])"
] ]
}, },
"execution_count": 24, "execution_count": 62,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
} }
], ],
"source": [ "source": [
"# Affichage des meilleures paramètres calculés par lmfit\n", "coeffs"
"coeffs_sine = result.params.valuesdict()\n",
"coeffs_sine"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 25, "execution_count": 63,
"metadata": { "metadata": {},
"hideCode": true,
"hidePrompt": true
},
"outputs": [], "outputs": [],
"source": [ "source": [
"def fitted_sine_curve(time):\n", "def fitted_curve(time):\n",
" year_CO2 = coeffs_sine.get('amp')*np.sin(time*coeffs_sine.get('freq') + coeffs_sine.get('shift'))\n", " CO2_1y = coeffs.get('amp')*np.sin(2*np.pi*coeffs.get('freq')*time) + coeffs.get('shift')\n",
" return year_CO2" " return np.round(CO2_1y, 2)"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 26, "execution_count": null,
"metadata": { "metadata": {},
"hideCode": true, "outputs": [],
"hidePrompt": true "source": []
}, },
{
"cell_type": "code",
"execution_count": 64,
"metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"# Ajout d'une colonne dans le dataframe data\n", "# Ajout d'une colonne dans le dataframe data\n",
"data['year_CO2'] = pd.Series(fitted_sine_curve(time), index=data.index)" "data_1y['fit'] = pd.Series(fitted_curve(time_1y), index=data_1y.index)"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 31, "execution_count": 65,
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
{ {
"data": { "data": {
"text/plain": [ "text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x7f8521627908>" "<matplotlib.axes._subplots.AxesSubplot at 0x7f5b08dc5a20>"
] ]
}, },
"execution_count": 31, "execution_count": 65,
"metadata": {}, "metadata": {},
"output_type": "execute_result" "output_type": "execute_result"
}, },
{ {
"data": { "data": {
"image/png": "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\n", "image/png": "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\n",
"text/plain": [ "text/plain": [
"<Figure size 432x288 with 1 Axes>" "<Figure size 432x288 with 1 Axes>"
] ]
...@@ -985,47 +1027,48 @@ ...@@ -985,47 +1027,48 @@
} }
], ],
"source": [ "source": [
"simul = data['trend_fit'] + data['year_CO2']\n", "data_1y.plot()"
"simul.plot()"
] ]
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": null,
"metadata": { "metadata": {},
"hideCode": true,
"hidePrompt": true
},
"outputs": [], "outputs": [],
"source": [] "source": []
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": null,
"metadata": { "metadata": {},
"hideCode": true,
"hidePrompt": true
},
"outputs": [], "outputs": [],
"source": [] "source": []
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": null,
"metadata": { "metadata": {},
"hideCode": true,
"hidePrompt": true
},
"outputs": [], "outputs": [],
"source": [] "source": []
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": null,
"metadata": { "metadata": {},
"hideCode": true, "outputs": [],
"hidePrompt": true "source": []
}, },
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [], "outputs": [],
"source": [] "source": []
} }
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment