870 lines
724 KiB
Plaintext
870 lines
724 KiB
Plaintext
|
{
|
|||
|
|
"cells": [
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "aea917db",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"# Darts: Quickstart\n",
|
|||
|
|
"https://unit8co.github.io/darts/quickstart/00-quickstart.html "
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 29,
|
|||
|
|
"id": "initial_id",
|
|||
|
|
"metadata": {
|
|||
|
|
"ExecuteTime": {
|
|||
|
|
"end_time": "2024-04-15T10:48:48.119781Z",
|
|||
|
|
"start_time": "2024-04-15T10:48:46.349468Z"
|
|||
|
|
},
|
|||
|
|
"collapsed": true
|
|||
|
|
},
|
|||
|
|
"outputs": [],
|
|||
|
|
"source": [
|
|||
|
|
"%matplotlib inline\n",
|
|||
|
|
"\n",
|
|||
|
|
"import pandas as pd\n",
|
|||
|
|
"import numpy as np\n",
|
|||
|
|
"import matplotlib.pyplot as plt\n",
|
|||
|
|
"\n",
|
|||
|
|
"from darts import TimeSeries\n",
|
|||
|
|
"from darts.datasets import AirPassengersDataset\n",
|
|||
|
|
"\n",
|
|||
|
|
"series = AirPassengersDataset().load()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "f7794de6",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"# Разделение данных"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 3,
|
|||
|
|
"id": "dd77e711c2745843",
|
|||
|
|
"metadata": {
|
|||
|
|
"ExecuteTime": {
|
|||
|
|
"end_time": "2024-03-31T10:28:06.776388100Z",
|
|||
|
|
"start_time": "2024-03-31T10:28:06.396651400Z"
|
|||
|
|
},
|
|||
|
|
"collapsed": false
|
|||
|
|
},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 3,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"# по процентам\n",
|
|||
|
|
"series1, series2 = series.split_before(0.75)\n",
|
|||
|
|
"series1.plot()\n",
|
|||
|
|
"series2.plot()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 4,
|
|||
|
|
"id": "425207ef",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 4,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGvCAYAAACJsNWPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/TGe4hAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB8BUlEQVR4nO3dd3xUVfo/8M9k0ntCQiCFEHoTpCgCIlXpEpAOK0QEV3+7fFcXQV1ZaTZQFHVtCMTVIALSorFRZOlSpSMkEJKQkEAK6ZNk5vfHcO/cm0ySuVNTPu/Xyxd35t7ce+YIzMNznnOOSqfT6UBERETkIE6ObgARERE1bgxGiIiIyKEYjBAREZFDMRghIiIih2IwQkRERA7FYISIiIgcisEIERERORSDESIiInIoBiN1kFarxbVr16DVah3dlHqB/aUM+8t07Ctl2F/KsL8MGIwQERGRQzEYISIiIodiMEJEREQOxWCEiIiIHIrBCBERETkUgxEiIiJyKAYjRERE5FAMRoiIiMihGIwQERGRQzEYISIiIodiMEJEREQOxWCEiIiIHIrBCBERETkUg5EGomXLlnj//fdNvv63336DSqVCbm6uzdpERERkCmdHN6AxGzhwIO6//35FQUR1jh07Bi8vL5Ov79u3L9LT0+Hn52fxs4mIyHFy83XYth8Y0hNoEaJydHPMwmCkDtPpdCgvL4ezc+3/m4KDgxXd29XVFc2aNTO3aUREVEfMfUeHzXuBjpHAuS8BJ6f6F5BwmMZBZs2ahX379mH16tVQqVRQqVSIjY2FSqXCjz/+iMcffxweHh44cOAAEhMTMXbsWISEhMDb2xsPPPAAdu3aJbtf5WEalUqFL774AuPGjYOnpyfatm2LnTt3iucrD9PExsbC398fP//8Mzp27Ahvb28MHz4c6enp4s+Ul5dj3rx58Pf3R5MmTbBw4ULMnDkT0dHRtuwqIiKqwekr+l8vJgMnLju2LeZqkJmRXr16ISMjw+7PbdasGY4fP27StatXr8aff/6JLl26YOnSpQCA8+fPAwBeeeUVvPjii+jTpw+aNGmClJQUjBw5Eq+//jrc3Nzw3//+F2PGjMHly5fRokWLap+xZMkSrFixAitXrsSHH36I6dOnIzk5GYGBgUavLyoqwjvvvIOvvvoKTk5OmDFjBubPn4+4uDgAwNtvv424uDisX78eHTt2xOrVq7F9+3YMGjRISTcREZEV3S0yHO84oMMDHetfZqRBBiMZGRlIS0tzdDNq5OfnB1dXV3h6eorDJZcuXQIALF68GPfffz8iIyPh5OSEwMBAdOvWTfzZZcuWYdu2bdi5cyf+9re/VfuMWbNmYerUqQCAN954Ax988AF+//13DB8+3Oj1ZWVl+PTTT9G6dWsAwN/+9jcxUAKADz/8EC+//DLGjRsHAPjoo4+QkJBgQS8QEZGl7hYajrcfAJbPcVxbzNUggxFH1UJY67m9evVCeXm5+LqgoACLFy/GDz/8gPT0dJSXl6O4uBg3btyo8T5du3YVj728vODr64vMzMxqr/f09BQDEQBo3ry5eH1eXh5u3bqFBx98UDyvVqvRs2dPaLVaxZ+RiIgsV1auQ3Gp4fX5a0Bimg6tw+pXdqRBBiOmDpXUVV5eXsjLyxNfz58/H7/++iveeecdtGnTBh4eHpgwYQI0Gk2N93FxcZG9VqlUNQYOxq7X6XRmfAIiIrKH/KKq7+04ALww2f5tsQQLWB3I1dUVFRUVtV538OBBzJo1C+PGjcN9992HZs2a4fr167ZvoISfnx9CQkJw7Ngx8b2KigqcPHnSru0gIiID6RCNYMeB+vePyAaZGakvWrZsiaNHj+L69evw9vauNmvRtm1bbN26FWPGjIFKpcKiRYscMjTy97//HW+++SbatGmDDh064MMPP0ROTg5UqvqVDiQiaijuGsmMHDgL3M7VIci//vzdzMyIA82fPx9qtRqdOnVCcHBwtTUgq1atQkBAAPr27YsxY8Zg2LBh6NGjh51bCyxcuBBTp07Fk08+iT59+sDb2xvDhg2Du7u73dtCRETyzIjTvW90rRb4/rBj2mMulY5FAXWOVqtFcnKyOJumrtJqtejYsSMmTZqEZcuWObQd9aG/6gr2l+nYV8qwv5SxRn8lHNZh1EL91/hjDwC/3BtJj+4PbHu9/vw/4DANmSw5ORm//PILBgwYgNLSUnz00Ue4du0apk2b5uimERE1SnmSzMiwB1U4fVWHzBzg59+B4lIdPNzqx1BN/QmbyOGcnJwQGxuLBx54AP369cPZs2exa9cudOzY0dFNIyJqlKTDNAE++uwIABSXAn+mOKZN5jArM/Lll19i06ZNyM/PR3h4ONasWQMvLy/Exsbi66+/hlarxdixYzFv3jyxuPH8+fNYtmwZUlJS0LlzZyxZsgTNmze36och24qIiMDBgwcd3QwiIrpHWsDq6wk09Te8Njbtt65SnBnZtGkTDh8+jLVr12Lfvn1YsmQJXFxccODAAWzevBmxsbHYtGkTDh06hB07dgAANBoNFixYgClTpmDPnj3o1q0bFi1aZPUPQ0RE1JjcLTSUffp6AT6ehnP1KRhRlBmpqKjAunXr8MUXX4irjbZt2xYAkJCQgHHjxiE8PBwAMGPGDMTHxyM6OhonTpyAi4uLuKHa7NmzMWTIEKSlpSEsLKzKczQaTZUFvZydneHq6qr4A9ZHwrRdrmxqGvaXMuwv07GvlGF/KWON/pLWjHh76ODtIT2ng1br+DkqphTnKgpGMjMzUVJSgl27dmHDhg3w9vbGX/7yF4wbNw7Xrl3DsGHDxGvbtGmDxMREAEBSUpIYtACAu7s7wsPDkZSUZDQYWb9+PdasWSN7b+LEiZg0aZKS5tZ7KSn1aMCvDmB/KcP+Mh37Shn2lzKW9NfNW00AeAMACvLSoCl2B9AEAHA95Q6Skwus0ELLREVF1XqN4mCkoKAAN27cwM6dO5GSkoJnn30WLVu2RFFREby8vMRrvby8UFxcDAAoLi6WnRPOFxUZzyHFxMRg+vTp8oY2ssxISkoKIiIiOD3OBOwvZdhfpmNfKcP+UsYa/aWV/FiHNmHIlHytunk0QWRkEwtbaR+KghE3NzcAwJw5c+Du7o62bdvisccew8GDB+Hp6YnCQkO+qLCwEB4e+nyRh4eH7Jxw3tPTE8a4uro2msCjJk5OTvwDrQD7Sxn2l+nYV8qwv5SxpL/yiwxDPP4+Kvh5AYB+aKagRAUnpwY4tTcyMhIuLi6y5b+F46ioKFy9elV8PzExUdwBtlWrVrJzJSUlSE1NRatWrSxqfGPXsmVLvP/+++JrlUqF7du3V3v99evXoVKpcPr0aYuea637EBGRZYSpvSoV4OVeuYDV8fUiplIUjHh4eGDIkCFYu3YtNBoNrl27hl9//RX9+vXDyJEjsXXrVqSmpuLOnTuIi4vDyJEjAQA9e/ZEaWkpduzYAY1Gg3Xr1qFjx45G60XIfOnp6RgxYoRV7zlr1iyx8FgQERGB9PR0dOnSxarPIiIiZYRgxMcTcHJSwUdSwNpgZ9MA+v1Jli5diqFDh8Lf3x9//etf0b17dwDAhAkTMHPmTGi1WkRHR2Ps2LEA9MMuK1euxLJly7BixQp06tTJocuHN1TCDCdbU6vVdnsWERFVT1hnxPdeRqS+Tu1VPEjl4+ODlStX4n//+x927tyJ8ePHi+diYmKwe/du7N27F//3f/8nG87p3LkzNm7ciIMHD2LNmjWNfsGzzz//HKGhoVWmdI0dOxazZ89GcnIyoqOjERISAm9vbzzwwAPYtWtXjfesPEzz+++/o3v37nB3d0evXr1w6tQp2fUVFRWYPXs2oqKi4OHhgfbt22P16tXi+cWLF+PLL7/Ejh07oFKpoFKp8Ntvvxkdptm3bx8efPBBuLm5oXnz5njppZdQXl4unh84cCDmzZuHBQsWIDAwEM2aNcPixYuVdxwREYmEqb1++gk1jScYIeuYOHEi7ty5g71794rvZWdn46effsK0adNQVFSEESNGYPfu3Th16hSGDx+OMWPGVLuzb2UFBQUYPXo0OnXqhBM
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"# по дате\n",
|
|||
|
|
"train, val = series.split_before(pd.Timestamp(\"19580101\"))\n",
|
|||
|
|
"train.plot(label=\"training\")\n",
|
|||
|
|
"val.plot(label=\"validation\")"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "2457080a",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"# Простая обработка"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 5,
|
|||
|
|
"id": "2aa69bb5273f2cf2",
|
|||
|
|
"metadata": {
|
|||
|
|
"ExecuteTime": {
|
|||
|
|
"end_time": "2024-03-28T11:33:00.911470500Z",
|
|||
|
|
"start_time": "2024-03-28T11:33:00.615384400Z"
|
|||
|
|
},
|
|||
|
|
"collapsed": false
|
|||
|
|
},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 5,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"series_noise = TimeSeries.from_times_and_values(\n",
|
|||
|
|
" series.time_index, np.random.randn(len(series))\n",
|
|||
|
|
")\n",
|
|||
|
|
"(series / 2 + 20 * series_noise - 10).plot()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 6,
|
|||
|
|
"id": "61b68acb9eeb8982",
|
|||
|
|
"metadata": {
|
|||
|
|
"ExecuteTime": {
|
|||
|
|
"end_time": "2024-03-28T11:33:03.886439200Z",
|
|||
|
|
"start_time": "2024-03-28T11:33:03.561839700Z"
|
|||
|
|
},
|
|||
|
|
"collapsed": false
|
|||
|
|
},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 6,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"series.map(np.log).plot()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 7,
|
|||
|
|
"id": "31f0b660fb1bbb41",
|
|||
|
|
"metadata": {
|
|||
|
|
"ExecuteTime": {
|
|||
|
|
"end_time": "2024-03-28T12:00:34.236946800Z",
|
|||
|
|
"start_time": "2024-03-28T12:00:33.929114600Z"
|
|||
|
|
},
|
|||
|
|
"collapsed": false
|
|||
|
|
},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 7,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"series.map(np.sqrt).stack(series).plot()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 28,
|
|||
|
|
"id": "11c98812823b3e9b",
|
|||
|
|
"metadata": {
|
|||
|
|
"ExecuteTime": {
|
|||
|
|
"end_time": "2024-03-31T10:46:17.602162900Z",
|
|||
|
|
"start_time": "2024-03-31T10:46:17.108542400Z"
|
|||
|
|
},
|
|||
|
|
"collapsed": false
|
|||
|
|
},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 28,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"series.map(lambda ts, x: x / ts.days_in_month).stack(series).plot()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 9,
|
|||
|
|
"id": "9235dd3d796f2581",
|
|||
|
|
"metadata": {
|
|||
|
|
"collapsed": false
|
|||
|
|
},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='time'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 9,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"#Восстановление пропущенных значений\n",
|
|||
|
|
"\n",
|
|||
|
|
"from darts.utils.missing_values import fill_missing_values, extract_subseries\n",
|
|||
|
|
"\n",
|
|||
|
|
"values = np.arange(50, step=0.5)\n",
|
|||
|
|
"values[10:30] = np.nan\n",
|
|||
|
|
"values[60:95] = np.nan\n",
|
|||
|
|
"series_ = TimeSeries.from_values(values)\n",
|
|||
|
|
"\n",
|
|||
|
|
"\n",
|
|||
|
|
"fill_missing_values(series_).plot(label=\"without missing values\")\n",
|
|||
|
|
"(series_).plot(label=\"with missing values (shifted below)\")\n",
|
|||
|
|
"# extract_subseries(series_, mode=\"any\") # выбирает существующие отрезки"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 35,
|
|||
|
|
"id": "aa5e2e40",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 35,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"# проверка восстановления более сложного ВР -> восстановление значений линейно между последними значениями\n",
|
|||
|
|
"series_ = series.copy()\n",
|
|||
|
|
"values_ = series_.all_values(copy=False)\n",
|
|||
|
|
"values_[10:30] = np.nan\n",
|
|||
|
|
"values_[60:102] = np.nan\n",
|
|||
|
|
"\n",
|
|||
|
|
"\n",
|
|||
|
|
"fill_missing_values(series_).plot(label=\"without missing values\")\n",
|
|||
|
|
"(series).plot(label=\"with missing values (shifted below)\")"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "c3ac3894",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"# Forecasting "
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 10,
|
|||
|
|
"id": "f7fd1585",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 10,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"#наивное прогнозирование\n",
|
|||
|
|
"from darts.models import NaiveSeasonal\n",
|
|||
|
|
"\n",
|
|||
|
|
"naive_model = NaiveSeasonal(K=1)\n",
|
|||
|
|
"naive_model.fit(train)\n",
|
|||
|
|
"naive_forecast = naive_model.predict(36)\n",
|
|||
|
|
"\n",
|
|||
|
|
"#подбор помог сильно повысить точность благодаря сильно выраженной сезонности \n",
|
|||
|
|
"naive_model_podbor = NaiveSeasonal(K=12)\n",
|
|||
|
|
"naive_model_podbor.fit(train)\n",
|
|||
|
|
"naive_forecast_podbor = naive_model_podbor.predict(36)\n",
|
|||
|
|
"\n",
|
|||
|
|
"series.plot(label=\"actual\")\n",
|
|||
|
|
"naive_forecast.plot(label=\"naive forecast (K=1)\")\n",
|
|||
|
|
"naive_forecast_podbor.plot(label=\"naive forecast (K=12)\")"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 11,
|
|||
|
|
"id": "ab2da4d5",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 1000x500 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"from darts.utils.statistics import plot_acf, check_seasonality\n",
|
|||
|
|
"\n",
|
|||
|
|
"plot_acf(train, m=12, alpha=0.05, max_lag=24)"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 12,
|
|||
|
|
"id": "560889da",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"name": "stdout",
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"There is seasonality of order 12.\n"
|
|||
|
|
]
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"# проверка сезонности\n",
|
|||
|
|
"for m in range(2, 25):\n",
|
|||
|
|
" is_seasonal, period = check_seasonality(train, m=m, alpha=0.05)\n",
|
|||
|
|
" if is_seasonal:\n",
|
|||
|
|
" train_period = period\n",
|
|||
|
|
" print(\"There is seasonality of order {}.\".format(period))"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 13,
|
|||
|
|
"id": "898483cd",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 13,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGvCAYAAACJsNWPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/TGe4hAAAACXBIWXMAAA9hAAAPYQGoP6dpAACO00lEQVR4nO3dd3xT5f4H8E9GR7o3nZQyC2UKyhYZUtlDpqBQUVR+XpwX0Asq4kBRFDcitF6tIHsogjLkggxZBQERaEsHUFq6m7ZJm5zfH+E8OSdN2iTN6Pi+X6/7uifJafLkgJxPv8+ScBzHgRBCCCHESaTObgAhhBBCmjcKI4QQQghxKgojhBBCCHEqCiOEEEIIcSoKI4QQQghxKgojhBBCCHEqCiOEEEIIcSoKI4QQQghxKgojDZBWq0V6ejq0Wq2zm9Io0PWyDF0v89G1sgxdL8vQ9dKjMEIIIYQQp6IwQgghhBCnojBCCCGEEKeiMEIIIYQQp6IwQgghhBCnojBCCCGEEKeiMEIIIYQQp6IwQgghhBCnojBCCCGEEKeiMEIIIYQQp6IwQgghhBCnojBCCCGEEKeiMEIIIYQQp6IwQmqYPXs2xo8f7+xmEEIIaSYojDRSb7zxBrp37+7sZhBCCHGyoqIiJCYmIjMz09lNsRqFEUIIIaQRmzt3Lh5//HE89NBD0Gq1zm6OVSiMONGePXswYMAA+Pn5ITAwEKNHj0Zqaip7PTs7G9OnT0dAQAA8PT3Rq1cvnDhxAklJSVi6dCnOnTsHiUQCiUSCpKQkXL9+HRKJBCkpKew9ioqKIJFI8PvvvwMANBoN5syZg5iYGCgUCnTo0AGrVq1y8DcnhBBiK/y/+X///TdOnz7t3MZYSe7sBthDr169kJOT4/DPDQ0NxalTp8w+X6lU4sUXX0TXrl1RVlaG1157DRMmTMCZM2egVCoxbtw4REREYOfOnQgNDcWZM2eg1WoxdepUXLhwAXv27MG+ffsAAL6+vrh9+3adn6nVahEZGYlNmzYhMDAQR48exdy5cxEWFoYpU6ZY/d0JIYQ4R0lJCTvesWMH7r33Xie2xjpNMozk5OTgxo0bzm5GnR5++GHR43Xr1iE4OBiXLl3CTz/9hLy8PJw8eRIBAQEAgLZt27Jzvby8IJfLERoaatFnuri4YOnSpexxTEwMjh07ho0bN1IYIYSQRkgYRrZv34633nrLia2xTpMMI5beoJ31uVevXsVrr72GEydO4M6dO6yvLzMzE5cuXUKPHj1YELGlzz//HOvWrUNmZiYqKiqgVqtpMCwhhDRCVVVVqKioYI8vXryI1NRUtGnTxomtslyTDCOWdJU405gxYxAdHY01a9YgPDwcWq0WnTt3hlqthru7u8XvJ5XqhgBxHMeeq6qqEp2zYcMGvPzyy/jwww/Rt29feHt7Y8WKFThx4kT9vgwhhBCHKy0trfHcjh078OKLLzqhNdajAaxOkp+fj3/++QeLFy/G0KFD0bFjRxQWFrLXY2NjkZKSgoKCAqM/7+rqCo1GI3ouODgYAHDr1i32nHAwKwD88ccf6NevH+bNm4cePXqgbdu2okGzhBBCGg9hFw1vx44dTmhJ/VAYcRJ/f38EBgbi66+/xrVr13DgwAFRkh0zZgxCQ0Mxfvx4/PHHH0hLS8OWLVtw7NgxAECrVq2Qnp6OlJQU3LlzByqVCgqFAn369MHy5cvx999/49ChQ1i8eLHoc9u1a4dTp05h7969uHLlCpYsWYKTJ0869LsTQgixDWNh5MiRI7hz544TWmM9CiNOIpVKsWHDBpw+fRqdO3fGCy+8gBUrVrDXXV1dsWfPHoSEhGDkyJHo0qULli9fDplMBkA3+PWhhx7C4MGDERwcjPXr1wPQDYKtrq5Gz5498fzzz9cYyPTUU09h4sSJmDp1Knr37o38/HzMmzfPcV+cEEKIzQjDCN9Vr9Vq8dNPPzmrSVaRcMIBBqRB0Gq1yMjIQHR0NPvLRUyj62UZul7mo2tlGbpelrHF9dq9ezdGjRoFABg+fDh+/fVXAMD48eOxbds2m7XV3uhvCyGEENJIFRcXs+P4+HiEhIQAAPbu3SuaZdPQURghhBBCGilhN42/vz+GDx8OAKioqMCVK1ec1SyLWRVGvv32W4waNQr3338/HnnkESiVSgBAUlIShg0bhiFDhmDVqlWiKaYXL17EtGnT0L9/f8ydO1c044MQQgghlhOGER8fH1YZAYxP+22oLA4jGzduxLFjx7B27VocOnQIS5cuhYuLC44cOYJNmzYhKSkJGzduxNGjR9n0IrVajQULFmDatGk4cOAAunXrhiVLltj8yxBCCCHNiWEY8fb2Zo8bUxixaNEzjUaDdevW4ZtvvmGrjbZr1w6AbhDNhAkTEBkZCQCYOXMmdu3ahfHjx+P06dNwcXHB+PHjAQBz5szB0KFDcePGDURERNT4HLVaDbVaLW6oXA5XV1eLv2BjxK/E2lh3X3Q0ul6WoetlPrpWlqHrZRlbXC/hmBEvLy94eXmJXmsIfxbmDM61KIzk5uaisrIS+/btww8//AAvLy88+uijmDBhAtLT0xEfH8/OFS6mlZaWxkILALi7uyMyMhJpaWlGw0hiYiLWrFkjem7y5MnNbu+UrKwsZzehUaHrZRm6Xuaja2UZul6Wqc/1unnzJjsuKysT/SJ//fp1ZGRk1KttthATE1PnORaHkbKyMmRmZmLnzp3IysrCM888g1atWqG8vByenp7sXE9PTzaSt6KiQvQa/3p5ebnRz0lISMCMGTPEDW1mlZGsrCxERUXR9Dgz0PWyDF0v89G1sgxdL8vY4noJKx+xsbHIzc1lj93c3BAdHV3vdjqCRWHEzc0NAPDkk0/C3d0d7dq1w/Dhw/HHH3/Aw8ODDWQFAKVSCYVCAQBQKBSi1/jXPTw8jH6Oq6trswketZFKpfQftAXoelmGrpf56FpZhq6XZepzvYTjQvz8/ODr68sel5WVNZo/B4taGR0dDRcXF0gkEvYcfxwTE4Nr166x54W7BrZu3Vr0WmVlJbKzs9G6det6NZ4QQghpzvgBrBKJBJ6eno12AKtFYUShUGDo0KFYu3Yt1Go10tPT8dtvv6F///4YOXIktm7diuzsbOTn5yM5ORkjR44EAPTs2RMqlQo7duyAWq3GunXr0LFjR6PjRYj5kpKS4OfnZ/fP2b59O9q2bQuZTIbnn3/e7p/XEKjVarRt2xZHjx51dlPq1KdPH2zZssXZzSCEOAEfRry9vSGVSptHGAGAhQsXoqioCMOGDcNzzz2Hp59+Gj169MCAAQMwadIkzJo1C5MmTUKfPn0wbtw4ALpulxUrVmD9+vUYPHgwzp49i2XLltn8yzQ3U6dOdciiNk899RQmTZqErKysRv3n9sYbb6B79+5mnfvVV18hJiYG/fr1Y89JJBJs376dPa6qqsL06dMRERGBCxcuWNyeW7du4ZFHHkH79u0hlUqNBr01a9Zg4MCB8Pf3h7+/P4YNG4Y///xTdM7ixYuxaNGiBjFqnhDiWHwY8fHxAYBGG0YsGjMC6L6ocEM3oYSEBCQkJBh9LS4uDhs2bLD040gtFAoFG5djL2VlZcjNzUV8fDzCw8Otfh+1Wt1oxgFxHIfPPvsMb775pslzysvL8fDDD+Pq1as4cuSIWaPFDalUKgQHB2Px4sX46KOPjJ7z+++/Y/r06ejXrx/c3d3x3nvvYfjw4bh48SKrLI4YMQJPPPEEfvnlF7ZHBSGkeeCn9vJjRRprGGkcI1uaoAceeADz58/HggULEBAQgNDQULzxxhuicz766CN06dIFnp6eiIqKwrx581BWVsZeF3bTXLlyBRKJBJcvX67xHvzYHQC4cOECRowYAS8vL7Ro0QKPPvqoya2mf//9d/YXe8iQIZBIJPj9998BAFu2bEFcXBzc3NzQqlUrfPjhh6KfbdWqFZYtW4bHHnsMPj4+mDt3LgDd1tYDBw6EQqFAVFQU5s+fLxrcrFKpsHDhQkRFRcHNzQ1t27bF2rVrAejWuZkzZw5iYmKgUCjQoUMHrFq1qkab77vvPnh6esLPzw/9+/d
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"# использование поиска сезонности для автоматической подстановки\n",
|
|||
|
|
"seasonal_model = NaiveSeasonal(K=train_period)\n",
|
|||
|
|
"seasonal_model.fit(train)\n",
|
|||
|
|
"seasonal_forecast = seasonal_model.predict(36)\n",
|
|||
|
|
"\n",
|
|||
|
|
"series.plot(label=\"actual\")\n",
|
|||
|
|
"seasonal_forecast.plot(label=\"naive forecast (K=12)\")"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 14,
|
|||
|
|
"id": "bf029400",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 14,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGvCAYAAACJsNWPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/TGe4hAAAACXBIWXMAAA9hAAAPYQGoP6dpAACUF0lEQVR4nO3dd3gU5doG8Htrsum9h4RegnREuhRBmhRpCgpIsx0+G2JDBdSjoCg2RATiUVABQUBRVJoUUVroNT2QkJBC+m6yO98fk3l3Jtkku5stKc/vus51ZndmZ94dYubOW2Ucx3EghBBCCHESubMLQAghhJCmjcIIIYQQQpyKwgghhBBCnIrCCCGEEEKcisIIIYQQQpyKwgghhBBCnIrCCCGEEEKcisIIIYQQQpyKwkg9ZDAYkJiYCIPB4OyiNAh0vyxD98t8dK8sQ/fLMnS/jCiMEEIIIcSpKIwQQgghxKkojBBCCCHEqSiMEEIIIcSpKIwQQgghxKkojBBCCCHEqSiMEEIIIcSpKIwQQgghxKkojBBCCCHEqSiMEEIIIcSpKIwQQgghxKkojBBCCCHEqSiMEEIIIcSpKIwQQgghxKkojNQDWVlZUKvVKCoqQllZGTw9PXHjxg22Pzo6GjKZDDKZDO7u7ujWrRu2bNnixBITQgipL/Ly8rBhwwakpKQ4uyhWozBSD/z999/o3Lkz3N3dcerUKfj5+SE8PFxyzNKlS5Geno7Tp0+jZ8+emDJlCo4ePeqkEjufTqdzdhEIIaRemDdvHh577DHcf//9MBgMzi6OVSiM1ANHjx5F3759AQCHDx9Gnz59qhzj6emJkJAQtGnTBp999hk0Gg127doFvV6P2bNno3nz5tBoNGjbti1WrVol+eyBAwdw9913w93dHT4+Pujbty+Sk5MBAGfOnMGgQYPg6ekJLy8vdO/eHSdOnGCfPXz4MPr37w+NRoPIyEgsWLAARUVFbH90dDTeeecdPPbYY/D09ESzZs3w5ZdfVvl+Xbp0gaurK3r06IGffvoJMpkMcXFx7Jjz589jxIgR8PDwQHBwMB555BHcvn2b7b/33nvx9NNP45lnnkFAQACGDx8OjuPw5ptvIjo6Gu3atUNERAQWLFhg/T8EIYQ0QMLv0kuXLuHkyZPOLYyVlM4ugD306NEDGRkZDr9uSEiI5EFek5SUFHTq1AkAUFxcDIVCgdjYWJSUlEAmk+HXX3/FtGnTsHr16iqfVSqVUKlU0Ol0MBgMiIiIwJYtW+Dv74+jR49i3rx5CA0NxeTJk1FeXo5x48Zh7ty5+O6776DT6fDvv/9CJpMBAKZNm4auXbti9erVUCgUiIuLg0qlAgDEx8fj/vvvx1tvvYX169cjKysLTz/9NJ5++mls2LCBleeDDz7AsmXL8Morr2Dr1q144oknMHDgQLRt2xb5+fkYM2YMRo4ciU2bNiE5ORnPPPOM5Pvk5eVh8ODBmDNnDj788EOUlJRg0aJFmDx5Mvbt28eO+/rrr/HEE0/gyJEjAIAff/wRH374ITZt2gRvb28olUqcO3fO/H8wQghpBPLz89n2jh070LNnTyeWxkpcIxQeHs4BcPj/wsPDzS5jWVkZl5iYyJ05c4ZTqVTcmTNnuOvXr3MeHh7c/v37ub/++ou7desWx3EcFxUVxX344Yccx3GcVqvl3nnnHQ4A9/PPP5s891NPPcU9+OCDHMdxXHZ2NgeAO3DggMljPT09udjYWJP7Zs+ezc2bN0/y3qFDhzi5XM6VlJSwsk2fPp3tNxgMXFBQELd69WqO4zhu9erVnL+/Pzue4zhu7dq1HADu9OnTHMdx3LJly7hhw4ZJrpOamsoB4K5cucJxHMcNHDiQ69q1q+SYDz74gGvTpg1XWlrKJSQkcHq93uT3IFJ6vZ7ul5noXlmG7pdlbHW/NBoNew7FxMTYqHSO1ShrRkJCQur9dZVKJaKjo7F582b07NkTnTp1wpEjRxAcHIwBAwYgOTkZAQEB7PhFixbhtddeQ2lpKTw8PPDuu+9i1KhRAIDPPvsM69evR0pKCkpKSqDT6dClSxcAgJ+fH2bOnInhw4fjvvvuw9ChQzF58mSEhoYCAJ577jnMmTMH33zzDYYOHYpJkyahZcuWAPgmnLNnz2Ljxo2sHBzHwWAwIDExEe3btwcAVsMDADKZDCEhIcjMzAQAXLlyBZ06dYKrqys75u6775bcizNnzmD//v3w8PCocp/i4+PRpk0bAED37t0l+yZNmoSPPvoIrVq1Qt++fTFp0iSMHTsWSmWj/LEmhJAqysrKUFJSwl5fuHAB8fHx7Pd4Q9Eof2ub21TiTDExMUhOTkZZWRkMBgM8PDxQXl6O8vJyeHl5ISwsDJcvX2bHL1y4EDNnzmR9KoRmlu+//x4vvPACPvjgA/Tu3Ruenp5YsWIF/vnnH/bZDRs2YMGCBfjtt9/www8/4LXXXsMff/yBe+65B2+++SYefvhh/PLLL/j111/xxhtv4Pvvv8f48eNRWFiI+fPnm+yH0axZM7YtNOsIZDKZRZ2oCgsLMWbMGLz33ntV9gmhCQDc3d0l+yIjI3HlyhX8/vvv2L59O55++ml88MEHOHjwYJUyEUJIY1RQUFDlvR07duC5555zQmms1yjDSEOwe/dulJWVYciQIVi+fDm6d++OqVOnYubMmRg2bBirWRAEBASgVatWVc5z5MgR9OnTB08++SR7Lz4+vspxXbt2RdeuXfHyyy+jd+/e2LRpE+655x4AQJs2bdCmTRs8++yzeOihh7BhwwaMHz8e3bp1w8WLF01e11xt27bFt99+C61WCxcXFwDA8ePHJcd069YNP/74I6Kjoy2u1dBoNBgzZgw6deqERYsWoUOHDjh37hy6detmdZkJIaShEPcXETTEMEKjaZwkKioKHh4euHXrFsaOHYvIyEhcuHABDz74IFq1alVlaG91WrdujRMnTmDPnj24evUqFi9eLHnYJyYm4uWXX8bff/+N5ORk/P7777h27Rrat2+PkpISPP300zhw4ACSk5Nx5MgRHD9+nDW/LFq0CEePHsXTTz+NuLg4XLt2DTt27MDTTz9t9vd8+OGHYTAYMG/ePFy6dAl79uzB+++/DwCsduepp55CTk4OHnroIRw/fhzx8fHYs2cPZs2aBb1eX+25Y2NjsW7dOpw/fx4pKSnYuHEjNBoNoqKizC4fIYQ0ZKbCyOHDhyWjERsCCiNOdODAAfTs2ROurq74999/ERERIWmWMMf8+fMxYcIETJkyBb169UJ2draklsTNzQ2XL1/Ggw8+iDZt2mDevHl46qmnMH/+fCgUCmRnZ+PRRx9FmzZtMHnyZIwYMQJLliwBwPcFOXjwIK5evYr+/fuja9eueP311xEWFmZ2+by8vLBr1y7ExcWhS5cuePXVV/H6668DAOtHEhYWhiNHjkCv12PYsGG466678Mwzz8DHxwdyefU/oj4+Pli7di369++PkSNHYu/evdi1axf8/f0tuoeEENJQicOI8PvSYDDg559/dlaRrCLjOI5zdiGIlMFgQHJyMqKiomp8GDdUGzduxKxZs3Dnzh1oNJo6n6+x3y9bo/tlPrpXlqH7ZRlb3K/du3ezwQzDhg3D77//DgAYN24ctm/fbrOy2hv1GSF297///Q8tWrRAeHg4zpw5w+YQsUUQIYSQpuzOnTtse/jw4YiLi0NmZib27NmDkpKSBvN7lqIrsbuMjAxMnz4d7du3x7PPPotJkyZVmaWVEEKI5cTNNL6+vhg2bBgAoKSkBFevXnVWsSxmVRj5+uuvMWrUKAwYMAAPP/wwmx48NjYWQ4cOxeDBg7Fq1SqIW4AuXLiAqVOnom/fvpg3bx7S09Nt8w1Ivffiiy8iKSkJpaWlSExMxIcffgg3NzdnF4sQQho8cRjx8vJCUFAQe21q2G99ZXEY2bx5M/7++2+sW7cOBw8exJIlS6BSqXD48GFs2bIFsbGx2Lx5M44ePYodO3YA4Bc1e/HFFzF16lTs27cPnTt3xuLFi23+ZQghhJCmpHIY8fT0ZK8bUhixqM+IXq/H+vXr8dVXX7HZRlu3bg2
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"# Добавляем линейный тренд (drift)\n",
|
|||
|
|
"from darts.models import NaiveDrift\n",
|
|||
|
|
"\n",
|
|||
|
|
"drift_model = NaiveDrift()\n",
|
|||
|
|
"drift_model.fit(train)\n",
|
|||
|
|
"drift_forecast = drift_model.predict(36)\n",
|
|||
|
|
"\n",
|
|||
|
|
"combined_forecast = drift_forecast + seasonal_forecast - train.last_value()\n",
|
|||
|
|
"\n",
|
|||
|
|
"series.plot()\n",
|
|||
|
|
"combined_forecast.plot(label=\"combined\")\n",
|
|||
|
|
"drift_forecast.plot(label=\"drift\")"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "d73c6968",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"## Метрики качества"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 15,
|
|||
|
|
"id": "d33d080f",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"name": "stdout",
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"Mean absolute percentage error for the combined naive drift + seasonal: 5.66%.\n"
|
|||
|
|
]
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"from darts.metrics import mape\n",
|
|||
|
|
"\n",
|
|||
|
|
"print(\n",
|
|||
|
|
" f\"Mean absolute percentage error for the combined naive drift + seasonal: {mape(series, combined_forecast):.2f}%.\"\n",
|
|||
|
|
")"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 16,
|
|||
|
|
"id": "5fb8943a",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"name": "stdout",
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"model ExponentialSmoothing() obtains MAPE: 5.11%\n"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"name": "stdout",
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"model TBATS() obtains MAPE: 5.87%\n",
|
|||
|
|
"model AutoARIMA() obtains MAPE: 11.65%\n",
|
|||
|
|
"model Theta() obtains MAPE: 8.15%\n"
|
|||
|
|
]
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"from darts.models import ExponentialSmoothing, TBATS, AutoARIMA, Theta\n",
|
|||
|
|
"\n",
|
|||
|
|
"def eval_model(model):\n",
|
|||
|
|
" model.fit(train)\n",
|
|||
|
|
" forecast = model.predict(len(val))\n",
|
|||
|
|
" print(f\"model {model} obtains MAPE: {mape(val, forecast):.2f}%\")\n",
|
|||
|
|
"\n",
|
|||
|
|
"eval_model(ExponentialSmoothing())\n",
|
|||
|
|
"eval_model(TBATS())\n",
|
|||
|
|
"eval_model(AutoARIMA())\n",
|
|||
|
|
"eval_model(Theta())"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "255b5f32",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"# Theta method\n",
|
|||
|
|
"Данный метод не рассматривался для экономии времени"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "8bccf5f0",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"# Модели машинного обучения\n",
|
|||
|
|
"Пример моделей:\n",
|
|||
|
|
"- RegressionModel can wrap around any sklearn-compatible regression model to produce forecasts (it has its own section below).\n",
|
|||
|
|
"- RNNModel is a flexible RNN implementation, which can be used like DeepAR.\n",
|
|||
|
|
"- NBEATSModel implements the N-BEATS model.\n",
|
|||
|
|
"- TFTModel implements the Temporal Fusion Transformer model.\n",
|
|||
|
|
"- TCNModel implements temporal convolutional networks.\n",
|
|||
|
|
"- …"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 17,
|
|||
|
|
"id": "22339b4b",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 17,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"from darts.datasets import AirPassengersDataset, MonthlyMilkDataset\n",
|
|||
|
|
"\n",
|
|||
|
|
"series_air = AirPassengersDataset().load().astype(np.float32)\n",
|
|||
|
|
"series_milk = MonthlyMilkDataset().load().astype(np.float32)\n",
|
|||
|
|
"\n",
|
|||
|
|
"# set aside last 36 months of each series as validation set:\n",
|
|||
|
|
"train_air, val_air = series_air[:-36], series_air[-36:]\n",
|
|||
|
|
"train_milk, val_milk = series_milk[:-36], series_milk[-36:]\n",
|
|||
|
|
"\n",
|
|||
|
|
"train_air.plot()\n",
|
|||
|
|
"val_air.plot()\n",
|
|||
|
|
"train_milk.plot()\n",
|
|||
|
|
"val_milk.plot()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 21,
|
|||
|
|
"id": "28e3676d",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 21,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 640x480 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"# масштабирование\n",
|
|||
|
|
"from darts.dataprocessing.transformers import Scaler\n",
|
|||
|
|
"\n",
|
|||
|
|
"scaler = Scaler()\n",
|
|||
|
|
"train_air_scaled, train_milk_scaled = scaler.fit_transform([train_air, train_milk])\n",
|
|||
|
|
"\n",
|
|||
|
|
"train_air_scaled.plot()\n",
|
|||
|
|
"train_milk_scaled.plot()"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "markdown",
|
|||
|
|
"id": "4536efeb",
|
|||
|
|
"metadata": {},
|
|||
|
|
"source": [
|
|||
|
|
"## Пример: N-BEATS\n",
|
|||
|
|
"\n",
|
|||
|
|
"Нужные параметры для настройки\n",
|
|||
|
|
"- `input_chunk_length`: this is the “lookback window” of the model - i.e., how many time steps of history the neural network takes as input to produce its output in a forward pass.\n",
|
|||
|
|
"\n",
|
|||
|
|
"- `output_chunk_length`: this is the “forward window” of the model - i.e., how many time steps of future values the neural network outputs in a forward pass."
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 22,
|
|||
|
|
"id": "21cac240",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"name": "stderr",
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"GPU available: False, used: False\n",
|
|||
|
|
"TPU available: False, using: 0 TPU cores\n",
|
|||
|
|
"HPU available: False, using: 0 HPUs\n",
|
|||
|
|
"\n",
|
|||
|
|
" | Name | Type | Params | Mode \n",
|
|||
|
|
"-------------------------------------------------------------\n",
|
|||
|
|
"0 | criterion | MSELoss | 0 | train\n",
|
|||
|
|
"1 | train_criterion | MSELoss | 0 | train\n",
|
|||
|
|
"2 | val_criterion | MSELoss | 0 | train\n",
|
|||
|
|
"3 | train_metrics | MetricCollection | 0 | train\n",
|
|||
|
|
"4 | val_metrics | MetricCollection | 0 | train\n",
|
|||
|
|
"5 | stacks | ModuleList | 6.2 M | train\n",
|
|||
|
|
"-------------------------------------------------------------\n",
|
|||
|
|
"6.2 M Trainable params\n",
|
|||
|
|
"1.4 K Non-trainable params\n",
|
|||
|
|
"6.2 M Total params\n",
|
|||
|
|
"24.787 Total estimated model params size (MB)\n"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"application/vnd.jupyter.widget-view+json": {
|
|||
|
|
"model_id": "ce63a7a3fdfa4c8fa5065c81374a9ac3",
|
|||
|
|
"version_major": 2,
|
|||
|
|
"version_minor": 0
|
|||
|
|
},
|
|||
|
|
"text/plain": [
|
|||
|
|
"Training: | | 0/? [00:00<?, ?it/s]"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"name": "stderr",
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"`Trainer.fit` stopped: `max_epochs=50` reached.\n"
|
|||
|
|
]
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"from darts.models import NBEATSModel\n",
|
|||
|
|
"\n",
|
|||
|
|
"model = NBEATSModel(input_chunk_length=24, output_chunk_length=12, random_state=42)\n",
|
|||
|
|
"\n",
|
|||
|
|
"model.fit([train_air_scaled, train_milk_scaled], epochs=50, verbose=True);"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"cell_type": "code",
|
|||
|
|
"execution_count": 23,
|
|||
|
|
"id": "5e2f9580",
|
|||
|
|
"metadata": {},
|
|||
|
|
"outputs": [
|
|||
|
|
{
|
|||
|
|
"name": "stderr",
|
|||
|
|
"output_type": "stream",
|
|||
|
|
"text": [
|
|||
|
|
"GPU available: False, used: False\n",
|
|||
|
|
"TPU available: False, using: 0 TPU cores\n",
|
|||
|
|
"HPU available: False, using: 0 HPUs\n"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"application/vnd.jupyter.widget-view+json": {
|
|||
|
|
"model_id": "c9a6fdcbc4a24d419bb730cf849ab034",
|
|||
|
|
"version_major": 2,
|
|||
|
|
"version_minor": 0
|
|||
|
|
},
|
|||
|
|
"text/plain": [
|
|||
|
|
"Predicting: | | 0/? [00:00<?, ?it/s]"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Axes: xlabel='Month'>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"execution_count": 23,
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "execute_result"
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
"data": {
|
|||
|
|
"image/png": "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
|
|||
|
|
"text/plain": [
|
|||
|
|
"<Figure size 1000x600 with 1 Axes>"
|
|||
|
|
]
|
|||
|
|
},
|
|||
|
|
"metadata": {},
|
|||
|
|
"output_type": "display_data"
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"source": [
|
|||
|
|
"pred_air, pred_milk = model.predict(series=[train_air_scaled, train_milk_scaled], n=36)\n",
|
|||
|
|
"\n",
|
|||
|
|
"# scale back:\n",
|
|||
|
|
"pred_air, pred_milk = scaler.inverse_transform([pred_air, pred_milk])\n",
|
|||
|
|
"\n",
|
|||
|
|
"plt.figure(figsize=(10, 6))\n",
|
|||
|
|
"series_air.plot(label=\"actual (air)\")\n",
|
|||
|
|
"series_milk.plot(label=\"actual (milk)\")\n",
|
|||
|
|
"pred_air.plot(label=\"forecast (air)\")\n",
|
|||
|
|
"pred_milk.plot(label=\"forecast (milk)\")"
|
|||
|
|
]
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
"metadata": {
|
|||
|
|
"kernelspec": {
|
|||
|
|
"display_name": "Python 3",
|
|||
|
|
"language": "python",
|
|||
|
|
"name": "python3"
|
|||
|
|
},
|
|||
|
|
"language_info": {
|
|||
|
|
"codemirror_mode": {
|
|||
|
|
"name": "ipython",
|
|||
|
|
"version": 3
|
|||
|
|
},
|
|||
|
|
"file_extension": ".py",
|
|||
|
|
"mimetype": "text/x-python",
|
|||
|
|
"name": "python",
|
|||
|
|
"nbconvert_exporter": "python",
|
|||
|
|
"pygments_lexer": "ipython3",
|
|||
|
|
"version": "3.11.9"
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
"nbformat": 4,
|
|||
|
|
"nbformat_minor": 5
|
|||
|
|
}
|