Skip to content

Instantly share code, notes, and snippets.

@ngupta23

ngupta23/sktime_preprocesing.ipynb

Created March 9, 2022 13:03
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save ngupta23/891c6d4e17df7fe538008eb34a99c044 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save ngupta23/891c6d4e17df7fe538008eb34a99c044 to your computer and use it in GitHub Desktop.
Download ZIP
pycaret_ts_preprocesing.ipynb
Display the source blob
Display the rendered blob
Raw
sktime_preprocesing.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "pycaret_ts_preprocesing.ipynb",
"provenance": [],
"collapsed_sections": [],
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/gist/ngupta23/891c6d4e17df7fe538008eb34a99c044/sktime_preprocesing.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"source": [
"# Installation"
],
"metadata": {
"id": "AS6S5gWAK5Cd"
}
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "iqlsAF56kHx9",
"outputId": "0ecf564e-2d0b-4a15-ec90-3e032c4187be"
},
"source": [
"!pip install sktime\n",
"!pip install pmdarima"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Requirement already satisfied: sktime in /usr/local/lib/python3.7/dist-packages (0.10.1)\n",
"Requirement already satisfied: numba>=0.53 in /usr/local/lib/python3.7/dist-packages (from sktime) (0.55.1)\n",
"Requirement already satisfied: scipy<1.8.0 in /usr/local/lib/python3.7/dist-packages (from sktime) (1.4.1)\n",
"Requirement already satisfied: statsmodels>=0.12.1 in /usr/local/lib/python3.7/dist-packages (from sktime) (0.13.2)\n",
"Requirement already satisfied: pandas<1.5.0,>=1.1.0 in /usr/local/lib/python3.7/dist-packages (from sktime) (1.3.5)\n",
"Requirement already satisfied: deprecated>=1.2.13 in /usr/local/lib/python3.7/dist-packages (from sktime) (1.2.13)\n",
"Requirement already satisfied: scikit-learn>=0.24.0 in /usr/local/lib/python3.7/dist-packages (from sktime) (1.0.2)\n",
"Requirement already satisfied: numpy<1.22,>=1.21.0 in /usr/local/lib/python3.7/dist-packages (from sktime) (1.21.5)\n",
"Requirement already satisfied: wrapt<2,>=1.10 in /usr/local/lib/python3.7/dist-packages (from deprecated>=1.2.13->sktime) (1.13.3)\n",
"Requirement already satisfied: setuptools in /usr/local/lib/python3.7/dist-packages (from numba>=0.53->sktime) (57.4.0)\n",
"Requirement already satisfied: llvmlite<0.39,>=0.38.0rc1 in /usr/local/lib/python3.7/dist-packages (from numba>=0.53->sktime) (0.38.0)\n",
"Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas<1.5.0,>=1.1.0->sktime) (2018.9)\n",
"Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas<1.5.0,>=1.1.0->sktime) (2.8.2)\n",
"Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas<1.5.0,>=1.1.0->sktime) (1.15.0)\n",
"Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.24.0->sktime) (3.1.0)\n",
"Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.24.0->sktime) (1.1.0)\n",
"Requirement already satisfied: packaging>=21.3 in /usr/local/lib/python3.7/dist-packages (from statsmodels>=0.12.1->sktime) (21.3)\n",
"Requirement already satisfied: patsy>=0.5.2 in /usr/local/lib/python3.7/dist-packages (from statsmodels>=0.12.1->sktime) (0.5.2)\n",
"Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=21.3->statsmodels>=0.12.1->sktime) (3.0.7)\n",
"Requirement already satisfied: pmdarima in /usr/local/lib/python3.7/dist-packages (1.8.5)\n",
"Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.4.1)\n",
"Requirement already satisfied: pandas>=0.19 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.3.5)\n",
"Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.1.0)\n",
"Requirement already satisfied: statsmodels!=0.12.0,>=0.11 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (0.13.2)\n",
"Requirement already satisfied: setuptools!=50.0.0,>=38.6.0 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (57.4.0)\n",
"Requirement already satisfied: Cython!=0.29.18,>=0.29 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (0.29.28)\n",
"Requirement already satisfied: urllib3 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.24.3)\n",
"Requirement already satisfied: numpy>=1.19.3 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.21.5)\n",
"Requirement already satisfied: scikit-learn>=0.22 in /usr/local/lib/python3.7/dist-packages (from pmdarima) (1.0.2)\n",
"Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.19->pmdarima) (2.8.2)\n",
"Requirement already satisfied: pytz>=2017.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.19->pmdarima) (2018.9)\n",
"Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.7/dist-packages (from python-dateutil>=2.7.3->pandas>=0.19->pmdarima) (1.15.0)\n",
"Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.22->pmdarima) (3.1.0)\n",
"Requirement already satisfied: packaging>=21.3 in /usr/local/lib/python3.7/dist-packages (from statsmodels!=0.12.0,>=0.11->pmdarima) (21.3)\n",
"Requirement already satisfied: patsy>=0.5.2 in /usr/local/lib/python3.7/dist-packages (from statsmodels!=0.12.0,>=0.11->pmdarima) (0.5.2)\n",
"Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging>=21.3->statsmodels!=0.12.0,>=0.11->pmdarima) (3.0.7)\n"
]
}
]
},
{
"cell_type": "markdown",
"source": [
"# Basic Example (sktime)"
],
"metadata": {
"id": "DXPMkmuTGWzG"
}
},
{
"cell_type": "code",
"metadata": {
"id": "bVWedmaakob2",
"colab": {
"base_uri": "https://localhost:8080/"
},
"outputId": "717adcbb-4325-4d01-fcc5-03be1170e880"
},
"source": [
"#### Step 1: Load data and simulate missing value ----\n",
"import numpy as np\n",
"from sktime.datasets import load_airline\n",
"y = load_airline()\n",
"y[2:10] = np.nan\n",
"y[70:80] = np.nan"
],
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
"/usr/local/lib/python3.7/dist-packages/pandas/core/series.py:1070: FutureWarning: Slicing a positional slice with .loc is not supported, and will raise TypeError in a future version. Use .loc with labels or .iloc with positions instead.\n",
" self.loc[key] = value\n"
]
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "EjsWc5TYkund"
},
"source": [
"#### Step 2: Create pipeline with preprocessing ----\n",
"from sktime.forecasting.compose import ForecastingPipeline\n",
"from sktime.transformations.series.impute import Imputer\n",
"from sktime.transformations.series.boxcox import LogTransformer\n",
"from sktime.forecasting.compose import TransformedTargetForecaster\n",
"from sktime.transformations.series.detrend import Deseasonalizer\n",
"from sktime.forecasting.arima import ARIMA\n",
"\n",
"# Preprocessing here works only on the y-values\n",
"forecaster = TransformedTargetForecaster(\n",
" [ \n",
" (\"impute\", Imputer()),\n",
" (\"log\", LogTransformer()),\n",
" (\"deseasonalize\", Deseasonalizer(model=\"multiplicative\", sp=12)),\n",
" (\"model\", ARIMA()),\n",
" ]\n",
")\n",
"\n",
"# Preprocessing here works only on the X values\n",
"pipe = ForecastingPipeline(\n",
" [\n",
" (\"impute\", Imputer()),\n",
" (\"log\", LogTransformer()),\n",
" (\"forecast\", forecaster)\n",
" ]\n",
")"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "KVnHO-bJkwBj",
"outputId": "7781db12-a0c0-4d2d-abe2-1eaf571c9393"
},
"source": [
"#### Step 3: Train and Predict ----\n",
"pipe.fit(y, X=None, fh=np.arange(1,13))\n",
"pipe.predict(X=None)"
],
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"1961-01 433.542916\n",
"1961-02 420.520173\n",
"1961-03 477.924732\n",
"1961-04 458.315857\n",
"1961-05 457.981903\n",
"1961-06 515.257830\n",
"1961-07 552.394610\n",
"1961-08 551.075341\n",
"1961-09 483.507705\n",
"1961-10 418.166156\n",
"1961-11 369.678009\n",
"1961-12 413.889702\n",
"Freq: M, dtype: float64"
]
},
"metadata": {},
"execution_count": 4
}
]
},
{
"cell_type": "markdown",
"source": [
"# More complex example (for pycaret)"
],
"metadata": {
"id": "CLFlYY42pHxj"
}
},
{
"cell_type": "markdown",
"source": [
"## Helper Functions (Internal to PyCaret)"
],
"metadata": {
"id": "2tuHJGzu-uDJ"
}
},
{
"cell_type": "code",
"metadata": {
"id": "9e6d4eVNlQ9D"
},
"source": [
"# https://www.sktime.org/en/stable/api_reference/auto_generated/sktime.transformations.series.compose.OptionalPassthrough.html\n",
"from sktime.transformations.series.compose import OptionalPassthrough\n"
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"def _get_imputer(impute):\n",
" if impute is None:\n",
" passthrough = True\n",
" method = \"drift\" # Placeholder\n",
" else:\n",
" allowed_values = [\n",
" \"drift\", \"linear\", \"nearest\", \"constant\", \"mean\", \n",
" \"median\", \"backfill\", \"bfill\", \"pad\", \"ffill\", \"random\"\n",
" ]\n",
" if impute not in allowed_values:\n",
" raise ValueError(f\"Impute value '{impute}' not allowed.\")\n",
" passthrough = False\n",
" method = impute\n",
" imputer = OptionalPassthrough(Imputer(method=method), passthrough=passthrough)\n",
" return imputer"
],
"metadata": {
"id": "2vc2SF8itM1s"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"from sktime.transformations.series.boxcox import BoxCoxTransformer\n",
"from sktime.transformations.series.boxcox import LogTransformer\n",
"from sktime.transformations.series.exponent import SqrtTransformer\n",
"from sktime.transformations.series.exponent import ExponentTransformer\n",
"from sktime.transformations.series.cos import CosineTransformer\n",
"\n",
"def _get_transformer(transform):\n",
" if transform is None:\n",
" transformer = None\n",
" else:\n",
" allowed_values = [\"box-cox\", \"log\", \"sqrt\", \"exp\", \"cos\"]\n",
" if transform not in allowed_values:\n",
" raise ValueError(f\"Transform value '{transform}' not allowed.\")\n",
" passthrough = False\n",
"\n",
" if transform == \"box-cox\":\n",
" transformer = OptionalPassthrough(BoxCoxTransformer(), passthrough=passthrough)\n",
" elif transform == \"log\":\n",
" transformer = OptionalPassthrough(LogTransformer(), passthrough=passthrough)\n",
" elif transform == \"sqrt\":\n",
" transformer = OptionalPassthrough(SqrtTransformer(), passthrough=passthrough)\n",
" elif transform == \"exp\":\n",
" transformer = OptionalPassthrough(ExponentTransformer(), passthrough=passthrough)\n",
" elif transform == \"cos\":\n",
" transformer = OptionalPassthrough(CosineTransformer(), passthrough=passthrough)\n",
"\n",
" return transformer"
],
"metadata": {
"id": "JLzn8gjct9Mi"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"from sktime.transformations.series.adapt import TabularToSeriesAdaptor\n",
"from sklearn.preprocessing import StandardScaler\n",
"\n",
"def _get_scaler(scale):\n",
" if scale is None:\n",
" scaler = None\n",
" else:\n",
" allowed_values = [\"z-score\"]\n",
" if scale not in allowed_values:\n",
" raise ValueError(f\"Scale value '{scale}' not allowed.\")\n",
" passthrough = False\n",
"\n",
" if scale == \"z-score\":\n",
" scaler = OptionalPassthrough(TabularToSeriesAdaptor(StandardScaler()))\n",
" \n",
" return scaler"
],
"metadata": {
"id": "BgnDf6OTt_sh"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"def _get_pipe():\n",
" \"\"\"Uses global variable for now. Will be fixed in PyCaret to use Class Attributes \"\"\"\n",
" imputer_target = _get_imputer(impute=imputate_target)\n",
" imputer_exogenous = _get_imputer(impute=imputate_exogenous)\n",
" transformer_target = _get_transformer(transform=transform_target)\n",
" transformer_exogenous = _get_transformer(transform=transform_exogenous)\n",
" scaler_target = _get_scaler(scale=scale_target)\n",
" scaler_exogenous = _get_scaler(scale=scale_exogenous)\n",
"\n",
" # print(imputer_target)\n",
" # print(imputer_exogenous)\n",
" # print(transformer_target)\n",
" # print(transformer_exogenous)\n",
" # print(scaler_target)\n",
" # print(scaler_exogenous)\n",
"\n",
" target_steps = []\n",
" target_steps.extend([(\"imputer\", imputer_target)])\n",
" if transformer_target is not None:\n",
" target_steps.extend([(\"transformer\", transformer_target)])\n",
" if scaler_target is not None:\n",
" target_steps.extend([(\"scaler\", scaler_target)])\n",
" target_steps.extend([(\"model\", model)])\n",
"\n",
" exog_steps = []\n",
"\n",
" exog_steps.extend([(\"imputer\", imputer_exogenous)])\n",
" if transformer_exogenous is not None:\n",
" exog_steps.extend([(\"transformer\", transformer_exogenous)])\n",
" if scaler_exogenous is not None:\n",
" exog_steps.extend([(\"scaler\", scaler_exogenous)])\n",
"\n",
" exog_steps.extend([(\"forecaster\", TransformedTargetForecaster(target_steps))])\n",
"\n",
" from sktime.forecasting.compose import ForecastingPipeline\n",
" pipe = ForecastingPipeline(exog_steps)\n",
" return pipe"
],
"metadata": {
"id": "o6K3O-pd0B-N"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"## User Inputs in Setup\n",
"\n",
"### Imputation\n",
"\n",
"**Proposed Arguments: `imputate_target` and `imputate_exogenous`**\n",
"\n",
"**Options**:\n",
"* None: No imputation is done, else specify any value allowed by sktime\n",
"* \"drift\" : drift/trend values by sktime.PolynomialTrendForecaster()\n",
"* \"linear\" : linear interpolation, by pd.Series.interpolate()\n",
"* \"nearest\" : use nearest value, by pd.Series.interpolate()\n",
"* \"constant\" : same constant value (given in arg value) for all NaN\n",
"* \"mean\" : pd.Series.mean()\n",
"* \"median\" : pd.Series.median()\n",
"* \"backfill\" ot \"bfill\" : adapted from pd.Series.fillna()\n",
"* \"pad\" or \"ffill\" : adapted from pd.Series.fillna()\n",
"* \"random\" : random values between pd.Series.min() and .max()\n",
"* \"forecaster\" : use an sktime Forecaster, given in arg forecaster (TODO: Maybe skip for now)\n",
"\n",
"### Transformation \n",
"\n",
"**Proposed Arguments: `transform_target` and `transform_exogenous`**\n",
"\n",
"**Options**:\n",
"* None\n",
"* \"box-cox\"\n",
"* \"log\"\n",
"* \"sqrt\"\n",
"* \"exp\"\n",
"* \"cos\"\n",
"* NOTE: yeo-johnson is not suppoted by sktime yet\n",
"\n",
"### Scaling\n",
"\n",
"**Proposed Arguments: `scale_target` and `scale_exogenous`**\n",
"\n",
"**Options**:\n",
"* None\n",
"* \"z-score\"\n",
"\n"
],
"metadata": {
"id": "Fg62t0ND_fzR"
}
},
{
"cell_type": "markdown",
"source": [
"## Exponential Smoothing\n",
"\n",
"NOTE: This does not support missing values, hence will fail without imputation"
],
"metadata": {
"id": "ogSjUZ6xFNIl"
}
},
{
"cell_type": "markdown",
"source": [
"### Without imputation"
],
"metadata": {
"id": "cYBabe_6IYYU"
}
},
{
"cell_type": "code",
"source": [
"imputate_target = None\n",
"imputate_exogenous = None\n",
"\n",
"transform_target = None \n",
"transform_exogenous = None\n",
"\n",
"scale_target = None\n",
"scale_exogenous = None\n",
"\n",
"# e.g. From `create_model(\"exp_smooth\")`\n",
"# Does not handle missing data ----\n",
"from sktime.forecasting.exp_smoothing import ExponentialSmoothing\n",
"model = ExponentialSmoothing()\n",
"\n",
"pipe = _get_pipe()\n",
"pipe"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "sauV03hD_CTJ",
"outputId": "df26f9ec-4406-4400-bfac-c2717908da73"
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"ForecastingPipeline(steps=[('imputer',\n",
" OptionalPassthrough(passthrough=True,\n",
" transformer=Imputer())),\n",
" ('forecaster',\n",
" TransformedTargetForecaster(steps=[('imputer',\n",
" OptionalPassthrough(passthrough=True,\n",
" transformer=Imputer())),\n",
" ('model',\n",
" ExponentialSmoothing())]))])"
]
},
"metadata": {},
"execution_count": 10
}
]
},
{
"cell_type": "code",
"source": [
"pipe.fit(y)\n",
"predictions = pipe.predict(fh=np.arange(1, 13))\n",
"print(predictions)\n",
"\n",
"from sktime.utils.plotting import plot_series\n",
"_ = plot_series(y, predictions)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 525
},
"id": "3bVSgbnM-aEV",
"outputId": "3d290b21-0845-49ae-e02c-b55f2ad193fd"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stderr",
"text": [
"/usr/local/lib/python3.7/dist-packages/statsmodels/tsa/holtwinters/model.py:917: ConvergenceWarning: Optimization failed to converge. Check mle_retvals.\n",
" ConvergenceWarning,\n"
]
},
{
"output_type": "stream",
"name": "stdout",
"text": [
"1961-01 NaN\n",
"1961-02 NaN\n",
"1961-03 NaN\n",
"1961-04 NaN\n",
"1961-05 NaN\n",
"1961-06 NaN\n",
"1961-07 NaN\n",
"1961-08 NaN\n",
"1961-09 NaN\n",
"1961-10 NaN\n",
"1961-11 NaN\n",
"1961-12 NaN\n",
"Freq: M, dtype: float64\n"
]
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1152x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"source": [
""
],
"metadata": {
"id": "vd26USbnLJMn"
}
},
{
"cell_type": "markdown",
"source": [
"### With imputation"
],
"metadata": {
"id": "KfTdF3h7FPI9"
}
},
{
"cell_type": "code",
"source": [
"imputate_target = \"drift\"\n",
"imputate_exogenous = None\n",
"\n",
"transform_target = None \n",
"transform_exogenous = None\n",
"\n",
"scale_target = None\n",
"scale_exogenous = None\n",
"\n",
"# e.g. From `create_model(\"exp_smooth\")`\n",
"# Does not handle missing data ----\n",
"from sktime.forecasting.exp_smoothing import ExponentialSmoothing\n",
"model = ExponentialSmoothing()\n",
"\n",
"pipe = _get_pipe()\n",
"pipe"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "AjjxhR_fDIJH",
"outputId": "b92516a8-f190-4932-bfce-05c187bf4739"
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"ForecastingPipeline(steps=[('imputer',\n",
" OptionalPassthrough(passthrough=True,\n",
" transformer=Imputer())),\n",
" ('forecaster',\n",
" TransformedTargetForecaster(steps=[('imputer',\n",
" OptionalPassthrough(transformer=Imputer())),\n",
" ('model',\n",
" ExponentialSmoothing())]))])"
]
},
"metadata": {},
"execution_count": 12
}
]
},
{
"cell_type": "code",
"source": [
"pipe.fit(y)\n",
"predictions = pipe.predict(fh=np.arange(1, 13))\n",
"print(predictions)\n",
"\n",
"from sktime.utils.plotting import plot_series\n",
"_ = plot_series(y, predictions)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 491
},
"id": "ucwHKoLhFWfP",
"outputId": "ef432b1b-1c40-42b9-e646-37b4536a19fc"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"1961-01 431.791781\n",
"1961-02 431.791781\n",
"1961-03 431.791781\n",
"1961-04 431.791781\n",
"1961-05 431.791781\n",
"1961-06 431.791781\n",
"1961-07 431.791781\n",
"1961-08 431.791781\n",
"1961-09 431.791781\n",
"1961-10 431.791781\n",
"1961-11 431.791781\n",
"1961-12 431.791781\n",
"Freq: M, dtype: float64\n"
]
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1152x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"source": [
"### With Box-Cox Transformaton"
],
"metadata": {
"id": "slNDf6TqIhts"
}
},
{
"cell_type": "code",
"source": [
"imputate_target = \"drift\"\n",
"imputate_exogenous = None\n",
"\n",
"transform_target = \"box-cox\" \n",
"transform_exogenous = None\n",
"\n",
"scale_target = None\n",
"scale_exogenous = None\n",
"\n",
"# e.g. From `create_model(\"exp_smooth\")`\n",
"# Does not handle missing data ----\n",
"from sktime.forecasting.exp_smoothing import ExponentialSmoothing\n",
"model = ExponentialSmoothing()\n",
"\n",
"pipe = _get_pipe()\n",
"pipe"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "FUjuXfGOHnKG",
"outputId": "19d61da7-7cd1-445e-f09e-a246307e8801"
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"ForecastingPipeline(steps=[('imputer',\n",
" OptionalPassthrough(passthrough=True,\n",
" transformer=Imputer())),\n",
" ('forecaster',\n",
" TransformedTargetForecaster(steps=[('imputer',\n",
" OptionalPassthrough(transformer=Imputer())),\n",
" ('transformer',\n",
" OptionalPassthrough(transformer=BoxCoxTransformer())),\n",
" ('model',\n",
" ExponentialSmoothing())]))])"
]
},
"metadata": {},
"execution_count": 14
}
]
},
{
"cell_type": "code",
"source": [
"pipe.fit(y)\n",
"predictions = pipe.predict(fh=np.arange(1, 13))\n",
"print(predictions)\n",
"\n",
"from sktime.utils.plotting import plot_series\n",
"_ = plot_series(y, predictions)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 491
},
"id": "FQruszeRIpIN",
"outputId": "86975a13-fd31-4ff6-dafc-0df3b3f79268"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"1961-01 431.999999\n",
"1961-02 431.999999\n",
"1961-03 431.999999\n",
"1961-04 431.999999\n",
"1961-05 431.999999\n",
"1961-06 431.999999\n",
"1961-07 431.999999\n",
"1961-08 431.999999\n",
"1961-09 431.999999\n",
"1961-10 431.999999\n",
"1961-11 431.999999\n",
"1961-12 431.999999\n",
"Freq: M, dtype: float64\n"
]
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1152x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"source": [
"### With Scaling"
],
"metadata": {
"id": "inUjhjIQIucs"
}
},
{
"cell_type": "code",
"source": [
"imputate_target = \"drift\"\n",
"imputate_exogenous = None\n",
"\n",
"transform_target = None \n",
"transform_exogenous = None\n",
"\n",
"scale_target = \"z-score\"\n",
"scale_exogenous = None\n",
"\n",
"# e.g. From `create_model(\"exp_smooth\")`\n",
"# Does not handle missing data ----\n",
"from sktime.forecasting.exp_smoothing import ExponentialSmoothing\n",
"model = ExponentialSmoothing()\n",
"\n",
"pipe = _get_pipe()\n",
"pipe"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "0nw7P4ysIq0M",
"outputId": "205d5a51-ee9f-497e-c9ad-193486c22495"
},
"execution_count": null,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"ForecastingPipeline(steps=[('imputer',\n",
" OptionalPassthrough(passthrough=True,\n",
" transformer=Imputer())),\n",
" ('forecaster',\n",
" TransformedTargetForecaster(steps=[('imputer',\n",
" OptionalPassthrough(transformer=Imputer())),\n",
" ('scaler',\n",
" OptionalPassthrough(transformer=TabularToSeriesAdaptor(transformer=StandardScaler()))),\n",
" ('model',\n",
" ExponentialSmoothing())]))])"
]
},
"metadata": {},
"execution_count": 16
}
]
},
{
"cell_type": "code",
"source": [
"pipe.fit(y)\n",
"predictions = pipe.predict(fh=np.arange(1, 13))\n",
"print(predictions)\n",
"\n",
"from sktime.utils.plotting import plot_series\n",
"_ = plot_series(y, predictions)"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 491
},
"id": "BcAWFxvSI3Y9",
"outputId": "61725534-b2ed-42ac-92b7-1f2021914279"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"1961-01 431.999999\n",
"1961-02 431.999999\n",
"1961-03 431.999999\n",
"1961-04 431.999999\n",
"1961-05 431.999999\n",
"1961-06 431.999999\n",
"1961-07 431.999999\n",
"1961-08 431.999999\n",
"1961-09 431.999999\n",
"1961-10 431.999999\n",
"1961-11 431.999999\n",
"1961-12 431.999999\n",
"Freq: M, dtype: float64\n"
]
},
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1152x288 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"source": [
"# Open Questions\n",
"\n",
"1. Is everyone OK with argument names to be added to setup (slightly different from regression/classification, but more appropraite naming)?\n",
"2. Should `OptionalPassthrough` with `passthrough=True` be used when `imputate` is not required by user? Same for `transform` and `scale`?\n",
" - Does this offer any advantages in tuning? \n",
" - Does it offer the user additional flexibility when then use the exported model outside pycaret?\n",
"3. What about categorical variables\n",
" - I dont think classical time series will support categorial variables (need to check).\n",
" - But reduced regression may support this (need to check).\n",
" - How will the setup arguments change if this needs to be suppoted in the future? Do we need to plan for it now?\n",
" - Should [TransformerWrapper](https://github.com/pycaret/pycaret/blob/fcd9ad809699ce5a6e2f76a8f5b0ba0be55d02db/pycaret/internal/preprocess/preprocessor.py#L252) be used for the transformers as done in regression/classification? How does this impact the pipeline? Can users still use the models without installing pycaret?"
],
"metadata": {
"id": "DzX7jOeOLnTM"
}
},
{
"cell_type": "code",
"source": [
""
],
"metadata": {
"id": "5WYAY2ZnLn3r"
},
"execution_count": null,
"outputs": []
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment