{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "040c997b",
   "metadata": {},
   "source": [
    "# The Easiest Way to Create an Interactive Dashboard in Python\n",
    "\n",
    "This notebook is an updated version of the original notebook supporting the blog post\n",
    "\n",
    "**[The Easiest Way to Create an Interactive Dashboard in Python](https://towardsdatascience.com/the-easiest-way-to-create-an-interactive-dashboard-in-python-77440f2511d1)**. Turn Pandas pipelines into a\n",
    "dashboard using [`param.rx`](https://param.holoviz.org/user_guide/Reactive_Expressions.html) and [panel.ReactiveExpr](https://panel.holoviz.org/reference/panes/ReactiveExpr.html).\n",
    "\n",
    "by *Sophia Yang* and *Marc Skov Madsen*.\n",
    "\n",
    "![Data App](https://github.com/sophiamyang/hvplot_interactive/blob/main/assets/easy-dataframe-dashboards.gif?raw=true)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abf60b58",
   "metadata": {},
   "source": [
    "## Import and configure packages\n",
    "\n",
    "Please note that in **Colab** you will need to `!pip install panel hvplot`. In VS Code notebooks you will need to install `!pip install panel hvplot jupyter_bokeh`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5cb553bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install panel==1.3.6 hvplot==0.9.1 # colab\n",
    "# !pip install panel==1.3.6 hvplot==0.9.1 jupyter_bokeh==3.0.7 # vscode"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f1112af",
   "metadata": {},
   "outputs": [],
   "source": [
    "import param\n",
    "import panel as pn\n",
    "\n",
    "pn.extension('tabulator', sizing_mode=\"stretch_width\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5adb11a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "import hvplot.pandas\n",
    "import holoviews as hv\n",
    "\n",
    "hv.extension('bokeh')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef4ccd06",
   "metadata": {},
   "source": [
    "## Define Color Palette"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ce8c5f82",
   "metadata": {},
   "outputs": [],
   "source": [
    "PALETTE = [\"#ff6f69\", \"#ffcc5c\", \"#88d8b0\", ]\n",
    "pn.Row(\n",
    "    pn.layout.HSpacer(height=50, styles={\"background\": PALETTE[0]}),\n",
    "    pn.layout.HSpacer(height=50, styles={\"background\": PALETTE[1]}),\n",
    "    pn.layout.HSpacer(height=50, styles={\"background\": PALETTE[2]}),\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7698a74c",
   "metadata": {},
   "source": [
    "## Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "09d60a44",
   "metadata": {},
   "outputs": [],
   "source": [
    "from bokeh.sampledata.autompg import autompg_clean as df\n",
    "df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1bfed521",
   "metadata": {},
   "source": [
    "## Define DataFrame Pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4a42b6e5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_data(df, cylinders=4, mfr=['ford','chevrolet'], yaxis=\"hp\"):\n",
    "    return (\n",
    "        df[\n",
    "            (df.cyl == cylinders) & \n",
    "            (df.mfr.isin(mfr))\n",
    "        ]\n",
    "        .groupby(['origin', 'mpg'])[yaxis].mean()\n",
    "        .to_frame()\n",
    "        .reset_index()\n",
    "        .sort_values(by='mpg')\n",
    "    )\n",
    "\n",
    "get_data(df).head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f928f0ba",
   "metadata": {},
   "source": [
    "## Make DataFrame Pipeline Interactive"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "835b20da",
   "metadata": {},
   "source": [
    "Define [Panel widgets](https://panel.holoviz.org/reference/index.html#widgets)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "956fa985",
   "metadata": {},
   "outputs": [],
   "source": [
    "cylinders = pn.widgets.IntSlider(name='Cylinders', start=4, end=8, step=2)\n",
    "mfr = pn.widgets.ToggleGroup(\n",
    "    name='MFR',\n",
    "    options=['ford', 'chevrolet', 'honda', 'toyota', 'audi'], \n",
    "    value=['ford', 'chevrolet', 'honda', 'toyota', 'audi'],\n",
    "    button_type='primary', button_style=\"outline\")\n",
    "yaxis = pn.widgets.RadioButtonGroup(\n",
    "    name='Y axis', \n",
    "    options=['hp', 'weight'],\n",
    "    button_type='primary', button_style=\"outline\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "743aa619",
   "metadata": {},
   "source": [
    "Combine pipeline and widgets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b8883956",
   "metadata": {},
   "outputs": [],
   "source": [
    "ipipeline=param.rx(get_data)(df, cylinders, mfr, yaxis)\n",
    "\n",
    "ipipeline.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bff77abf",
   "metadata": {},
   "source": [
    "## Pipe to Tabulator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d262cb39",
   "metadata": {},
   "outputs": [],
   "source": [
    "itable = pn.widgets.Tabulator(ipipeline, pagination='remote', page_size=10)\n",
    "itable"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b21344a1",
   "metadata": {},
   "source": [
    "Check out the [Tabulator Reference Guide](https://panel.holoviz.org/reference/widgets/Tabulator.html) for more inspiration."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "246398c8",
   "metadata": {},
   "source": [
    "## Pipe to hvPlot and HoloViews"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fcc805e",
   "metadata": {},
   "source": [
    "First we will create the interactive plot with [hvPlot](https://hvplot.holoviz.org/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "24423cca",
   "metadata": {},
   "outputs": [],
   "source": [
    "ihvplot = ipipeline.hvplot(x='mpg', y=yaxis, by='origin', color=PALETTE, line_width=6, height=400, responsive=True)\n",
    "ihvplot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2b5271e",
   "metadata": {},
   "source": [
    "The we will put it in a HoloViews pane"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b958a6e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "iplot = pn.pane.HoloViews(ihvplot)\n",
    "iplot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c943ca4",
   "metadata": {},
   "source": [
    "Check out the [HoloViews Reference Guide](https://panel.holoviz.org/reference/panes/HoloViews.html) for more inspiration"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "345db224",
   "metadata": {},
   "source": [
    "## Layout using Template\n",
    "\n",
    "Here we use the [FastListTemplate](https://panel.holoviz.org/reference/templates/FastListTemplate.html#templates-gallery-fastlisttemplate)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d224e0e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "template = pn.template.FastListTemplate(\n",
    "    title='The easiest way to create a dashboard', \n",
    "    sidebar=[cylinders, 'Manufacturers', mfr, 'Y axis' , yaxis],\n",
    "    main=[iplot, itable],\n",
    "    accent_base_color=\"#88d8b0\",\n",
    "    header_background=\"#88d8b0\",\n",
    "    # main_layout=None, # Use this if you want a gray sidebar and white main area\n",
    ")\n",
    "template.servable(); # Add semicolon because templates don't render well in a notebook"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8831435d",
   "metadata": {},
   "source": [
    "To *serve the notebook* run `panel serve the_easiest_way_to_create_dashboard.ipynb`."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}