{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a8ec0623-c2b2-481c-b1d0-a605d19d0062",
   "metadata": {},
   "source": [
    "# Dataframe Plot\n",
    "\n",
    "The dataframe plot is used to visualize data logged across multiple experiments as\n",
    "dataframes on a single plot. The dataframes logged to each experiment must have\n",
    "matching schema in order to properly visualize them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "712796d5-81d6-4382-8844-b523d5812c9d",
   "metadata": {},
   "outputs": [],
   "source": [
    "import random\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import plotly.express as px\n",
    "\n",
    "from rubicon_ml import Rubicon\n",
    "from rubicon_ml.viz import DataframePlot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b75b715-279f-4d37-931f-2197886f1c6c",
   "metadata": {},
   "source": [
    "First, we'll create a few experiments and log a dataframe to each one."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "59ef9397-23ae-4aec-a0a7-9532ea07d41a",
   "metadata": {},
   "outputs": [],
   "source": [
    "DISPLAY_DFS = False\n",
    "\n",
    "rubicon = Rubicon(persistence=\"memory\", auto_git_enabled=True)\n",
    "project = rubicon.get_or_create_project(\"plot comparison\")\n",
    "\n",
    "num_experiments_to_log = 6\n",
    "data_ranges = [\n",
    "    (random.randint(0, 15000), random.randint(0, 15000))\n",
    "    for _ in range(num_experiments_to_log)\n",
    "]\n",
    "dates = pd.date_range(start=\"1/1/2010\", end=\"12/1/2020\", freq=\"MS\")\n",
    "\n",
    "for start, stop in data_ranges:\n",
    "    data = np.array([list(dates), np.linspace(start, stop, len(dates))])\n",
    "    data_df = pd.DataFrame.from_records(\n",
    "        data.T,\n",
    "        columns=[\"calendar month\", \"open accounts\"],\n",
    "    )\n",
    "\n",
    "    dataframe = project.log_experiment().log_dataframe(data_df, name=\"open accounts\")\n",
    "    \n",
    "    if DISPLAY_DFS:\n",
    "        print(f\"dataframe {dataframe.id}\")\n",
    "        display(data_df.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a067df5-2706-4038-b3f3-57fdfb9981af",
   "metadata": {},
   "source": [
    "Now, we can instantiate the `DataframePlot` object with the experiments we just logged. We\n",
    "also need to provide the name of the dataframe we're plotting. Optionally, provide a Plotly\n",
    "express plotting function as `plotting_func` to visualize the dataframes with any of Plotly\n",
    "express' available options.\n",
    "\n",
    "More on the available Plotly express visualizations can be found in [the Plotly express\n",
    "documentation](https://plotly.com/python/plotly-express/).\n",
    "\n",
    "We can view the plot right in the notebook with `show`. The Dash application\n",
    "itself will be running on http://127.0.0.1:8050/ when running locally. Use the\n",
    "`serve` command to launch the server directly without rendering the widget in the\n",
    "current Python interpreter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0790c5b6-7398-4bf4-a923-8252cc8cd033",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dash is running on http://127.0.0.1:8050/\n"
     ]
    }
   ],
   "source": [
    "DataframePlot(\n",
    "    experiments=project.experiments(),\n",
    "    dataframe_name=\"open accounts\",\n",
    "    x=\"calendar month\",\n",
    "    y=\"open accounts\",\n",
    "    plotting_func=px.line,\n",
    ").show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65a143a5-24d3-44c1-b8b7-85e29b6d515d",
   "metadata": {},
   "source": [
    "![dataframe-plot](dataframe-plot.png \"dataframe plot\")"
   ]
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}