{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "\n# Functional  depth \n\nSample usage of Depth for functional data.\nIt will plot samples and dataset based on depth notions.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "from depth.model.DepthFunc import DepthFunc \nimport numpy as np\nimport pandas as pd\nfrom matplotlib import pyplot as plt"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "## Creating dataset and samples\nnp.random.seed(2801)\nn_cases = 100\nrows = []\n# fig,(ax1,ax2,ax3)=plt.subplots(1,3,figsize=(15,5))\nfor case in range(1, n_cases + 1):\n    n_points = 100#np.random.randint(3, 16)\n    timestamps = pd.to_datetime(\"2024-01-01\") + pd.to_timedelta(\n        np.sort(np.random.randint(0, 1000, n_points)), unit='s'\n    )\n    baseVal = np.linspace(0,7,n_points)+np.random.normal(0,.2,n_points)\n    bruit = np.random.normal(0,1,1)\n    value_1 = np.cos(baseVal)+bruit\n    value_2 = np.sin(baseVal)+bruit*0.5\n    value_3 = np.cos(baseVal)*np.sin(baseVal**1.2)+bruit*0.5#+np.sin(baseVal)#+decVal\n    \n    # value_2 = np.random.rand(n_points)\n    # value_3 = np.random.rand(n_points)\n    value_3[-1] = None\n    for t, v1, v2, v3 in zip(timestamps, value_1, value_2, value_3):\n        rows.append([case, t, v1, v2, v3])\n    # ax1.plot(timestamps,value_1)\n    # ax2.plot(timestamps,value_2)\n    # ax3.plot(timestamps,value_3)\ndf = pd.DataFrame(rows, columns=[\"case_id\", \"timestamp\", \"value_1\", \"value_2\", \"value_3\"])\n\ndf.head(10)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "fig,(ax1,ax2,ax3)=plt.subplots(1,3,figsize=(15,5))\nfor case in range(1, n_cases + 1):\n    ax1.plot(df[df.case_id==case].value_1.values)\n    ax2.plot(df[df.case_id==case].value_2.values)\n    ax3.plot(df[df.case_id==case].value_3.values)\nplt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Create model and load dataset for depth computation \n\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "model=DepthFunc().load_dataset(df,interpolate_grid=False)\nDpth=model.projection_based_func_depth(df,notion=\"projection\",output_option=\"lowest_depth\",NRandom=1000)\nprint(\"depth of first 10 functions:\" , Dpth)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "from matplotlib import cm\nfig,(ax1,ax2,ax3)=plt.subplots(1,3,figsize=(15,5))\nfor case in range(1, n_cases + 1):\n    ax1.plot(df[df.case_id==case].value_1.values,c=cm.plasma((Dpth[case-1]-Dpth.min())/(Dpth.max()-Dpth.min())))\n    ax2.plot(df[df.case_id==case].value_2.values,c=cm.plasma((Dpth[case-1]-Dpth.min())/(Dpth.max()-Dpth.min())))\n    ax3.plot(df[df.case_id==case].value_3.values,c=cm.plasma((Dpth[case-1]-Dpth.min())/(Dpth.max()-Dpth.min())))\n\nplt.show()"
      ]
    }
  ],
  "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": "3.12.0"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}