ML_Poject_EDA_RF_SHFvsDHF.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ML_Poject_EDA_RF_SHFvsDHF.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyMq5EdRQGL1PtL7ANTZvbMR",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/JARACH-209/AI-ML_Lab/blob/master/ML_Poject_EDA_RF_SHFvsDHF.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "na5YHKv9dyyE"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9_GmjszLeLED"
      },
      "source": [
        "df = pd.read_csv(\"/content/dat.csv\")\n",
        "df.drop(columns='Unnamed: 0',inplace=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "E2CQGIdnkeeY"
      },
      "source": [
        "df.drop(columns=['NYHA.cardiac.function.classification','DestinationDischarge','admission.ward','admission.way','inpatient.number','consciousness','eye.opening','erythrocyte.sedimentation.rate','ageCat'],inplace=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CAbTwzA0lt7R"
      },
      "source": [
        "df.drop(columns=['apolipoprotein.A','apolipoprotein.B','verbal.response','movement','respiratory.support.','oxygen.inhalation','discharge.department'],inplace=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZqQSCqzOjEwe",
        "outputId": "de274552-4c3f-4e47-a3c0-c19f7de060f7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "df.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2008, 150)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ILTX5pwHnoE7",
        "outputId": "7f5dd73d-2b2d-4a26-c12f-62210496e540",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "len(df.loc[df['congestive.heart.failure'] == 1])    # 1872 Patients had congestive heart failure"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1872"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UQ8o1DMCp4yG",
        "outputId": "ddfb6758-fe8c-49a5-dc13-f795b906870d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "chf_corr = sorted(abs(df.corr()['congestive.heart.failure']).items(),key = lambda kv: kv[1],reverse=True)[:10]\n",
        "chf_corr"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('congestive.heart.failure', 1.0),\n",
              " ('leukemia', nan),\n",
              " ('EA', 0.5304830043529579),\n",
              " ('CCI.score', 0.3018254879254003),\n",
              " ('cerebrovascular.disease', 0.1571027713704445),\n",
              " ('myoglobin', 0.1492432208217126),\n",
              " ('moderate.to.severe.chronic.kidney.disease', 0.13600221538391113),\n",
              " ('death.within.6.months', 0.10906740032518758),\n",
              " ('tricuspid.valve.return.pressure', 0.08922311994442841),\n",
              " ('LVEF', 0.08622530444045182)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CWSanCMVqusQ"
      },
      "source": [
        "* EA > 0 : 393   (Important for Diastolic HF)\n",
        "    * EMS : 980\n",
        "    * AMS : 550\n",
        "* MAP ( Mean Arterial Pressure) >110 : 319\n",
        "    * Mean MAP with EF > 50 : 98.13\n",
        "    * Mean MAP with EF < 40 : 92.44\n",
        "    * Number of patients with MAP > 110\n",
        "        * EF > 40 : 101\n",
        "        * EF < 40 : 19\n",
        "        * EF > 50 : 70\n",
        "* LVEF > 0 : 635\n",
        "* MyoInfa : 143"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gA5oqzI8qQ3_",
        "outputId": "368146f0-6e35-4c93-9a4f-385605d568bc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "len(df.loc[df['congestive.heart.failure']==1].loc[df['LVEF'] >50]['mitral.valve.EMS'].loc[df['EA']>0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "111"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "co8WSHPINXoQ",
        "outputId": "024d0f77-ee18-4085-eaee-93f1d26ca38e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "len(df.loc[df['LVEF']>60].loc[df['congestive.heart.failure']==1].loc[df.map>110])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "33"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0EiuPjmXMn-n",
        "outputId": "98993a1a-73c9-47ec-8f88-81ad18103f0e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "len(df.loc[df['congestive.heart.failure']==1].loc[df['LVEF'] <40].loc[df.map>110])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "19"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8-OnnkijNuMy",
        "outputId": "f975d9e4-baa3-4866-e11c-56d71375670e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "#Map of people having CHF and EF<45\n",
        "len(df.loc[df['congestive.heart.failure']==1].loc[df.LVEF <45].loc[df.map > 110])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "28"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AocsIzRlhiSf"
      },
      "source": [
        "ra_corr = sorted(abs(df.corr().EA).items(),key = lambda kv: kv[1],reverse=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KrxtaL6riXSl",
        "outputId": "e916cb4f-f2c5-4356-fcf3-9d8313edf816",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "ra_corr"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('connective.tissue.disease', nan),\n",
              " ('hemiplegia', nan),\n",
              " ('leukemia', nan),\n",
              " ('malignant.lymphoma', nan),\n",
              " ('EA', 1.0),\n",
              " ('tricuspid.valve.return.pressure', 1.0),\n",
              " ('congestive.heart.failure', 0.5304830043529579),\n",
              " ('time.of.death..days.from.admission.', 0.454265836538509),\n",
              " ('lipoprotein', 0.3433058584167551),\n",
              " ('re.admission.time..days.from.admission.', 0.2482452130013647),\n",
              " ('time.to.emergency.department.within.6.months', 0.2482452130013647),\n",
              " ('LVEF', 0.20585386911196504),\n",
              " ('myocardial.infarction', 0.175962824013606),\n",
              " ('myoglobin', 0.15239589749649518),\n",
              " ('white.globulin.ratio', 0.13349144592028936),\n",
              " ('prothrombin.activity', 0.12367700439262552),\n",
              " ('high.sensitivity.protein', 0.1152849487736945),\n",
              " ('serum.magnesium', 0.11482094094793562),\n",
              " ('systolic.blood.pressure', 0.10825591601906842),\n",
              " ('calcium', 0.10760239364947947),\n",
              " ('CCI.score', 0.10080333599499468),\n",
              " ('hemoglobin', 0.09226729558460069),\n",
              " ('mean.platelet.volume', 0.09055019213200781),\n",
              " ('eosinophil.count', 0.08799769646707814),\n",
              " ('hematocrit', 0.0837045248245729),\n",
              " ('eosinophil.ratio', 0.08260781031868895),\n",
              " ('return.to.emergency.department.within.6.months', 0.0791564582000524),\n",
              " ('glomerular.filtration.rate', 0.07458571463559802),\n",
              " ('map', 0.07445970671062938),\n",
              " ('mean.hemoglobin.concentration', 0.07392428343845732),\n",
              " ('left.ventricular.end.diastolic.diameter.LV', 0.0721600255588804),\n",
              " ('red.blood.cell', 0.06544780033475707),\n",
              " ('respiration', 0.0651224581030743),\n",
              " ('mitral.valve.AMS', 0.06292745564776678),\n",
              " ('solid.tumor', 0.0626836179646844),\n",
              " ('moderate.to.severe.chronic.kidney.disease', 0.0625428914636967),\n",
              " ('peripheral.vascular.disease', 0.060481179636034596),\n",
              " ('cystatin', 0.05326389792708659),\n",
              " ('mean.hemoglobin.volume', 0.052024346262985084),\n",
              " ('neutrophil.ratio', 0.04737433544063276),\n",
              " ('cerebrovascular.disease', 0.04711474302789805),\n",
              " ('pulse', 0.04693087706363554),\n",
              " ('re.admission.within.28.days', 0.045069758296294955),\n",
              " ('urea', 0.04286222219318756),\n",
              " ('peptic.ulcer.disease', 0.04037365223784272),\n",
              " ('creatinine.enzymatic.method', 0.040343996928470384),\n",
              " ('white.blood.cell', 0.037853948026178925),\n",
              " ('uric.acid', 0.037192892816090714),\n",
              " ('lymphocyte.count', 0.035610875181005794),\n",
              " ('platelet.hematocrit', 0.034685180939034166),\n",
              " ('death.within.3.months', 0.03411723847474325),\n",
              " ('death.within.6.months', 0.03411723847474325),\n",
              " ('coefficient.of.variation.of.red.blood.cell.distribution.width',\n",
              "  0.03337896548155749),\n",
              " ('acute.renal.failure', 0.032144135387277395),\n",
              " ('mitral.valve.EMS', 0.03092504294218211),\n",
              " ('dementia', 0.03081284697749085),\n",
              " ('monocyte.count', 0.029086643272184496),\n",
              " ('diastolic.blood.pressure', 0.02863648091585222),\n",
              " ('monocyte.ratio', 0.026750124124643997),\n",
              " ('mean.corpuscular.volume', 0.026508019254732947),\n",
              " ('platelet.distribution.width', 0.025947499343812713),\n",
              " ('liver.disease', 0.025494276559444248),\n",
              " ('body.temperature', 0.025159173330005873),\n",
              " ('standard.deviation.of.red.blood.cell.distribution.width',\n",
              "  0.024179447912854972),\n",
              " ('Chronic.obstructive.pulmonary.disease', 0.023482312153792567),\n",
              " ('death.within.28.days', 0.020957413161451038),\n",
              " ('basophil.ratio', 0.01970384411877793),\n",
              " ('visit.times', 0.017887152406775447),\n",
              " ('re.admission.within.3.months', 0.017311581025752122),\n",
              " ('basophil.count', 0.015825788371714267),\n",
              " ('tricuspid.valve.return.velocity', 0.01476777670749877),\n",
              " ('BMI', 0.011785564287354855),\n",
              " ('fio2', 0.01155464012288527),\n",
              " ('platelet', 0.010230190480538503),\n",
              " ('AIDS', 0.008247501463102288),\n",
              " ('weight', 0.007958183735066654),\n",
              " ('diabetes', 0.00710890761650827),\n",
              " ('height', 0.002206647362390603),\n",
              " ('re.admission.within.6.months', 0.0006580569073554909),\n",
              " ('cholinesterase', nan),\n",
              " ('homocysteine', 0.1931912764981171),\n",
              " ('body.temperature.blood.gas', nan),\n",
              " ('oxygen.saturation', 0.10954893470407615),\n",
              " ('partial.oxygen.pressure', 0.10827839751173139),\n",
              " ('reduced.hemoglobin', 0.10497050965831607),\n",
              " ('indirect.bilirubin', 0.10390443428993136),\n",
              " ('total.bilirubin', 0.09889320656214035),\n",
              " ('hematocrit.blood.gas', 0.09875733993361299),\n",
              " ('oxyhemoglobin', 0.09628899041455388),\n",
              " ('total.hemoglobin', 0.09534115355351062),\n",
              " ('glutamyltranspeptidase', 0.0912196837527966),\n",
              " ('globulin', 0.0882817289126106),\n",
              " ('glutamic.oxaliplatin', 0.08601929942674916),\n",
              " ('glucose.blood.gas', 0.08293326785280894),\n",
              " ('fucosidase', 0.07977633349496167),\n",
              " ('low.density.lipoprotein.cholesterol', 0.07929199976660231),\n",
              " ('sodium', 0.07854858797891846),\n",
              " ('Inorganic.Phosphorus', 0.0763962030448065),\n",
              " ('direct.bilirubin', 0.07476845846841224),\n",
              " ('cholesterol', 0.07437047463684578),\n",
              " ('sodium.ion', 0.07266836450305955),\n",
              " ('international.normalized.ratio', 0.06263034771177681),\n",
              " ('prothrombin.time.ratio', 0.06225804918867597),\n",
              " ('chloride.ion', 0.061775388349061565),\n",
              " ('nucleotidase', 0.06157534397079375),\n",
              " ('partial.pressure.of.carbon.dioxide', 0.060988857812738016),\n",
              " ('fibrinogen', 0.06078975879899065),\n",
              " ('pH', 0.05941576976047362),\n",
              " ('thrombin.time', 0.0521264752921943),\n",
              " ('alkaline.phosphatase', 0.0466677570397859),\n",
              " ('total.protein', 0.04635049020626975),\n",
              " ('high.density.lipoprotein.cholesterol', 0.04333850122781657),\n",
              " ('chloride', 0.04308072101106242),\n",
              " ('methemoglobin', 0.04265409664011644),\n",
              " ('brain.natriuretic.peptide', 0.04201117708637026),\n",
              " ('albumin', 0.03434029006111753),\n",
              " ('D.dimer', 0.03285371740076686),\n",
              " ('potassium.ion', 0.032757926041196275),\n",
              " ('hydroxybutyrate.dehydrogenase', 0.03129101913571191),\n",
              " ('hydroxybutyrate.dehydrogenase.to.lactate.dehydrogenase',\n",
              "  0.030707521603874743),\n",
              " ('total.carbon.dioxide', 0.03030941349575646),\n",
              " ('carboxyhemoglobin', 0.030149216821131676),\n",
              " ('high.sensitivity.troponin', 0.0289839500139981),\n",
              " ('lactate.dehydrogenase', 0.028462856019056543),\n",
              " ('measured.bicarbonate', 0.02832754241130626),\n",
              " ('total.bile.acid', 0.027985558941147896),\n",
              " ('neutrophil.count', 0.02534977613224184),\n",
              " ('glutamic.pyruvic.transaminase', 0.02488759852502932),\n",
              " ('anion.gap', 0.02470475923735134),\n",
              " ('lactate', 0.023980925636223588),\n",
              " ('carbon.dioxide.binding.capacity', 0.02224602694208919),\n",
              " ('free.calcium', 0.02197305810580405),\n",
              " ('creatine.kinase.isoenzyme', 0.02164650400220669),\n",
              " ('triglyceride', 0.01949150308366104),\n",
              " ('GCS', 0.015691937064130645),\n",
              " ('standard.residual.base', 0.013131684191624042),\n",
              " ('dischargeDay', 0.011912823766710308),\n",
              " ('creatine.kinase.isoenzyme.to.creatine.kinase', 0.011688862682759885),\n",
              " ('activated.partial.thromboplastin.time', 0.01157935318368123),\n",
              " ('standard.bicarbonate', 0.007041312509961021),\n",
              " ('potassium', 0.005236568919044453),\n",
              " ('glutamic.oxaloacetic.transaminase', 0.0047799795883488915),\n",
              " ('creatine.kinase', 0.004034969504504124),\n",
              " ('measured.residual.base', 0.0009666715415627601)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hL7vMWXYUZ5R"
      },
      "source": [
        "------"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m3ML1DfyOqEf"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XE3Ozk2uO7WH"
      },
      "source": [
        "temp_s = \"pulse respiration systolic.blood.pressure diastolic.blood.pressure map weight height BMI type.of.heart.failure NYHA.cardiac.function.classification Killip.grade myocardial.infarction congestive.heart.failure peripheral.vascular.disease cerebrovascular.disease dementia Chronic.obstructive.pulmonary.disease connective.tissue.disease peptic.ulcer.disease diabetes moderate.to.severe.chronic.kidney.disease hemiplegia leukemia\"\n",
        "temp = \"liver.disease AIDS CCI.score type.II.respiratory.failure consciousness fio2 acute.renal.failure LVEF left.ventricular.end.diastolic.diameter.LV mitral.valve.EMS mitral.valve.AMS EA tricuspid.valve.return.velocity tricuspid.valve.return.pressure\"\n",
        "temp = temp.split(\" \")\n",
        "feature_names = temp_s.split(\" \")\n",
        "feature_names.extend(temp)\n",
        "\n",
        "# feature names to be used"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vaNLxFzrLWAS"
      },
      "source": [
        "df_og = df\n",
        "df = df[feature_names]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2qJp7aenRL7o",
        "outputId": "67d567f8-635a-4131-f9f4-7d7ba0e2342e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "df.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 2008 entries, 0 to 2007\n",
            "Data columns (total 37 columns):\n",
            " #   Column                                      Non-Null Count  Dtype  \n",
            "---  ------                                      --------------  -----  \n",
            " 0   pulse                                       2008 non-null   int64  \n",
            " 1   respiration                                 2008 non-null   int64  \n",
            " 2   systolic.blood.pressure                     2008 non-null   int64  \n",
            " 3   diastolic.blood.pressure                    2008 non-null   int64  \n",
            " 4   map                                         2008 non-null   float64\n",
            " 5   weight                                      2008 non-null   float64\n",
            " 6   height                                      2008 non-null   float64\n",
            " 7   BMI                                         2008 non-null   float64\n",
            " 8   type.of.heart.failure                       2008 non-null   object \n",
            " 9   NYHA.cardiac.function.classification        2008 non-null   object \n",
            " 10  Killip.grade                                2008 non-null   object \n",
            " 11  myocardial.infarction                       2008 non-null   int64  \n",
            " 12  congestive.heart.failure                    2008 non-null   int64  \n",
            " 13  peripheral.vascular.disease                 2008 non-null   int64  \n",
            " 14  cerebrovascular.disease                     2008 non-null   int64  \n",
            " 15  dementia                                    2008 non-null   int64  \n",
            " 16  Chronic.obstructive.pulmonary.disease       2008 non-null   int64  \n",
            " 17  connective.tissue.disease                   2008 non-null   int64  \n",
            " 18  peptic.ulcer.disease                        2006 non-null   float64\n",
            " 19  diabetes                                    2008 non-null   int64  \n",
            " 20  moderate.to.severe.chronic.kidney.disease   2006 non-null   float64\n",
            " 21  hemiplegia                                  2008 non-null   int64  \n",
            " 22  leukemia                                    2008 non-null   int64  \n",
            " 23  liver.disease                               2007 non-null   float64\n",
            " 24  AIDS                                        2008 non-null   int64  \n",
            " 25  CCI.score                                   2003 non-null   float64\n",
            " 26  type.II.respiratory.failure                 2008 non-null   object \n",
            " 27  consciousness                               2008 non-null   object \n",
            " 28  fio2                                        2008 non-null   int64  \n",
            " 29  acute.renal.failure                         2008 non-null   int64  \n",
            " 30  LVEF                                        635 non-null    float64\n",
            " 31  left.ventricular.end.diastolic.diameter.LV  1311 non-null   float64\n",
            " 32  mitral.valve.EMS                            980 non-null    float64\n",
            " 33  mitral.valve.AMS                            550 non-null    float64\n",
            " 34  EA                                          393 non-null    float64\n",
            " 35  tricuspid.valve.return.velocity             790 non-null    float64\n",
            " 36  tricuspid.valve.return.pressure             182 non-null    float64\n",
            "dtypes: float64(15), int64(17), object(5)\n",
            "memory usage: 580.6+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0jIdu3WxZ6P_"
      },
      "source": [
        "df.dropna(subset= feature_names[:29], inplace = True)       #dropping all the null values except index >29"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DD0zyJbMR3hn",
        "outputId": "a944adac-a140-4e4c-a6dd-68787f577b96",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "df.loc[df[\"CCI.score\"].isnull() == True].index.to_list()        #finds the index of null values"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[467, 743, 817, 1003, 1497]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S6_90Ax-b1TT"
      },
      "source": [
        "df.drop(columns = \"consciousness\",inplace = True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BJJODZ5naao7",
        "outputId": "69542365-49b7-46f9-f5b0-ed763112d828",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "df.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 2003 entries, 0 to 2007\n",
            "Data columns (total 36 columns):\n",
            " #   Column                                      Non-Null Count  Dtype  \n",
            "---  ------                                      --------------  -----  \n",
            " 0   pulse                                       2003 non-null   int64  \n",
            " 1   respiration                                 2003 non-null   int64  \n",
            " 2   systolic.blood.pressure                     2003 non-null   int64  \n",
            " 3   diastolic.blood.pressure                    2003 non-null   int64  \n",
            " 4   map                                         2003 non-null   float64\n",
            " 5   weight                                      2003 non-null   float64\n",
            " 6   height                                      2003 non-null   float64\n",
            " 7   BMI                                         2003 non-null   float64\n",
            " 8   type.of.heart.failure                       2003 non-null   object \n",
            " 9   NYHA.cardiac.function.classification        2003 non-null   object \n",
            " 10  Killip.grade                                2003 non-null   object \n",
            " 11  myocardial.infarction                       2003 non-null   int64  \n",
            " 12  congestive.heart.failure                    2003 non-null   int64  \n",
            " 13  peripheral.vascular.disease                 2003 non-null   int64  \n",
            " 14  cerebrovascular.disease                     2003 non-null   int64  \n",
            " 15  dementia                                    2003 non-null   int64  \n",
            " 16  Chronic.obstructive.pulmonary.disease       2003 non-null   int64  \n",
            " 17  connective.tissue.disease                   2003 non-null   int64  \n",
            " 18  peptic.ulcer.disease                        2003 non-null   float64\n",
            " 19  diabetes                                    2003 non-null   int64  \n",
            " 20  moderate.to.severe.chronic.kidney.disease   2003 non-null   float64\n",
            " 21  hemiplegia                                  2003 non-null   int64  \n",
            " 22  leukemia                                    2003 non-null   int64  \n",
            " 23  liver.disease                               2003 non-null   float64\n",
            " 24  AIDS                                        2003 non-null   int64  \n",
            " 25  CCI.score                                   2003 non-null   float64\n",
            " 26  type.II.respiratory.failure                 2003 non-null   object \n",
            " 27  fio2                                        2003 non-null   int64  \n",
            " 28  acute.renal.failure                         2003 non-null   int64  \n",
            " 29  LVEF                                        633 non-null    float64\n",
            " 30  left.ventricular.end.diastolic.diameter.LV  1306 non-null   float64\n",
            " 31  mitral.valve.EMS                            976 non-null    float64\n",
            " 32  mitral.valve.AMS                            546 non-null    float64\n",
            " 33  EA                                          390 non-null    float64\n",
            " 34  tricuspid.valve.return.velocity             789 non-null    float64\n",
            " 35  tricuspid.valve.return.pressure             182 non-null    float64\n",
            "dtypes: float64(15), int64(17), object(4)\n",
            "memory usage: 579.0+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hj3kP0O8a2qJ"
      },
      "source": [
        "df[\"type.of.heart.failure\"].unique()\n",
        "replace_values = {'Both' : 3, 'Right': 1, 'Left': 2}\n",
        "df.replace({'type.of.heart.failure' : replace_values}, inplace =True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-1toIag7bVvL"
      },
      "source": [
        "df['Killip.grade'].unique()\n",
        "replace_values = {'III' : 3, 'I': 1, 'II': 2,'IV': 4}\n",
        "df.replace({'Killip.grade' : replace_values}, inplace =True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jEok7PShbnZx"
      },
      "source": [
        "df['NYHA.cardiac.function.classification'].unique()\n",
        "replace_values = {'III' : 3, 'I': 1, 'II': 2,'IV': 4}\n",
        "df.replace({'NYHA.cardiac.function.classification' : replace_values}, inplace =True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0MUNBwBTbujI"
      },
      "source": [
        "df['type.II.respiratory.failure'].unique()\n",
        "df['type.II.respiratory.failure'].replace(to_replace = 'NonTypeII', value = 0, inplace = True)\n",
        "df['type.II.respiratory.failure'].replace(to_replace = 'TypeII', value = 1, inplace = True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f7ewVYnqd-f4",
        "outputId": "11a69709-2930-4b73-ed63-f667bd6497e5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "df.info()       # All the data is ordinal or continuous now"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 2003 entries, 0 to 2007\n",
            "Data columns (total 36 columns):\n",
            " #   Column                                      Non-Null Count  Dtype  \n",
            "---  ------                                      --------------  -----  \n",
            " 0   pulse                                       2003 non-null   int64  \n",
            " 1   respiration                                 2003 non-null   int64  \n",
            " 2   systolic.blood.pressure                     2003 non-null   int64  \n",
            " 3   diastolic.blood.pressure                    2003 non-null   int64  \n",
            " 4   map                                         2003 non-null   float64\n",
            " 5   weight                                      2003 non-null   float64\n",
            " 6   height                                      2003 non-null   float64\n",
            " 7   BMI                                         2003 non-null   float64\n",
            " 8   type.of.heart.failure                       2003 non-null   int64  \n",
            " 9   NYHA.cardiac.function.classification        2003 non-null   int64  \n",
            " 10  Killip.grade                                2003 non-null   int64  \n",
            " 11  myocardial.infarction                       2003 non-null   int64  \n",
            " 12  congestive.heart.failure                    2003 non-null   int64  \n",
            " 13  peripheral.vascular.disease                 2003 non-null   int64  \n",
            " 14  cerebrovascular.disease                     2003 non-null   int64  \n",
            " 15  dementia                                    2003 non-null   int64  \n",
            " 16  Chronic.obstructive.pulmonary.disease       2003 non-null   int64  \n",
            " 17  connective.tissue.disease                   2003 non-null   int64  \n",
            " 18  peptic.ulcer.disease                        2003 non-null   float64\n",
            " 19  diabetes                                    2003 non-null   int64  \n",
            " 20  moderate.to.severe.chronic.kidney.disease   2003 non-null   float64\n",
            " 21  hemiplegia                                  2003 non-null   int64  \n",
            " 22  leukemia                                    2003 non-null   int64  \n",
            " 23  liver.disease                               2003 non-null   float64\n",
            " 24  AIDS                                        2003 non-null   int64  \n",
            " 25  CCI.score                                   2003 non-null   float64\n",
            " 26  type.II.respiratory.failure                 2003 non-null   int64  \n",
            " 27  fio2                                        2003 non-null   int64  \n",
            " 28  acute.renal.failure                         2003 non-null   int64  \n",
            " 29  LVEF                                        633 non-null    float64\n",
            " 30  left.ventricular.end.diastolic.diameter.LV  1306 non-null   float64\n",
            " 31  mitral.valve.EMS                            976 non-null    float64\n",
            " 32  mitral.valve.AMS                            546 non-null    float64\n",
            " 33  EA                                          390 non-null    float64\n",
            " 34  tricuspid.valve.return.velocity             789 non-null    float64\n",
            " 35  tricuspid.valve.return.pressure             182 non-null    float64\n",
            "dtypes: float64(15), int64(21)\n",
            "memory usage: 579.0 KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KyLMbyhGiMbM"
      },
      "source": [
        "df['brain.natriuretic.peptide'] = df_og['brain.natriuretic.peptide']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B-ZjRlVNeOPG"
      },
      "source": [
        "### Attempt to predict CHF with limited data\n",
        "\n",
        "Preserving :\n",
        "* LVEF\n",
        "* LV End Diastolic Diameter\n",
        "* brain.natriuretic.preptide"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AvlNc874edLF",
        "outputId": "b21e36a1-d4bf-4346-ba73-c98aa34bc181",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "len(df.loc[df['brain.natriuretic.peptide']>0].loc[df['LVEF'] > 0].loc[df['left.ventricular.end.diastolic.diameter.LV']>0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "587"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xsJXvjbSi70R"
      },
      "source": [
        "df_chf = df.dropna(subset = [\"brain.natriuretic.peptide\",'LVEF','left.ventricular.end.diastolic.diameter.LV'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xU3JQZq0lnk8"
      },
      "source": [
        "df_chf.dropna(axis=1,inplace = True)   #droping the whole features having null values"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mQ2vckA8yuEx"
      },
      "source": [
        "df_chf.columns"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WwFz9GnimeEa"
      },
      "source": [
        "Y = df_chf['congestive.heart.failure'].to_numpy()\n",
        "X = pd.DataFrame(df_chf.drop(columns = ['dementia','type.of.heart.failure','weight','height','congestive.heart.failure','peptic.ulcer.disease'])).to_numpy()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8gHHrSu_tjbc"
      },
      "source": [
        "df_X = pd.DataFrame(df_chf.drop(columns = ['dementia','type.of.heart.failure','weight','height','congestive.heart.failure','peptic.ulcer.disease']))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6LxClOiTnFjy",
        "outputId": "af5f770a-26dd-4334-e5e2-25d4a2b09e39",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "X.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(587, 26)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 87
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WPWxYwQwnH3q",
        "outputId": "a1b965a1-804f-485b-c05a-a5eb7ba5470d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "Y.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(587,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 88
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U2mjpQ--p39v"
      },
      "source": [
        "from sklearn import preprocessing\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "scaler = preprocessing.StandardScaler().fit(X)\n",
        "X = scaler.transform(X)\n",
        "\n",
        "Y.reshape(-1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ic0wegZBt2za",
        "outputId": "37c98f55-da22-42a3-dfe9-a0c51ef70792",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "Y"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 91
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yjVDrPDcBkh5"
      },
      "source": [
        "df_chf.to_csv(\"All_CHF_Dataframe.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JrCDYiWYE_cq"
      },
      "source": [
        "Two paths :\n",
        "* Predict Type of HF without LVEF\n",
        "\n",
        "        Class 0 : HFrEF ; EF < 40\n",
        "        Class 1 : HFpEF ; EF > 40\n",
        "\n",
        "* Predcit CHF without LVEF"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5UDAWjSUdfu0"
      },
      "source": [
        "Predicting the type HF"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yg7o3_80fOQJ"
      },
      "source": [
        "df_lvef = df_chf\n",
        "df_lvef['LVEF'].clip(40,41,inplace = True)         #Classifying the LVEF in two categories"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9fSNsyK6ih3K"
      },
      "source": [
        "* EF <= 40 as 0\n",
        "* EF > 40 as 1"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jhbBvdWveMc6"
      },
      "source": [
        "df_lvef['LVEF'].replace(to_replace=40,value=int(0),inplace=True)\n",
        "df_lvef['LVEF'].replace(to_replace=41,value=int(1),inplace=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RMeOl2SFj7Dh"
      },
      "source": [
        "We drop the CHF variable as well because all are 1.\n",
        "Also we wish to diagnose the early signs hence it is a target variable."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_WyKg_0BdfLb"
      },
      "source": [
        "Y = df_lvef['LVEF'].to_numpy()\n",
        "X = pd.DataFrame(df_lvef.drop(columns = ['LVEF','dementia','type.of.heart.failure','weight','height','congestive.heart.failure','peptic.ulcer.disease'])).to_numpy()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "y9wsRwJjmOnX"
      },
      "source": [
        "from sklearn import preprocessing\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wYjpw4GUm4OV"
      },
      "source": [
        "Y = Y.astype(int)\n",
        "Y = Y.reshape(-1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YGRuhIu_naoL"
      },
      "source": [
        "x_train, x_test, y_train, y_test = train_test_split(X,Y,test_size = 0.148)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nXLZSY3IutLc"
      },
      "source": [
        "scaler = preprocessing.StandardScaler().fit(x_train)\n",
        "x_train = scaler.transform(x_train)\n",
        "scaler = preprocessing.StandardScaler().fit(x_test)\n",
        "x_test = scaler.transform(x_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dgEjrlmctvg2"
      },
      "source": [
        "RandomSearch CV"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eOwHLs8nsSPg"
      },
      "source": [
        "from sklearn.model_selection import RandomizedSearchCV# Number of trees in random forest\n",
        "n_estimators = [int(x) for x in np.linspace(start = 500, stop = 2000, num = 10)]\n",
        "# Number of features to consider at every split\n",
        "max_features = ['auto', 'sqrt']\n",
        "# Maximum number of levels in tree\n",
        "max_depth = [int(x) for x in np.linspace(10, 110, num = 11)]\n",
        "max_depth.append(None)\n",
        "# Minimum number of samples required to split a node\n",
        "min_samples_split = [2, 3,5]\n",
        "# Minimum number of samples required at each leaf node\n",
        "min_samples_leaf = [ 2, 3, 4]\n",
        "# Method of selecting samples for training each tree\n",
        "bootstrap = [True, False]# Create the random grid\n",
        "random_grid = {'n_estimators': n_estimators,\n",
        "               'max_features': max_features,\n",
        "               'max_depth': max_depth,\n",
        "               'min_samples_split': min_samples_split,\n",
        "               'min_samples_leaf': min_samples_leaf,\n",
        "               'bootstrap': bootstrap} "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ry12AETJttCY",
        "outputId": "371ed99d-6cd6-4c46-acf0-efc59e2b9a1f",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "clf = RandomForestClassifier()\n",
        "rf_rscv = RandomizedSearchCV(estimator=clf,param_distributions = random_grid,n_iter = 100,cv = 5)\n",
        "rf_rscv.fit(x_train,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomizedSearchCV(cv=5, error_score=nan,\n",
              "                   estimator=RandomForestClassifier(bootstrap=True,\n",
              "                                                    ccp_alpha=0.0,\n",
              "                                                    class_weight=None,\n",
              "                                                    criterion='gini',\n",
              "                                                    max_depth=None,\n",
              "                                                    max_features='auto',\n",
              "                                                    max_leaf_nodes=None,\n",
              "                                                    max_samples=None,\n",
              "                                                    min_impurity_decrease=0.0,\n",
              "                                                    min_impurity_split=None,\n",
              "                                                    min_samples_leaf=1,\n",
              "                                                    min_samples_split=2,\n",
              "                                                    min_weight_fraction_leaf=0.0,\n",
              "                                                    n_estimators=100,\n",
              "                                                    n_jobs...\n",
              "                   param_distributions={'bootstrap': [True, False],\n",
              "                                        'max_depth': [10, 20, 30, 40, 50, 60,\n",
              "                                                      70, 80, 90, 100, 110,\n",
              "                                                      None],\n",
              "                                        'max_features': ['auto', 'sqrt'],\n",
              "                                        'min_samples_leaf': [2, 3, 4],\n",
              "                                        'min_samples_split': [2, 3, 5],\n",
              "                                        'n_estimators': [500, 666, 833, 1000,\n",
              "                                                         1166, 1333, 1500, 1666,\n",
              "                                                         1833, 2000]},\n",
              "                   pre_dispatch='2*n_jobs', random_state=None, refit=True,\n",
              "                   return_train_score=False, scoring=None, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pDE3Kr1R90IL",
        "outputId": "4da2494a-8e3e-4fab-a1e0-db8b786ef099",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "rf_rscv.best_params_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'bootstrap': True,\n",
              " 'max_depth': 60,\n",
              " 'max_features': 'auto',\n",
              " 'min_samples_leaf': 3,\n",
              " 'min_samples_split': 3,\n",
              " 'n_estimators': 1333}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_L5now6H-Nyz"
      },
      "source": [
        "Grid Search CV"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H92LbROUsCtl",
        "outputId": "2c16458d-953b-447f-c8a4-aa6dd627ebea",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "from sklearn.model_selection import GridSearchCV# Create the parameter grid based on the results of random search \n",
        "param_grid = {\n",
        "    'bootstrap': [True],\n",
        "    'max_depth': [50, 60,70,80],\n",
        "    'max_features': ['auto','sqrt'],\n",
        "    'min_samples_leaf': [2,3,4],\n",
        "    'min_samples_split': [2,3,4],\n",
        "    'n_estimators': [900,1000,1333,1400]\n",
        "}\n",
        "rf = RandomForestClassifier()# Instantiate the grid search model\n",
        "grid_search = GridSearchCV(estimator = rf, param_grid = param_grid, \n",
        "                          cv = 3, n_jobs = -1, verbose = 2)\n",
        "grid_search.fit(x_train,y_train)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Fitting 3 folds for each of 288 candidates, totalling 864 fits\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:   54.8s\n",
            "[Parallel(n_jobs=-1)]: Done 158 tasks      | elapsed:  3.8min\n",
            "[Parallel(n_jobs=-1)]: Done 361 tasks      | elapsed:  8.7min\n",
            "[Parallel(n_jobs=-1)]: Done 644 tasks      | elapsed: 15.5min\n",
            "[Parallel(n_jobs=-1)]: Done 864 out of 864 | elapsed: 20.8min finished\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GridSearchCV(cv=3, error_score=nan,\n",
              "             estimator=RandomForestClassifier(bootstrap=True, ccp_alpha=0.0,\n",
              "                                              class_weight=None,\n",
              "                                              criterion='gini', max_depth=None,\n",
              "                                              max_features='auto',\n",
              "                                              max_leaf_nodes=None,\n",
              "                                              max_samples=None,\n",
              "                                              min_impurity_decrease=0.0,\n",
              "                                              min_impurity_split=None,\n",
              "                                              min_samples_leaf=1,\n",
              "                                              min_samples_split=2,\n",
              "                                              min_weight_fraction_leaf=0.0,\n",
              "                                              n_estimators=100, n_jobs=None,\n",
              "                                              oob_score=False,\n",
              "                                              random_state=None, verbose=0,\n",
              "                                              warm_start=False),\n",
              "             iid='deprecated', n_jobs=-1,\n",
              "             param_grid={'bootstrap': [True], 'max_depth': [50, 60, 70, 80],\n",
              "                         'max_features': ['auto', 'sqrt'],\n",
              "                         'min_samples_leaf': [2, 3, 4],\n",
              "                         'min_samples_split': [2, 3, 4],\n",
              "                         'n_estimators': [900, 1000, 1333, 1400]},\n",
              "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
              "             scoring=None, verbose=2)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8tDxxNgf_l2j",
        "outputId": "fcc2590d-65bb-492d-a193-9b1e8d6292f9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "grid_search.best_params_"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'bootstrap': True,\n",
              " 'max_depth': 80,\n",
              " 'max_features': 'sqrt',\n",
              " 'min_samples_leaf': 4,\n",
              " 'min_samples_split': 3,\n",
              " 'n_estimators': 1333}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xG9FROjlrH23"
      },
      "source": [
        "clf = RandomForestClassifier(n_estimators=1333,bootstrap=True,max_depth=80, max_features='sqrt', min_samples_split= 4, min_samples_leaf=3).fit(x_train,y_train)\n",
        "y_predicted = clf.predict(x_test)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "is44yepoxlCq",
        "outputId": "9c0ea2e6-4e3b-4466-8bb6-41de6e277d62",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "y_train_pred = clf.predict(x_train)\n",
        "print(\"Accuracy Test %0.2f\" %(accuracy_score(y_test,y_predicted)*100))\n",
        "print(\"Accuracy Train %0.2f\" %(accuracy_score(y_train,y_train_pred)*100))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy Test 88.51\n",
            "Accuracy Train 94.40\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TymGjDOgub1V"
      },
      "source": [
        "Visualizations"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ajwNcQGBud-J",
        "outputId": "0476a834-a60f-4f40-bda0-355628d24cef",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "df_xtest = pd.DataFrame(x_train)\n",
        "df_xtrain.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(500, 25)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tGO2plrEs_Sq",
        "outputId": "1edde336-2819-47c1-b8c8-316988ff96a2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369
        }
      },
      "source": [
        "plt.style.use('ggplot')\n",
        "fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(15, 5))\n",
        "axes[0].scatter(df_xtrain.iloc[:,23],df_xtrain.iloc[:,4],c=y_train_pred,cmap='brg')\n",
        "axes[1].scatter(df_xtrain.iloc[:,23],df_xtrain.iloc[:,4],c=y_train,cmap='brg')\n",
        "axes[0].title.set_text(\"Predictions\")\n",
        "axes[1].title.set_text(\"Actual Plot\")\n",
        "fig.tight_layout()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1UDju_kW8C45"
      },
      "source": [
        "Importand Features :\n",
        "* BNP\n",
        "* left.ventricular.end.diastolic.diameter.LV"
      ]
    }
  ]
}