mlessentials/Lab03/Exercise3.02/Exercise3_02.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Qq90W1oIrv1j"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import altair as alt\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1375,
     "status": "ok",
     "timestamp": 1572999445313,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "lX4RqHx3rfSd",
    "outputId": "735616a9-a3e3-447e-90c9-8a7b5ac64072"
   },
   "outputs": [],
   "source": [
    "file_url = '../bank-full.csv'\n",
    "bankData = pd.read_csv(file_url, sep=\";\")\n",
    "bankData.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1371,
     "status": "ok",
     "timestamp": 1572999445314,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "Rz80uwq9AVhf",
    "outputId": "013dcd08-9f5c-4305-913e-c66057a3c881"
   },
   "outputs": [],
   "source": [
    "filter_mask = bankData['y'] == 'yes'\n",
    "bankSub1 = bankData[filter_mask].groupby('age')['y'].agg(agegrp='count').reset_index()\n",
    "bankSub1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 368
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1364,
     "status": "ok",
     "timestamp": 1572999445314,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "0gLl4ExhDThV",
    "outputId": "ff8f65de-47fe-4318-ebf9-76ebebdc0ec0"
   },
   "outputs": [],
   "source": [
    "# Visualising the relationship using altair\n",
    "alt.Chart(bankSub1).mark_line().encode(x='age', y='agegrp')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1358,
     "status": "ok",
     "timestamp": 1572999445315,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "nEWi-O5nCHtt",
    "outputId": "8f8ed75e-d711-48a2-a2fe-b92a6d06ee31"
   },
   "outputs": [],
   "source": [
    "# Getting another perspective\n",
    "ageTot = bankData.groupby('age')['y'].agg(ageTot='count').reset_index()\n",
    "ageTot.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1353,
     "status": "ok",
     "timestamp": 1572999445316,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "AcVBmjvLCk3O",
    "outputId": "5ccf99b4-488f-4de8-aa99-fc154fbbe408"
   },
   "outputs": [],
   "source": [
    "# Getting all the details in one place\n",
    "ageProp = bankData.groupby(['age','y'])['y'].agg(ageCat='count').reset_index()\n",
    "ageProp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1347,
     "status": "ok",
     "timestamp": 1572999445316,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "q0j3PhBtCv_O",
    "outputId": "03a37cb9-4af7-44b6-9f20-8de6abcdee31"
   },
   "outputs": [],
   "source": [
    "# Merging both the data frames\n",
    "ageComb = pd.merge(ageProp, ageTot,left_on = ['age'], right_on = ['age'])\n",
    "ageComb['catProp'] = (ageComb.ageCat/ageComb.ageTot)*100\n",
    "ageComb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 418
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1342,
     "status": "ok",
     "timestamp": 1572999445317,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "p-n6MerCEMyf",
    "outputId": "c6b83817-4f8e-474d-99ef-6d3282d43c82"
   },
   "outputs": [],
   "source": [
    "# Visualising the relationship using altair\n",
    "alt.Chart(ageComb).mark_line().encode(x='age', y='catProp').facet(column='y')"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Exercise3.02_updated.ipynb",
   "provenance": []
  },
  "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": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}