mlessentials/Lab10/Activity10.1/Activity10_1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "lPeZ4zc4NZ3G"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import altair as alt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "aPjMg_BfNjk7"
   },
   "outputs": [],
   "source": [
    "file_url = '../dataset/churn.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "6ynf1L9mOm0p"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(file_url)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "5BH1588M5Xiu",
    "outputId": "d9c77a45-9b3c-4f0e-b1e3-9e5ca93cf5cc"
   },
   "outputs": [],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 340
    },
    "colab_type": "code",
    "id": "lGNVxXRa5ml7",
    "outputId": "c3ab5acb-1c0e-430a-b373-f9c549565401"
   },
   "outputs": [],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 224
    },
    "colab_type": "code",
    "id": "v9a1K2z45yj8",
    "outputId": "4e12e44e-aae0-47a3-b045-c3e35338a974"
   },
   "outputs": [],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 224
    },
    "colab_type": "code",
    "id": "j8HTP8ev6JfU",
    "outputId": "adbb749c-35e4-414b-e873-2a1c57833089"
   },
   "outputs": [],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 224
    },
    "colab_type": "code",
    "id": "jAe01B6c6ouX",
    "outputId": "f464197a-49c6-422b-b1b0-18d5253d397d"
   },
   "outputs": [],
   "source": [
    "df.sample(n=5, random_state=8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 317
    },
    "colab_type": "code",
    "id": "_YRYPtRW7FrQ",
    "outputId": "3c0f9d06-3892-49cf-f50a-c258434dc36c"
   },
   "outputs": [],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "id": "Vu1Cj1Ad7lvn",
    "outputId": "286dc11a-cb89-458b-a6ea-d834557b0599"
   },
   "outputs": [],
   "source": [
    "for col_name in ['churn', 'internationalplan', 'voicemailplan']:\n",
    "  print(f\"{col_name}: {df[col_name].unique()}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "lWJ8lXbX7m8u"
   },
   "outputs": [],
   "source": [
    "chart1 = alt.Chart(df).mark_bar().encode(\n",
    "    alt.X(\"churn\"), \n",
    "    y='count()'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "9HIBNaD79_X7"
   },
   "outputs": [],
   "source": [
    "chart2 = alt.Chart(df).mark_bar().encode(\n",
    "    alt.X(\"internationalplan\"), \n",
    "    y='count()'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "EXEXUCR8-I8v"
   },
   "outputs": [],
   "source": [
    "chart3 = alt.Chart(df).mark_bar().encode(\n",
    "    alt.X(\"voicemailplan\"), \n",
    "    y='count()'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 377
    },
    "colab_type": "code",
    "id": "6QNfvoXD-UR0",
    "outputId": "15358e7d-23c6-48d5-82b5-415ccc717762"
   },
   "outputs": [],
   "source": [
    "chart1 | chart2 | chart3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "HIIQZYLg_aIM"
   },
   "outputs": [],
   "source": [
    "num_df = df.select_dtypes(include='number')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 119
    },
    "colab_type": "code",
    "id": "YIfkkZpI_gyt",
    "outputId": "5035d229-3e77-4945-8406-b43c44bb6b41"
   },
   "outputs": [],
   "source": [
    "num_cols = num_df.columns\n",
    "num_cols.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "AZ2zHyKjCFo-"
   },
   "outputs": [],
   "source": [
    "charts = alt.Chart(df).mark_bar().encode(\n",
    "    alt.X(alt.repeat(\"column\"), type='quantitative', bin=True),\n",
    "    y='count()'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 390
    },
    "colab_type": "code",
    "id": "O74hCOKP-tBC",
    "outputId": "d5ff5894-285e-41c0-c59c-fefd56a36a97"
   },
   "outputs": [],
   "source": [
    "charts.repeat(\n",
    "    column=list(num_cols)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 375
    },
    "colab_type": "code",
    "id": "0AEba_79GRDO",
    "outputId": "cfc83547-6f3a-42d4-c0cc-27f2dacbe186"
   },
   "outputs": [],
   "source": [
    "alt.Chart(df).mark_boxplot().encode(\n",
    "    x='churn',\n",
    "    y='totaldaycalls'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 375
    },
    "colab_type": "code",
    "id": "Ui2womT0GGra",
    "outputId": "f62243a9-7aa8-4e84-f016-9b3afc982763"
   },
   "outputs": [],
   "source": [
    "alt.Chart(df).mark_boxplot().encode(\n",
    "    x='churn',\n",
    "    y='numbervmailmessages'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Activity10_1.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
}