mlessentials/Lab12/Exercise12.02/Exercise12_02.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "HEiOAwQPW0qb"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import altair as alt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "j7whidfaYjns"
   },
   "outputs": [],
   "source": [
    "file_url = '../Dataset/ames_iowa_housing.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "nFMz2jNVt-xy"
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv(file_url)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 253
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2600,
     "status": "ok",
     "timestamp": 1574650583600,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "Lp9Bk6AWlK0n",
    "outputId": "eb425c9c-2266-4707-e855-badde94603ee"
   },
   "outputs": [],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 2595,
     "status": "ok",
     "timestamp": 1574650583601,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "QjJwGSZCMMPb",
    "outputId": "75b24e8f-4b91-4c4b-98d9-bff706550e7d"
   },
   "outputs": [],
   "source": [
    "df['YearBuilt'].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 401,
     "output_embedded_package_id": "1g6Glb2CHeLNvlz3dXLyEkUS9Zh0Ouk89"
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4198,
     "status": "ok",
     "timestamp": 1574650585210,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "hh-DdotZmXd_",
    "outputId": "03a39169-6555-4734-f77e-77febec0afc8"
   },
   "outputs": [],
   "source": [
    "alt.Chart(df).mark_circle().encode(alt.X('YearBuilt:O'), y='count()')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "AsklvQuJm4pO"
   },
   "outputs": [],
   "source": [
    "year_built = df['YearBuilt'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "H5tmt8o7nQFY"
   },
   "outputs": [],
   "source": [
    "decade_list = [year - (year % 10) for year in year_built]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "zy4lCGRQpvGZ"
   },
   "outputs": [],
   "source": [
    "decade_built = sorted(set(decade_list))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 272
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4181,
     "status": "ok",
     "timestamp": 1574650585213,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "wYvKuXJLnv9X",
    "outputId": "765d14c3-4e51-4dce-f58c-1ac80e70fe4f"
   },
   "outputs": [],
   "source": [
    "decade_built"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "NZMRmR_1n2Sc"
   },
   "outputs": [],
   "source": [
    "df['DecadeBuilt'] = pd.cut(df['YearBuilt'], bins=decade_built)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 4173,
     "status": "ok",
     "timestamp": 1574650585214,
     "user": {
      "displayName": "Anthony So",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AAuE7mCYY-iGjUIqBSnlLoszfZTN7rU7FRNg05Rdt9Ii3A=s64",
      "userId": "11809607246124237079"
     },
     "user_tz": -660
    },
    "id": "xJwp63DyLPu_",
    "outputId": "c8b63e9e-4b63-4bbf-d246-3a28e28312a2"
   },
   "outputs": [],
   "source": [
    "df[['YearBuilt', 'DecadeBuilt']].head()"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Exercise12_02.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
}