mlessentials/Lab11/Exercise11.02/Exercise11_02.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "HEiOAwQPW0qb"
   },
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "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": 221
    },
    "colab_type": "code",
    "id": "VFZSIuUjddvZ",
    "outputId": "fc2622b9-88ca-4495-bb23-32c5fe6f43bc"
   },
   "outputs": [],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "mWw1dgZTgh5r"
   },
   "outputs": [],
   "source": [
    "df['Id'] = df['Id'].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "L9012mdDoiID"
   },
   "outputs": [],
   "source": [
    "df['MSSubClass'] = df['MSSubClass'].astype('category')\n",
    "df['OverallQual'] = df['OverallQual'].astype('category')\n",
    "df['OverallCond'] = df['OverallCond'].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "id": "uiLlKYp1p1Lp",
    "outputId": "b5c505a5-47ba-429b-fb39-a5a8df4636c1"
   },
   "outputs": [],
   "source": [
    "for col_name in ['Id', 'MSSubClass', 'OverallQual', 'OverallCond']:\n",
    "  print(col_name)\n",
    "  print(df[col_name].cat.categories)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "eLSewbSud4vl"
   },
   "outputs": [],
   "source": [
    "obj_df = df.select_dtypes(include='object')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 187
    },
    "colab_type": "code",
    "id": "U3FmCwWrqu0-",
    "outputId": "29157bcd-57d3-41b8-875d-4126aa21bd88"
   },
   "outputs": [],
   "source": [
    "obj_cols = obj_df.columns\n",
    "obj_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "colab_type": "code",
    "id": "Dc7pw1ZufBdL",
    "outputId": "ac425683-1c23-4551-b2ae-6f7960eae694"
   },
   "outputs": [],
   "source": [
    "for col_name in obj_cols:\n",
    "  print(col_name)\n",
    "  print(df[col_name].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "6OgiuQVwfQ-l"
   },
   "outputs": [],
   "source": [
    "for col_name in obj_cols:\n",
    "  df[col_name] = df[col_name].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 221
    },
    "colab_type": "code",
    "id": "0z64JDTcfaXv",
    "outputId": "643b14fd-a3bb-4322-e3eb-1fc9526f27a0"
   },
   "outputs": [],
   "source": [
    "df.dtypes"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "Exercise11.02.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
}