Lifesat.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "yTijF097rKrf",
      "metadata": {
        "id": "yTijF097rKrf"
      },
      "source": [
        "**Chapter 1 – The Machine Learning landscape**\n",
        "\n",
        "_This is the code used to generate some of the figures in chapter 1._\n",
        "\n",
        "<table align=\"left\">\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://colab.research.google.com/github/ColstonBod-oy/handson-ml-exercises/blob/main/Lifesat.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
        "  </td>\n",
        "</table>"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "tXI4Pv-IrKrn",
      "metadata": {
        "id": "tXI4Pv-IrKrn"
      },
      "source": [
        "# Setup"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "Sr7oYRGerKro",
      "metadata": {
        "id": "Sr7oYRGerKro"
      },
      "source": [
        "First, let's make sure this notebook works well in both python 2 and 3, import a few common modules, ensure MatplotLib plots figures inline and prepare a function to save the figures:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "cDYDS4QrrKrp",
      "metadata": {
        "id": "cDYDS4QrrKrp"
      },
      "outputs": [],
      "source": [
        "# To support both python 2 and python 3\n",
        "from __future__ import division, print_function, unicode_literals\n",
        "\n",
        "# Common imports\n",
        "import numpy as np\n",
        "import os\n",
        "\n",
        "# to make this notebook's output stable across runs\n",
        "np.random.seed(42)\n",
        "\n",
        "# To plot pretty figures\n",
        "%matplotlib inline\n",
        "import matplotlib as mpl\n",
        "import matplotlib.pyplot as plt\n",
        "mpl.rc('axes', labelsize=14)\n",
        "mpl.rc('xtick', labelsize=12)\n",
        "mpl.rc('ytick', labelsize=12)\n",
        "\n",
        "# Where to save the figures\n",
        "PROJECT_ROOT_DIR = \".\"\n",
        "CHAPTER_ID = \"fundamentals\"\n",
        "IMAGES_PATH = os.path.join(PROJECT_ROOT_DIR, \"images\", CHAPTER_ID)\n",
        "os.makedirs(IMAGES_PATH, exist_ok=True)\n",
        "\n",
        "def save_fig(fig_id, tight_layout=True, fig_extension=\"png\", resolution=300):\n",
        "    path = os.path.join(IMAGES_PATH, fig_id + \".\" + fig_extension)\n",
        "    print(\"Saving figure\", fig_id)\n",
        "    if tight_layout:\n",
        "        plt.tight_layout()\n",
        "    plt.savefig(path, format=fig_extension, dpi=resolution)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "n7539L9grKrr",
      "metadata": {
        "id": "n7539L9grKrr"
      },
      "source": [
        "# Code example 1-1"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "ZHmeDY-OrKrr",
      "metadata": {
        "id": "ZHmeDY-OrKrr"
      },
      "source": [
        "This function just merges the OECD's life satisfaction data and the IMF's GDP per capita data. It's a bit too long and boring and it's not specific to Machine Learning, which is why I left it out of the book."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "kxqmyn-5rKrs",
      "metadata": {
        "id": "kxqmyn-5rKrs"
      },
      "outputs": [],
      "source": [
        "def prepare_country_stats(oecd_bli, gdp_per_capita):\n",
        "    oecd_bli = oecd_bli[oecd_bli[\"INEQUALITY\"]==\"TOT\"]\n",
        "    oecd_bli = oecd_bli.pivot(index=\"Country\", columns=\"Indicator\", values=\"Value\")\n",
        "    gdp_per_capita.rename(columns={\"2015\": \"GDP per capita\"}, inplace=True)\n",
        "    gdp_per_capita.set_index(\"Country\", inplace=True)\n",
        "    full_country_stats = pd.merge(left=oecd_bli, right=gdp_per_capita,\n",
        "                                  left_index=True, right_index=True)\n",
        "    full_country_stats.sort_values(by=\"GDP per capita\", inplace=True)\n",
        "    remove_indices = [0, 1, 6, 8, 33, 34, 35]\n",
        "    keep_indices = list(set(range(36)) - set(remove_indices))\n",
        "    return full_country_stats[[\"GDP per capita\", 'Life satisfaction']].iloc[keep_indices]"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "iVKE42b7rKru",
      "metadata": {
        "id": "iVKE42b7rKru"
      },
      "source": [
        "The code in the book expects the data files to be located in the current directory. I just tweaked it here to fetch the files in datasets/lifesat."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "A2S-sy_yrKrv",
      "metadata": {
        "id": "A2S-sy_yrKrv"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "datapath = os.path.join(\"datasets\", \"lifesat\", \"\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "kjyyfUNxrKrv",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kjyyfUNxrKrv",
        "outputId": "48cb4839-8bfd-43eb-c880-185a6b0a660b"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading oecd_bli_2015.csv\n",
            "Downloading gdp_per_capita.csv\n"
          ]
        }
      ],
      "source": [
        "# Download the data\n",
        "import urllib.request\n",
        "DOWNLOAD_ROOT = \"https://raw.githubusercontent.com/ColstonBod-oy/handson-ml-exercises/main/\"\n",
        "os.makedirs(datapath, exist_ok=True)\n",
        "for filename in (\"oecd_bli_2015.csv\", \"gdp_per_capita.csv\"):\n",
        "    print(\"Downloading\", filename)\n",
        "    url = DOWNLOAD_ROOT + \"datasets/lifesat/\" + filename\n",
        "    urllib.request.urlretrieve(url, datapath + filename)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "YVIVhWsOrKrx",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        },
        "id": "YVIVhWsOrKrx",
        "outputId": "759057ae-2de7-42ac-9ca6-af29ec49eb59"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[5.76666667]]\n"
          ]
        }
      ],
      "source": [
        "# Code example\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import sklearn.neighbors\n",
        "\n",
        "# Load the data\n",
        "oecd_bli = pd.read_csv(datapath + \"oecd_bli_2015.csv\", thousands=',')\n",
        "gdp_per_capita = pd.read_csv(datapath + \"gdp_per_capita.csv\",thousands=',',delimiter='\\t',\n",
        "                             encoding='latin1', na_values=\"n/a\")\n",
        "\n",
        "# Prepare the data\n",
        "country_stats = prepare_country_stats(oecd_bli, gdp_per_capita)\n",
        "X = np.c_[country_stats[\"GDP per capita\"]]\n",
        "y = np.c_[country_stats[\"Life satisfaction\"]]\n",
        "\n",
        "# Visualize the data\n",
        "country_stats.plot(kind='scatter', x=\"GDP per capita\", y='Life satisfaction')\n",
        "plt.show()\n",
        "\n",
        "# Select a k-neighbors regression model\n",
        "model = sklearn.neighbors.KNeighborsRegressor(n_neighbors=3)\n",
        "\n",
        "# Train the model\n",
        "model.fit(X, y)\n",
        "\n",
        "# Make a prediction for Cyprus\n",
        "X_new = np.array([[22587.0]])  # Cyprus' GDP per capita\n",
        "print(model.predict(X_new)) # outputs [[ 5.76666667]]"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "2qfvAdY916pd",
      "metadata": {
        "id": "2qfvAdY916pd"
      },
      "source": [
        "# Save and load model"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "wDheht9fZ2kE",
      "metadata": {
        "id": "wDheht9fZ2kE"
      },
      "source": [
        "The code below saves the trained Scikit-Learn model along with the features and data types used on the training set. After that, the files were then loaded to be used again on the test set."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "wiQswD2aZxKe",
      "metadata": {
        "id": "wiQswD2aZxKe"
      },
      "source": [
        "**Note**: There's a lot of unnecessary code here, which just shows the process of how data could be saved and reused in production. Additional comments were also provided to show how the code would look like if the test set (`X_new`) was a Pandas DataFrame that contains a bigger test data. You could visit https://stackoverflow.com/questions/34864695/saving-prediction-results-to-csv for the original code."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "bUvBUPMw2PUw",
      "metadata": {
        "id": "bUvBUPMw2PUw"
      },
      "outputs": [],
      "source": [
        "#from sklearn.externals import joblib # deprecated, use import joblib instead\n",
        "import joblib\n",
        "\n",
        "# Create model path\n",
        "modelpath = os.path.join(\"models\", \"lifesat\", \"\")\n",
        "\n",
        "# Save the model\n",
        "joblib.dump(model, modelpath + \"lifesat_model.pkl\") # DIFF\n",
        "\n",
        "# Don't do if X is already a pandas.DataFrame\n",
        "X_train = country_stats[\"GDP per capita\"].to_frame() # X is a numpy.ndarray\n",
        "\n",
        "# Save columns in order\n",
        "pd.DataFrame(X_train.columns).to_csv(datapath + \"lifesat_feature_list.csv\", index = None) # X.columns\n",
        "\n",
        "# Save data types of train set\n",
        "pd.DataFrame(X_train.dtypes).reset_index().to_csv(datapath + \"lifesat_data_types.csv\", index = None) # X.dtypes"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "481c5f03",
      "metadata": {
        "id": "481c5f03"
      },
      "outputs": [],
      "source": [
        "# Load data\n",
        "feature_list = pd.read_csv(datapath + \"lifesat_feature_list.csv\")\n",
        "feature_list = pd.Index(list(feature_list[\"0\"]))\n",
        "\n",
        "# Don't do if X_new is already a pandas.DataFrame\n",
        "X_test = pd.DataFrame(X_new, columns=[\"GDP per capita\"], index=[\"Country\"]) # X_new is a numpy.ndarray\n",
        "\n",
        "add_cols = list(feature_list.difference(X_test.columns)) # X_new.columns\n",
        "drop_cols = list(X_test.columns.difference(feature_list)) # X_new.columns.difference\n",
        "\n",
        "for col in add_cols:\n",
        "    X_test[col] = np.nan # X_new[col]\n",
        "\n",
        "for col in drop_cols:\n",
        "    X_test = X_test.drop(col, axis = 1) # X_new.drop\n",
        "\n",
        "# Reorder columns\n",
        "X_test = X_test[feature_list] # X_new[feature_list]\n",
        "\n",
        "types = pd.read_csv(datapath + \"lifesat_data_types.csv\")\n",
        "for i in range(len(types)):\n",
        "    X_test[types.iloc[i,0]] = X_test[types.iloc[i,0]].astype(types.iloc[i,1]) # X_new[types.iloc[i,0]]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "95c7b812",
      "metadata": {
        "id": "95c7b812"
      },
      "outputs": [],
      "source": [
        "# Load model\n",
        "model_loaded = joblib.load(modelpath + \"lifesat_model.pkl\") # DIFF\n",
        "\n",
        "# Don't do if model was fitted with feature names\n",
        "X_test = np.c_[X_test[\"GDP per capita\"]] # KNeighborsRegressor was fitted without feature names\n",
        "\n",
        "# Make predictions\n",
        "predictions = model_loaded.predict(X_test) # X_new\n",
        "\n",
        "# Don't do if X_test is already a pandas.DataFrame\n",
        "X_test = pd.DataFrame(X_test, columns=[\"GDP per capita\"], index=[\"Country\"]) # X_test was converted to a numpy.ndarray"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "9be4fa35",
      "metadata": {
        "id": "9be4fa35"
      },
      "outputs": [],
      "source": [
        "# Save prediction results\n",
        "resultpath = os.path.join(\"results\", \"lifesat\", \"\")\n",
        "res = pd.DataFrame([[\"Cyprus\", predictions.item(0)]]) # pd.DataFrame(predictions)\n",
        "res.columns = [\"Country\", \"Prediction\"] # [\"Prediction\"]\n",
        "res.set_index(X_test.index[0], inplace=True) # res.index = X_new.index # it's important for comparison \n",
        "res.to_csv(resultpath + \"lifesat_prediction_results.csv\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T7VWUJ_T3Ggk"
      },
      "source": [
        "# Save and load model using Google Drive"
      ],
      "id": "T7VWUJ_T3Ggk"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Fr4ZdEZ43axY"
      },
      "source": [
        "**Note**: This is only meant to be used when trying to save models while using Google Colab; otherwise, use `joblib` to save locally."
      ],
      "id": "Fr4ZdEZ43axY"
    },
    {
      "cell_type": "code",
      "source": [
        "# Mount the drive\n",
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive')"
      ],
      "metadata": {
        "id": "d8wXfPyT5N9K",
        "outputId": "a79d7eb2-02db-445c-a347-07b9bddf16d8",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "id": "d8wXfPyT5N9K",
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/gdrive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Save the model\n",
        "import torch\n",
        "model_save_name = 'model.pt'\n",
        "path = F\"/content/gdrive/My Drive/{model_save_name}\" \n",
        "torch.save(model, path)"
      ],
      "metadata": {
        "id": "U1_WGIwZ5vcI"
      },
      "id": "U1_WGIwZ5vcI",
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load model\n",
        "model = torch.load(path)\n",
        "model"
      ],
      "metadata": {
        "id": "d-Os-Gna6suv",
        "outputId": "118e4558-6b82-4108-ee69-ae6bf00c35d6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "id": "d-Os-Gna6suv",
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "KNeighborsRegressor(n_neighbors=3)"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "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": "3.9.12"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}