boosted_trees.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "boosted_trees.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "7765UFHoyGx6"
      },
      "source": [
        "##### Copyright 2019 The TensorFlow Authors."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "cellView": "form",
        "colab_type": "code",
        "id": "KVtTDrUNyL7x",
        "colab": {}
      },
      "source": [
        "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "xPYxZMrWyA0N"
      },
      "source": [
        "# Boosted trees using Estimators"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "p_vOREjRx-Y0"
      },
      "source": [
        "<table class=\"tfo-notebook-buttons\" align=\"left\">\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://www.tensorflow.org/tutorials/estimator/boosted_trees\"><img src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" />View on TensorFlow.org</a>\n",
        "  </td>\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/docs/blob/master/site/en/tutorials/estimator/boosted_trees.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
        "  </td>\n",
        "  <td>\n",
        "    <a target=\"_blank\" href=\"https://github.com/tensorflow/docs/blob/master/site/en/tutorials/estimator/boosted_trees.ipynb\"><img src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\">View source on GitHub</a>\n",
        "  </td>\n",
        "</table>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "dW3r7qVxzqN5"
      },
      "source": [
        "This tutorial is an end-to-end walkthrough of training a Gradient Boosting  model using decision trees with the `tf.estimator` API. Boosted Trees models are among the most popular and effective machine learning approaches for both regression and classification. It is an ensemble technique that combines the predictions from several (think 10s, 100s or even 1000s) tree models.\n",
        "\n",
        "Boosted Trees models are popular with many machine learning practitioners as they can achieve impressive performance with minimal hyperparameter tuning."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "eylrTPAN3rJV"
      },
      "source": [
        "## Load the titanic dataset\n",
        "You will be using the titanic dataset, where the (rather morbid) goal is to predict passenger survival, given characteristics such as gender, age, class, etc."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "KuhAiPfZ3rJW",
        "colab": {}
      },
      "source": [
        "from __future__ import absolute_import, division, print_function, unicode_literals\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from IPython.display import clear_output\n",
        "from matplotlib import pyplot as plt\n",
        "\n",
        "# Load dataset.\n",
        "dftrain = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/train.csv')\n",
        "dfeval = pd.read_csv('https://storage.googleapis.com/tf-datasets/titanic/eval.csv')\n",
        "y_train = dftrain.pop('survived')\n",
        "y_eval = dfeval.pop('survived')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "NFtnFm1T0kMf",
        "colab": {}
      },
      "source": [
        "try:\n",
        "  %tensorflow_version 2.x\n",
        "except Exception:\n",
        "  pass\n",
        "import tensorflow as tf\n",
        "tf.random.set_seed(123)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "3ioodHdVJVdA"
      },
      "source": [
        "The dataset consists of a training set and an evaluation set:\n",
        "\n",
        "* `dftrain` and `y_train` are the *training set*—the data the model uses to learn.\n",
        "* The model is tested against the *eval set*, `dfeval`, and `y_eval`.\n",
        "\n",
        "For training you will use the following features:\n",
        "\n",
        "\n",
        "<table>\n",
        "  <tr>\n",
        "    <th>Feature Name</th>\n",
        "    <th>Description</th>\n",
        "  </tr>\n",
        "  <tr>\n",
        "    <td>sex</td>\n",
        "    <td>Gender of passenger</td>\n",
        "  </tr>\n",
        "  <tr>\n",
        "    <td>age</td>\n",
        "    <td>Age of passenger</td>\n",
        "  </tr>\n",
        "    <tr>\n",
        "    <td>n_siblings_spouses</td>\n",
        "    <td># siblings and partners aboard</td>\n",
        "  </tr>\n",
        "    <tr>\n",
        "    <td>parch</td>\n",
        "    <td># of parents and children aboard</td>\n",
        "  </tr>\n",
        "    <tr>\n",
        "    <td>fare</td>\n",
        "    <td>Fare passenger paid.</td>\n",
        "  </tr>\n",
        "    <tr>\n",
        "    <td>class</td>\n",
        "    <td>Passenger's class on ship</td>\n",
        "  </tr>\n",
        "    <tr>\n",
        "    <td>deck</td>\n",
        "    <td>Which deck passenger was on</td>\n",
        "  </tr>\n",
        "    <tr>\n",
        "    <td>embark_town</td>\n",
        "    <td>Which town passenger embarked from</td>\n",
        "  </tr>\n",
        "    <tr>\n",
        "    <td>alone</td>\n",
        "    <td>If passenger was alone</td>\n",
        "  </tr>\n",
        "</table>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "AoPiWsJALr-k"
      },
      "source": [
        "## Explore the data"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "slcat1yzmzw5"
      },
      "source": [
        "Let's first preview some of the data and create summary statistics on the training set."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "15PLelXBlxEW",
        "outputId": "b300d1e6-b3c2-476e-c7a2-a5a00d3ef03f",
        "colab": {
          "height": 204
        }
      },
      "source": [
        "dftrain.head()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>sex</th>\n",
              "      <th>age</th>\n",
              "      <th>n_siblings_spouses</th>\n",
              "      <th>parch</th>\n",
              "      <th>fare</th>\n",
              "      <th>class</th>\n",
              "      <th>deck</th>\n",
              "      <th>embark_town</th>\n",
              "      <th>alone</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>male</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>Third</td>\n",
              "      <td>unknown</td>\n",
              "      <td>Southampton</td>\n",
              "      <td>n</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>female</td>\n",
              "      <td>38.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>First</td>\n",
              "      <td>C</td>\n",
              "      <td>Cherbourg</td>\n",
              "      <td>n</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>female</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>Third</td>\n",
              "      <td>unknown</td>\n",
              "      <td>Southampton</td>\n",
              "      <td>y</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>female</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>First</td>\n",
              "      <td>C</td>\n",
              "      <td>Southampton</td>\n",
              "      <td>n</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>male</td>\n",
              "      <td>28.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.4583</td>\n",
              "      <td>Third</td>\n",
              "      <td>unknown</td>\n",
              "      <td>Queenstown</td>\n",
              "      <td>y</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "      sex   age  n_siblings_spouses  parch  ...  class     deck  embark_town alone\n",
              "0    male  22.0                   1      0  ...  Third  unknown  Southampton     n\n",
              "1  female  38.0                   1      0  ...  First        C    Cherbourg     n\n",
              "2  female  26.0                   0      0  ...  Third  unknown  Southampton     y\n",
              "3  female  35.0                   1      0  ...  First        C  Southampton     n\n",
              "4    male  28.0                   0      0  ...  Third  unknown   Queenstown     y\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 0
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "j2hiM4ETmqP0",
        "outputId": "e8b8cc92-0a70-44bd-b7a6-54b16bb0ed63",
        "colab": {
          "height": 297
        }
      },
      "source": [
        "dftrain.describe()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>age</th>\n",
              "      <th>n_siblings_spouses</th>\n",
              "      <th>parch</th>\n",
              "      <th>fare</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>627.000000</td>\n",
              "      <td>627.000000</td>\n",
              "      <td>627.000000</td>\n",
              "      <td>627.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>29.631308</td>\n",
              "      <td>0.545455</td>\n",
              "      <td>0.379585</td>\n",
              "      <td>34.385399</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>12.511818</td>\n",
              "      <td>1.151090</td>\n",
              "      <td>0.792999</td>\n",
              "      <td>54.597730</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.750000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>23.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>7.895800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>28.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>15.045800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>35.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>31.387500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>80.000000</td>\n",
              "      <td>8.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>512.329200</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              age  n_siblings_spouses       parch        fare\n",
              "count  627.000000          627.000000  627.000000  627.000000\n",
              "mean    29.631308            0.545455    0.379585   34.385399\n",
              "std     12.511818            1.151090    0.792999   54.597730\n",
              "min      0.750000            0.000000    0.000000    0.000000\n",
              "25%     23.000000            0.000000    0.000000    7.895800\n",
              "50%     28.000000            0.000000    0.000000   15.045800\n",
              "75%     35.000000            1.000000    0.000000   31.387500\n",
              "max     80.000000            8.000000    5.000000  512.329200"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 0
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "-IR0e8V-LyJ4"
      },
      "source": [
        "There are 627 and 264 examples in the training and evaluation sets, respectively."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "_1NwYqGwDjFf",
        "outputId": "b9901780-2cbc-4731-c040-21b79a245c3f",
        "colab": {
          "height": 34
        }
      },
      "source": [
        "dftrain.shape[0], dfeval.shape[0]"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(627, 264)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 0
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "28UFJ4KSMK3V"
      },
      "source": [
        "The majority of passengers are in their 20's and 30's."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "CaVDmZtuDfux",
        "outputId": "df4752dc-9976-4db1-9f09-6a057c1b6f46",
        "colab": {
          "height": 279
        }
      },
      "source": [
        "dftrain.age.hist(bins=20)\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYMAAAEGCAYAAACHGfl5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAwCAKAMAgCgDAIAoAwCAKAMAgKT/Az96X9CTJZCZAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd15df1c150>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "1pifWiCoMbR5"
      },
      "source": [
        "There are approximately twice as male passengers as female passengers aboard."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "-WazAq30MO5J",
        "outputId": "77a2208b-a25f-4ee8-a31e-8cb3fb1a5f41",
        "colab": {
          "height": 274
        }
      },
      "source": [
        "dftrain.sex.value_counts().plot(kind='barh')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZcAAAEBCAYAAAC0WehTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd15cb6afd0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "7_XkxrpmmVU_"
      },
      "source": [
        "The majority of passengers were in the \"third\" class."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "zZ3PvVy4l4gI",
        "outputId": "147730b6-c79a-4864-9a40-b40e97383e8c",
        "colab": {
          "height": 274
        }
      },
      "source": [
        "dftrain['class'].value_counts().plot(kind='barh')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZoAAAEBCAYAAABBp2PjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAwAw9X+gI1LNFj+FVgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd15cb7aed0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "HM5SlwlxmZMT"
      },
      "source": [
        "Most passengers embarked from Southampton."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "RVTSrdr4mZaC",
        "outputId": "3512101a-0913-48d3-d4b0-6e0fdc2ed34d",
        "colab": {
          "height": 274
        }
      },
      "source": [
        "dftrain['embark_town'].value_counts().plot(kind='barh')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbwAAAEBCAYAAAAD5BB0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRESMoMITEREjqPBERMQIKjwRETHC/wNAPMBcj8HpnQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd15cfd73d0>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "aTn1niLPob3x"
      },
      "source": [
        "Females have a much higher chance of surviving vs. males. This will clearly be a predictive feature for the model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "Eh3KW5oYkaNS",
        "outputId": "830984b0-0fbc-4d69-b107-9d4097b0549f",
        "colab": {
          "height": 292
        }
      },
      "source": [
        "pd.concat([dftrain, y_train], axis=1).groupby('sex').survived.mean().plot(kind='barh').set_xlabel('% survive')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAagAAAETCAYAAABjv5J2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAIxEoAAARiJQAAAjESgAgJH+HwQ8SRjqGoVPAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd15d1a8850>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "krkRHuMp3rJn"
      },
      "source": [
        "## Create feature columns and input functions\n",
        "The Gradient Boosting estimator can utilize both numeric and categorical features. Feature columns work with all TensorFlow estimators and their purpose is to define the features used for modeling. Additionally they provide some feature engineering capabilities like one-hot-encoding, normalization, and bucketization. In this tutorial, the fields in `CATEGORICAL_COLUMNS` are transformed from categorical columns to one-hot-encoded columns ([indicator column](https://www.tensorflow.org/api_docs/python/tf/feature_column/indicator_column)):"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "upaNWxcF3rJn",
        "colab": {}
      },
      "source": [
        "fc = tf.feature_column\n",
        "CATEGORICAL_COLUMNS = ['sex', 'n_siblings_spouses', 'parch', 'class', 'deck',\n",
        "                       'embark_town', 'alone']\n",
        "NUMERIC_COLUMNS = ['age', 'fare']\n",
        "\n",
        "def one_hot_cat_column(feature_name, vocab):\n",
        "  return tf.feature_column.indicator_column(\n",
        "      tf.feature_column.categorical_column_with_vocabulary_list(feature_name,\n",
        "                                                 vocab))\n",
        "feature_columns = []\n",
        "for feature_name in CATEGORICAL_COLUMNS:\n",
        "  # Need to one-hot encode categorical features.\n",
        "  vocabulary = dftrain[feature_name].unique()\n",
        "  feature_columns.append(one_hot_cat_column(feature_name, vocabulary))\n",
        "\n",
        "for feature_name in NUMERIC_COLUMNS:\n",
        "  feature_columns.append(tf.feature_column.numeric_column(feature_name,\n",
        "                                           dtype=tf.float32))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "74GNtFpStSAz"
      },
      "source": [
        "You can view the transformation that a feature column produces. For example, here is the output when using the `indicator_column` on a single example:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "Eaq79D9FtmF8",
        "outputId": "d6db9dc1-a533-46c7-b78a-8ce367056a59",
        "colab": {
          "height": 51
        }
      },
      "source": [
        "example = dict(dftrain.head(1))\n",
        "class_fc = tf.feature_column.indicator_column(tf.feature_column.categorical_column_with_vocabulary_list('class', ('First', 'Second', 'Third')))\n",
        "print('Feature value: \"{}\"'.format(example['class'].iloc[0]))\n",
        "print('One-hot encoded: ', tf.keras.layers.DenseFeatures([class_fc])(example).numpy())"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Feature value: \"Third\"\n",
            "One-hot encoded:  [[ 0.  0.  1.]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "YbCUn3nCusC3"
      },
      "source": [
        "Additionally, you can view all of the feature column transformations together:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "omIYcsVws3g0",
        "outputId": "165d15ee-4e7f-481d-f623-106f5dbc9d23",
        "colab": {
          "height": 102
        }
      },
      "source": [
        "tf.keras.layers.DenseFeatures(feature_columns)(example).numpy()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[ 22.  ,   1.  ,   0.  ,   1.  ,   0.  ,   0.  ,   1.  ,   0.  ,\n",
              "          0.  ,   0.  ,   0.  ,   0.  ,   0.  ,   0.  ,   1.  ,   0.  ,\n",
              "          0.  ,   0.  ,   7.25,   1.  ,   0.  ,   0.  ,   0.  ,   0.  ,\n",
              "          0.  ,   0.  ,   1.  ,   0.  ,   0.  ,   0.  ,   0.  ,   0.  ,\n",
              "          1.  ,   0.  ]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 0
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "-UOlROp33rJo"
      },
      "source": [
        "Next you need to create the input functions. These will specify how data will be read into our model for both training and inference. You will use the `from_tensor_slices` method in the [`tf.data`](https://www.tensorflow.org/api_docs/python/tf/data) API to read in data directly from Pandas. This is suitable for smaller, in-memory datasets. For larger datasets, the tf.data API supports a variety of file formats (including [csv](https://www.tensorflow.org/api_docs/python/tf/data/experimental/make_csv_dataset)) so that you can process datasets that do not fit in memory."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "9dquwCQB3rJp",
        "colab": {}
      },
      "source": [
        "# Use entire batch since this is such a small dataset.\n",
        "NUM_EXAMPLES = len(y_train)\n",
        "\n",
        "def make_input_fn(X, y, n_epochs=None, shuffle=True):\n",
        "  def input_fn():\n",
        "    dataset = tf.data.Dataset.from_tensor_slices((dict(X), y))\n",
        "    if shuffle:\n",
        "      dataset = dataset.shuffle(NUM_EXAMPLES)\n",
        "    # For training, cycle thru dataset as many times as need (n_epochs=None).\n",
        "    dataset = dataset.repeat(n_epochs)\n",
        "    # In memory training doesn't use batching.\n",
        "    dataset = dataset.batch(NUM_EXAMPLES)\n",
        "    return dataset\n",
        "  return input_fn\n",
        "\n",
        "# Training and evaluation input functions.\n",
        "train_input_fn = make_input_fn(dftrain, y_train)\n",
        "eval_input_fn = make_input_fn(dfeval, y_eval, shuffle=False, n_epochs=1)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "HttfNNlN3rJr"
      },
      "source": [
        "## Train and evaluate the model\n",
        "\n",
        "Below you will do the following steps:\n",
        "\n",
        "1. Initialize the model, specifying the features and hyperparameters.\n",
        "2. Feed the training data to the model using the `train_input_fn` and train the model using the `train` function.\n",
        "3. You will assess model performance using the evaluation set—in this example, the `dfeval` DataFrame. You will verify that the predictions match the labels from the `y_eval` array.\n",
        "\n",
        "Before training a Boosted Trees model, let's first train a linear classifier (logistic regression model). It is best practice to start with simpler model to establish a benchmark."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "JPOGpmmq3rJr",
        "outputId": "0b6bd66c-90fb-40b6-ed15-7130ff63627e",
        "colab": {
          "height": 221
        }
      },
      "source": [
        "linear_est = tf.estimator.LinearClassifier(feature_columns)\n",
        "\n",
        "# Train model.\n",
        "linear_est.train(train_input_fn, max_steps=100)\n",
        "\n",
        "# Evaluation.\n",
        "result = linear_est.evaluate(eval_input_fn)\n",
        "clear_output()\n",
        "print(pd.Series(result))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "accuracy                  0.765152\n",
            "accuracy_baseline         0.625000\n",
            "auc                       0.832844\n",
            "auc_precision_recall      0.789631\n",
            "average_loss              0.478908\n",
            "global_step             100.000000\n",
            "label/mean                0.375000\n",
            "loss                      0.478908\n",
            "precision                 0.703297\n",
            "prediction/mean           0.350790\n",
            "recall                    0.646465\n",
            "dtype: float64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "BarkNXwA3rJu"
      },
      "source": [
        "Next let's train a Boosted Trees model. For boosted trees, regression (`BoostedTreesRegressor`) and classification (`BoostedTreesClassifier`) are supported. Since the goal is to predict a class - survive or not survive, you will use the `BoostedTreesClassifier`.\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "tgEzMtlw3rJu",
        "outputId": "2ac2d0f9-a7e9-46aa-d8f5-569a65c32251",
        "colab": {
          "height": 221
        }
      },
      "source": [
        "# Since data fits into memory, use entire dataset per layer. It will be faster.\n",
        "# Above one batch is defined as the entire dataset.\n",
        "n_batches = 1\n",
        "est = tf.estimator.BoostedTreesClassifier(feature_columns,\n",
        "                                          n_batches_per_layer=n_batches)\n",
        "\n",
        "# The model will stop training once the specified number of trees is built, not\n",
        "# based on the number of steps.\n",
        "est.train(train_input_fn, max_steps=100)\n",
        "\n",
        "# Eval.\n",
        "result = est.evaluate(eval_input_fn)\n",
        "clear_output()\n",
        "print(pd.Series(result))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "accuracy                  0.829545\n",
            "accuracy_baseline         0.625000\n",
            "auc                       0.872788\n",
            "auc_precision_recall      0.857807\n",
            "average_loss              0.411839\n",
            "global_step             100.000000\n",
            "label/mean                0.375000\n",
            "loss                      0.411839\n",
            "precision                 0.793478\n",
            "prediction/mean           0.381942\n",
            "recall                    0.737374\n",
            "dtype: float64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "hEflwznXvuMP"
      },
      "source": [
        "Now you can use the train model to make predictions on a passenger from the evaluation set. TensorFlow models are optimized to make predictions on a batch, or collection, of examples at once. Earlier,  the `eval_input_fn` is  defined using the entire evaluation set."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "6zmIjTr73rJ4",
        "outputId": "f934b588-cc13-4b5b-e9d1-5253606fa40d",
        "colab": {
          "height": 289
        }
      },
      "source": [
        "pred_dicts = list(est.predict(eval_input_fn))\n",
        "probs = pd.Series([pred['probabilities'][1] for pred in pred_dicts])\n",
        "\n",
        "probs.plot(kind='hist', bins=20, title='predicted probabilities')\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEQCAYAAABIqvhxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd15dfa9390>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "mBUaNN1BzJHG"
      },
      "source": [
        "Finally you can also look at the receiver operating characteristic (ROC) of the results, which will give us a better idea of the tradeoff between the true positive rate and false positive rate."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "NzxghvVz3rJ6",
        "outputId": "98bcd7d1-35b8-4593-b22f-ca020c2323fa",
        "colab": {
          "height": 307
        }
      },
      "source": [
        "from sklearn.metrics import roc_curve\n",
        "\n",
        "fpr, tpr, _ = roc_curve(y_eval, probs)\n",
        "plt.plot(fpr, tpr)\n",
        "plt.title('ROC curve')\n",
        "plt.xlabel('false positive rate')\n",
        "plt.ylabel('true positive rate')\n",
        "plt.xlim(0,)\n",
        "plt.ylim(0,)\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZMAAAEiCAYAAAA8ij+xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niIiMxjAhIiKjMUyIiMho/w/TWJJiO+X3vQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<matplotlib.figure.Figure at 0x7fd15c907f50>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}