pymc3.10-classification-using-BART.ipynb

pymc3-10-bart-experiments.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "pymc3.10-classification-using-BART.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPKjf9aieEeS571emG2sBTJ",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/ShokuninSan/f2696379614933913e68b0327e2fd71d/pymc3-10-bart-experiments.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Vi0lNfG_mLqQ",
        "outputId": "584563bf-2ca8-4fd0-d7e3-7ce026ccd4a3"
      },
      "source": [
        "!pip install pymc3==3.10"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: pymc3==3.10 in /usr/local/lib/python3.6/dist-packages (3.10.0)\n",
            "Requirement already satisfied: scipy>=0.18.1 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (1.4.1)\n",
            "Requirement already satisfied: theano-pymc==1.0.11 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (1.0.11)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (3.7.4.3)\n",
            "Requirement already satisfied: numpy>=1.13.0 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (1.18.5)\n",
            "Requirement already satisfied: arviz>=0.9.0 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (0.10.0)\n",
            "Requirement already satisfied: patsy>=0.5.1 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (0.5.1)\n",
            "Requirement already satisfied: dill in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (0.3.3)\n",
            "Requirement already satisfied: pandas>=0.18.0 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (1.1.5)\n",
            "Requirement already satisfied: fastprogress>=0.2.0 in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (1.0.0)\n",
            "Requirement already satisfied: contextvars; python_version < \"3.7\" in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (2.4)\n",
            "Requirement already satisfied: dataclasses; python_version < \"3.7\" in /usr/local/lib/python3.6/dist-packages (from pymc3==3.10) (0.8)\n",
            "Requirement already satisfied: matplotlib>=3.0 in /usr/local/lib/python3.6/dist-packages (from arviz>=0.9.0->pymc3==3.10) (3.2.2)\n",
            "Requirement already satisfied: xarray>=0.16.1 in /usr/local/lib/python3.6/dist-packages (from arviz>=0.9.0->pymc3==3.10) (0.16.2)\n",
            "Requirement already satisfied: netcdf4 in /usr/local/lib/python3.6/dist-packages (from arviz>=0.9.0->pymc3==3.10) (1.5.5)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.6/dist-packages (from arviz>=0.9.0->pymc3==3.10) (20.7)\n",
            "Requirement already satisfied: setuptools>=38.4 in /usr/local/lib/python3.6/dist-packages (from arviz>=0.9.0->pymc3==3.10) (50.3.2)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from patsy>=0.5.1->pymc3==3.10) (1.15.0)\n",
            "Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.18.0->pymc3==3.10) (2.8.1)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.18.0->pymc3==3.10) (2018.9)\n",
            "Requirement already satisfied: immutables>=0.9 in /usr/local/lib/python3.6/dist-packages (from contextvars; python_version < \"3.7\"->pymc3==3.10) (0.14)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=3.0->arviz>=0.9.0->pymc3==3.10) (0.10.0)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=3.0->arviz>=0.9.0->pymc3==3.10) (1.3.1)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=3.0->arviz>=0.9.0->pymc3==3.10) (2.4.7)\n",
            "Requirement already satisfied: cftime in /usr/local/lib/python3.6/dist-packages (from netcdf4->arviz>=0.9.0->pymc3==3.10) (1.3.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aI2ce03DDc9u"
      },
      "source": [
        "import theano\n",
        "import numpy as np\n",
        "import pymc3 as pm\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.datasets import load_breast_cancer\n",
        "from sklearn.model_selection import train_test_split"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zxqywt4lah7Q"
      },
      "source": [
        "X, Y = load_breast_cancer(return_X_y=True)"
      ],
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jy_YzYUboKWp"
      },
      "source": [
        "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, stratify=Y)"
      ],
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hn8t_culayDW",
        "outputId": "66c68b36-4575-4e95-def6-2ea99812242f"
      },
      "source": [
        "X_train.shape, Y_train.shape, X_test.shape, Y_test.shape"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((426, 30), (426,), (143, 30), (143,))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JmnrHgOrwpjV"
      },
      "source": [
        "## Using `BART`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VJABwZPvEtu6"
      },
      "source": [
        "X_shared = theano.shared(X_train)"
      ],
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 196
        },
        "id": "Ic6uQSGPaz11",
        "outputId": "0dccd868-5d5b-41b0-d133-42e333bb6e1f"
      },
      "source": [
        "with pm.Model() as model:\n",
        "    \n",
        "    x = pm.BART('x', X_shared.get_value(), Y_train)\n",
        "    y = pm.Bernoulli('y', p=pm.math.sigmoid(x), observed=Y_train)\n",
        "\n",
        "    trace = pm.sample()"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/pymc3/sampling.py:468: FutureWarning: In an upcoming release, pm.sample will return an `arviz.InferenceData` object instead of a `MultiTrace` by default. You can pass return_inferencedata=True or return_inferencedata=False to be safe and silence this warning.\n",
            "  FutureWarning,\n",
            "The BART model is experimental. Use with caution.\n",
            "Sequential sampling (2 chains in 1 job)\n",
            "PGBART: [x]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "    <div>\n",
              "        <style>\n",
              "            /* Turns off some styling */\n",
              "            progress {\n",
              "                /* gets rid of default border in Firefox and Opera. */\n",
              "                border: none;\n",
              "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "                background-size: auto;\n",
              "            }\n",
              "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "                background: #F44336;\n",
              "            }\n",
              "        </style>\n",
              "      <progress value='2000' class='' max='2000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      100.00% [2000/2000 05:08<00:00 Sampling chain 0, 0 divergences]\n",
              "    </div>\n",
              "    "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "    <div>\n",
              "        <style>\n",
              "            /* Turns off some styling */\n",
              "            progress {\n",
              "                /* gets rid of default border in Firefox and Opera. */\n",
              "                border: none;\n",
              "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "                background-size: auto;\n",
              "            }\n",
              "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "                background: #F44336;\n",
              "            }\n",
              "        </style>\n",
              "      <progress value='2000' class='' max='2000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      100.00% [2000/2000 05:08<00:00 Sampling chain 1, 0 divergences]\n",
              "    </div>\n",
              "    "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Sampling 2 chains for 1_000 tune and 1_000 draw iterations (2_000 + 2_000 draws total) took 617 seconds.\n",
            "The estimated number of effective samples is smaller than 200 for some parameters.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RaCv3m3iFuaX"
      },
      "source": [
        "X_shared.set_value(X_test)"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 37
        },
        "id": "-ZB5MU5ltkHv",
        "outputId": "39018170-798a-4d8d-941a-0ea14c60cdd1"
      },
      "source": [
        "with model:\n",
        "    ppc = pm.sample_posterior_predictive(trace)"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "\n",
              "    <div>\n",
              "        <style>\n",
              "            /* Turns off some styling */\n",
              "            progress {\n",
              "                /* gets rid of default border in Firefox and Opera. */\n",
              "                border: none;\n",
              "                /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
              "                background-size: auto;\n",
              "            }\n",
              "            .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
              "                background: #F44336;\n",
              "            }\n",
              "        </style>\n",
              "      <progress value='2000' class='' max='2000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      100.00% [2000/2000 00:01<00:00]\n",
              "    </div>\n",
              "    "
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s29AtTLFlrVK"
      },
      "source": [
        "posterior = ppc.get('y')[:, :X_test.shape[0]] # maybe it's the the value of `shape` which isn't correct?"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GeuruOfSc5n3",
        "outputId": "71bf018c-af32-4d70-f8c5-e21bbff04ca7"
      },
      "source": [
        "posterior.shape"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2000, 143)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mcE6Fh7ADjdw",
        "outputId": "8c25196e-52e4-4d5f-a4c0-8156f13f810a"
      },
      "source": [
        "print(classification_report(Y_test, np.round(posterior.mean(0))))"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.32      0.15      0.21        53\n",
            "           1       0.62      0.81      0.70        90\n",
            "\n",
            "    accuracy                           0.57       143\n",
            "   macro avg       0.47      0.48      0.45       143\n",
            "weighted avg       0.51      0.57      0.52       143\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tNI9UQK_sEbh"
      },
      "source": [
        "## Using sklearn's `RandomForest`"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "adSl86ZvHlI5"
      },
      "source": [
        "from sklearn.ensemble import RandomForestClassifier"
      ],
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z6rNQ38isOEE"
      },
      "source": [
        "rf = RandomForestClassifier()"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZOa48lOtsQov",
        "outputId": "9ae8795a-1d14-4246-91cc-c5e92263013e"
      },
      "source": [
        "rf.fit(X_train, Y_train)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,\n",
              "                       criterion='gini', max_depth=None, max_features='auto',\n",
              "                       max_leaf_nodes=None, max_samples=None,\n",
              "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
              "                       min_samples_leaf=1, min_samples_split=2,\n",
              "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
              "                       n_jobs=None, oob_score=False, random_state=None,\n",
              "                       verbose=0, warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KFEYb--EsS6c"
      },
      "source": [
        "y_pred = rf.predict(X_test)"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WR8kaEcTsYuV",
        "outputId": "a9973957-834f-4747-ad1d-08a16d0223c9"
      },
      "source": [
        "print(classification_report(Y_test, y_pred))"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       1.00      0.91      0.95        53\n",
            "           1       0.95      1.00      0.97        90\n",
            "\n",
            "    accuracy                           0.97       143\n",
            "   macro avg       0.97      0.95      0.96       143\n",
            "weighted avg       0.97      0.97      0.96       143\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hwszgI7Jsdoo"
      },
      "source": [
        ""
      ],
      "execution_count": 38,
      "outputs": []
    }
  ]
}