{ "nbformat": 4, "nbformat_minor": 0, "metadata": { "colab": { "name": "decisionTreeTutorial1.ipynb", "version": "0.3.2", "provenance": [] }, "kernelspec": { "name": "python3", "display_name": "Python 3" } }, "cells": [ { "metadata": { "id": "XD8qqVh82jpk", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "from sklearn.model_selection import train_test_split\n", "from sklearn.tree import DecisionTreeClassifier\n", "import numpy as np\n", "import pandas as pd" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "p-JobDUrUqj8", "colab_type": "text" }, "cell_type": "markdown", "source": [ "# Obteniendo los datos" ] }, { "metadata": { "id": "rM8Hi3Sa6KvU", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 196 }, "outputId": "d802c9e1-ec4f-44bf-a4da-0b9726ebf6ea" }, "cell_type": "code", "source": [ "!wget https://raw.githubusercontent.com/susanli2016/Machine-Learning-with-Python/master/diabetes.csv" ], "execution_count": 39, "outputs": [ { "output_type": "stream", "text": [ "--2018-10-14 23:42:51-- https://raw.githubusercontent.com/susanli2016/Machine-Learning-with-Python/master/diabetes.csv\n", "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.0.133, 151.101.64.133, 151.101.128.133, ...\n", "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.0.133|:443... connected.\n", "HTTP request sent, awaiting response... 200 OK\n", "Length: 23875 (23K) [text/plain]\n", "Saving to: ‘diabetes.csv’\n", "\n", "\rdiabetes.csv 0%[ ] 0 --.-KB/s \rdiabetes.csv 100%[===================>] 23.32K --.-KB/s in 0.008s \n", "\n", "2018-10-14 23:42:51 (2.91 MB/s) - ‘diabetes.csv’ saved [23875/23875]\n", "\n" ], "name": "stdout" } ] }, { "metadata": { "id": "AQH3mqGw6mb4", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "dataset = pd.read_csv(\"diabetes.csv\")" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "zNf_2A507Ry1", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 191 }, "outputId": "03a6cdaf-fb39-40d9-ef6e-5095fa79a29a" }, "cell_type": "code", "source": [ "dataset.head()" ], "execution_count": 42, "outputs": [ { "output_type": "execute_result", "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
PregnanciesGlucoseBloodPressureSkinThicknessInsulinBMIDiabetesPedigreeFunctionAgeOutcome
061487235033.60.627501
11856629026.60.351310
28183640023.30.672321
318966239428.10.167210
40137403516843.12.288331
\n", "
" ], "text/plain": [ " Pregnancies Glucose BloodPressure SkinThickness Insulin BMI \\\n", "0 6 148 72 35 0 33.6 \n", "1 1 85 66 29 0 26.6 \n", "2 8 183 64 0 0 23.3 \n", "3 1 89 66 23 94 28.1 \n", "4 0 137 40 35 168 43.1 \n", "\n", " DiabetesPedigreeFunction Age Outcome \n", "0 0.627 50 1 \n", "1 0.351 31 0 \n", "2 0.672 32 1 \n", "3 0.167 21 0 \n", "4 2.288 33 1 " ] }, "metadata": { "tags": [] }, "execution_count": 42 } ] }, { "metadata": { "id": "_Zw_D0fL9-aP", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 33 }, "outputId": "03e55a50-dcfa-454f-a64f-ac218a9d0541" }, "cell_type": "code", "source": [ "dataset.shape" ], "execution_count": 47, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "(768, 9)" ] }, "metadata": { "tags": [] }, "execution_count": 47 } ] }, { "metadata": { "id": "T1DwasNFADFx", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "features = dataset.drop([\"Outcome\"], axis=1)\n", "X = np.array(features)\n", "y = np.array(dataset[\"Outcome\"])" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "Dl-ZKkbwDOTV", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "X_train, X_val, y_train, y_val = train_test_split(X, y, random_state=0, test_size=0.20)" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "3pJRQQifDYBy", "colab_type": "text" }, "cell_type": "markdown", "source": [ "# Creando el modelo" ] }, { "metadata": { "id": "XmSdOAfKDaQe", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "tree = DecisionTreeClassifier()" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "EoZ311RoDfOu", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 115 }, "outputId": "bfd247ab-16de-4392-d432-4aa2b18c3fd1" }, "cell_type": "code", "source": [ "tree.fit(X_train, y_train)" ], "execution_count": 141, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n", " max_features=None, max_leaf_nodes=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, presort=False, random_state=None,\n", " splitter='best')" ] }, "metadata": { "tags": [] }, "execution_count": 141 } ] }, { "metadata": { "id": "ovi5dSkDDg31", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 33 }, "outputId": "a3eac31a-73c7-4b4e-cdb9-de80e40c2dce" }, "cell_type": "code", "source": [ "tree.tree_.max_depth" ], "execution_count": 142, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "15" ] }, "metadata": { "tags": [] }, "execution_count": 142 } ] }, { "metadata": { "id": "hoOUCb75Di4f", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "validation_prediction = tree.predict(X_val)\n", "training_prediction = tree.predict(X_train)" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "4AmBMrzuEAJK", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 50 }, "outputId": "05759b6c-5b00-4332-b10d-bdf66ae09530" }, "cell_type": "code", "source": [ "print('Exactitud training data: ', accuracy_score(y_true=y_train, y_pred=training_prediction))\n", "print('Exactitud validation data: ', accuracy_score(y_true=y_val, y_pred=validation_prediction))" ], "execution_count": 144, "outputs": [ { "output_type": "stream", "text": [ "Exactitud training data: 1.0\n", "Exactitud validation data: 0.7922077922077922\n" ], "name": "stdout" } ] }, { "metadata": { "id": "4V2KjHFbEF3f", "colab_type": "text" }, "cell_type": "markdown", "source": [ "# Mostrando el arbol de forma visual" ] }, { "metadata": { "id": "mN_EHt_6Eawq", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "!apt-get install graphviz" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "TboG7v8IELj7", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 82 }, "outputId": "d41e8aa3-9e1f-48b3-d6fe-48d350745ced" }, "cell_type": "code", "source": [ "!pip install graphviz" ], "execution_count": 57, "outputs": [ { "output_type": "stream", "text": [ "Collecting graphviz\n", " Downloading https://files.pythonhosted.org/packages/47/87/313cd4ea4f75472826acb74c57f94fc83e04ba93e4ccf35656f6b7f502e2/graphviz-0.9-py2.py3-none-any.whl\n", "Installing collected packages: graphviz\n", "Successfully installed graphviz-0.9\n" ], "name": "stdout" } ] }, { "metadata": { "id": "yELy96AHENvr", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "import graphviz \n", "from sklearn.tree import export_graphviz" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "AvbMS7gkEPbn", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "feature_names = features.columns" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "CY6Dq_akESiT", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "dot_data = export_graphviz(tree, out_file=None, \n", " feature_names=feature_names, \n", " class_names=True, \n", " filled=True, rounded=True, \n", " special_characters=True) \n", "graph = graphviz.Source(dot_data)" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "qz0HXbLQTEY4", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 2416 }, "outputId": "4094cfa9-f95c-4d7d-cbe5-84d817842427" }, "cell_type": "code", "source": [ "graph" ], "execution_count": 147, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "" ], "image/svg+xml": "\n\n\n\n\n\nTree\n\n\n\n0\n\nGlucose ≤ 123.5\ngini = 0.461\nsamples = 614\nvalue = [393, 221]\nclass = y\n0\n\n\n\n1\n\nAge ≤ 28.5\ngini = 0.301\nsamples = 352\nvalue = [287, 65]\nclass = y\n0\n\n\n\n0->1\n\n\nTrue\n\n\n\n100\n\nBMI ≤ 30.05\ngini = 0.482\nsamples = 262\nvalue = [106, 156]\nclass = y\n1\n\n\n\n0->100\n\n\nFalse\n\n\n\n2\n\nBMI ≤ 30.95\ngini = 0.162\nsamples = 202\nvalue = [184, 18]\nclass = y\n0\n\n\n\n1->2\n\n\n\n\n\n39\n\nBMI ≤ 26.35\ngini = 0.43\nsamples = 150\nvalue = [103, 47]\nclass = y\n0\n\n\n\n1->39\n\n\n\n\n\n3\n\nPregnancies ≤ 7.0\ngini = 0.036\nsamples = 110\nvalue = [108, 2]\nclass = y\n0\n\n\n\n2->3\n\n\n\n\n\n10\n\nBloodPressure ≤ 53.0\ngini = 0.287\nsamples = 92\nvalue = [76, 16]\nclass = y\n0\n\n\n\n2->10\n\n\n\n\n\n4\n\nDiabetesPedigreeFunction ≤ 0.672\ngini = 0.018\nsamples = 109\nvalue = [108, 1]\nclass = y\n0\n\n\n\n3->4\n\n\n\n\n\n9\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n3->9\n\n\n\n\n\n5\n\ngini = 0.0\nsamples = 99\nvalue = [99, 0]\nclass = y\n0\n\n\n\n4->5\n\n\n\n\n\n6\n\nDiabetesPedigreeFunction ≤ 0.697\ngini = 0.18\nsamples = 10\nvalue = [9, 1]\nclass = y\n0\n\n\n\n4->6\n\n\n\n\n\n7\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n6->7\n\n\n\n\n\n8\n\ngini = 0.0\nsamples = 9\nvalue = [9, 0]\nclass = y\n0\n\n\n\n6->8\n\n\n\n\n\n11\n\nDiabetesPedigreeFunction ≤ 0.508\ngini = 0.444\nsamples = 6\nvalue = [2, 4]\nclass = y\n1\n\n\n\n10->11\n\n\n\n\n\n14\n\nDiabetesPedigreeFunction ≤ 1.272\ngini = 0.24\nsamples = 86\nvalue = [74, 12]\nclass = y\n0\n\n\n\n10->14\n\n\n\n\n\n12\n\ngini = 0.0\nsamples = 4\nvalue = [0, 4]\nclass = y\n1\n\n\n\n11->12\n\n\n\n\n\n13\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n11->13\n\n\n\n\n\n15\n\nDiabetesPedigreeFunction ≤ 0.501\ngini = 0.225\nsamples = 85\nvalue = [74, 11]\nclass = y\n0\n\n\n\n14->15\n\n\n\n\n\n38\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n14->38\n\n\n\n\n\n16\n\nBMI ≤ 45.35\ngini = 0.135\nsamples = 55\nvalue = [51, 4]\nclass = y\n0\n\n\n\n15->16\n\n\n\n\n\n27\n\nBloodPressure ≤ 69.0\ngini = 0.358\nsamples = 30\nvalue = [23, 7]\nclass = y\n0\n\n\n\n15->27\n\n\n\n\n\n17\n\nInsulin ≤ 36.5\ngini = 0.105\nsamples = 54\nvalue = [51, 3]\nclass = y\n0\n\n\n\n16->17\n\n\n\n\n\n26\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n16->26\n\n\n\n\n\n18\n\nInsulin ≤ 34.0\ngini = 0.266\nsamples = 19\nvalue = [16, 3]\nclass = y\n0\n\n\n\n17->18\n\n\n\n\n\n25\n\ngini = 0.0\nsamples = 35\nvalue = [35, 0]\nclass = y\n0\n\n\n\n17->25\n\n\n\n\n\n19\n\nGlucose ≤ 111.5\ngini = 0.198\nsamples = 18\nvalue = [16, 2]\nclass = y\n0\n\n\n\n18->19\n\n\n\n\n\n24\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n18->24\n\n\n\n\n\n20\n\ngini = 0.0\nsamples = 13\nvalue = [13, 0]\nclass = y\n0\n\n\n\n19->20\n\n\n\n\n\n21\n\nBMI ≤ 34.5\ngini = 0.48\nsamples = 5\nvalue = [3, 2]\nclass = y\n0\n\n\n\n19->21\n\n\n\n\n\n22\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n21->22\n\n\n\n\n\n23\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n21->23\n\n\n\n\n\n28\n\nGlucose ≤ 88.5\ngini = 0.492\nsamples = 16\nvalue = [9, 7]\nclass = y\n0\n\n\n\n27->28\n\n\n\n\n\n37\n\ngini = 0.0\nsamples = 14\nvalue = [14, 0]\nclass = y\n0\n\n\n\n27->37\n\n\n\n\n\n29\n\ngini = 0.0\nsamples = 7\nvalue = [7, 0]\nclass = y\n0\n\n\n\n28->29\n\n\n\n\n\n30\n\nDiabetesPedigreeFunction ≤ 0.908\ngini = 0.346\nsamples = 9\nvalue = [2, 7]\nclass = y\n1\n\n\n\n28->30\n\n\n\n\n\n31\n\nGlucose ≤ 98.5\ngini = 0.219\nsamples = 8\nvalue = [1, 7]\nclass = y\n1\n\n\n\n30->31\n\n\n\n\n\n36\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n30->36\n\n\n\n\n\n32\n\nBMI ≤ 37.4\ngini = 0.444\nsamples = 3\nvalue = [1, 2]\nclass = y\n1\n\n\n\n31->32\n\n\n\n\n\n35\n\ngini = 0.0\nsamples = 5\nvalue = [0, 5]\nclass = y\n1\n\n\n\n31->35\n\n\n\n\n\n33\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n32->33\n\n\n\n\n\n34\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n32->34\n\n\n\n\n\n40\n\ngini = 0.0\nsamples = 29\nvalue = [29, 0]\nclass = y\n0\n\n\n\n39->40\n\n\n\n\n\n41\n\nGlucose ≤ 99.5\ngini = 0.475\nsamples = 121\nvalue = [74, 47]\nclass = y\n0\n\n\n\n39->41\n\n\n\n\n\n42\n\nGlucose ≤ 28.5\ngini = 0.337\nsamples = 42\nvalue = [33, 9]\nclass = y\n0\n\n\n\n41->42\n\n\n\n\n\n59\n\nBMI ≤ 27.55\ngini = 0.499\nsamples = 79\nvalue = [41, 38]\nclass = y\n0\n\n\n\n41->59\n\n\n\n\n\n43\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n42->43\n\n\n\n\n\n44\n\nAge ≤ 42.5\ngini = 0.289\nsamples = 40\nvalue = [33, 7]\nclass = y\n0\n\n\n\n42->44\n\n\n\n\n\n45\n\nDiabetesPedigreeFunction ≤ 1.16\ngini = 0.137\nsamples = 27\nvalue = [25, 2]\nclass = y\n0\n\n\n\n44->45\n\n\n\n\n\n52\n\nBMI ≤ 30.85\ngini = 0.473\nsamples = 13\nvalue = [8, 5]\nclass = y\n0\n\n\n\n44->52\n\n\n\n\n\n46\n\nDiabetesPedigreeFunction ≤ 0.171\ngini = 0.074\nsamples = 26\nvalue = [25, 1]\nclass = y\n0\n\n\n\n45->46\n\n\n\n\n\n51\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n45->51\n\n\n\n\n\n47\n\nAge ≤ 32.0\ngini = 0.375\nsamples = 4\nvalue = [3, 1]\nclass = y\n0\n\n\n\n46->47\n\n\n\n\n\n50\n\ngini = 0.0\nsamples = 22\nvalue = [22, 0]\nclass = y\n0\n\n\n\n46->50\n\n\n\n\n\n48\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n47->48\n\n\n\n\n\n49\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n47->49\n\n\n\n\n\n53\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n52->53\n\n\n\n\n\n54\n\nSkinThickness ≤ 21.5\ngini = 0.397\nsamples = 11\nvalue = [8, 3]\nclass = y\n0\n\n\n\n52->54\n\n\n\n\n\n55\n\ngini = 0.0\nsamples = 5\nvalue = [5, 0]\nclass = y\n0\n\n\n\n54->55\n\n\n\n\n\n56\n\nBloodPressure ≤ 75.0\ngini = 0.5\nsamples = 6\nvalue = [3, 3]\nclass = y\n0\n\n\n\n54->56\n\n\n\n\n\n57\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n56->57\n\n\n\n\n\n58\n\ngini = 0.0\nsamples = 3\nvalue = [0, 3]\nclass = y\n1\n\n\n\n56->58\n\n\n\n\n\n60\n\ngini = 0.0\nsamples = 5\nvalue = [0, 5]\nclass = y\n1\n\n\n\n59->60\n\n\n\n\n\n61\n\nDiabetesPedigreeFunction ≤ 0.179\ngini = 0.494\nsamples = 74\nvalue = [41, 33]\nclass = y\n0\n\n\n\n59->61\n\n\n\n\n\n62\n\ngini = 0.0\nsamples = 8\nvalue = [8, 0]\nclass = y\n0\n\n\n\n61->62\n\n\n\n\n\n63\n\nPregnancies ≤ 6.5\ngini = 0.5\nsamples = 66\nvalue = [33, 33]\nclass = y\n0\n\n\n\n61->63\n\n\n\n\n\n64\n\nPregnancies ≤ 1.5\ngini = 0.483\nsamples = 44\nvalue = [26, 18]\nclass = y\n0\n\n\n\n63->64\n\n\n\n\n\n87\n\nAge ≤ 39.0\ngini = 0.434\nsamples = 22\nvalue = [7, 15]\nclass = y\n1\n\n\n\n63->87\n\n\n\n\n\n65\n\nDiabetesPedigreeFunction ≤ 0.893\ngini = 0.475\nsamples = 18\nvalue = [7, 11]\nclass = y\n1\n\n\n\n64->65\n\n\n\n\n\n74\n\nAge ≤ 34.5\ngini = 0.393\nsamples = 26\nvalue = [19, 7]\nclass = y\n0\n\n\n\n64->74\n\n\n\n\n\n66\n\nDiabetesPedigreeFunction ≤ 0.2\ngini = 0.391\nsamples = 15\nvalue = [4, 11]\nclass = y\n1\n\n\n\n65->66\n\n\n\n\n\n73\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n65->73\n\n\n\n\n\n67\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n66->67\n\n\n\n\n\n68\n\nBloodPressure ≤ 92.0\ngini = 0.26\nsamples = 13\nvalue = [2, 11]\nclass = y\n1\n\n\n\n66->68\n\n\n\n\n\n69\n\nGlucose ≤ 101.0\ngini = 0.153\nsamples = 12\nvalue = [1, 11]\nclass = y\n1\n\n\n\n68->69\n\n\n\n\n\n72\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n68->72\n\n\n\n\n\n70\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n69->70\n\n\n\n\n\n71\n\ngini = 0.0\nsamples = 11\nvalue = [0, 11]\nclass = y\n1\n\n\n\n69->71\n\n\n\n\n\n75\n\nBloodPressure ≤ 63.0\ngini = 0.133\nsamples = 14\nvalue = [13, 1]\nclass = y\n0\n\n\n\n74->75\n\n\n\n\n\n80\n\nBloodPressure ≤ 74.5\ngini = 0.5\nsamples = 12\nvalue = [6, 6]\nclass = y\n0\n\n\n\n74->80\n\n\n\n\n\n76\n\nSkinThickness ≤ 25.5\ngini = 0.5\nsamples = 2\nvalue = [1, 1]\nclass = y\n0\n\n\n\n75->76\n\n\n\n\n\n79\n\ngini = 0.0\nsamples = 12\nvalue = [12, 0]\nclass = y\n0\n\n\n\n75->79\n\n\n\n\n\n77\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n76->77\n\n\n\n\n\n78\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n76->78\n\n\n\n\n\n81\n\ngini = 0.0\nsamples = 4\nvalue = [0, 4]\nclass = y\n1\n\n\n\n80->81\n\n\n\n\n\n82\n\nBMI ≤ 32.65\ngini = 0.375\nsamples = 8\nvalue = [6, 2]\nclass = y\n0\n\n\n\n80->82\n\n\n\n\n\n83\n\nBMI ≤ 29.3\ngini = 0.444\nsamples = 3\nvalue = [1, 2]\nclass = y\n1\n\n\n\n82->83\n\n\n\n\n\n86\n\ngini = 0.0\nsamples = 5\nvalue = [5, 0]\nclass = y\n0\n\n\n\n82->86\n\n\n\n\n\n84\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n83->84\n\n\n\n\n\n85\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n83->85\n\n\n\n\n\n88\n\ngini = 0.0\nsamples = 8\nvalue = [0, 8]\nclass = y\n1\n\n\n\n87->88\n\n\n\n\n\n89\n\nDiabetesPedigreeFunction ≤ 0.587\ngini = 0.5\nsamples = 14\nvalue = [7, 7]\nclass = y\n0\n\n\n\n87->89\n\n\n\n\n\n90\n\nGlucose ≤ 109.0\ngini = 0.42\nsamples = 10\nvalue = [7, 3]\nclass = y\n0\n\n\n\n89->90\n\n\n\n\n\n99\n\ngini = 0.0\nsamples = 4\nvalue = [0, 4]\nclass = y\n1\n\n\n\n89->99\n\n\n\n\n\n91\n\nGlucose ≤ 107.0\ngini = 0.5\nsamples = 6\nvalue = [3, 3]\nclass = y\n0\n\n\n\n90->91\n\n\n\n\n\n98\n\ngini = 0.0\nsamples = 4\nvalue = [4, 0]\nclass = y\n0\n\n\n\n90->98\n\n\n\n\n\n92\n\nBMI ≤ 35.4\ngini = 0.48\nsamples = 5\nvalue = [3, 2]\nclass = y\n0\n\n\n\n91->92\n\n\n\n\n\n97\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n91->97\n\n\n\n\n\n93\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n92->93\n\n\n\n\n\n94\n\nDiabetesPedigreeFunction ≤ 0.221\ngini = 0.444\nsamples = 3\nvalue = [1, 2]\nclass = y\n1\n\n\n\n92->94\n\n\n\n\n\n95\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n94->95\n\n\n\n\n\n96\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n94->96\n\n\n\n\n\n101\n\nAge ≤ 26.0\ngini = 0.438\nsamples = 74\nvalue = [50, 24]\nclass = y\n0\n\n\n\n100->101\n\n\n\n\n\n132\n\nGlucose ≤ 157.5\ngini = 0.418\nsamples = 188\nvalue = [56, 132]\nclass = y\n1\n\n\n\n100->132\n\n\n\n\n\n102\n\ngini = 0.0\nsamples = 19\nvalue = [19, 0]\nclass = y\n0\n\n\n\n101->102\n\n\n\n\n\n103\n\nAge ≤ 60.5\ngini = 0.492\nsamples = 55\nvalue = [31, 24]\nclass = y\n0\n\n\n\n101->103\n\n\n\n\n\n104\n\nGlucose ≤ 151.5\ngini = 0.5\nsamples = 47\nvalue = [23, 24]\nclass = y\n1\n\n\n\n103->104\n\n\n\n\n\n131\n\ngini = 0.0\nsamples = 8\nvalue = [8, 0]\nclass = y\n0\n\n\n\n103->131\n\n\n\n\n\n105\n\nGlucose ≤ 125.5\ngini = 0.469\nsamples = 32\nvalue = [20, 12]\nclass = y\n0\n\n\n\n104->105\n\n\n\n\n\n124\n\nBMI ≤ 27.1\ngini = 0.32\nsamples = 15\nvalue = [3, 12]\nclass = y\n1\n\n\n\n104->124\n\n\n\n\n\n106\n\nSkinThickness ≤ 27.0\ngini = 0.375\nsamples = 8\nvalue = [2, 6]\nclass = y\n1\n\n\n\n105->106\n\n\n\n\n\n111\n\nBloodPressure ≤ 73.0\ngini = 0.375\nsamples = 24\nvalue = [18, 6]\nclass = y\n0\n\n\n\n105->111\n\n\n\n\n\n107\n\ngini = 0.0\nsamples = 5\nvalue = [0, 5]\nclass = y\n1\n\n\n\n106->107\n\n\n\n\n\n108\n\nAge ≤ 43.0\ngini = 0.444\nsamples = 3\nvalue = [2, 1]\nclass = y\n0\n\n\n\n106->108\n\n\n\n\n\n109\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n108->109\n\n\n\n\n\n110\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n108->110\n\n\n\n\n\n112\n\nBloodPressure ≤ 64.5\ngini = 0.5\nsamples = 8\nvalue = [4, 4]\nclass = y\n0\n\n\n\n111->112\n\n\n\n\n\n117\n\nBMI ≤ 28.0\ngini = 0.219\nsamples = 16\nvalue = [14, 2]\nclass = y\n0\n\n\n\n111->117\n\n\n\n\n\n113\n\nAge ≤ 28.5\ngini = 0.32\nsamples = 5\nvalue = [4, 1]\nclass = y\n0\n\n\n\n112->113\n\n\n\n\n\n116\n\ngini = 0.0\nsamples = 3\nvalue = [0, 3]\nclass = y\n1\n\n\n\n112->116\n\n\n\n\n\n114\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n113->114\n\n\n\n\n\n115\n\ngini = 0.0\nsamples = 4\nvalue = [4, 0]\nclass = y\n0\n\n\n\n113->115\n\n\n\n\n\n118\n\ngini = 0.0\nsamples = 9\nvalue = [9, 0]\nclass = y\n0\n\n\n\n117->118\n\n\n\n\n\n119\n\nBMI ≤ 29.55\ngini = 0.408\nsamples = 7\nvalue = [5, 2]\nclass = y\n0\n\n\n\n117->119\n\n\n\n\n\n120\n\nAge ≤ 30.5\ngini = 0.444\nsamples = 3\nvalue = [1, 2]\nclass = y\n1\n\n\n\n119->120\n\n\n\n\n\n123\n\ngini = 0.0\nsamples = 4\nvalue = [4, 0]\nclass = y\n0\n\n\n\n119->123\n\n\n\n\n\n121\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n120->121\n\n\n\n\n\n122\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n120->122\n\n\n\n\n\n125\n\ngini = 0.0\nsamples = 8\nvalue = [0, 8]\nclass = y\n1\n\n\n\n124->125\n\n\n\n\n\n126\n\nBMI ≤ 29.1\ngini = 0.49\nsamples = 7\nvalue = [3, 4]\nclass = y\n1\n\n\n\n124->126\n\n\n\n\n\n127\n\nAge ≤ 36.5\ngini = 0.375\nsamples = 4\nvalue = [3, 1]\nclass = y\n0\n\n\n\n126->127\n\n\n\n\n\n130\n\ngini = 0.0\nsamples = 3\nvalue = [0, 3]\nclass = y\n1\n\n\n\n126->130\n\n\n\n\n\n128\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n127->128\n\n\n\n\n\n129\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n127->129\n\n\n\n\n\n133\n\nAge ≤ 28.5\ngini = 0.482\nsamples = 116\nvalue = [47, 69]\nclass = y\n1\n\n\n\n132->133\n\n\n\n\n\n202\n\nInsulin ≤ 595.0\ngini = 0.219\nsamples = 72\nvalue = [9, 63]\nclass = y\n1\n\n\n\n132->202\n\n\n\n\n\n134\n\nBloodPressure ≤ 73.0\ngini = 0.49\nsamples = 42\nvalue = [24, 18]\nclass = y\n0\n\n\n\n133->134\n\n\n\n\n\n157\n\nDiabetesPedigreeFunction ≤ 0.429\ngini = 0.428\nsamples = 74\nvalue = [23, 51]\nclass = y\n1\n\n\n\n133->157\n\n\n\n\n\n135\n\nDiabetesPedigreeFunction ≤ 0.186\ngini = 0.463\nsamples = 22\nvalue = [8, 14]\nclass = y\n1\n\n\n\n134->135\n\n\n\n\n\n148\n\nBloodPressure ≤ 89.0\ngini = 0.32\nsamples = 20\nvalue = [16, 4]\nclass = y\n0\n\n\n\n134->148\n\n\n\n\n\n136\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n135->136\n\n\n\n\n\n137\n\nGlucose ≤ 147.0\ngini = 0.388\nsamples = 19\nvalue = [5, 14]\nclass = y\n1\n\n\n\n135->137\n\n\n\n\n\n138\n\nInsulin ≤ 365.0\ngini = 0.231\nsamples = 15\nvalue = [2, 13]\nclass = y\n1\n\n\n\n137->138\n\n\n\n\n\n145\n\nGlucose ≤ 154.5\ngini = 0.375\nsamples = 4\nvalue = [3, 1]\nclass = y\n0\n\n\n\n137->145\n\n\n\n\n\n139\n\nPregnancies ≤ 3.5\ngini = 0.133\nsamples = 14\nvalue = [1, 13]\nclass = y\n1\n\n\n\n138->139\n\n\n\n\n\n144\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n138->144\n\n\n\n\n\n140\n\ngini = 0.0\nsamples = 11\nvalue = [0, 11]\nclass = y\n1\n\n\n\n139->140\n\n\n\n\n\n141\n\nBMI ≤ 33.6\ngini = 0.444\nsamples = 3\nvalue = [1, 2]\nclass = y\n1\n\n\n\n139->141\n\n\n\n\n\n142\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n141->142\n\n\n\n\n\n143\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n141->143\n\n\n\n\n\n146\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n145->146\n\n\n\n\n\n147\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n145->147\n\n\n\n\n\n149\n\nPregnancies ≤ 4.5\ngini = 0.117\nsamples = 16\nvalue = [15, 1]\nclass = y\n0\n\n\n\n148->149\n\n\n\n\n\n154\n\nDiabetesPedigreeFunction ≤ 0.302\ngini = 0.375\nsamples = 4\nvalue = [1, 3]\nclass = y\n1\n\n\n\n148->154\n\n\n\n\n\n150\n\ngini = 0.0\nsamples = 14\nvalue = [14, 0]\nclass = y\n0\n\n\n\n149->150\n\n\n\n\n\n151\n\nGlucose ≤ 146.5\ngini = 0.5\nsamples = 2\nvalue = [1, 1]\nclass = y\n0\n\n\n\n149->151\n\n\n\n\n\n152\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n151->152\n\n\n\n\n\n153\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n151->153\n\n\n\n\n\n155\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n154->155\n\n\n\n\n\n156\n\ngini = 0.0\nsamples = 3\nvalue = [0, 3]\nclass = y\n1\n\n\n\n154->156\n\n\n\n\n\n158\n\nBMI ≤ 45.55\ngini = 0.499\nsamples = 38\nvalue = [18, 20]\nclass = y\n1\n\n\n\n157->158\n\n\n\n\n\n187\n\nInsulin ≤ 333.5\ngini = 0.239\nsamples = 36\nvalue = [5, 31]\nclass = y\n1\n\n\n\n157->187\n\n\n\n\n\n159\n\nPregnancies ≤ 1.5\ngini = 0.496\nsamples = 33\nvalue = [18, 15]\nclass = y\n0\n\n\n\n158->159\n\n\n\n\n\n186\n\ngini = 0.0\nsamples = 5\nvalue = [0, 5]\nclass = y\n1\n\n\n\n158->186\n\n\n\n\n\n160\n\ngini = 0.0\nsamples = 3\nvalue = [0, 3]\nclass = y\n1\n\n\n\n159->160\n\n\n\n\n\n161\n\nBMI ≤ 37.25\ngini = 0.48\nsamples = 30\nvalue = [18, 12]\nclass = y\n0\n\n\n\n159->161\n\n\n\n\n\n162\n\nBMI ≤ 36.25\ngini = 0.499\nsamples = 21\nvalue = [10, 11]\nclass = y\n1\n\n\n\n161->162\n\n\n\n\n\n181\n\nGlucose ≤ 146.5\ngini = 0.198\nsamples = 9\nvalue = [8, 1]\nclass = y\n0\n\n\n\n161->181\n\n\n\n\n\n163\n\nPregnancies ≤ 8.5\ngini = 0.499\nsamples = 19\nvalue = [10, 9]\nclass = y\n0\n\n\n\n162->163\n\n\n\n\n\n180\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n162->180\n\n\n\n\n\n164\n\nDiabetesPedigreeFunction ≤ 0.266\ngini = 0.49\nsamples = 14\nvalue = [6, 8]\nclass = y\n1\n\n\n\n163->164\n\n\n\n\n\n175\n\nBloodPressure ≤ 73.0\ngini = 0.32\nsamples = 5\nvalue = [4, 1]\nclass = y\n0\n\n\n\n163->175\n\n\n\n\n\n165\n\nBMI ≤ 31.65\ngini = 0.496\nsamples = 11\nvalue = [6, 5]\nclass = y\n0\n\n\n\n164->165\n\n\n\n\n\n174\n\ngini = 0.0\nsamples = 3\nvalue = [0, 3]\nclass = y\n1\n\n\n\n164->174\n\n\n\n\n\n166\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n165->166\n\n\n\n\n\n167\n\nAge ≤ 67.5\ngini = 0.444\nsamples = 9\nvalue = [6, 3]\nclass = y\n0\n\n\n\n165->167\n\n\n\n\n\n168\n\nGlucose ≤ 126.5\ngini = 0.375\nsamples = 8\nvalue = [6, 2]\nclass = y\n0\n\n\n\n167->168\n\n\n\n\n\n173\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n167->173\n\n\n\n\n\n169\n\nAge ≤ 34.5\ngini = 0.444\nsamples = 3\nvalue = [1, 2]\nclass = y\n1\n\n\n\n168->169\n\n\n\n\n\n172\n\ngini = 0.0\nsamples = 5\nvalue = [5, 0]\nclass = y\n0\n\n\n\n168->172\n\n\n\n\n\n170\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n169->170\n\n\n\n\n\n171\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n169->171\n\n\n\n\n\n176\n\nPregnancies ≤ 9.5\ngini = 0.5\nsamples = 2\nvalue = [1, 1]\nclass = y\n0\n\n\n\n175->176\n\n\n\n\n\n179\n\ngini = 0.0\nsamples = 3\nvalue = [3, 0]\nclass = y\n0\n\n\n\n175->179\n\n\n\n\n\n177\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n176->177\n\n\n\n\n\n178\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n176->178\n\n\n\n\n\n182\n\ngini = 0.0\nsamples = 7\nvalue = [7, 0]\nclass = y\n0\n\n\n\n181->182\n\n\n\n\n\n183\n\nSkinThickness ≤ 17.5\ngini = 0.5\nsamples = 2\nvalue = [1, 1]\nclass = y\n0\n\n\n\n181->183\n\n\n\n\n\n184\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n183->184\n\n\n\n\n\n185\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n183->185\n\n\n\n\n\n188\n\nPregnancies ≤ 0.5\ngini = 0.165\nsamples = 33\nvalue = [3, 30]\nclass = y\n1\n\n\n\n187->188\n\n\n\n\n\n199\n\nDiabetesPedigreeFunction ≤ 0.581\ngini = 0.444\nsamples = 3\nvalue = [2, 1]\nclass = y\n0\n\n\n\n187->199\n\n\n\n\n\n189\n\nBloodPressure ≤ 60.0\ngini = 0.48\nsamples = 5\nvalue = [2, 3]\nclass = y\n1\n\n\n\n188->189\n\n\n\n\n\n194\n\nBMI ≤ 40.05\ngini = 0.069\nsamples = 28\nvalue = [1, 27]\nclass = y\n1\n\n\n\n188->194\n\n\n\n\n\n190\n\ngini = 0.0\nsamples = 2\nvalue = [0, 2]\nclass = y\n1\n\n\n\n189->190\n\n\n\n\n\n191\n\nBloodPressure ≤ 85.0\ngini = 0.444\nsamples = 3\nvalue = [2, 1]\nclass = y\n0\n\n\n\n189->191\n\n\n\n\n\n192\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n191->192\n\n\n\n\n\n193\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n191->193\n\n\n\n\n\n195\n\ngini = 0.0\nsamples = 22\nvalue = [0, 22]\nclass = y\n1\n\n\n\n194->195\n\n\n\n\n\n196\n\nBMI ≤ 40.7\ngini = 0.278\nsamples = 6\nvalue = [1, 5]\nclass = y\n1\n\n\n\n194->196\n\n\n\n\n\n197\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n196->197\n\n\n\n\n\n198\n\ngini = 0.0\nsamples = 5\nvalue = [0, 5]\nclass = y\n1\n\n\n\n196->198\n\n\n\n\n\n200\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n199->200\n\n\n\n\n\n201\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n199->201\n\n\n\n\n\n203\n\nDiabetesPedigreeFunction ≤ 0.307\ngini = 0.182\nsamples = 69\nvalue = [7, 62]\nclass = y\n1\n\n\n\n202->203\n\n\n\n\n\n220\n\nDiabetesPedigreeFunction ≤ 0.412\ngini = 0.444\nsamples = 3\nvalue = [2, 1]\nclass = y\n0\n\n\n\n202->220\n\n\n\n\n\n204\n\nBMI ≤ 31.4\ngini = 0.401\nsamples = 18\nvalue = [5, 13]\nclass = y\n1\n\n\n\n203->204\n\n\n\n\n\n213\n\nAge ≤ 48.0\ngini = 0.075\nsamples = 51\nvalue = [2, 49]\nclass = y\n1\n\n\n\n203->213\n\n\n\n\n\n205\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n204->205\n\n\n\n\n\n206\n\nGlucose ≤ 179.5\ngini = 0.305\nsamples = 16\nvalue = [3, 13]\nclass = y\n1\n\n\n\n204->206\n\n\n\n\n\n207\n\nGlucose ≤ 177.0\ngini = 0.42\nsamples = 10\nvalue = [3, 7]\nclass = y\n1\n\n\n\n206->207\n\n\n\n\n\n212\n\ngini = 0.0\nsamples = 6\nvalue = [0, 6]\nclass = y\n1\n\n\n\n206->212\n\n\n\n\n\n208\n\nBMI ≤ 45.1\ngini = 0.219\nsamples = 8\nvalue = [1, 7]\nclass = y\n1\n\n\n\n207->208\n\n\n\n\n\n211\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n207->211\n\n\n\n\n\n209\n\ngini = 0.0\nsamples = 7\nvalue = [0, 7]\nclass = y\n1\n\n\n\n208->209\n\n\n\n\n\n210\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n208->210\n\n\n\n\n\n214\n\ngini = 0.0\nsamples = 42\nvalue = [0, 42]\nclass = y\n1\n\n\n\n213->214\n\n\n\n\n\n215\n\nAge ≤ 50.5\ngini = 0.346\nsamples = 9\nvalue = [2, 7]\nclass = y\n1\n\n\n\n213->215\n\n\n\n\n\n216\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n215->216\n\n\n\n\n\n217\n\nBMI ≤ 45.95\ngini = 0.219\nsamples = 8\nvalue = [1, 7]\nclass = y\n1\n\n\n\n215->217\n\n\n\n\n\n218\n\ngini = 0.0\nsamples = 7\nvalue = [0, 7]\nclass = y\n1\n\n\n\n217->218\n\n\n\n\n\n219\n\ngini = 0.0\nsamples = 1\nvalue = [1, 0]\nclass = y\n0\n\n\n\n217->219\n\n\n\n\n\n221\n\ngini = 0.0\nsamples = 1\nvalue = [0, 1]\nclass = y\n1\n\n\n\n220->221\n\n\n\n\n\n222\n\ngini = 0.0\nsamples = 2\nvalue = [2, 0]\nclass = y\n0\n\n\n\n220->222\n\n\n\n\n\n" }, "metadata": { "tags": [] }, "execution_count": 147 } ] }, { "metadata": { "id": "tea6hwnGQn_J", "colab_type": "text" }, "cell_type": "markdown", "source": [ "# Creando el segundo modelo" ] }, { "metadata": { "id": "stmDtlUJEXnE", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "tree = DecisionTreeClassifier(min_samples_leaf=10, max_depth=8, min_samples_split=50)" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "DBGRIi_kQuIK", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 115 }, "outputId": "b20dba3f-da76-455a-f4d0-e94b575d8ecd" }, "cell_type": "code", "source": [ "tree.fit(X_train, y_train)" ], "execution_count": 149, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=8,\n", " max_features=None, max_leaf_nodes=None,\n", " min_impurity_decrease=0.0, min_impurity_split=None,\n", " min_samples_leaf=10, min_samples_split=50,\n", " min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n", " splitter='best')" ] }, "metadata": { "tags": [] }, "execution_count": 149 } ] }, { "metadata": { "id": "qLLVhArzQxu2", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "validation_prediction = tree.predict(X_val)\n", "training_prediction = tree.predict(X_train)" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "ckMKznJbQzPB", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 50 }, "outputId": "88e6a9ae-1dc7-4381-c145-8b971c6e3808" }, "cell_type": "code", "source": [ "print('Exactitud training data: ', accuracy_score(y_true=y_train, y_pred=training_prediction))\n", "print('Exactitud validation data: ', accuracy_score(y_true=y_val, y_pred=validation_prediction))" ], "execution_count": 151, "outputs": [ { "output_type": "stream", "text": [ "Exactitud training data: 0.7964169381107492\n", "Exactitud validation data: 0.8116883116883117\n" ], "name": "stdout" } ] }, { "metadata": { "id": "k5EtFykfR3Fg", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "dot_data = export_graphviz(tree, out_file=None, \n", " feature_names=feature_names, \n", " class_names=True, \n", " filled=True, rounded=True, \n", " special_characters=True) \n", "graph = graphviz.Source(dot_data)" ], "execution_count": 0, "outputs": [] }, { "metadata": { "id": "5UnNfVLaSPM8", "colab_type": "code", "colab": { "base_uri": "https://localhost:8080/", "height": 1048 }, "outputId": "b31527c0-86d7-4b6b-bcb5-e25718d787c7" }, "cell_type": "code", "source": [ "graph" ], "execution_count": 153, "outputs": [ { "output_type": "execute_result", "data": { "text/plain": [ "" ], "image/svg+xml": "\n\n\n\n\n\nTree\n\n\n\n0\n\nGlucose ≤ 123.5\ngini = 0.461\nsamples = 614\nvalue = [393, 221]\nclass = y\n0\n\n\n\n1\n\nAge ≤ 28.5\ngini = 0.301\nsamples = 352\nvalue = [287, 65]\nclass = y\n0\n\n\n\n0->1\n\n\nTrue\n\n\n\n22\n\nBMI ≤ 30.05\ngini = 0.482\nsamples = 262\nvalue = [106, 156]\nclass = y\n1\n\n\n\n0->22\n\n\nFalse\n\n\n\n2\n\nBMI ≤ 30.95\ngini = 0.162\nsamples = 202\nvalue = [184, 18]\nclass = y\n0\n\n\n\n1->2\n\n\n\n\n\n13\n\nBMI ≤ 26.35\ngini = 0.43\nsamples = 150\nvalue = [103, 47]\nclass = y\n0\n\n\n\n1->13\n\n\n\n\n\n3\n\nGlucose ≤ 106.5\ngini = 0.036\nsamples = 110\nvalue = [108, 2]\nclass = y\n0\n\n\n\n2->3\n\n\n\n\n\n6\n\nBloodPressure ≤ 64.5\ngini = 0.287\nsamples = 92\nvalue = [76, 16]\nclass = y\n0\n\n\n\n2->6\n\n\n\n\n\n4\n\ngini = 0.0\nsamples = 83\nvalue = [83, 0]\nclass = y\n0\n\n\n\n3->4\n\n\n\n\n\n5\n\ngini = 0.137\nsamples = 27\nvalue = [25, 2]\nclass = y\n0\n\n\n\n3->5\n\n\n\n\n\n7\n\ngini = 0.404\nsamples = 32\nvalue = [23, 9]\nclass = y\n0\n\n\n\n6->7\n\n\n\n\n\n8\n\nBloodPressure ≤ 81.0\ngini = 0.206\nsamples = 60\nvalue = [53, 7]\nclass = y\n0\n\n\n\n6->8\n\n\n\n\n\n9\n\nBMI ≤ 33.7\ngini = 0.147\nsamples = 50\nvalue = [46, 4]\nclass = y\n0\n\n\n\n8->9\n\n\n\n\n\n12\n\ngini = 0.42\nsamples = 10\nvalue = [7, 3]\nclass = y\n0\n\n\n\n8->12\n\n\n\n\n\n10\n\ngini = 0.305\nsamples = 16\nvalue = [13, 3]\nclass = y\n0\n\n\n\n9->10\n\n\n\n\n\n11\n\ngini = 0.057\nsamples = 34\nvalue = [33, 1]\nclass = y\n0\n\n\n\n9->11\n\n\n\n\n\n14\n\ngini = 0.0\nsamples = 29\nvalue = [29, 0]\nclass = y\n0\n\n\n\n13->14\n\n\n\n\n\n15\n\nGlucose ≤ 99.5\ngini = 0.475\nsamples = 121\nvalue = [74, 47]\nclass = y\n0\n\n\n\n13->15\n\n\n\n\n\n16\n\ngini = 0.337\nsamples = 42\nvalue = [33, 9]\nclass = y\n0\n\n\n\n15->16\n\n\n\n\n\n17\n\nDiabetesPedigreeFunction ≤ 0.22\ngini = 0.499\nsamples = 79\nvalue = [41, 38]\nclass = y\n0\n\n\n\n15->17\n\n\n\n\n\n18\n\ngini = 0.375\nsamples = 16\nvalue = [12, 4]\nclass = y\n0\n\n\n\n17->18\n\n\n\n\n\n19\n\nPregnancies ≤ 6.5\ngini = 0.497\nsamples = 63\nvalue = [29, 34]\nclass = y\n1\n\n\n\n17->19\n\n\n\n\n\n20\n\ngini = 0.493\nsamples = 41\nvalue = [23, 18]\nclass = y\n0\n\n\n\n19->20\n\n\n\n\n\n21\n\ngini = 0.397\nsamples = 22\nvalue = [6, 16]\nclass = y\n1\n\n\n\n19->21\n\n\n\n\n\n23\n\nAge ≤ 26.0\ngini = 0.438\nsamples = 74\nvalue = [50, 24]\nclass = y\n0\n\n\n\n22->23\n\n\n\n\n\n28\n\nGlucose ≤ 157.5\ngini = 0.418\nsamples = 188\nvalue = [56, 132]\nclass = y\n1\n\n\n\n22->28\n\n\n\n\n\n24\n\ngini = 0.0\nsamples = 19\nvalue = [19, 0]\nclass = y\n0\n\n\n\n23->24\n\n\n\n\n\n25\n\nAge ≤ 54.5\ngini = 0.492\nsamples = 55\nvalue = [31, 24]\nclass = y\n0\n\n\n\n23->25\n\n\n\n\n\n26\n\ngini = 0.494\nsamples = 38\nvalue = [17, 21]\nclass = y\n1\n\n\n\n25->26\n\n\n\n\n\n27\n\ngini = 0.291\nsamples = 17\nvalue = [14, 3]\nclass = y\n0\n\n\n\n25->27\n\n\n\n\n\n29\n\nAge ≤ 28.5\ngini = 0.482\nsamples = 116\nvalue = [47, 69]\nclass = y\n1\n\n\n\n28->29\n\n\n\n\n\n34\n\nDiabetesPedigreeFunction ≤ 0.307\ngini = 0.219\nsamples = 72\nvalue = [9, 63]\nclass = y\n1\n\n\n\n28->34\n\n\n\n\n\n30\n\ngini = 0.49\nsamples = 42\nvalue = [24, 18]\nclass = y\n0\n\n\n\n29->30\n\n\n\n\n\n31\n\nDiabetesPedigreeFunction ≤ 0.429\ngini = 0.428\nsamples = 74\nvalue = [23, 51]\nclass = y\n1\n\n\n\n29->31\n\n\n\n\n\n32\n\ngini = 0.499\nsamples = 38\nvalue = [18, 20]\nclass = y\n1\n\n\n\n31->32\n\n\n\n\n\n33\n\ngini = 0.239\nsamples = 36\nvalue = [5, 31]\nclass = y\n1\n\n\n\n31->33\n\n\n\n\n\n35\n\ngini = 0.401\nsamples = 18\nvalue = [5, 13]\nclass = y\n1\n\n\n\n34->35\n\n\n\n\n\n36\n\nInsulin ≤ 257.0\ngini = 0.137\nsamples = 54\nvalue = [4, 50]\nclass = y\n1\n\n\n\n34->36\n\n\n\n\n\n37\n\ngini = 0.05\nsamples = 39\nvalue = [1, 38]\nclass = y\n1\n\n\n\n36->37\n\n\n\n\n\n38\n\ngini = 0.32\nsamples = 15\nvalue = [3, 12]\nclass = y\n1\n\n\n\n36->38\n\n\n\n\n\n" }, "metadata": { "tags": [] }, "execution_count": 153 } ] }, { "metadata": { "id": "BbZe3mXGSQON", "colab_type": "code", "colab": {} }, "cell_type": "code", "source": [ "" ], "execution_count": 0, "outputs": [] } ] }