decision tree and random forest udemy project.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7b353960",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ca872097",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('loan_data.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4b550d81",
   "metadata": {},
   "outputs": [],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f7dbf86c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "63cccb44",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "41ad65bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6579126f",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "for col in df.columns:\n",
    "    print(df[col].value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "8177a72c",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "df[df['credit.policy']==1]['fico'].hist(alpha=0.5,color='blue',\n",
    "                                              bins=30,label='Credit.Policy=1')\n",
    "df[df['credit.policy']==0]['fico'].hist(alpha=0.5,color='red',\n",
    "                                              bins=30,label='Credit.Policy=0')\n",
    "plt.legend()\n",
    "plt.xlabel('FICO')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "faab974c",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(10,6))\n",
    "df[df['not.fully.paid']==1]['fico'].hist(alpha=0.5,color='blue',\n",
    "                                              bins=30,label='not.fully.paid=1')\n",
    "df[df['not.fully.paid']==0]['fico'].hist(alpha=0.5,color='red',\n",
    "                                              bins=30,label='not.fully.paid=0')\n",
    "plt.legend()\n",
    "plt.xlabel('FICO')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "e5546eed",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(11,7))\n",
    "sns.countplot(x='purpose',hue='not.fully.paid',data=df,palette='Set1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1bda1336",
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.jointplot(x='fico',y='int.rate',data=df,color='purple')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "aa8167ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(11,7))\n",
    "sns.lmplot(y='int.rate',x='fico',data=df,hue='credit.policy',\n",
    "           col='not.fully.paid',palette='Set1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1fd2a75f",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.info()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c68f1e73",
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_feats = ['purpose']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "29445fec",
   "metadata": {},
   "outputs": [],
   "source": [
    "final_data = pd.get_dummies(df,columns=cat_feats,drop_first=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6cbd5e9e",
   "metadata": {},
   "outputs": [],
   "source": [
    "final_data.info()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "fdc1d30c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "f1889cf4",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = final_data.drop('not.fully.paid',axis=1)\n",
    "y = final_data['not.fully.paid']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.30, random_state=101)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "09d3c857",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "6992194b",
   "metadata": {},
   "outputs": [],
   "source": [
    "dtree = DecisionTreeClassifier()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "4448e8cb",
   "metadata": {},
   "outputs": [],
   "source": [
    "dtree.fit(X_train,y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "632d18c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = dtree.predict(X_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6e6b2e46",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report,confusion_matrix\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "8e0dab89",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(classification_report(y_test,predictions))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "3513eb29",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "bf14c2b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfc = RandomForestClassifier(n_estimators=600)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "68c93450",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfc.fit(X_train,y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8c5a8549",
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions = rfc.predict(X_test)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f5111ede",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report,confusion_matrix\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3d2ae52a",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(classification_report(y_test,predictions))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "24447247",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(confusion_matrix(y_test,predictions))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5a374d44",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}