Deploy your machine learning model using streamlit.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "afb5efdd-fbde-436d-a8e6-be1c2b00321f",
   "metadata": {},
   "source": [
    "## Deploy your Machine Learning Model using Streamlit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "b3f910ba-69fc-4ac7-b8bb-b17368835f5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "variables = ['APARTMENTS_AVG','LIVINGAPARTMENTS_AVG','OCCUPATION_TYPE']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "e5532721-263e-4fd1-aa4b-c87740d4bf71",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "path = ''\n",
    "df = pd.read_csv(path + '/application_train.csv')\n",
    "df = df.sample(frac=0.1) # Take some records just to build a toy model\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(df[variables],df['TARGET'],random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81fe2b64-2558-42bd-ab37-273668fa45e4",
   "metadata": {},
   "source": [
    "### Data pre-processing: fillna for numeric variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "827fd584-c958-4f8a-a452-307354b459f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Impute with the mean of the training data\n",
    "# Keep the same mean to impute the test data or any future data\n",
    "APARTMENTS_AVG_MEAN = X_train['APARTMENTS_AVG'].mean()\n",
    "APARTMENTS_AVG_MAX = X_train['APARTMENTS_AVG'].mean()\n",
    "APARTMENTS_AVG_MIN = X_train['APARTMENTS_AVG'].min()\n",
    "LIVINGAPARTMENTS_AVG_MEAN = X_train['LIVINGAPARTMENTS_AVG'].mean()\n",
    "LIVINGAPARTMENTS_AVG_MAX = X_train['LIVINGAPARTMENTS_AVG'].max()\n",
    "LIVINGAPARTMENTS_AVG_MIN = X_train['LIVINGAPARTMENTS_AVG'].min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "f8545ccc-155b-4da9-b29d-b6a41ffc8424",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 1.0)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "LIVINGAPARTMENTS_AVG_MIN, LIVINGAPARTMENTS_AVG_MAX"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "fae750e8-38e9-457c-9ed3-29d6226f574b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.0, 0.11768214539007016)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "APARTMENTS_AVG_MIN, APARTMENTS_AVG_MAX"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "0e8e6621-85e9-4d28-a849-4af9aaa8377e",
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_values = {'APARTMENTS_AVG': APARTMENTS_AVG_MEAN, 'LIVINGAPARTMENTS_AVG': LIVINGAPARTMENTS_AVG_MEAN}\n",
    "X_train = X_train.fillna(value=mean_values)\n",
    "X_test = X_test.fillna(value=mean_values)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff73a035-b14f-4d31-8a2f-493561e67911",
   "metadata": {},
   "source": [
    "### Data pre-processing: categorical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "3715ebe4-be65-49dd-8889-7c21cc5e7cd6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Laborers                 4117\n",
       "Sales staff              2315\n",
       "Core staff               2107\n",
       "Managers                 1577\n",
       "Drivers                  1420\n",
       "High skill tech staff     860\n",
       "Accountants               742\n",
       "Medicine staff            633\n",
       "Security staff            530\n",
       "Cooking staff             456\n",
       "Cleaning staff            350\n",
       "Private service staff     185\n",
       "Low-skill Laborers        160\n",
       "Secretaries               111\n",
       "Waiters/barmen staff       91\n",
       "Realty agents              59\n",
       "HR staff                   41\n",
       "IT staff                   40\n",
       "Name: OCCUPATION_TYPE, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train['OCCUPATION_TYPE'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ce3551cb-318e-406a-83c0-96cb97e289f0",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "7f33af60-d4ac-430f-b1d8-ae41736f9700",
   "metadata": {},
   "outputs": [],
   "source": [
    "OCCUPATION_list  = ['Laborers','Sales staff','Core staff','Managers','Drivers','High skill tech staff',\n",
    "'Accountants','Medicine staff','Security staff','Cooking staff','Cleaning staff']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "ed91d9d3-fb38-4d55-bbf2-1c1b008da7a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train['OCCUPATION_GRP'] = np.where(X_train['OCCUPATION_TYPE'].isin(OCCUPATION_list), X_train['OCCUPATION_TYPE'], 'OTHER')\n",
    "X_test['OCCUPATION_GRP'] = np.where(X_test['OCCUPATION_TYPE'].isin(OCCUPATION_list), X_test['OCCUPATION_TYPE'], 'OTHER')\n",
    "X_train = X_train.drop('OCCUPATION_TYPE',axis=1)\n",
    "X_test = X_test.drop('OCCUPATION_TYPE',axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "b4b4e4c7-50b0-43fa-9ccb-5b907cae3e15",
   "metadata": {},
   "outputs": [],
   "source": [
    "#X_train['OCCUPATION_GRP'] = X_train['OCCUPATION_GRP'].str.replace(' ','_')\n",
    "#X_train['OCCUPATION_GRP'] = X_train['OCCUPATION_GRP'].str.replace('/','_')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "42dcbda5-cb91-4718-9f9d-eb706574f881",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "OTHER                    7956\n",
       "Laborers                 4117\n",
       "Sales staff              2315\n",
       "Core staff               2107\n",
       "Managers                 1577\n",
       "Drivers                  1420\n",
       "High skill tech staff     860\n",
       "Accountants               742\n",
       "Medicine staff            633\n",
       "Security staff            530\n",
       "Cooking staff             456\n",
       "Cleaning staff            350\n",
       "Name: OCCUPATION_GRP, dtype: int64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train['OCCUPATION_GRP'].value_counts(dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "1fd5876b-e3e1-4a0b-ae1a-9ce02bb200bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get_Dummy for One-Hot #\n",
    "def getDummy(df,var):\n",
    "    df[var] = df[var].str.replace(' ','_')\n",
    "    df[var] = df[var].str.replace('/','_')\n",
    "    dummies = pd.get_dummies(df[var])\n",
    "    df2 = pd.concat([df, dummies], axis=1)\n",
    "    df2 = df2.drop([var], axis=1)\n",
    "    return(df2)\n",
    "    \n",
    "X_train = getDummy(X_train,'OCCUPATION_GRP')\n",
    "X_test = getDummy(X_test,'OCCUPATION_GRP')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "0f1d9231-80f5-479f-869c-e4bca54e9162",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['APARTMENTS_AVG', 'Accountants', 'Cleaning_staff', 'Cooking_staff',\n",
       "       'Core_staff', 'Drivers', 'High_skill_tech_staff',\n",
       "       'LIVINGAPARTMENTS_AVG', 'Laborers', 'Managers', 'Medicine_staff',\n",
       "       'OTHER', 'Sales_staff', 'Security_staff'], dtype=object)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sort(X_train.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "6444f0f2-faaa-429d-85dd-c63b3e975f57",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(max_depth=6, min_samples_leaf=5)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier # for classification\n",
    "\n",
    "# First, specify the model. \n",
    "dtree = DecisionTreeClassifier(min_samples_leaf = 5, max_depth = 6)\n",
    "# Then, train the model.\n",
    "dtree.fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "a241854b-3584-4065-9da1-6d8130658992",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='feature'>"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = pd.DataFrame({'feature': X_train.columns, 'importance': np.round(dtree.feature_importances_,3)})\n",
    "importances = importances.sort_values('importance',ascending=False)\n",
    "importances.plot.bar(x='feature', figsize=(4,3),fontsize=6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "52ad2a29-3f2f-4e7b-a0e4-d3458568214a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.06263063, 0.06263063, 0.06263063, 0.10874704, 0.06263063])"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Predict a few records\n",
    "predictions = dtree.predict_proba(X_test[0:5])[:,1]\n",
    "predictions"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bc6318f-bca8-4d37-9a44-3f422b387450",
   "metadata": {},
   "source": [
    "## Save the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "7d7ef156-3083-45b7-bdb3-17de6f206f34",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.06263063])"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pickle \n",
    "# save the model to disk\n",
    "modelname = path + '/toymodel.pkl'\n",
    "pickle.dump(dtree, open(modelname, 'wb'))\n",
    " \n",
    "# load the model from disk\n",
    "loaded_model = pickle.load(open(modelname, 'rb'))\n",
    "predictions = loaded_model.predict_proba(X_test[0:1])[:,1]\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "5997a6e5-1fd1-4bc4-96ca-00a4944db0c0",
   "metadata": {},
   "outputs": [
    {
     "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>APARTMENTS_AVG</th>\n",
       "      <th>LIVINGAPARTMENTS_AVG</th>\n",
       "      <th>Accountants</th>\n",
       "      <th>Cleaning_staff</th>\n",
       "      <th>Cooking_staff</th>\n",
       "      <th>Core_staff</th>\n",
       "      <th>Drivers</th>\n",
       "      <th>High_skill_tech_staff</th>\n",
       "      <th>Laborers</th>\n",
       "      <th>Managers</th>\n",
       "      <th>Medicine_staff</th>\n",
       "      <th>OTHER</th>\n",
       "      <th>Sales_staff</th>\n",
       "      <th>Security_staff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>154376</th>\n",
       "      <td>0.0186</td>\n",
       "      <td>0.101631</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        APARTMENTS_AVG  LIVINGAPARTMENTS_AVG  Accountants  Cleaning_staff  \\\n",
       "154376          0.0186              0.101631            0               0   \n",
       "\n",
       "        Cooking_staff  Core_staff  Drivers  High_skill_tech_staff  Laborers  \\\n",
       "154376              0           0        0                      0         0   \n",
       "\n",
       "        Managers  Medicine_staff  OTHER  Sales_staff  Security_staff  \n",
       "154376         0               0      1            0               0  "
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test[0:1]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}