AlphabetSoupCharity_Optimization2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>EIN</th>\n",
       "      <th>NAME</th>\n",
       "      <th>APPLICATION_TYPE</th>\n",
       "      <th>AFFILIATION</th>\n",
       "      <th>CLASSIFICATION</th>\n",
       "      <th>USE_CASE</th>\n",
       "      <th>ORGANIZATION</th>\n",
       "      <th>STATUS</th>\n",
       "      <th>INCOME_AMT</th>\n",
       "      <th>SPECIAL_CONSIDERATIONS</th>\n",
       "      <th>ASK_AMT</th>\n",
       "      <th>IS_SUCCESSFUL</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10520599</td>\n",
       "      <td>BLUE KNIGHTS MOTORCYCLE CLUB</td>\n",
       "      <td>T10</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C1000</td>\n",
       "      <td>ProductDev</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10531628</td>\n",
       "      <td>AMERICAN CHESAPEAKE CLUB CHARITABLE TR</td>\n",
       "      <td>T3</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C2000</td>\n",
       "      <td>Preservation</td>\n",
       "      <td>Co-operative</td>\n",
       "      <td>1</td>\n",
       "      <td>1-9999</td>\n",
       "      <td>N</td>\n",
       "      <td>108590</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10547893</td>\n",
       "      <td>ST CLOUD PROFESSIONAL FIREFIGHTERS</td>\n",
       "      <td>T5</td>\n",
       "      <td>CompanySponsored</td>\n",
       "      <td>C3000</td>\n",
       "      <td>ProductDev</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10553066</td>\n",
       "      <td>SOUTHSIDE ATHLETIC ASSOCIATION</td>\n",
       "      <td>T3</td>\n",
       "      <td>CompanySponsored</td>\n",
       "      <td>C2000</td>\n",
       "      <td>Preservation</td>\n",
       "      <td>Trust</td>\n",
       "      <td>1</td>\n",
       "      <td>10000-24999</td>\n",
       "      <td>N</td>\n",
       "      <td>6692</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10556103</td>\n",
       "      <td>GENETIC RESEARCH INSTITUTE OF THE DESERT</td>\n",
       "      <td>T3</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C1000</td>\n",
       "      <td>Heathcare</td>\n",
       "      <td>Trust</td>\n",
       "      <td>1</td>\n",
       "      <td>100000-499999</td>\n",
       "      <td>N</td>\n",
       "      <td>142590</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        EIN                                      NAME APPLICATION_TYPE  \\\n",
       "0  10520599              BLUE KNIGHTS MOTORCYCLE CLUB              T10   \n",
       "1  10531628    AMERICAN CHESAPEAKE CLUB CHARITABLE TR               T3   \n",
       "2  10547893        ST CLOUD PROFESSIONAL FIREFIGHTERS               T5   \n",
       "3  10553066            SOUTHSIDE ATHLETIC ASSOCIATION               T3   \n",
       "4  10556103  GENETIC RESEARCH INSTITUTE OF THE DESERT               T3   \n",
       "\n",
       "        AFFILIATION CLASSIFICATION      USE_CASE  ORGANIZATION  STATUS  \\\n",
       "0       Independent          C1000    ProductDev   Association       1   \n",
       "1       Independent          C2000  Preservation  Co-operative       1   \n",
       "2  CompanySponsored          C3000    ProductDev   Association       1   \n",
       "3  CompanySponsored          C2000  Preservation         Trust       1   \n",
       "4       Independent          C1000     Heathcare         Trust       1   \n",
       "\n",
       "      INCOME_AMT SPECIAL_CONSIDERATIONS  ASK_AMT  IS_SUCCESSFUL  \n",
       "0              0                      N     5000              1  \n",
       "1         1-9999                      N   108590              1  \n",
       "2              0                      N     5000              0  \n",
       "3    10000-24999                      N     6692              1  \n",
       "4  100000-499999                      N   142590              1  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Import our dependencies\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import pandas as pd\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Dense, Input\n",
    "from tensorflow.keras.callbacks import ModelCheckpoint\n",
    "import os\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow import keras\n",
    "import tensorflow as tf\n",
    "import keras_tuner as kt\n",
    "from kerastuner.tuners import Hyperband\n",
    "from kerastuner import HyperParameters\n",
    "\n",
    "\n",
    "#  Import and read the charity_data.csv.\n",
    "import pandas as pd \n",
    "application_df = pd.read_csv(\"https://static.bc-edx.com/data/dl-1-2/m21/lms/starter/charity_data.csv\")\n",
    "application_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Saving the column NAME"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>NAME</th>\n",
       "      <th>APPLICATION_TYPE</th>\n",
       "      <th>AFFILIATION</th>\n",
       "      <th>CLASSIFICATION</th>\n",
       "      <th>USE_CASE</th>\n",
       "      <th>ORGANIZATION</th>\n",
       "      <th>STATUS</th>\n",
       "      <th>INCOME_AMT</th>\n",
       "      <th>SPECIAL_CONSIDERATIONS</th>\n",
       "      <th>ASK_AMT</th>\n",
       "      <th>IS_SUCCESSFUL</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BLUE KNIGHTS MOTORCYCLE CLUB</td>\n",
       "      <td>T10</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C1000</td>\n",
       "      <td>ProductDev</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AMERICAN CHESAPEAKE CLUB CHARITABLE TR</td>\n",
       "      <td>T3</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C2000</td>\n",
       "      <td>Preservation</td>\n",
       "      <td>Co-operative</td>\n",
       "      <td>1</td>\n",
       "      <td>1-9999</td>\n",
       "      <td>N</td>\n",
       "      <td>108590</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ST CLOUD PROFESSIONAL FIREFIGHTERS</td>\n",
       "      <td>T5</td>\n",
       "      <td>CompanySponsored</td>\n",
       "      <td>C3000</td>\n",
       "      <td>ProductDev</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SOUTHSIDE ATHLETIC ASSOCIATION</td>\n",
       "      <td>T3</td>\n",
       "      <td>CompanySponsored</td>\n",
       "      <td>C2000</td>\n",
       "      <td>Preservation</td>\n",
       "      <td>Trust</td>\n",
       "      <td>1</td>\n",
       "      <td>10000-24999</td>\n",
       "      <td>N</td>\n",
       "      <td>6692</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GENETIC RESEARCH INSTITUTE OF THE DESERT</td>\n",
       "      <td>T3</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C1000</td>\n",
       "      <td>Heathcare</td>\n",
       "      <td>Trust</td>\n",
       "      <td>1</td>\n",
       "      <td>100000-499999</td>\n",
       "      <td>N</td>\n",
       "      <td>142590</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34294</th>\n",
       "      <td>THE LIONS CLUB OF HONOLULU KAMEHAMEHA</td>\n",
       "      <td>T4</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C1000</td>\n",
       "      <td>ProductDev</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34295</th>\n",
       "      <td>INTERNATIONAL ASSOCIATION OF LIONS CLUBS</td>\n",
       "      <td>T4</td>\n",
       "      <td>CompanySponsored</td>\n",
       "      <td>C3000</td>\n",
       "      <td>ProductDev</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34296</th>\n",
       "      <td>PTA HAWAII CONGRESS</td>\n",
       "      <td>T3</td>\n",
       "      <td>CompanySponsored</td>\n",
       "      <td>C2000</td>\n",
       "      <td>Preservation</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34297</th>\n",
       "      <td>AMERICAN FEDERATION OF GOVERNMENT EMPLOYEES LO...</td>\n",
       "      <td>T5</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C3000</td>\n",
       "      <td>ProductDev</td>\n",
       "      <td>Association</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>5000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34298</th>\n",
       "      <td>WATERHOUSE CHARITABLE TR</td>\n",
       "      <td>T3</td>\n",
       "      <td>Independent</td>\n",
       "      <td>C1000</td>\n",
       "      <td>Preservation</td>\n",
       "      <td>Co-operative</td>\n",
       "      <td>1</td>\n",
       "      <td>1M-5M</td>\n",
       "      <td>N</td>\n",
       "      <td>36500179</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>34299 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    NAME APPLICATION_TYPE  \\\n",
       "0                           BLUE KNIGHTS MOTORCYCLE CLUB              T10   \n",
       "1                 AMERICAN CHESAPEAKE CLUB CHARITABLE TR               T3   \n",
       "2                     ST CLOUD PROFESSIONAL FIREFIGHTERS               T5   \n",
       "3                         SOUTHSIDE ATHLETIC ASSOCIATION               T3   \n",
       "4               GENETIC RESEARCH INSTITUTE OF THE DESERT               T3   \n",
       "...                                                  ...              ...   \n",
       "34294              THE LIONS CLUB OF HONOLULU KAMEHAMEHA               T4   \n",
       "34295           INTERNATIONAL ASSOCIATION OF LIONS CLUBS               T4   \n",
       "34296                                PTA HAWAII CONGRESS               T3   \n",
       "34297  AMERICAN FEDERATION OF GOVERNMENT EMPLOYEES LO...               T5   \n",
       "34298                           WATERHOUSE CHARITABLE TR               T3   \n",
       "\n",
       "            AFFILIATION CLASSIFICATION      USE_CASE  ORGANIZATION  STATUS  \\\n",
       "0           Independent          C1000    ProductDev   Association       1   \n",
       "1           Independent          C2000  Preservation  Co-operative       1   \n",
       "2      CompanySponsored          C3000    ProductDev   Association       1   \n",
       "3      CompanySponsored          C2000  Preservation         Trust       1   \n",
       "4           Independent          C1000     Heathcare         Trust       1   \n",
       "...                 ...            ...           ...           ...     ...   \n",
       "34294       Independent          C1000    ProductDev   Association       1   \n",
       "34295  CompanySponsored          C3000    ProductDev   Association       1   \n",
       "34296  CompanySponsored          C2000  Preservation   Association       1   \n",
       "34297       Independent          C3000    ProductDev   Association       1   \n",
       "34298       Independent          C1000  Preservation  Co-operative       1   \n",
       "\n",
       "          INCOME_AMT SPECIAL_CONSIDERATIONS   ASK_AMT  IS_SUCCESSFUL  \n",
       "0                  0                      N      5000              1  \n",
       "1             1-9999                      N    108590              1  \n",
       "2                  0                      N      5000              0  \n",
       "3        10000-24999                      N      6692              1  \n",
       "4      100000-499999                      N    142590              1  \n",
       "...              ...                    ...       ...            ...  \n",
       "34294              0                      N      5000              0  \n",
       "34295              0                      N      5000              0  \n",
       "34296              0                      N      5000              0  \n",
       "34297              0                      N      5000              1  \n",
       "34298          1M-5M                      N  36500179              0  \n",
       "\n",
       "[34299 rows x 11 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drop the non-beneficial ID column, 'EIN'.\n",
    "application_df.drop(columns= ['EIN'], axis=1, inplace=True)\n",
    "application_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NAME                      19568\n",
       "APPLICATION_TYPE             17\n",
       "AFFILIATION                   6\n",
       "CLASSIFICATION               71\n",
       "USE_CASE                      5\n",
       "ORGANIZATION                  4\n",
       "STATUS                        2\n",
       "INCOME_AMT                    9\n",
       "SPECIAL_CONSIDERATIONS        2\n",
       "ASK_AMT                    8747\n",
       "IS_SUCCESSFUL                 2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Determine the number of unique values in each column.\n",
    "application_df.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "APPLICATION_TYPE\n",
       "T3     27037\n",
       "T4      1542\n",
       "T6      1216\n",
       "T5      1173\n",
       "T19     1065\n",
       "T8       737\n",
       "T7       725\n",
       "T10      528\n",
       "T9       156\n",
       "T13       66\n",
       "T12       27\n",
       "T2        16\n",
       "T25        3\n",
       "T14        3\n",
       "T29        2\n",
       "T15        2\n",
       "T17        1\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Look at APPLICATION_TYPE value counts for binning\n",
    "app_type_count = application_df['APPLICATION_TYPE'].value_counts()\n",
    "app_type_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "APPLICATION_TYPE\n",
       "T3       27037\n",
       "T4        1542\n",
       "T6        1216\n",
       "T5        1173\n",
       "T19       1065\n",
       "T8         737\n",
       "T7         725\n",
       "T10        528\n",
       "Other      276\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Choose a cutoff value and create a list of application types to be replaced\n",
    "# use the variable name `application_types_to_replace`\n",
    "application_types_to_replace = list(app_type_count[app_type_count < 500].index)\n",
    "\n",
    "\n",
    "# Replace in dataframe\n",
    "for app in application_types_to_replace:\n",
    "    application_df['APPLICATION_TYPE'] = application_df['APPLICATION_TYPE'].replace(app,\"Other\")\n",
    "\n",
    "# Check to make sure binning was successful\n",
    "application_df['APPLICATION_TYPE'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CLASSIFICATION\n",
       "C1000    17326\n",
       "C2000     6074\n",
       "C1200     4837\n",
       "C3000     1918\n",
       "C2100     1883\n",
       "C7000      777\n",
       "C1700      287\n",
       "C4000      194\n",
       "C5000      116\n",
       "C1270      114\n",
       "C2700      104\n",
       "C2800       95\n",
       "C7100       75\n",
       "C1300       58\n",
       "C1280       50\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Look at CLASSIFICATION value counts for binning\n",
    "classification_count = application_df['CLASSIFICATION'].value_counts()\n",
    "classification_count.head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CLASSIFICATION\n",
       "C1000    17326\n",
       "C2000     6074\n",
       "C1200     4837\n",
       "C3000     1918\n",
       "C2100     1883\n",
       "C7000      777\n",
       "C1700      287\n",
       "C4000      194\n",
       "C5000      116\n",
       "C1270      114\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# You may find it helpful to look at CLASSIFICATION value counts >1\n",
    "classification_greater1 = classification_count.loc[classification_count > 1]\n",
    "classification_greater1.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cutoff value changed. < 1000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CLASSIFICATION\n",
       "C1000    17326\n",
       "C2000     6074\n",
       "C1200     4837\n",
       "Other     2261\n",
       "C3000     1918\n",
       "C2100     1883\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Choose a cutoff value and create a list of classifications to be replaced\n",
    "# use the variable name `classifications_to_replace`\n",
    "classifications_to_replace = list(classification_count[classification_count < 1000].index)\n",
    "\n",
    "\n",
    "# Replace in dataframe\n",
    "for cls in classifications_to_replace:\n",
    "    application_df['CLASSIFICATION'] = application_df['CLASSIFICATION'].replace(cls,\"Other\")\n",
    "    \n",
    "# Check to make sure binning was successful\n",
    "application_df['CLASSIFICATION'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Column NAME reintroduced."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NAME\n",
       "PARENT BOOSTER USA INC                                                  1260\n",
       "TOPS CLUB INC                                                            765\n",
       "UNITED STATES BOWLING CONGRESS INC                                       700\n",
       "WASHINGTON STATE UNIVERSITY                                              492\n",
       "AMATEUR ATHLETIC UNION OF THE UNITED STATES INC                          408\n",
       "                                                                        ... \n",
       "ST LOUIS SLAM WOMENS FOOTBALL                                              1\n",
       "AIESEC ALUMNI IBEROAMERICA CORP                                            1\n",
       "WEALLBLEEDRED ORG INC                                                      1\n",
       "AMERICAN SOCIETY FOR STANDARDS IN MEDIUMSHIP & PSYCHICAL INVESTIGATI       1\n",
       "WATERHOUSE CHARITABLE TR                                                   1\n",
       "Name: count, Length: 19568, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Look at NAME value counts for binning\n",
    "name_counts = application_df['NAME'].value_counts()\n",
    "name_counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NAME\n",
       "Other                                                                 28123\n",
       "PARENT BOOSTER USA INC                                                 1260\n",
       "TOPS CLUB INC                                                           765\n",
       "UNITED STATES BOWLING CONGRESS INC                                      700\n",
       "WASHINGTON STATE UNIVERSITY                                             492\n",
       "AMATEUR ATHLETIC UNION OF THE UNITED STATES INC                         408\n",
       "PTA TEXAS CONGRESS                                                      368\n",
       "SOROPTIMIST INTERNATIONAL OF THE AMERICAS INC                           331\n",
       "ALPHA PHI SIGMA                                                         313\n",
       "TOASTMASTERS INTERNATIONAL                                              293\n",
       "MOST WORSHIPFUL STRINGER FREE AND ACCEPTED MASONS                       287\n",
       "LITTLE LEAGUE BASEBALL INC                                              277\n",
       "INTERNATIONAL ASSOCIATION OF LIONS CLUBS                                266\n",
       "MOMS CLUB                                                               210\n",
       "INTERNATIONAL ASSOCIATION OF SHEET METAL AIR RAIL & TRANSPORTATION      206\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Choose a cutoff value and create a list of names to be replaced\n",
    "names_to_replace = list(name_counts[name_counts < 200].index)\n",
    "\n",
    "# Replace in dataframe\n",
    "for name in names_to_replace:\n",
    "    application_df['NAME'] = application_df['NAME'].replace(\n",
    "        name, \"Other\")\n",
    "\n",
    "# Check to make sure binning was successful\n",
    "application_df['NAME'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>STATUS</th>\n",
       "      <th>ASK_AMT</th>\n",
       "      <th>IS_SUCCESSFUL</th>\n",
       "      <th>NAME_ALPHA PHI SIGMA</th>\n",
       "      <th>NAME_AMATEUR ATHLETIC UNION OF THE UNITED STATES INC</th>\n",
       "      <th>NAME_INTERNATIONAL ASSOCIATION OF LIONS CLUBS</th>\n",
       "      <th>NAME_INTERNATIONAL ASSOCIATION OF SHEET METAL AIR RAIL &amp; TRANSPORTATION</th>\n",
       "      <th>NAME_LITTLE LEAGUE BASEBALL INC</th>\n",
       "      <th>NAME_MOMS CLUB</th>\n",
       "      <th>NAME_MOST WORSHIPFUL STRINGER FREE AND ACCEPTED MASONS</th>\n",
       "      <th>...</th>\n",
       "      <th>INCOME_AMT_1-9999</th>\n",
       "      <th>INCOME_AMT_10000-24999</th>\n",
       "      <th>INCOME_AMT_100000-499999</th>\n",
       "      <th>INCOME_AMT_10M-50M</th>\n",
       "      <th>INCOME_AMT_1M-5M</th>\n",
       "      <th>INCOME_AMT_25000-99999</th>\n",
       "      <th>INCOME_AMT_50M+</th>\n",
       "      <th>INCOME_AMT_5M-10M</th>\n",
       "      <th>SPECIAL_CONSIDERATIONS_N</th>\n",
       "      <th>SPECIAL_CONSIDERATIONS_Y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>5000</td>\n",
       "      <td>1</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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>108590</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>5000</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>...</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>6692</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>142590</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 59 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   STATUS  ASK_AMT  IS_SUCCESSFUL  NAME_ALPHA PHI SIGMA  \\\n",
       "0       1     5000              1                     0   \n",
       "1       1   108590              1                     0   \n",
       "2       1     5000              0                     0   \n",
       "3       1     6692              1                     0   \n",
       "4       1   142590              1                     0   \n",
       "\n",
       "   NAME_AMATEUR ATHLETIC UNION OF THE UNITED STATES INC  \\\n",
       "0                                                  0      \n",
       "1                                                  0      \n",
       "2                                                  0      \n",
       "3                                                  0      \n",
       "4                                                  0      \n",
       "\n",
       "   NAME_INTERNATIONAL ASSOCIATION OF LIONS CLUBS  \\\n",
       "0                                              0   \n",
       "1                                              0   \n",
       "2                                              0   \n",
       "3                                              0   \n",
       "4                                              0   \n",
       "\n",
       "   NAME_INTERNATIONAL ASSOCIATION OF SHEET METAL AIR RAIL & TRANSPORTATION  \\\n",
       "0                                                  0                         \n",
       "1                                                  0                         \n",
       "2                                                  0                         \n",
       "3                                                  0                         \n",
       "4                                                  0                         \n",
       "\n",
       "   NAME_LITTLE LEAGUE BASEBALL INC  NAME_MOMS CLUB  \\\n",
       "0                                0               0   \n",
       "1                                0               0   \n",
       "2                                0               0   \n",
       "3                                0               0   \n",
       "4                                0               0   \n",
       "\n",
       "   NAME_MOST WORSHIPFUL STRINGER FREE AND ACCEPTED MASONS  ...  \\\n",
       "0                                                  0       ...   \n",
       "1                                                  0       ...   \n",
       "2                                                  0       ...   \n",
       "3                                                  0       ...   \n",
       "4                                                  0       ...   \n",
       "\n",
       "   INCOME_AMT_1-9999  INCOME_AMT_10000-24999  INCOME_AMT_100000-499999  \\\n",
       "0                  0                       0                         0   \n",
       "1                  1                       0                         0   \n",
       "2                  0                       0                         0   \n",
       "3                  0                       1                         0   \n",
       "4                  0                       0                         1   \n",
       "\n",
       "   INCOME_AMT_10M-50M  INCOME_AMT_1M-5M  INCOME_AMT_25000-99999  \\\n",
       "0                   0                 0                       0   \n",
       "1                   0                 0                       0   \n",
       "2                   0                 0                       0   \n",
       "3                   0                 0                       0   \n",
       "4                   0                 0                       0   \n",
       "\n",
       "   INCOME_AMT_50M+  INCOME_AMT_5M-10M  SPECIAL_CONSIDERATIONS_N  \\\n",
       "0                0                  0                         1   \n",
       "1                0                  0                         1   \n",
       "2                0                  0                         1   \n",
       "3                0                  0                         1   \n",
       "4                0                  0                         1   \n",
       "\n",
       "   SPECIAL_CONSIDERATIONS_Y  \n",
       "0                         0  \n",
       "1                         0  \n",
       "2                         0  \n",
       "3                         0  \n",
       "4                         0  \n",
       "\n",
       "[5 rows x 59 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert categorical data to numeric with `pd.get_dummies`\n",
    "numeric_data = pd.get_dummies(application_df)\n",
    "numeric_data = numeric_data.astype(int)\n",
    "numeric_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split our preprocessed data into our features and target arrays\n",
    "y = numeric_data ['IS_SUCCESSFUL']\n",
    "X = numeric_data.drop(columns='IS_SUCCESSFUL', axis=1)\n",
    "\n",
    "# Split the preprocessed data into a training and testing dataset\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,y, random_state=78)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a StandardScaler instances\n",
    "scaler = StandardScaler()\n",
    "\n",
    "# Fit the StandardScaler\n",
    "X_scaler = scaler.fit(X_train)\n",
    "\n",
    "# Scale the data\n",
    "X_train_scaled = X_scaler.transform(X_train)\n",
    "X_test_scaled = X_scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compile, Train and Evaluate the Model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Optimization.\n",
    "Adding a thir layer.\n",
    "Activation changed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def create_model(hp, input_features):\n",
    "    nn_model3 = tf.keras.Sequential([\n",
    "        # Specify the input shape in the Input layer\n",
    "        tf.keras.layers.Input(shape=(input_features,)),\n",
    "    ])\n",
    "\n",
    "    # Allow Keras Tuner to decide which activation function to use in hidden layers\n",
    "    activation_choice = hp.Choice('activation', values=['relu', 'tanh', 'sigmoid'])\n",
    "    \n",
    "    # Allow Keras Tuner to decide the number of neurons in the first layer\n",
    "    nn_model3.add(tf.keras.layers.Dense(\n",
    "        units=hp.Int('first_units', min_value=1, max_value=90, step=5),\n",
    "        activation=activation_choice))\n",
    "\n",
    "    # Allow Keras Tuner to decide the number of hidden layers and neurons in hidden layers\n",
    "    for i in range(hp.Int('num_layers', 1, 5)):\n",
    "        nn_model3.add(tf.keras.layers.Dense(\n",
    "            units=hp.Int(f'units_layer_{i}', min_value=1, max_value=30, step=5),\n",
    "            activation=activation_choice))\n",
    "    \n",
    "    nn_model3.add(tf.keras.layers.Dense(units=1, activation=\"sigmoid\"))\n",
    "\n",
    "    # Allow Keras Tuner to decide the learning rate\n",
    "    learning_rate = hp.Choice('learning_rate', values=[1e-2, 1e-3, 1e-4])\n",
    "    \n",
    "    # Compile the model\n",
    "    nn_model3.compile(\n",
    "        optimizer=tf.keras.optimizers.Adam(learning_rate=learning_rate),\n",
    "        loss=\"binary_crossentropy\",\n",
    "        metrics=[\"accuracy\"])\n",
    "    \n",
    "    return nn_model3\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Number of epoch increased."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "input_features = X_train_scaled.shape[1]\n",
    "\n",
    "def build_model(hp):\n",
    "    return create_model(hp, input_features)\n",
    "\n",
    "# Initialize the tuner\n",
    "tuner = Hyperband(\n",
    "    build_model,\n",
    "    objective='val_accuracy',\n",
    "    max_epochs=25,\n",
    "    factor=3,\n",
    "    directory='keras_tuner_dir',\n",
    "    project_name='keras_tuner_demo'\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 30 Complete [00h 00m 09s]\n",
      "val_accuracy: 0.7473275065422058\n",
      "\n",
      "Best val_accuracy So Far: 0.7484936714172363\n",
      "Total elapsed time: 00h 01m 48s\n"
     ]
    }
   ],
   "source": [
    "tuner.search(X_train_scaled, y_train, epochs=25, validation_split=0.2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 404us/step - accuracy: 0.7211 - loss: 0.5558\n",
      "Epoch 2/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 415us/step - accuracy: 0.7478 - loss: 0.5144\n",
      "Epoch 3/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 450us/step - accuracy: 0.7530 - loss: 0.5052\n",
      "Epoch 4/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 437us/step - accuracy: 0.7537 - loss: 0.5024\n",
      "Epoch 5/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 402us/step - accuracy: 0.7585 - loss: 0.4974\n",
      "Epoch 6/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 400us/step - accuracy: 0.7564 - loss: 0.4976\n",
      "Epoch 7/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 400us/step - accuracy: 0.7531 - loss: 0.5018\n",
      "Epoch 8/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 401us/step - accuracy: 0.7576 - loss: 0.5000\n",
      "Epoch 9/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 458us/step - accuracy: 0.7580 - loss: 0.4976\n",
      "Epoch 10/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 426us/step - accuracy: 0.7615 - loss: 0.4955\n",
      "Epoch 11/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 398us/step - accuracy: 0.7553 - loss: 0.4993\n",
      "Epoch 12/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 399us/step - accuracy: 0.7580 - loss: 0.4968\n",
      "Epoch 13/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 397us/step - accuracy: 0.7554 - loss: 0.4984\n",
      "Epoch 14/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 402us/step - accuracy: 0.7543 - loss: 0.4983\n",
      "Epoch 15/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 465us/step - accuracy: 0.7627 - loss: 0.4938\n",
      "Epoch 16/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 422us/step - accuracy: 0.7595 - loss: 0.4933\n",
      "Epoch 17/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 404us/step - accuracy: 0.7635 - loss: 0.4883\n",
      "Epoch 18/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 400us/step - accuracy: 0.7579 - loss: 0.4950\n",
      "Epoch 19/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 409us/step - accuracy: 0.7600 - loss: 0.4894\n",
      "Epoch 20/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 399us/step - accuracy: 0.7570 - loss: 0.4946\n",
      "Epoch 21/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 429us/step - accuracy: 0.7672 - loss: 0.4863\n",
      "Epoch 22/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 399us/step - accuracy: 0.7597 - loss: 0.4902\n",
      "Epoch 23/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 404us/step - accuracy: 0.7621 - loss: 0.4903\n",
      "Epoch 24/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 404us/step - accuracy: 0.7590 - loss: 0.4924\n",
      "Epoch 25/25\n",
      "\u001b[1m804/804\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 404us/step - accuracy: 0.7605 - loss: 0.4886\n"
     ]
    }
   ],
   "source": [
    "# Get the optimal hyperparameters\n",
    "best_hps = tuner.get_best_hyperparameters(num_trials=1)[0]\n",
    "\n",
    "# Build the model with the best hyperparameters\n",
    "model = tuner.hypermodel.build(best_hps)\n",
    "\n",
    "# Train it on the data for 25 epochs\n",
    "fit_model = model.fit(X_train_scaled, y_train, epochs=25)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "268/268 - 0s - 548us/step - accuracy: 0.7482 - loss: 0.5079\n",
      "Loss: 0.5079114437103271, Accuracy: 0.7482215762138367\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the model using the test data\n",
    "model_loss, model_accuracy = model.evaluate(X_test_scaled, y_test, verbose=2)\n",
    "\n",
    "# Print the loss and accuracy\n",
    "print(f\"Loss: {model_loss}, Accuracy: {model_accuracy}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A loss value of 50 indicates that the model works better than the 2 previous models.\n",
    "\n",
    "The accuracy percent shows that 74% of the model's predicted values align with the true values in the original dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "#nn_model3.save('Models/AlphabetSoupCharityOptimization.h5')\n",
    "#Warning: this file format is considered legacy. We recommend using instead the native Keras format,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the model using the native Keras format\n",
    "\n",
    "model.save('Models/AlphabetSoupCharityOptimization2.keras')\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A loss value of 51 indicates that the model works better, but can be still further optimized.\n",
    "\n",
    "The accuracy percent shows that 74% of the model's predicted values align with the true values in the original dataset."
   ]
  }
 ],
 "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": "3.11.5"
  },
  "vscode": {
   "interpreter": {
    "hash": "31f2aee4e71d21fbe5cf8b01ff0e069b9275f58929596ceb00d14d90e3e16cd6"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}