DefineAndSolveMLProblem.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lab 8: Define and Solve an ML Problem of Your Choosing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this lab assignment, you will follow the machine learning life cycle and implement a model to solve a machine learning problem of your choosing. You will select a data set and choose a predictive problem that the data set supports.  You will then inspect the data with your problem in mind and begin to formulate a  project plan. You will then implement the machine learning project plan. \n",
    "\n",
    "You will complete the following tasks:\n",
    "\n",
    "1. Build Your DataFrame\n",
    "2. Define Your ML Problem\n",
    "3. Perform exploratory data analysis to understand your data.\n",
    "4. Define Your Project Plan\n",
    "5. Implement Your Project Plan:\n",
    "    * Prepare your data for your model.\n",
    "    * Fit your model to the training data and evaluate your model.\n",
    "    * Improve your model's performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 1: Build Your DataFrame\n",
    "\n",
    "You will have the option to choose one of four data sets that you have worked with in this program:\n",
    "\n",
    "* The \"census\" data set that contains Census information from 1994: `censusData.csv`\n",
    "* Airbnb NYC \"listings\" data set: `airbnbListingsData.csv`\n",
    "* World Happiness Report (WHR) data set: `WHR2018Chapter2OnlineData.csv`\n",
    "* Book Review data set: `bookReviewsData.csv`\n",
    "\n",
    "Note that these are variations of the data sets that you have worked with in this program. For example, some do not include some of the preprocessing necessary for specific models. \n",
    "\n",
    "#### Load a Data Set and Save it as a Pandas DataFrame\n",
    "\n",
    "The code cell below contains filenames (path + filename) for each of the four data sets available to you.\n",
    "\n",
    "<b>Task:</b> In the code cell below, use the same method you have been using to load the data using `pd.read_csv()` and save it to DataFrame `df`. \n",
    "\n",
    "You can load each file as a new DataFrame to inspect the data before choosing your data set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<bound method NDFrame.head of                                                     name  \\\n",
       "0                                  Skylit Midtown Castle   \n",
       "1      Whole flr w/private bdrm, bath & kitchen(pls r...   \n",
       "2               Spacious Brooklyn Duplex, Patio + Garden   \n",
       "3                       Large Furnished Room Near B'way　   \n",
       "4                     Cozy Clean Guest Room - Family Apt   \n",
       "...                                                  ...   \n",
       "28017                            Astoria Luxury suite 2A   \n",
       "28018  Newly renovated suite in the heart of Williams...   \n",
       "28019      Perfect Room to Stay in Brooklyn! Near Metro!   \n",
       "28020       New Beautiful Modern One Bedroom in Brooklyn   \n",
       "28021    Large, modern, private 1 bedroom in beach condo   \n",
       "\n",
       "                                             description  \\\n",
       "0      Beautiful, spacious skylit studio in the heart...   \n",
       "1      Enjoy 500 s.f. top floor in 1899 brownstone, w...   \n",
       "2      We welcome you to stay in our lovely 2 br dupl...   \n",
       "3      Please don’t expect the luxury here just a bas...   \n",
       "4      Our best guests are seeking a safe, clean, spa...   \n",
       "...                                                  ...   \n",
       "28017  THIS LOVELY HOME IS THE SPACIOUS SUITE WITH PR...   \n",
       "28018  Just fully renovated from head to toe. On the ...   \n",
       "28019  Amazing and comfortable space in Brooklyn, sam...   \n",
       "28020  This stylish place to stay is perfect for a gr...   \n",
       "28021  Private bedroom on its own floor with very lar...   \n",
       "\n",
       "                                   neighborhood_overview    host_name  \\\n",
       "0      Centrally located in the heart of Manhattan ju...     Jennifer   \n",
       "1      Just the right mix of urban center and local n...  LisaRoxanne   \n",
       "2                                                    NaN      Rebecca   \n",
       "3        Theater district, many restaurants around here.     Shunichi   \n",
       "4      Our neighborhood is full of restaurants and ca...    MaryEllen   \n",
       "...                                                  ...          ...   \n",
       "28017                                                NaN        Vicky   \n",
       "28018                                                NaN       Samuel   \n",
       "28019                                                NaN       Carlos   \n",
       "28020                                                NaN        Lexia   \n",
       "28021  Beach, surf shop, stop and shop, Dunkin’ Donut...      Justine   \n",
       "\n",
       "                           host_location  \\\n",
       "0      New York, New York, United States   \n",
       "1      New York, New York, United States   \n",
       "2      Brooklyn, New York, United States   \n",
       "3      New York, New York, United States   \n",
       "4      New York, New York, United States   \n",
       "...                                  ...   \n",
       "28017    Queens, New York, United States   \n",
       "28018  New York, New York, United States   \n",
       "28019                                 US   \n",
       "28020  New York, New York, United States   \n",
       "28021                                 US   \n",
       "\n",
       "                                              host_about  host_response_rate  \\\n",
       "0      A New Yorker since 2000! My passion is creatin...                0.80   \n",
       "1      Laid-back Native New Yorker (formerly bi-coast...                0.09   \n",
       "2      Rebecca is an artist/designer, and Henoch is i...                1.00   \n",
       "3      I used to work for a financial industry but no...                1.00   \n",
       "4      Welcome to family life with my oldest two away...                 NaN   \n",
       "...                                                  ...                 ...   \n",
       "28017                                                NaN                1.00   \n",
       "28018  Hello, my name is Sam. I am a real estate prof...                0.91   \n",
       "28019                                                NaN                0.99   \n",
       "28020  I am a graphic designer, swell chaser and duri...                0.90   \n",
       "28021                                                NaN                 NaN   \n",
       "\n",
       "       host_acceptance_rate  host_is_superhost  host_listings_count  ...  \\\n",
       "0                      0.17               True                  8.0  ...   \n",
       "1                      0.69               True                  1.0  ...   \n",
       "2                      0.25               True                  1.0  ...   \n",
       "3                      1.00               True                  1.0  ...   \n",
       "4                       NaN               True                  1.0  ...   \n",
       "...                     ...                ...                  ...  ...   \n",
       "28017                  1.00               True                  8.0  ...   \n",
       "28018                  0.89               True                  0.0  ...   \n",
       "28019                  0.99               True                  6.0  ...   \n",
       "28020                  1.00               True                  3.0  ...   \n",
       "28021                   NaN               True                  0.0  ...   \n",
       "\n",
       "       review_scores_communication  review_scores_location  \\\n",
       "0                             4.79                    4.86   \n",
       "1                             4.80                    4.71   \n",
       "2                             5.00                    4.50   \n",
       "3                             4.42                    4.87   \n",
       "4                             4.95                    4.94   \n",
       "...                            ...                     ...   \n",
       "28017                         5.00                    3.00   \n",
       "28018                         5.00                    5.00   \n",
       "28019                         5.00                    5.00   \n",
       "28020                         5.00                    5.00   \n",
       "28021                         5.00                    5.00   \n",
       "\n",
       "       review_scores_value instant_bookable calculated_host_listings_count  \\\n",
       "0                     4.41            False                              3   \n",
       "1                     4.64            False                              1   \n",
       "2                     5.00            False                              1   \n",
       "3                     4.36            False                              1   \n",
       "4                     4.92            False                              1   \n",
       "...                    ...              ...                            ...   \n",
       "28017                 1.00             True                              8   \n",
       "28018                 5.00            False                              1   \n",
       "28019                 2.00             True                              7   \n",
       "28020                 5.00            False                              3   \n",
       "28021                 5.00             True                              1   \n",
       "\n",
       "       calculated_host_listings_count_entire_homes  \\\n",
       "0                                                3   \n",
       "1                                                1   \n",
       "2                                                1   \n",
       "3                                                0   \n",
       "4                                                0   \n",
       "...                                            ...   \n",
       "28017                                            0   \n",
       "28018                                            1   \n",
       "28019                                            0   \n",
       "28020                                            3   \n",
       "28021                                            0   \n",
       "\n",
       "       calculated_host_listings_count_private_rooms  \\\n",
       "0                                                 0   \n",
       "1                                                 0   \n",
       "2                                                 0   \n",
       "3                                                 1   \n",
       "4                                                 1   \n",
       "...                                             ...   \n",
       "28017                                             8   \n",
       "28018                                             0   \n",
       "28019                                             7   \n",
       "28020                                             0   \n",
       "28021                                             1   \n",
       "\n",
       "       calculated_host_listings_count_shared_rooms  reviews_per_month  \\\n",
       "0                                                0               0.33   \n",
       "1                                                0               4.86   \n",
       "2                                                0               0.02   \n",
       "3                                                0               3.68   \n",
       "4                                                0               0.87   \n",
       "...                                            ...                ...   \n",
       "28017                                            0               1.00   \n",
       "28018                                            0               2.00   \n",
       "28019                                            0               1.00   \n",
       "28020                                            0               1.00   \n",
       "28021                                            0               1.00   \n",
       "\n",
       "      n_host_verifications  \n",
       "0                        9  \n",
       "1                        6  \n",
       "2                        3  \n",
       "3                        4  \n",
       "4                        7  \n",
       "...                    ...  \n",
       "28017                    2  \n",
       "28018                    5  \n",
       "28019                    2  \n",
       "28020                    7  \n",
       "28021                    1  \n",
       "\n",
       "[28022 rows x 50 columns]>"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# File names of the four data sets\n",
    "adultDataSet_filename = os.path.join(os.getcwd(), \"data\", \"censusData.csv\")\n",
    "airbnbDataSet_filename = os.path.join(os.getcwd(), \"data\", \"airbnbListingsData.csv\")\n",
    "WHRDataSet_filename = os.path.join(os.getcwd(), \"data\", \"WHR2018Chapter2OnlineData.csv\")\n",
    "bookReviewDataSet_filename = os.path.join(os.getcwd(), \"data\", \"bookReviewsData.csv\")\n",
    "\n",
    "\n",
    "df = pd.read_csv(airbnbDataSet_filename)# YOUR CODE HERE\n",
    "\n",
    "df.head"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 2: Define Your ML Problem\n",
    "\n",
    "Next you will formulate your ML Problem. In the markdown cell below, answer the following questions:\n",
    "\n",
    "1. List the data set you have chosen.\n",
    "2. What will you be predicting? What is the label?\n",
    "3. Is this a supervised or unsupervised learning problem? Is this a clustering, classification or regression problem? Is it a binary classificaiton or multi-class classifiction problem?\n",
    "4. What are your features? (note: this list may change after your explore your data)\n",
    "5. Explain why this is an important problem. In other words, how would a company create value with a model that predicts this label?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset Chosen\n",
    "**Airbnb Listings Data Set**\n",
    "\n",
    "### Prediction Task\n",
    "**Prediction**: Whether an Airbnb listing has availability.\n",
    "**Label**: `has_availability`\n",
    "\n",
    "### Type of Learning Problem\n",
    "**Learning Type**: Supervised learning  \n",
    "**Problem Type**: Classification  \n",
    "**Classification Type**: Binary classification (True or False)\n",
    "\n",
    "### Features\n",
    "Initially considered features:\n",
    "- `host_response_rate`\n",
    "- `host_acceptance_rate`\n",
    "- `host_is_superhost`\n",
    "- `host_listings_count`\n",
    "- `review_scores_communication`\n",
    "- `review_scores_location`\n",
    "- `review_scores_value`\n",
    "- `instant_bookable`\n",
    "- `calculated_host_listings_count`\n",
    "- `calculated_host_listings_count_entire_homes`\n",
    "- `calculated_host_listings_count_private_rooms`\n",
    "- `calculated_host_listings_count_shared_rooms`\n",
    "- `reviews_per_month`\n",
    "- `n_host_verifications`\n",
    "\n",
    "Revised Complete Listed features:\n",
    "\n",
    "'host_response_rate', 'host_acceptance_rate', 'host_listings_count', 'host_total_listings_count', 'host_has_profile_pic', 'host_identity_verified', 'accommodates', 'bathrooms', 'bedrooms', 'beds', 'price', 'minimum_nights', 'maximum_nights', 'minimum_minimum_nights', 'maximum_minimum_nights', 'minimum_maximum_nights', 'maximum_maximum_nights', 'minimum_nights_avg_ntm', 'maximum_nights_avg_ntm', 'availability_30', 'availability_60', 'availability_90', 'availability_365', 'number_of_reviews', 'number_of_reviews_ltm', 'number_of_reviews_l30d', 'review_scores_rating', 'review_scores_cleanliness', 'review_scores_checkin', 'review_scores_communication', 'review_scores_location', 'review_scores_value', 'calculated_host_listings_count', 'calculated_host_listings_count_entire_homes', 'calculated_host_listings_count_private_rooms', 'calculated_host_listings_count_shared_rooms', 'reviews_per_month', 'n_host_verifications'\n",
    "\n",
    "### Importance of the Problem\n",
    "**Value Creation**:\n",
    "- **Improved Guest Experience**: Ensures guests see available listings, reducing frustration and enhancing satisfaction.\n",
    "- **Optimized Host Management**: Helps hosts manage listings better, adjusting strategies to maximize occupancy.\n",
    "- **Platform Efficiency**: Optimizes search results, leading to higher conversion rates and better resource utilization.\n",
    "- **Revenue Maximization**: Higher occupancy rates mean more revenue for both the platform and hosts.\n",
    "- **Strategic Insights**: Provides valuable insights into factors affecting availability, helping hosts improve listing performance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 3: Understand Your Data\n",
    "\n",
    "The next step is to perform exploratory data analysis. Inspect and analyze your data set with your machine learning problem in mind. Consider the following as you inspect your data:\n",
    "\n",
    "1. What data preparation techniques would you like to use? These data preparation techniques may include:\n",
    "\n",
    "    * addressing missingness, such as replacing missing values with means\n",
    "    * finding and replacing outliers\n",
    "    * renaming features and labels\n",
    "    * finding and replacing outliers\n",
    "    * performing feature engineering techniques such as one-hot encoding on categorical features\n",
    "    * selecting appropriate features and removing irrelevant features\n",
    "    * performing specific data cleaning and preprocessing techniques for an NLP problem\n",
    "    * addressing class imbalance in your data sample to promote fair AI\n",
    "    \n",
    "\n",
    "2. What machine learning model (or models) you would like to use that is suitable for your predictive problem and data?\n",
    "    * Are there other data preparation techniques that you will need to apply to build a balanced modeling data set for your problem and model? For example, will you need to scale your data?\n",
    " \n",
    " \n",
    "3. How will you evaluate and improve the model's performance?\n",
    "    * Are there specific evaluation metrics and methods that are appropriate for your model?\n",
    "    \n",
    "\n",
    "Think of the different techniques you have used to inspect and analyze your data in this course. These include using Pandas to apply data filters, using the Pandas `describe()` method to get insight into key statistics for each column, using the Pandas `dtypes` property to inspect the data type of each column, and using Matplotlib and Seaborn to detect outliers and visualize relationships between features and labels. If you are working on a classification problem, use techniques you have learned to determine if there is class imbalance.\n",
    "\n",
    "<b>Task</b>: Use the techniques you have learned in this course to inspect and analyze your data. You can import additional packages that you have used in this course that you will need to perform this task.\n",
    "\n",
    "<b>Note</b>: You can add code cells if needed by going to the <b>Insert</b> menu and clicking on <b>Insert Cell Below</b> in the drop-drown menu."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 4: Define Your Project Plan\n",
    "\n",
    "Now that you understand your data, in the markdown cell below, define your plan to implement the remaining phases of the machine learning life cycle (data preparation, modeling, evaluation) to solve your ML problem. Answer the following questions:\n",
    "\n",
    "* Do you have a new feature list? If so, what are the features that you chose to keep and remove after inspecting the data? \n",
    "* Explain different data preparation techniques that you will use to prepare your data for modeling.\n",
    "* What is your model (or models)?\n",
    "* Describe your plan to train your model, analyze its performance and then improve the model. That is, describe your model building, validation and selection plan to produce a model that generalizes well to new data. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plan for Implementing Remaining Phases of the Machine Learning Life Cycle\n",
    "\n",
    "#### Feature List\n",
    "After inspecting the data, I decided to keep the following features:\n",
    "- Numerical: `host_response_rate`, `host_acceptance_rate`, `host_listings_count`, `host_total_listings_count`, `accommodates`, `bathrooms`, `bedrooms`, `beds`, `price`, `minimum_nights`, `maximum_nights`, `availability_30`, `availability_60`, `availability_90`, `availability_365`, `number_of_reviews`, `review_scores_rating`, `review_scores_cleanliness`, `review_scores_checkin`, `review_scores_communication`, `review_scores_location`, `review_scores_value`, `calculated_host_listings_count`, `reviews_per_month`\n",
    "- One-hot encoded categorical: `host_is_superhost`, `instant_bookable`\n",
    "\n",
    "Removed features include columns with high missing values, irrelevant columns, and those dropped due to preprocessing.\n",
    "\n",
    "#### Data Preparation Techniques\n",
    "1. **Handling Missing Values**: Fill missing numeric values with the mean and categorical values with the mode.\n",
    "2. **Convert Percentages**: Convert percentage values to float.\n",
    "3. **Binary Conversion**: Convert boolean columns to integers.\n",
    "4. **Standardization**: Scale numerical features using StandardScaler.\n",
    "5. **Remove Constant Features**: Use `VarianceThreshold` to remove features with zero variance.\n",
    "6. **One-hot Encoding**: One-hot encode binary categorical variables.\n",
    "\n",
    "#### Model(s)\n",
    "1. **Logistic Regression**: For initial model building and evaluation.\n",
    "2. **Random Forest Classifier**: For improved performance with hyperparameter tuning.\n",
    "\n",
    "#### Plan for Model Training, Analysis, and Improvement\n",
    "1. **Model Building**:\n",
    "   - **Initial Model**: Train a Logistic Regression model to set a baseline.\n",
    "   - **Feature Selection**: Use `SelectKBest` to identify the top 10 features.\n",
    "2. **Model Validation**:\n",
    "   - **Cross-Validation**: Use 5-fold cross-validation to evaluate model performance.\n",
    "   - **Hyperparameter Tuning**: Use GridSearchCV to find the best hyperparameters for the Random Forest model.\n",
    "3. **Model Analysis**:\n",
    "   - **Performance Metrics**: Evaluate using accuracy, precision, recall, F1 score, and ROC AUC score.\n",
    "4. **Model Improvement**:\n",
    "   - **Feature Engineering**: Explore additional feature creation or transformation if needed.\n",
    "   - **Ensemble Methods**: Consider combining multiple models if individual model performance is insufficient.\n",
    "5. **Model Selection**:\n",
    "   - **Best Model Selection**: Select the model with the best cross-validation performance.\n",
    "   - **Final Evaluation**: Evaluate the final model on a separate test set to ensure generalizability. \n",
    "\n",
    "This plan will guide the preparation, modeling, and iterative improvement to build a robust model that generalizes well to new data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Part 5: Implement Your Project Plan\n",
    "\n",
    "<b>Task:</b> In the code cell below, import additional packages that you have used in this course that you will need to implement your project plan."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, roc_auc_score, roc_curve, precision_recall_curve\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.feature_selection import SelectKBest, f_classif, VarianceThreshold\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import roc_curve, auc, precision_recall_curve\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix, roc_curve, auc, precision_recall_curve\n",
    "import pickle"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Task:</b> Use the rest of this notebook to carry out your project plan. \n",
    "\n",
    "You will:\n",
    "\n",
    "1. Prepare your data for your model.\n",
    "2. Fit your model to the training data and evaluate your model.\n",
    "3. Improve your model's performance by performing model selection and/or feature selection techniques to find best model for your problem.\n",
    "\n",
    "Add code cells below and populate the notebook with commentary, code, analyses, results, and figures as you see fit. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Airbnb Listings Data Set:\n",
      "   host_response_rate  host_acceptance_rate  host_is_superhost  \\\n",
      "0                0.80                  0.17               True   \n",
      "1                0.09                  0.69               True   \n",
      "2                1.00                  0.25               True   \n",
      "3                1.00                  1.00               True   \n",
      "4                 NaN                   NaN               True   \n",
      "\n",
      "   host_listings_count  host_total_listings_count  host_has_profile_pic  \\\n",
      "0                  8.0                        8.0                  True   \n",
      "1                  1.0                        1.0                  True   \n",
      "2                  1.0                        1.0                  True   \n",
      "3                  1.0                        1.0                  True   \n",
      "4                  1.0                        1.0                  True   \n",
      "\n",
      "   host_identity_verified  accommodates  bathrooms  bedrooms  ...  \\\n",
      "0                    True             1        1.0       NaN  ...   \n",
      "1                    True             3        1.0       1.0  ...   \n",
      "2                    True             4        1.5       2.0  ...   \n",
      "3                    True             2        1.0       1.0  ...   \n",
      "4                    True             1        1.0       1.0  ...   \n",
      "\n",
      "   review_scores_communication  review_scores_location  review_scores_value  \\\n",
      "0                         4.79                    4.86                 4.41   \n",
      "1                         4.80                    4.71                 4.64   \n",
      "2                         5.00                    4.50                 5.00   \n",
      "3                         4.42                    4.87                 4.36   \n",
      "4                         4.95                    4.94                 4.92   \n",
      "\n",
      "   instant_bookable  calculated_host_listings_count  \\\n",
      "0             False                               3   \n",
      "1             False                               1   \n",
      "2             False                               1   \n",
      "3             False                               1   \n",
      "4             False                               1   \n",
      "\n",
      "   calculated_host_listings_count_entire_homes  \\\n",
      "0                                            3   \n",
      "1                                            1   \n",
      "2                                            1   \n",
      "3                                            0   \n",
      "4                                            0   \n",
      "\n",
      "   calculated_host_listings_count_private_rooms  \\\n",
      "0                                             0   \n",
      "1                                             0   \n",
      "2                                             0   \n",
      "3                                             1   \n",
      "4                                             1   \n",
      "\n",
      "   calculated_host_listings_count_shared_rooms  reviews_per_month  \\\n",
      "0                                            0               0.33   \n",
      "1                                            0               4.86   \n",
      "2                                            0               0.02   \n",
      "3                                            0               3.68   \n",
      "4                                            0               0.87   \n",
      "\n",
      "   n_host_verifications  \n",
      "0                     9  \n",
      "1                     6  \n",
      "2                     3  \n",
      "3                     4  \n",
      "4                     7  \n",
      "\n",
      "[5 rows x 41 columns]\n",
      "host_response_rate                              11843\n",
      "host_acceptance_rate                            11113\n",
      "host_is_superhost                                   0\n",
      "host_listings_count                                 0\n",
      "host_total_listings_count                           0\n",
      "host_has_profile_pic                                0\n",
      "host_identity_verified                              0\n",
      "accommodates                                        0\n",
      "bathrooms                                           0\n",
      "bedrooms                                         2918\n",
      "beds                                             1354\n",
      "price                                               0\n",
      "minimum_nights                                      0\n",
      "maximum_nights                                      0\n",
      "minimum_minimum_nights                              0\n",
      "maximum_minimum_nights                              0\n",
      "minimum_maximum_nights                              0\n",
      "maximum_maximum_nights                              0\n",
      "minimum_nights_avg_ntm                              0\n",
      "maximum_nights_avg_ntm                              0\n",
      "has_availability                                    0\n",
      "availability_30                                     0\n",
      "availability_60                                     0\n",
      "availability_90                                     0\n",
      "availability_365                                    0\n",
      "number_of_reviews                                   0\n",
      "number_of_reviews_ltm                               0\n",
      "number_of_reviews_l30d                              0\n",
      "review_scores_rating                                0\n",
      "review_scores_cleanliness                           0\n",
      "review_scores_checkin                               0\n",
      "review_scores_communication                         0\n",
      "review_scores_location                              0\n",
      "review_scores_value                                 0\n",
      "instant_bookable                                    0\n",
      "calculated_host_listings_count                      0\n",
      "calculated_host_listings_count_entire_homes         0\n",
      "calculated_host_listings_count_private_rooms        0\n",
      "calculated_host_listings_count_shared_rooms         0\n",
      "reviews_per_month                                   0\n",
      "n_host_verifications                                0\n",
      "dtype: int64\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 28022 entries, 0 to 28021\n",
      "Data columns (total 41 columns):\n",
      " #   Column                                        Non-Null Count  Dtype  \n",
      "---  ------                                        --------------  -----  \n",
      " 0   host_response_rate                            16179 non-null  float64\n",
      " 1   host_acceptance_rate                          16909 non-null  float64\n",
      " 2   host_is_superhost                             28022 non-null  bool   \n",
      " 3   host_listings_count                           28022 non-null  float64\n",
      " 4   host_total_listings_count                     28022 non-null  float64\n",
      " 5   host_has_profile_pic                          28022 non-null  bool   \n",
      " 6   host_identity_verified                        28022 non-null  bool   \n",
      " 7   accommodates                                  28022 non-null  int64  \n",
      " 8   bathrooms                                     28022 non-null  float64\n",
      " 9   bedrooms                                      25104 non-null  float64\n",
      " 10  beds                                          26668 non-null  float64\n",
      " 11  price                                         28022 non-null  float64\n",
      " 12  minimum_nights                                28022 non-null  int64  \n",
      " 13  maximum_nights                                28022 non-null  int64  \n",
      " 14  minimum_minimum_nights                        28022 non-null  float64\n",
      " 15  maximum_minimum_nights                        28022 non-null  float64\n",
      " 16  minimum_maximum_nights                        28022 non-null  float64\n",
      " 17  maximum_maximum_nights                        28022 non-null  float64\n",
      " 18  minimum_nights_avg_ntm                        28022 non-null  float64\n",
      " 19  maximum_nights_avg_ntm                        28022 non-null  float64\n",
      " 20  has_availability                              28022 non-null  bool   \n",
      " 21  availability_30                               28022 non-null  int64  \n",
      " 22  availability_60                               28022 non-null  int64  \n",
      " 23  availability_90                               28022 non-null  int64  \n",
      " 24  availability_365                              28022 non-null  int64  \n",
      " 25  number_of_reviews                             28022 non-null  int64  \n",
      " 26  number_of_reviews_ltm                         28022 non-null  int64  \n",
      " 27  number_of_reviews_l30d                        28022 non-null  int64  \n",
      " 28  review_scores_rating                          28022 non-null  float64\n",
      " 29  review_scores_cleanliness                     28022 non-null  float64\n",
      " 30  review_scores_checkin                         28022 non-null  float64\n",
      " 31  review_scores_communication                   28022 non-null  float64\n",
      " 32  review_scores_location                        28022 non-null  float64\n",
      " 33  review_scores_value                           28022 non-null  float64\n",
      " 34  instant_bookable                              28022 non-null  bool   \n",
      " 35  calculated_host_listings_count                28022 non-null  int64  \n",
      " 36  calculated_host_listings_count_entire_homes   28022 non-null  int64  \n",
      " 37  calculated_host_listings_count_private_rooms  28022 non-null  int64  \n",
      " 38  calculated_host_listings_count_shared_rooms   28022 non-null  int64  \n",
      " 39  reviews_per_month                             28022 non-null  float64\n",
      " 40  n_host_verifications                          28022 non-null  int64  \n",
      "dtypes: bool(5), float64(21), int64(15)\n",
      "memory usage: 7.8 MB\n",
      "None\n",
      "       host_response_rate  host_acceptance_rate  host_listings_count  \\\n",
      "count        16179.000000          16909.000000         28022.000000   \n",
      "mean             0.906901              0.791953            14.554778   \n",
      "std              0.227282              0.276732           120.721287   \n",
      "min              0.000000              0.000000             0.000000   \n",
      "25%              0.940000              0.680000             1.000000   \n",
      "50%              1.000000              0.910000             1.000000   \n",
      "75%              1.000000              1.000000             3.000000   \n",
      "max              1.000000              1.000000          3387.000000   \n",
      "\n",
      "       host_total_listings_count  accommodates     bathrooms      bedrooms  \\\n",
      "count               28022.000000  28022.000000  28022.000000  25104.000000   \n",
      "mean                   14.554778      2.874491      1.142174      1.329708   \n",
      "std                   120.721287      1.860251      0.421132      0.700726   \n",
      "min                     0.000000      1.000000      0.000000      1.000000   \n",
      "25%                     1.000000      2.000000      1.000000      1.000000   \n",
      "50%                     1.000000      2.000000      1.000000      1.000000   \n",
      "75%                     3.000000      4.000000      1.000000      1.000000   \n",
      "max                  3387.000000     16.000000      8.000000     12.000000   \n",
      "\n",
      "               beds         price  minimum_nights  ...  review_scores_checkin  \\\n",
      "count  26668.000000  28022.000000    28022.000000  ...           28022.000000   \n",
      "mean       1.629556    154.228749       18.689387  ...               4.814300   \n",
      "std        1.097104    140.816605       25.569151  ...               0.438603   \n",
      "min        1.000000     29.000000        1.000000  ...               0.000000   \n",
      "25%        1.000000     70.000000        2.000000  ...               4.810000   \n",
      "50%        1.000000    115.000000       30.000000  ...               4.960000   \n",
      "75%        2.000000    180.000000       30.000000  ...               5.000000   \n",
      "max       21.000000   1000.000000     1250.000000  ...               5.000000   \n",
      "\n",
      "       review_scores_communication  review_scores_location  \\\n",
      "count                 28022.000000            28022.000000   \n",
      "mean                      4.808041                4.750393   \n",
      "std                       0.464585                0.415717   \n",
      "min                       0.000000                0.000000   \n",
      "25%                       4.810000                4.670000   \n",
      "50%                       4.970000                4.880000   \n",
      "75%                       5.000000                5.000000   \n",
      "max                       5.000000                5.000000   \n",
      "\n",
      "       review_scores_value  calculated_host_listings_count  \\\n",
      "count         28022.000000                    28022.000000   \n",
      "mean              4.647670                        9.581900   \n",
      "std               0.518023                       32.227523   \n",
      "min               0.000000                        1.000000   \n",
      "25%               4.550000                        1.000000   \n",
      "50%               4.780000                        1.000000   \n",
      "75%               5.000000                        3.000000   \n",
      "max               5.000000                      421.000000   \n",
      "\n",
      "       calculated_host_listings_count_entire_homes  \\\n",
      "count                                 28022.000000   \n",
      "mean                                      5.562986   \n",
      "std                                      26.121426   \n",
      "min                                       0.000000   \n",
      "25%                                       0.000000   \n",
      "50%                                       1.000000   \n",
      "75%                                       1.000000   \n",
      "max                                     308.000000   \n",
      "\n",
      "       calculated_host_listings_count_private_rooms  \\\n",
      "count                                  28022.000000   \n",
      "mean                                       3.902077   \n",
      "std                                       17.972386   \n",
      "min                                        0.000000   \n",
      "25%                                        0.000000   \n",
      "50%                                        0.000000   \n",
      "75%                                        1.000000   \n",
      "max                                      359.000000   \n",
      "\n",
      "       calculated_host_listings_count_shared_rooms  reviews_per_month  \\\n",
      "count                                 28022.000000       28022.000000   \n",
      "mean                                      0.048283           1.758325   \n",
      "std                                       0.442459           4.446143   \n",
      "min                                       0.000000           0.010000   \n",
      "25%                                       0.000000           0.130000   \n",
      "50%                                       0.000000           0.510000   \n",
      "75%                                       0.000000           1.830000   \n",
      "max                                       8.000000         141.000000   \n",
      "\n",
      "       n_host_verifications  \n",
      "count          28022.000000  \n",
      "mean               5.169510  \n",
      "std                2.028497  \n",
      "min                1.000000  \n",
      "25%                4.000000  \n",
      "50%                5.000000  \n",
      "75%                7.000000  \n",
      "max               13.000000  \n",
      "\n",
      "[8 rows x 36 columns]\n"
     ]
    }
   ],
   "source": [
    "airbnb_df = df\n",
    "# Drop unnecessary columns\n",
    "airbnb_df.drop(columns=[\n",
    "    'name', 'room_type', 'neighbourhood_group_cleansed', 'amenities', 'host_location',\n",
    "    'description', 'neighborhood_overview', 'host_name', 'host_about'\n",
    "], inplace=True)\n",
    "\n",
    "# Display the first few rows of the DataFrame to confirm successful loading\n",
    "print(\"Airbnb Listings Data Set:\")\n",
    "print(airbnb_df.head())\n",
    "\n",
    "# Check for missing values\n",
    "print(airbnb_df.isnull().sum())\n",
    "\n",
    "# Get a summary of the dataset\n",
    "print(airbnb_df.info())\n",
    "print(airbnb_df.describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "# Fill missing numeric values with the mean\n",
    "for col in ['host_response_rate', 'host_acceptance_rate', 'reviews_per_month']:\n",
    "    airbnb_df[col].fillna(airbnb_df[col].mean(), inplace=True)\n",
    "\n",
    "# Drop rows with missing target values\n",
    "airbnb_df.dropna(subset=['has_availability'], inplace=True)\n",
    "\n",
    "# Fill other missing categorical values with the mode\n",
    "for col in ['host_is_superhost', 'instant_bookable']:\n",
    "    airbnb_df[col].fillna(airbnb_df[col].mode()[0], inplace=True)\n",
    "\n",
    "def convert_percentage_to_float(x):\n",
    "    return float(x.strip('%')) / 100 if isinstance(x, str) else x\n",
    "\n",
    "airbnb_df['host_response_rate'] = airbnb_df['host_response_rate'].apply(convert_percentage_to_float)\n",
    "airbnb_df['host_acceptance_rate'] = airbnb_df['host_acceptance_rate'].apply(convert_percentage_to_float)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Convert 'has_availability' to binary (True -> 1, False -> 0)\n",
    "airbnb_df['has_availability'] = airbnb_df['has_availability'].apply(lambda x: 1 if x else 0)\n",
    "\n",
    "# Convert boolean columns to integers\n",
    "bool_cols = ['host_has_profile_pic', 'host_identity_verified', 'instant_bookable']\n",
    "for col in bool_cols:\n",
    "    airbnb_df[col] = airbnb_df[col].astype(int)\n",
    "\n",
    "# One-hot encode binary categorical variables\n",
    "airbnb_df = pd.get_dummies(airbnb_df, columns=['host_is_superhost', 'instant_bookable'])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# List of numerical features to scale\n",
    "num_features = [\n",
    "    'host_response_rate', 'host_acceptance_rate', 'host_listings_count',\n",
    "    'host_total_listings_count', 'host_has_profile_pic',\n",
    "    'host_identity_verified', 'accommodates', 'bathrooms', 'bedrooms',\n",
    "    'beds', 'price', 'minimum_nights', 'maximum_nights',\n",
    "    'minimum_minimum_nights', 'maximum_minimum_nights',\n",
    "    'minimum_maximum_nights', 'maximum_maximum_nights',\n",
    "    'minimum_nights_avg_ntm', 'maximum_nights_avg_ntm',\n",
    "    'availability_30', 'availability_60', 'availability_90',\n",
    "    'availability_365', 'number_of_reviews', 'number_of_reviews_ltm',\n",
    "    'number_of_reviews_l30d', 'review_scores_rating',\n",
    "    'review_scores_cleanliness', 'review_scores_checkin',\n",
    "    'review_scores_communication', 'review_scores_location',\n",
    "    'review_scores_value', 'calculated_host_listings_count',\n",
    "    'calculated_host_listings_count_entire_homes',\n",
    "    'calculated_host_listings_count_private_rooms',\n",
    "    'calculated_host_listings_count_shared_rooms', 'reviews_per_month',\n",
    "    'n_host_verifications'\n",
    "]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "host_response_rate                              float64\n",
      "host_acceptance_rate                            float64\n",
      "host_listings_count                             float64\n",
      "host_total_listings_count                       float64\n",
      "host_has_profile_pic                            float64\n",
      "host_identity_verified                          float64\n",
      "accommodates                                    float64\n",
      "bathrooms                                       float64\n",
      "bedrooms                                        float64\n",
      "beds                                            float64\n",
      "price                                           float64\n",
      "minimum_nights                                  float64\n",
      "maximum_nights                                  float64\n",
      "minimum_minimum_nights                          float64\n",
      "maximum_minimum_nights                          float64\n",
      "minimum_maximum_nights                          float64\n",
      "maximum_maximum_nights                          float64\n",
      "minimum_nights_avg_ntm                          float64\n",
      "maximum_nights_avg_ntm                          float64\n",
      "has_availability                                  int64\n",
      "availability_30                                 float64\n",
      "availability_60                                 float64\n",
      "availability_90                                 float64\n",
      "availability_365                                float64\n",
      "number_of_reviews                               float64\n",
      "number_of_reviews_ltm                           float64\n",
      "number_of_reviews_l30d                          float64\n",
      "review_scores_rating                            float64\n",
      "review_scores_cleanliness                       float64\n",
      "review_scores_checkin                           float64\n",
      "review_scores_communication                     float64\n",
      "review_scores_location                          float64\n",
      "review_scores_value                             float64\n",
      "calculated_host_listings_count                  float64\n",
      "calculated_host_listings_count_entire_homes     float64\n",
      "calculated_host_listings_count_private_rooms    float64\n",
      "calculated_host_listings_count_shared_rooms     float64\n",
      "reviews_per_month                               float64\n",
      "n_host_verifications                            float64\n",
      "host_is_superhost_True                            uint8\n",
      "instant_bookable_0                                uint8\n",
      "instant_bookable_1                                uint8\n",
      "dtype: object\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Remove the target column from the numerical features list\n",
    "if 'has_availability' in num_features:\n",
    "    num_features.remove('has_availability')\n",
    "\n",
    "# Verify numerical features exist in DataFrame\n",
    "for feature in num_features:\n",
    "    if feature not in airbnb_df.columns:\n",
    "        print(f\"Warning: Feature '{feature}' not found in DataFrame\")\n",
    "\n",
    "# Standardize the numerical features\n",
    "scaler = StandardScaler()\n",
    "airbnb_df[num_features] = scaler.fit_transform(airbnb_df[num_features])\n",
    "\n",
    "# Check if all columns are numeric\n",
    "print(airbnb_df.dtypes)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DataFrame columns: Index(['host_response_rate', 'host_acceptance_rate', 'host_listings_count',\n",
      "       'host_total_listings_count', 'host_has_profile_pic',\n",
      "       'host_identity_verified', 'accommodates', 'bathrooms', 'bedrooms',\n",
      "       'beds', 'price', 'minimum_nights', 'maximum_nights',\n",
      "       'minimum_minimum_nights', 'maximum_minimum_nights',\n",
      "       'minimum_maximum_nights', 'maximum_maximum_nights',\n",
      "       'minimum_nights_avg_ntm', 'maximum_nights_avg_ntm', 'has_availability',\n",
      "       'availability_30', 'availability_60', 'availability_90',\n",
      "       'availability_365', 'number_of_reviews', 'number_of_reviews_ltm',\n",
      "       'number_of_reviews_l30d', 'review_scores_rating',\n",
      "       'review_scores_cleanliness', 'review_scores_checkin',\n",
      "       'review_scores_communication', 'review_scores_location',\n",
      "       'review_scores_value', 'calculated_host_listings_count',\n",
      "       'calculated_host_listings_count_entire_homes',\n",
      "       'calculated_host_listings_count_private_rooms',\n",
      "       'calculated_host_listings_count_shared_rooms', 'reviews_per_month',\n",
      "       'n_host_verifications', 'host_is_superhost_True', 'instant_bookable_0',\n",
      "       'instant_bookable_1'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Check the columns in the DataFrame\n",
    "print(\"DataFrame columns:\", airbnb_df.columns)\n",
    "\n",
    "# Define numerical and categorical columns\n",
    "numerical_cols = airbnb_df.select_dtypes(include=['int64', 'float64']).columns.tolist()\n",
    "categorical_cols = airbnb_df.select_dtypes(include=['object', 'bool', 'uint8']).columns.tolist()\n",
    "\n",
    "# Define the target column\n",
    "target_column = 'has_availability'\n",
    "\n",
    "# Remove the target column from numerical and categorical features list if it exists\n",
    "if target_column in numerical_cols:\n",
    "    numerical_cols.remove(target_column)\n",
    "if target_column in categorical_cols:\n",
    "    categorical_cols.remove(target_column)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Numerical columns before preprocessing: ['host_response_rate', 'host_acceptance_rate', 'host_listings_count', 'host_total_listings_count', 'host_has_profile_pic', 'host_identity_verified', 'accommodates', 'bathrooms', 'bedrooms', 'beds', 'price', 'minimum_nights', 'maximum_nights', 'minimum_minimum_nights', 'maximum_minimum_nights', 'minimum_maximum_nights', 'maximum_maximum_nights', 'minimum_nights_avg_ntm', 'maximum_nights_avg_ntm', 'availability_30', 'availability_60', 'availability_90', 'availability_365', 'number_of_reviews', 'number_of_reviews_ltm', 'number_of_reviews_l30d', 'review_scores_rating', 'review_scores_cleanliness', 'review_scores_checkin', 'review_scores_communication', 'review_scores_location', 'review_scores_value', 'calculated_host_listings_count', 'calculated_host_listings_count_entire_homes', 'calculated_host_listings_count_private_rooms', 'calculated_host_listings_count_shared_rooms', 'reviews_per_month', 'n_host_verifications']\n",
      "Categorical columns before preprocessing: ['host_is_superhost_True', 'instant_bookable_0', 'instant_bookable_1']\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Verify columns before preprocessing\n",
    "print(\"Numerical columns before preprocessing:\", numerical_cols)\n",
    "print(\"Categorical columns before preprocessing:\", categorical_cols)\n",
    "\n",
    "# Define preprocessing steps\n",
    "preprocessor = ColumnTransformer(\n",
    "    transformers=[\n",
    "        ('num', Pipeline(steps=[\n",
    "            ('imputer', SimpleImputer(strategy='mean')),\n",
    "            ('scaler', StandardScaler())\n",
    "        ]), numerical_cols),\n",
    "        ('cat', Pipeline(steps=[\n",
    "            ('imputer', SimpleImputer(strategy='most_frequent')),\n",
    "            ('onehot', OneHotEncoder(handle_unknown='ignore'))\n",
    "        ]), categorical_cols)\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Columns in X before transformation: Index(['host_response_rate', 'host_acceptance_rate', 'host_listings_count',\n",
      "       'host_total_listings_count', 'host_has_profile_pic',\n",
      "       'host_identity_verified', 'accommodates', 'bathrooms', 'bedrooms',\n",
      "       'beds', 'price', 'minimum_nights', 'maximum_nights',\n",
      "       'minimum_minimum_nights', 'maximum_minimum_nights',\n",
      "       'minimum_maximum_nights', 'maximum_maximum_nights',\n",
      "       'minimum_nights_avg_ntm', 'maximum_nights_avg_ntm', 'availability_30',\n",
      "       'availability_60', 'availability_90', 'availability_365',\n",
      "       'number_of_reviews', 'number_of_reviews_ltm', 'number_of_reviews_l30d',\n",
      "       'review_scores_rating', 'review_scores_cleanliness',\n",
      "       'review_scores_checkin', 'review_scores_communication',\n",
      "       'review_scores_location', 'review_scores_value',\n",
      "       'calculated_host_listings_count',\n",
      "       'calculated_host_listings_count_entire_homes',\n",
      "       'calculated_host_listings_count_private_rooms',\n",
      "       'calculated_host_listings_count_shared_rooms', 'reviews_per_month',\n",
      "       'n_host_verifications', 'host_is_superhost_True', 'instant_bookable_0',\n",
      "       'instant_bookable_1'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Define features and target variable\n",
    "X = airbnb_df.drop('has_availability', axis=1)\n",
    "y = airbnb_df['has_availability']\n",
    "\n",
    "# Verify columns before transformation\n",
    "print(\"Columns in X before transformation:\", X.columns)\n",
    "\n",
    "try:\n",
    "    X_transformed = preprocessor.fit_transform(X)\n",
    "except ValueError as e:\n",
    "    missing_columns = set(numerical_cols + categorical_cols) - set(X.columns)\n",
    "    raise ValueError(f\"The following columns are missing in the DataFrame: {missing_columns}\") from e\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initial X_transformed shape: (28022, 43)\n",
      "Shape after VarianceThreshold: (28022, 40)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Verify there are no NaN values in the transformed data\n",
    "if pd.DataFrame(X_transformed).isnull().sum().sum() > 0:\n",
    "    raise ValueError(\"There are still missing values in the transformed features.\")\n",
    "preprocessed_columns = numerical_cols + preprocessor.transformers_[1][1].named_steps['onehot'].get_feature_names_out(categorical_cols).tolist()\n",
    "\n",
    "# Verify transformation output\n",
    "print(\"Initial X_transformed shape:\", X_transformed.shape)\n",
    "\n",
    "# Remove constant features\n",
    "constant_filter = VarianceThreshold(threshold=0)\n",
    "X_transformed = constant_filter.fit_transform(X_transformed)\n",
    "# Update the feature names after removing constant features\n",
    "feature_names_after_variance_threshold = [f for i, f in enumerate(preprocessed_columns) if constant_filter.get_support()[i]]\n",
    "# Verify shape after removing constant features\n",
    "print(\"Shape after VarianceThreshold:\", X_transformed.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape after SelectKBest: (28022, 10)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Select top k features\n",
    "k = 10\n",
    "selector = SelectKBest(score_func=f_classif, k=k)\n",
    "\n",
    "if pd.DataFrame(X_transformed).isnull().sum().sum() > 0:\n",
    "    raise ValueError(\"There are still missing values in the transformed features before fitting SelectKBest.\")\n",
    "\n",
    "X_new = selector.fit_transform(X_transformed, y)\n",
    "\n",
    "# Update the feature names after feature selection\n",
    "selected_features = [f for i, f in enumerate(feature_names_after_variance_threshold) if selector.get_support()[i]]\n",
    "# Verify shape after feature selection\n",
    "print(\"Shape after SelectKBest:\", X_new.shape)\n",
    "\n",
    "# Split the data into training and testing sets with selected features\n",
    "X_train_new, X_test_new, y_train_new, y_test_new = train_test_split(X_new, y, test_size=0.2, random_state=42)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shapes after train-test split - X_train_new: (22417, 10) X_test_new: (5605, 10) y_train_new: (22417,) y_test_new: (5605,)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Verify shapes after train-test split\n",
    "print(\"Shapes after train-test split - X_train_new:\", X_train_new.shape, \"X_test_new:\", X_test_new.shape, \"y_train_new:\", y_train_new.shape, \"y_test_new:\", y_test_new.shape)\n",
    "\n",
    "# Define the model\n",
    "rf = RandomForestClassifier(random_state=42)\n",
    "\n",
    "# Define the parameter grid\n",
    "param_grid = {\n",
    "    'n_estimators': [50, 100, 200],\n",
    "    'max_depth': [None, 10, 20, 30],\n",
    "    'min_samples_split': [2, 5, 10],\n",
    "    'min_samples_leaf': [1, 2, 4]\n",
    "}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Starting GridSearchCV\n",
      "GridSearchCV completed\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "#confirm we reached here\n",
    "print(\"Starting GridSearchCV\")\n",
    "\n",
    "# Perform grid search\n",
    "try:\n",
    "    grid_search = GridSearchCV(estimator=rf, param_grid=param_grid, cv=5, scoring='roc_auc', n_jobs=-1)\n",
    "    grid_search.fit(X_train_new, y_train_new)\n",
    "    print(\"GridSearchCV completed\")\n",
    "except Exception as e:\n",
    "    print(\"Error during GridSearchCV:\", e)\n",
    "    raise\n",
    "\n",
    "# Get the best model\n",
    "best_rf = grid_search.best_estimator_\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model saved to best_rf_model.pkl\n",
      "Model loaded from best_rf_model.pkl\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Save the best model to a file\n",
    "with open('best_rf_model.pkl', 'wb') as model_file:\n",
    "    pickle.dump(best_rf, model_file)\n",
    "\n",
    "print(\"Model saved to best_rf_model.pkl\")\n",
    "\n",
    "# Load the model from the file\n",
    "with open('best_rf_model.pkl', 'rb') as model_file:\n",
    "    loaded_rf_model = pickle.load(model_file)\n",
    "\n",
    "print(\"Model loaded from best_rf_model.pkl\")\n",
    "\n",
    "# Verify the loaded model by making predictions\n",
    "loaded_y_pred_new = loaded_rf_model.predict(X_test_new)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Loaded Random Forest Classifier with Grid Search:\n",
      "Accuracy: 0.960392506690455\n",
      "Precision: 0.960392506690455\n",
      "Recall: 1.0\n",
      "F1 Score: 0.9797961412449946\n",
      "ROC AUC Score: 0.5\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Evaluate the loaded model\n",
    "print(\"Loaded Random Forest Classifier with Grid Search:\")\n",
    "print(\"Accuracy:\", accuracy_score(y_test_new, loaded_y_pred_new))\n",
    "print(\"Precision:\", precision_score(y_test_new, loaded_y_pred_new))\n",
    "print(\"Recall:\", recall_score(y_test_new, loaded_y_pred_new))\n",
    "print(\"F1 Score:\", f1_score(y_test_new, loaded_y_pred_new))\n",
    "print(\"ROC AUC Score:\", roc_auc_score(y_test_new, loaded_y_pred_new))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Plot confusion matrix\n",
    "conf_matrix = confusion_matrix(y_test_new, loaded_y_pred_new)\n",
    "plt.figure(figsize=(8, 6))\n",
    "sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues')\n",
    "plt.title('Confusion Matrix')\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('Actual')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Plot ROC curve\n",
    "fpr, tpr, thresholds = roc_curve(y_test_new, grid_search.predict_proba(X_test_new)[:, 1])\n",
    "roc_auc = auc(fpr, tpr)\n",
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(fpr, tpr, color='darkorange', lw=2, label=f'ROC curve (area = {roc_auc:.2f})')\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "plt.legend(loc='lower right')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Plot precision-recall curve\n",
    "precision, recall, _ = precision_recall_curve(y_test_new, grid_search.predict_proba(X_test_new)[:, 1])\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.plot(recall, precision, color='blue', lw=2)\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.title('Precision-Recall Curve')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Plot feature importance\n",
    "feature_importances = best_rf.feature_importances_\n",
    "\n",
    "plt.figure(figsize=(12, 8))\n",
    "plt.barh(range(len(feature_importances)), feature_importances, align='center')\n",
    "plt.yticks(range(len(feature_importances)), selected_features)\n",
    "plt.xlabel('Feature Importance')\n",
    "plt.title('Feature Importance from Random Forest')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Visualize model performance\n",
    "metrics = {\n",
    "    'Accuracy': accuracy_score(y_test_new, loaded_y_pred_new),\n",
    "    'Precision': precision_score(y_test_new, loaded_y_pred_new),\n",
    "    'Recall': recall_score(y_test_new, loaded_y_pred_new),\n",
    "    'F1 Score': f1_score(y_test_new, loaded_y_pred_new),\n",
    "    'ROC AUC': roc_auc_score(y_test_new, loaded_y_pred_new)\n",
    "}\n",
    "\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(metrics.keys(), metrics.values(), color=['blue', 'green', 'red', 'purple', 'orange'])\n",
    "plt.xlabel('Metric')\n",
    "plt.ylabel('Score')\n",
    "plt.ylim([0, 1])\n",
    "plt.title('Model Performance Metrics')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.9.19"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}