{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 📓 Interactive Notebook: Advanced Data Manipulation with Pandas & Seaborn\n",
    "\n",
    "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/aakashkh/data-patterns/blob/main/pandas_session/pandas_session.ipynb)\n",
    "\n",
    "Welcome! This notebook is the hands-on component of our 1-hour session: **\"Advanced Data Manipulation & EDA with Pandas: Preprocessing for Machine Learning\"**.\n",
    "\n",
    "### 🎯 Learning Objectives\n",
    "By the end of this notebook, you will be able to:\n",
    "1. **Clean and prepare data** for machine learning models (handling missing values, de-duplication, data type casting).\n",
    "2. **Encode categorical features** into numerical representations (`One-Hot Encoding` and `Ordinal Mapping`).\n",
    "3. **Perform advanced filtering and row transformations** using boolean indexing, lambda functions, and vectorized NumPy functions.\n",
    "4. **Integrate multiple data sources** using database-style merges (`pd.merge`).\n",
    "5. **Perform GroupBy aggregations and transforms** (`Split-Apply-Combine`) to engineer group-level features.\n",
    "6. **Perform Exploratory Data Analysis (EDA)** and visualize distributions, outliers, and feature correlations using Seaborn.\n",
    "7. **Extract numerical features from datetime columns** and build time-series features (lags, leads, rolling averages).\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🧠 Understanding the Ecosystem: Pandas Internals & Modern Alternates\n",
    "\n",
    "Before we write code, let's look at how Pandas works under the hood and why the industry is adopting newer data tools as dataset sizes scale.\n",
    "\n",
    "### 1. Pandas Internals: The Constraints\n",
    "* **In-Memory Design**: Pandas is designed to load your entire dataset into RAM. A common rule of thumb is that **you need 5x to 10x the size of your dataset in RAM** to perform transformations without running out of memory (OOM crashes).\n",
    "* **Single-Core / Single-Threaded**: The core engine of Pandas (built on top of older NumPy versions) runs on a single CPU core. Even if your computer has 8 or 16 cores, Pandas will leave 90%+ of your processor power idle during a long computation.\n",
    "\n",
    "### 2. Modern Alternates for Big Data & ML Pipelines\n",
    "When datasets grow beyond 10GB, loading them in Pandas becomes a bottleneck. Two major engines are changing how we do feature engineering for ML:\n",
    "\n",
    "* **Polars (Written in Rust)**:\n",
    "  * **Multi-Threaded**: Parallelizes execution across all available CPU cores by default.\n",
    "  * **Lazy Evaluation**: Instead of executing every line of code immediately (eagerly), Polars builds a query plan and optimizes it (e.g., combining filters, skipping unused columns) before processing the data.\n",
    "  * **Streaming**: Can process data in chunks (out-of-core), allowing you to run operations on datasets larger than your physical RAM.\n",
    "\n",
    "* **DuckDB (In-Process Columnar SQL)**:\n",
    "  * **Vectorized Query Engine**: Processes data in columns rather than rows, which is extremely fast for analytical aggregations.\n",
    "  * **Serverless**: Runs directly inside your Python process (like SQLite but optimized for analytics).\n",
    "  * **Pandas Integration**: DuckDB can query Pandas DataFrames or Parquet files directly using SQL without copying the data, making it an excellent tool for joining massive tables before feeding clean arrays to machine learning models.\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Libraries imported successfully!\n",
      "Pandas version:     3.0.0\n",
      "NumPy version:      2.4.1\n",
      "Seaborn version:    0.13.2\n",
      "Matplotlib version: 3.10.8\n"
     ]
    }
   ],
   "source": [
    "# 💻 Step 1: Import core packages and print versions\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Set style for plots\n",
    "sns.set_theme(style=\"whitegrid\")\n",
    "plt.rcParams[\"figure.figsize\"] = (10, 6)\n",
    "\n",
    "print(\"Libraries imported successfully!\")\n",
    "print(f\"Pandas version:     {pd.__version__}\")\n",
    "print(f\"NumPy version:      {np.__version__}\")\n",
    "print(f\"Seaborn version:    {sns.__version__}\")\n",
    "print(f\"Matplotlib version: {mpl.__version__}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🛠️ Step 2: Synthetic Data Generation\n",
    "To make this notebook completely self-contained and realistic, we will generate a synthetic e-commerce dataset. \n",
    "This dataset simulates retail transactions and customer profiles, complete with **missing values (NaNs), duplicate records, string variables, and raw dates**.\n",
    "\n",
    "> **Colab Tip**: Because this dataset is generated in-memory, you do not need to upload any external CSV files or mount Google Drive to run this notebook!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Transactions dataset generated. Shape: (205, 6)\n",
      "Customers dataset generated. Shape: (50, 5)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\aakashkhandelwal\\AppData\\Local\\Temp\\ipykernel_18864\\1921948098.py:11: Pandas4Warning: 'd' is deprecated and will be removed in a future version, please use 'D' instead.\n",
      "  'signup_date': pd.date_range(start='2025-01-01', periods=50, freq='d')\n"
     ]
    }
   ],
   "source": [
    "# Seed for reproducibility\n",
    "np.random.seed(42)\n",
    "n_samples = 200\n",
    "\n",
    "# 1. Create Customer Profiles DataFrame\n",
    "customers_data = {\n",
    "    'customer_id': [f\"CUST_{i:03d}\" for i in range(1, 51)],\n",
    "    'age': np.random.choice([22, 28, 35, 42, 50, np.nan], size=50, p=[0.2, 0.2, 0.2, 0.2, 0.1, 0.1]), # Injecting NaNs\n",
    "    'gender': np.random.choice(['M', 'F', 'Other'], size=50),\n",
    "    'membership': np.random.choice(['Bronze', 'Silver', 'Gold'], size=50, p=[0.5, 0.3, 0.2]),\n",
    "    'signup_date': pd.date_range(start='2025-01-01', periods=50, freq='d')\n",
    "}\n",
    "df_customers = pd.DataFrame(customers_data)\n",
    "\n",
    "# 2. Create Transactions DataFrame\n",
    "tx_dates = pd.date_range(start='2026-01-01', periods=n_samples, freq='h')\n",
    "transactions_data = {\n",
    "    'transaction_id': [f\"TX_{i:04d}\" for i in range(1, n_samples + 1)],\n",
    "    'customer_id': np.random.choice(df_customers['customer_id'], size=n_samples),\n",
    "    'timestamp': tx_dates,\n",
    "    'amount': np.random.normal(loc=120, scale=40, size=n_samples).round(2),\n",
    "    'category': np.random.choice(['Electronics', 'Clothing', 'Books', np.nan], size=n_samples, p=[0.4, 0.4, 0.15, 0.05]), # Injecting NaNs\n",
    "    'is_fraud': np.random.choice([0, 1], size=n_samples, p=[0.95, 0.05]) # Imbalanced target variable\n",
    "}\n",
    "df_transactions = pd.DataFrame(transactions_data)\n",
    "\n",
    "# Inject some duplicate rows into transactions\n",
    "duplicates = df_transactions.iloc[10:15]\n",
    "df_transactions = pd.concat([df_transactions, duplicates], ignore_index=True)\n",
    "\n",
    "print(f\"Transactions dataset generated. Shape: {df_transactions.shape}\")\n",
    "print(f\"Customers dataset generated. Shape: {df_customers.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🧹 Step 3: Data Cleaning & Pre-ML Formatting\n",
    "Before training models in Scikit-Learn, we must clean our data. If the data contains missing values (`NaN`) or duplicate rows, training will either crash or suffer from data leakage. Let's clean it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of duplicate rows in transactions: 5\n",
      "Shape after removing duplicates: (200, 6)\n",
      "\n",
      "Missing values in customer profiles:\n",
      "customer_id    0\n",
      "age            5\n",
      "gender         0\n",
      "membership     0\n",
      "signup_date    0\n",
      "dtype: int64\n",
      "\n",
      "Imputed missing age with median age: 28.0\n",
      "Missing age counts after imputation: 0\n"
     ]
    }
   ],
   "source": [
    "# 1. Check for Duplicate Rows\n",
    "dup_count = df_transactions.duplicated().sum()\n",
    "print(f\"Number of duplicate rows in transactions: {dup_count}\")\n",
    "\n",
    "# Remove duplicates\n",
    "df_transactions = df_transactions.drop_duplicates(keep='first')\n",
    "print(f\"Shape after removing duplicates: {df_transactions.shape}\")\n",
    "\n",
    "# 2. Check for Missing Values\n",
    "print(\"\\nMissing values in customer profiles:\")\n",
    "print(df_customers.isnull().sum())\n",
    "\n",
    "# Impute missing Customer Age with the Median\n",
    "median_age = df_customers['age'].median()\n",
    "df_customers['age'] = df_customers['age'].fillna(median_age)\n",
    "print(f\"\\nImputed missing age with median age: {median_age}\")\n",
    "print(f\"Missing age counts after imputation: {df_customers['age'].isnull().sum()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🏷️ Step 4: Categorical Encoding\n",
    "Machine Learning algorithms require numerical input. String classes like \"Bronze\" or \"F\" cannot be used in formulas directly. We will convert them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sample of encoded customer profiles:\n"
     ]
    },
    {
     "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>customer_id</th>\n",
       "      <th>membership</th>\n",
       "      <th>membership_encoded</th>\n",
       "      <th>gender_M</th>\n",
       "      <th>gender_Other</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CUST_001</td>\n",
       "      <td>Bronze</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CUST_002</td>\n",
       "      <td>Silver</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CUST_003</td>\n",
       "      <td>Silver</td>\n",
       "      <td>1</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CUST_004</td>\n",
       "      <td>Gold</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CUST_005</td>\n",
       "      <td>Bronze</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  customer_id membership  membership_encoded  gender_M  gender_Other\n",
       "0    CUST_001     Bronze                   0     False          True\n",
       "1    CUST_002     Silver                   1     False         False\n",
       "2    CUST_003     Silver                   1     False         False\n",
       "3    CUST_004       Gold                   2     False         False\n",
       "4    CUST_005     Bronze                   0     False         False"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1. Ordinal Encoding (Mapping ranked data to numerical scale)\n",
    "membership_map = {'Bronze': 0, 'Silver': 1, 'Gold': 2}\n",
    "df_customers['membership_encoded'] = df_customers['membership'].map(membership_map)\n",
    "\n",
    "# 2. One-Hot Encoding (Creating dummy variables for nominal features)\n",
    "# We drop the first column using drop_first=True to avoid the dummy variable trap (collinearity)\n",
    "df_customers_encoded = pd.get_dummies(df_customers, columns=['gender'], drop_first=True)\n",
    "\n",
    "print(\"Sample of encoded customer profiles:\")\n",
    "df_customers_encoded[['customer_id', 'membership', 'membership_encoded', 'gender_M', 'gender_Other']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 💡 Advanced Tip: Memory Optimization with Categoricals\n",
    "When preparing large datasets for machine learning, memory efficiency is key. By default, Pandas reads string columns as `object` data types, which are memory-heavy. Casting them to `category` types significantly compresses memory usage. Let's audit this using `.info(memory_usage='deep')`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- Memory before category type-casting ---\n",
      "<class 'pandas.DataFrame'>\n",
      "RangeIndex: 50 entries, 0 to 49\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype\n",
      "---  ------      --------------  -----\n",
      " 0   membership  50 non-null     str  \n",
      " 1   gender      50 non-null     str  \n",
      "dtypes: str(2)\n",
      "memory usage: 1.3 KB\n",
      "\n",
      "--- Memory after category type-casting ---\n",
      "<class 'pandas.DataFrame'>\n",
      "RangeIndex: 50 entries, 0 to 49\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype   \n",
      "---  ------      --------------  -----   \n",
      " 0   membership  50 non-null     category\n",
      " 1   gender      50 non-null     category\n",
      "dtypes: category(2)\n",
      "memory usage: 305.0 bytes\n"
     ]
    }
   ],
   "source": [
    "# Check memory usage of string categories before optimization\n",
    "print(\"--- Memory before category type-casting ---\")\n",
    "df_customers[['membership', 'gender']].info(memory_usage='deep')\n",
    "\n",
    "# Cast to category type\n",
    "df_mem_opt = df_customers.copy()\n",
    "df_mem_opt['membership'] = df_mem_opt['membership'].astype('category')\n",
    "df_mem_opt['gender'] = df_mem_opt['gender'].astype('category')\n",
    "\n",
    "print(\"\\n--- Memory after category type-casting ---\")\n",
    "df_mem_opt[['membership', 'gender']].info(memory_usage='deep')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🔍 Step 5: Advanced Filtering & Custom Feature Engineering\n",
    "Let's explore how to filter rows and engineer custom columns using standard filters, lambdas, and vectorized NumPy tools."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>transaction_id</th>\n",
       "      <th>amount</th>\n",
       "      <th>category</th>\n",
       "      <th>premium_fee</th>\n",
       "      <th>is_high_value</th>\n",
       "      <th>spend_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>TX_0001</td>\n",
       "      <td>47.92</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>2.3960</td>\n",
       "      <td>0</td>\n",
       "      <td>Low_Spend</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TX_0002</td>\n",
       "      <td>138.71</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>6.9355</td>\n",
       "      <td>0</td>\n",
       "      <td>Medium_Spend</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>TX_0003</td>\n",
       "      <td>57.94</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>2.8970</td>\n",
       "      <td>0</td>\n",
       "      <td>Low_Spend</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>TX_0004</td>\n",
       "      <td>99.58</td>\n",
       "      <td>Electronics</td>\n",
       "      <td>4.9790</td>\n",
       "      <td>0</td>\n",
       "      <td>Medium_Spend</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>TX_0005</td>\n",
       "      <td>90.68</td>\n",
       "      <td>Books</td>\n",
       "      <td>1.8136</td>\n",
       "      <td>0</td>\n",
       "      <td>Medium_Spend</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  transaction_id  amount     category  premium_fee  is_high_value  \\\n",
       "0        TX_0001   47.92  Electronics       2.3960              0   \n",
       "1        TX_0002  138.71  Electronics       6.9355              0   \n",
       "2        TX_0003   57.94  Electronics       2.8970              0   \n",
       "3        TX_0004   99.58  Electronics       4.9790              0   \n",
       "4        TX_0005   90.68        Books       1.8136              0   \n",
       "\n",
       "    spend_group  \n",
       "0     Low_Spend  \n",
       "1  Medium_Spend  \n",
       "2     Low_Spend  \n",
       "3  Medium_Spend  \n",
       "4  Medium_Spend  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1. Filtering using bitwise operators\n",
    "high_value_electronics = df_transactions[\n",
    "    (df_transactions['category'] == 'Electronics') & (df_transactions['amount'] > 150)\n",
    "]\n",
    "\n",
    "# 2. Row-wise Transformation using lambda & .apply()\n",
    "# (Creating a category-based premium flag)\n",
    "df_transactions['premium_fee'] = df_transactions.apply(\n",
    "    lambda row: row['amount'] * 0.05 if row['category'] == 'Electronics' else row['amount'] * 0.02,\n",
    "    axis=1\n",
    ")\n",
    "\n",
    "# 3. Fast Vectorized Categorization using np.where() and np.select()\n",
    "# Create an binary indicator column\n",
    "df_transactions['is_high_value'] = np.where(df_transactions['amount'] > 150, 1, 0)\n",
    "\n",
    "# Create multi-class categorical columns\n",
    "conditions = [\n",
    "    df_transactions['amount'] < 80,\n",
    "    (df_transactions['amount'] >= 80) & (df_transactions['amount'] < 150),\n",
    "    df_transactions['amount'] >= 150\n",
    "]\n",
    "choices = ['Low_Spend', 'Medium_Spend', 'High_Spend']\n",
    "df_transactions['spend_group'] = np.select(conditions, choices, default='Unknown')\n",
    "\n",
    "df_transactions[['transaction_id', 'amount', 'category', 'premium_fee', 'is_high_value', 'spend_group']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🤝 Step 6: Integrating Data Sources\n",
    "In real-world analytics, we need to join transaction event tables with user feature tables to train classification or regression models. We do this using `pd.merge()`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Merged DataFrame Shape: (200, 15)\n",
      "Merged columns:\n",
      "['transaction_id', 'customer_id', 'timestamp', 'amount', 'category', 'is_fraud', 'premium_fee', 'is_high_value', 'spend_group', 'age', 'membership', 'signup_date', 'membership_encoded', 'gender_M', 'gender_Other']\n"
     ]
    },
    {
     "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>transaction_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>amount</th>\n",
       "      <th>age</th>\n",
       "      <th>membership_encoded</th>\n",
       "      <th>gender_M</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>TX_0001</td>\n",
       "      <td>CUST_032</td>\n",
       "      <td>47.92</td>\n",
       "      <td>22.0</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TX_0002</td>\n",
       "      <td>CUST_032</td>\n",
       "      <td>138.71</td>\n",
       "      <td>22.0</td>\n",
       "      <td>2</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>TX_0003</td>\n",
       "      <td>CUST_024</td>\n",
       "      <td>57.94</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>TX_0004</td>\n",
       "      <td>CUST_041</td>\n",
       "      <td>99.58</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>TX_0005</td>\n",
       "      <td>CUST_049</td>\n",
       "      <td>90.68</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  transaction_id customer_id  amount   age  membership_encoded  gender_M\n",
       "0        TX_0001    CUST_032   47.92  22.0                   2     False\n",
       "1        TX_0002    CUST_032  138.71  22.0                   2     False\n",
       "2        TX_0003    CUST_024   57.94  28.0                   0     False\n",
       "3        TX_0004    CUST_041   99.58  22.0                   0     False\n",
       "4        TX_0005    CUST_049   90.68  35.0                   0     False"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# SQL-style Left Join to attach customer features to transaction records\n",
    "df_merged = pd.merge(df_transactions, df_customers_encoded, on='customer_id', how='left')\n",
    "\n",
    "print(f\"Merged DataFrame Shape: {df_merged.shape}\")\n",
    "print(\"Merged columns:\")\n",
    "print(df_merged.columns.tolist())\n",
    "df_merged[['transaction_id', 'customer_id', 'amount', 'age', 'membership_encoded', 'gender_M']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 📊 Step 6.5: GroupBy and Aggregations (Split-Apply-Combine)\n",
    "Aggregating customer features or transaction behaviors is crucial for model inputs (e.g., calculating user-level statistics or normalizing purchase amounts by membership level)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean transaction amount by membership level:\n",
      "membership\n",
      "Bronze    113.059903\n",
      "Gold      123.339535\n",
      "Silver    123.535741\n",
      "Name: amount, dtype: float64\n",
      "\n",
      "Multi-metric aggregations:\n",
      "                amount                       age\n",
      "                  mean     max        std median\n",
      "membership                                      \n",
      "Bronze      113.059903  205.73  42.116029   28.0\n",
      "Gold        123.339535  204.52  41.723574   28.0\n",
      "Silver      123.535741  247.48  46.608314   28.0\n"
     ]
    },
    {
     "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>customer_id</th>\n",
       "      <th>membership</th>\n",
       "      <th>amount</th>\n",
       "      <th>amount_diff_tier_mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>CUST_032</td>\n",
       "      <td>Gold</td>\n",
       "      <td>47.92</td>\n",
       "      <td>-75.419535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>CUST_032</td>\n",
       "      <td>Gold</td>\n",
       "      <td>138.71</td>\n",
       "      <td>15.370465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>CUST_024</td>\n",
       "      <td>Bronze</td>\n",
       "      <td>57.94</td>\n",
       "      <td>-55.119903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>CUST_041</td>\n",
       "      <td>Bronze</td>\n",
       "      <td>99.58</td>\n",
       "      <td>-13.479903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CUST_049</td>\n",
       "      <td>Bronze</td>\n",
       "      <td>90.68</td>\n",
       "      <td>-22.379903</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  customer_id membership  amount  amount_diff_tier_mean\n",
       "0    CUST_032       Gold   47.92             -75.419535\n",
       "1    CUST_032       Gold  138.71              15.370465\n",
       "2    CUST_024     Bronze   57.94             -55.119903\n",
       "3    CUST_041     Bronze   99.58             -13.479903\n",
       "4    CUST_049     Bronze   90.68             -22.379903"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1. Basic grouping and mean calculation\n",
    "group_means = df_merged.groupby('membership')['amount'].mean()\n",
    "print(\"Mean transaction amount by membership level:\")\n",
    "print(group_means)\n",
    "\n",
    "# 2. Multi-column aggregation using .agg()\n",
    "agg_results = df_merged.groupby('membership').agg({\n",
    "    'amount': ['mean', 'max', 'std'],\n",
    "    'age': 'median'\n",
    "})\n",
    "print(\"\\nMulti-metric aggregations:\")\n",
    "print(agg_results)\n",
    "\n",
    "# 3. Advanced Feature Engineering using .transform()\n",
    "# (Subtracting group mean to center the purchase amount per customer tier)\n",
    "df_merged['amount_diff_tier_mean'] = df_merged['amount'] - df_merged.groupby('membership')['amount'].transform('mean')\n",
    "df_merged[['customer_id', 'membership', 'amount', 'amount_diff_tier_mean']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 📅 Step 7: Dates and Time Series Preprocessing\n",
    "Timestamps must be parsed and split into numerical factors to enable machine learning models to capture temporal trends. \n",
    "Additionally, we will construct lagged features (previous values) and rolling average features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>amount</th>\n",
       "      <th>lag_amount_1h</th>\n",
       "      <th>lag_amount_2h</th>\n",
       "      <th>target_lead_1h</th>\n",
       "      <th>rolling_mean_3h</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>timestamp</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2026-01-01 02:00:00</th>\n",
       "      <td>57.94</td>\n",
       "      <td>138.71</td>\n",
       "      <td>47.92</td>\n",
       "      <td>99.58</td>\n",
       "      <td>81.523333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-01-01 03:00:00</th>\n",
       "      <td>99.58</td>\n",
       "      <td>57.94</td>\n",
       "      <td>138.71</td>\n",
       "      <td>90.68</td>\n",
       "      <td>98.743333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-01-01 04:00:00</th>\n",
       "      <td>90.68</td>\n",
       "      <td>99.58</td>\n",
       "      <td>57.94</td>\n",
       "      <td>127.81</td>\n",
       "      <td>82.733333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-01-01 05:00:00</th>\n",
       "      <td>127.81</td>\n",
       "      <td>90.68</td>\n",
       "      <td>99.58</td>\n",
       "      <td>97.76</td>\n",
       "      <td>106.023333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-01-01 06:00:00</th>\n",
       "      <td>97.76</td>\n",
       "      <td>127.81</td>\n",
       "      <td>90.68</td>\n",
       "      <td>114.89</td>\n",
       "      <td>105.416667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     amount  lag_amount_1h  lag_amount_2h  target_lead_1h  \\\n",
       "timestamp                                                                   \n",
       "2026-01-01 02:00:00   57.94         138.71          47.92           99.58   \n",
       "2026-01-01 03:00:00   99.58          57.94         138.71           90.68   \n",
       "2026-01-01 04:00:00   90.68          99.58          57.94          127.81   \n",
       "2026-01-01 05:00:00  127.81          90.68          99.58           97.76   \n",
       "2026-01-01 06:00:00   97.76         127.81          90.68          114.89   \n",
       "\n",
       "                     rolling_mean_3h  \n",
       "timestamp                             \n",
       "2026-01-01 02:00:00        81.523333  \n",
       "2026-01-01 03:00:00        98.743333  \n",
       "2026-01-01 04:00:00        82.733333  \n",
       "2026-01-01 05:00:00       106.023333  \n",
       "2026-01-01 06:00:00       105.416667  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1. Parse dates and extract numerical features\n",
    "df_merged['timestamp'] = pd.to_datetime(df_merged['timestamp'])\n",
    "df_merged['hour'] = df_merged['timestamp'].dt.hour\n",
    "df_merged['day_of_week'] = df_merged['timestamp'].dt.dayofweek\n",
    "df_merged['is_weekend'] = np.where(df_merged['day_of_week'] >= 5, 1, 0)\n",
    "\n",
    "# 2. Build Time Series Shift Features (Lag/Lead)\n",
    "# For timeline alignment, we set index to timestamp and sort it\n",
    "df_ts = df_merged.set_index('timestamp').sort_index()\n",
    "\n",
    "# Lag features (creating features from past hours)\n",
    "df_ts['lag_amount_1h'] = df_ts['amount'].shift(1)\n",
    "df_ts['lag_amount_2h'] = df_ts['amount'].shift(2)\n",
    "\n",
    "# Lead feature (creating target column for tomorrow's prediction)\n",
    "df_ts['target_lead_1h'] = df_ts['amount'].shift(-1)\n",
    "\n",
    "# 3. Rolling Window Aggregations (Smoothing trends)\n",
    "df_ts['rolling_mean_3h'] = df_ts['amount'].rolling(window=3).mean()\n",
    "\n",
    "# Drop boundary NaNs generated by shifting\n",
    "df_ts = df_ts.dropna(subset=['lag_amount_1h', 'lag_amount_2h', 'target_lead_1h', 'rolling_mean_3h'])\n",
    "\n",
    "df_ts[['amount', 'lag_amount_1h', 'lag_amount_2h', 'target_lead_1h', 'rolling_mean_3h']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 🧠 Challenge Exercise: Volatility Tracking\n",
    "Can you calculate a **14-day rolling standard deviation** of the transaction `amount` to measure price volatility over time?\n",
    "\n",
    "Write your code in the cell below. When you are done, you can double-click the **Solution** markdown cell below to reveal the answer!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Write your code here to calculate 14-day rolling standard deviation:\n",
    "# Hint: use .rolling() with a window parameter of '14d' or 14 on the datetime-indexed df_ts"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 🔍 Click here to see the Solution!\n",
    "<details>\n",
    "<summary>Reveal Code</summary>\n",
    "\n",
    "```python\n",
    "# Calculate a 14-day rolling standard deviation\n",
    "df_ts['rolling_std_14d'] = df_ts['amount'].rolling(window='14d').std()\n",
    "print(df_ts[['amount', 'rolling_std_14d']].head(10))\n",
    "```\n",
    "</details>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 📊 Step 8: Exploratory Data Analysis & Seaborn Visualizations\n",
    "\n",
    "Before building models, we inspect our features and targets. Here, we address three critical preprocessing checkpoints:\n",
    "1. **Target Class Imbalance & The Accuracy Paradox**: We count target occurrences with `value_counts(normalize=True)`. If a dataset is heavily imbalanced (e.g., 99% legitimate, 1% fraud), a model that simply predicts 'Legitimate' for everything will have **99% accuracy**, but fails to catch any fraud. This is the **Accuracy Paradox**; it warns us to use Precision, Recall, F1-Score, and ROC-AUC metrics instead of simple accuracy during evaluation in the next class.\n",
    "2. **Outlier Detection**: We use boxplots to inspect ranges. Outliers skew models that minimize squared errors, so we must identify them (noting that boxplot whiskers extend to the minimum and maximum of the *non-outlier* data points, while outliers are plotted as individual points beyond).\n",
    "3. **Collinearity Check**: We plot correlation matrices to identify redundant features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fraud (Target) distribution (%):\n",
      "is_fraud\n",
      "0    95.94\n",
      "1     4.06\n",
      "Name: proportion, dtype: float64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 1. Target Imbalance Analysis\n",
    "fraud_dist = df_ts['is_fraud'].value_counts(normalize=True) * 100\n",
    "print(\"Fraud (Target) distribution (%):\")\n",
    "print(fraud_dist.round(2))\n",
    "\n",
    "# Create subplots\n",
    "fig, axes = plt.subplots(1, 2, figsize=(16, 6))\n",
    "\n",
    "# Boxplot for Outlier Detection across Groups\n",
    "sns.boxplot(\n",
    "    data=df_ts, \n",
    "    x='membership', \n",
    "    y='amount', \n",
    "    hue='is_fraud', \n",
    "    palette='muted', \n",
    "    ax=axes[0]\n",
    ")\n",
    "axes[0].set_title(\"Transaction Amounts by Membership & Fraud Status\")\n",
    "axes[0].set_xlabel(\"Membership Level\")\n",
    "axes[0].set_ylabel(\"Amount ($)\")\n",
    "\n",
    "# Correlation Heatmap for Feature Selection (Checking collinear features)\n",
    "# Isolating numeric variables\n",
    "numeric_cols = df_ts.select_dtypes(include=[np.number]).columns.tolist()\n",
    "# Drop ID indices and constants from correlation inspection\n",
    "cols_to_corr = [col for col in numeric_cols if col not in ['is_high_value']]\n",
    "corr_matrix = df_ts[cols_to_corr].corr()\n",
    "\n",
    "sns.heatmap(\n",
    "    corr_matrix, \n",
    "    annot=True, \n",
    "    cmap='coolwarm', \n",
    "    fmt='.2f', \n",
    "    linewidths=0.5, \n",
    "    ax=axes[1]\n",
    ")\n",
    "axes[1].set_title(\"Correlation Matrix for Feature Selection\")\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🚀 Step 9: Scikit-Learn Handover Split ($X$ and $y$)\n",
    "To conclude the session, let's perform the final step: splitting the clean DataFrame into a feature matrix $X$ and a target vector $y$, ready to be passed directly to `scikit-learn` estimators."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- Final Preprocessed Shapes for Scikit-Learn ---\n",
      "Feature Matrix X Shape: (197, 14) (Rows, Features)\n",
      "Target Vector y Shape: (197,) (Rows,)\n",
      "\n",
      "Features list (X):\n",
      "['amount', 'premium_fee', 'is_high_value', 'age', 'membership_encoded', 'gender_M', 'gender_Other', 'amount_diff_tier_mean', 'hour', 'day_of_week', 'is_weekend', 'lag_amount_1h', 'lag_amount_2h', 'rolling_mean_3h']\n",
      "\n",
      "First 3 rows of feature matrix X:\n"
     ]
    },
    {
     "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>amount</th>\n",
       "      <th>premium_fee</th>\n",
       "      <th>is_high_value</th>\n",
       "      <th>age</th>\n",
       "      <th>membership_encoded</th>\n",
       "      <th>gender_M</th>\n",
       "      <th>gender_Other</th>\n",
       "      <th>amount_diff_tier_mean</th>\n",
       "      <th>hour</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>is_weekend</th>\n",
       "      <th>lag_amount_1h</th>\n",
       "      <th>lag_amount_2h</th>\n",
       "      <th>rolling_mean_3h</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>timestamp</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2026-01-01 02:00:00</th>\n",
       "      <td>57.94</td>\n",
       "      <td>2.8970</td>\n",
       "      <td>0</td>\n",
       "      <td>28.0</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>-55.119903</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>138.71</td>\n",
       "      <td>47.92</td>\n",
       "      <td>81.523333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-01-01 03:00:00</th>\n",
       "      <td>99.58</td>\n",
       "      <td>4.9790</td>\n",
       "      <td>0</td>\n",
       "      <td>22.0</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>-13.479903</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>57.94</td>\n",
       "      <td>138.71</td>\n",
       "      <td>98.743333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2026-01-01 04:00:00</th>\n",
       "      <td>90.68</td>\n",
       "      <td>1.8136</td>\n",
       "      <td>0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>-22.379903</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>99.58</td>\n",
       "      <td>57.94</td>\n",
       "      <td>82.733333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     amount  premium_fee  is_high_value   age  \\\n",
       "timestamp                                                       \n",
       "2026-01-01 02:00:00   57.94       2.8970              0  28.0   \n",
       "2026-01-01 03:00:00   99.58       4.9790              0  22.0   \n",
       "2026-01-01 04:00:00   90.68       1.8136              0  35.0   \n",
       "\n",
       "                     membership_encoded  gender_M  gender_Other  \\\n",
       "timestamp                                                         \n",
       "2026-01-01 02:00:00                   0     False          True   \n",
       "2026-01-01 03:00:00                   0     False          True   \n",
       "2026-01-01 04:00:00                   0     False          True   \n",
       "\n",
       "                     amount_diff_tier_mean  hour  day_of_week  is_weekend  \\\n",
       "timestamp                                                                   \n",
       "2026-01-01 02:00:00             -55.119903     2            3           0   \n",
       "2026-01-01 03:00:00             -13.479903     3            3           0   \n",
       "2026-01-01 04:00:00             -22.379903     4            3           0   \n",
       "\n",
       "                     lag_amount_1h  lag_amount_2h  rolling_mean_3h  \n",
       "timestamp                                                           \n",
       "2026-01-01 02:00:00         138.71          47.92        81.523333  \n",
       "2026-01-01 03:00:00          57.94         138.71        98.743333  \n",
       "2026-01-01 04:00:00          99.58          57.94        82.733333  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Drop ID labels and string/date objects that are not encoded\n",
    "features_to_drop = [\n",
    "    'transaction_id',\n",
    "    'customer_id', \n",
    "    'category', # Raw string categorical - encoded counterpart is needed if we kept it\n",
    "    'membership', # Raw string categorical - encoded counterpart is membership_encoded\n",
    "    'signup_date', # Raw datetime string\n",
    "    'is_fraud', # Target label must be dropped from feature space X\n",
    "    'target_lead_1h', # Target label for regression forecasting\n",
    "    'spend_group' # Created for demonstration - dropping to keep X numerical\n",
    "]\n",
    "\n",
    "# Split into Feature Matrix X and Target vector y\n",
    "X = df_ts.drop(columns=features_to_drop)\n",
    "y = df_ts['is_fraud'] # Binary target for classification\n",
    "\n",
    "print(\"--- Final Preprocessed Shapes for Scikit-Learn ---\")\n",
    "print(f\"Feature Matrix X Shape: {X.shape} (Rows, Features)\")\n",
    "print(f\"Target Vector y Shape: {y.shape} (Rows,)\")\n",
    "print(\"\\nFeatures list (X):\")\n",
    "print(X.columns.tolist())\n",
    "print(\"\\nFirst 3 rows of feature matrix X:\")\n",
    "X.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 🏆 Step 10: Advanced Best Practices & Pandas 3.0 Roadmap\n",
    "\n",
    "Let's review core best practices to write high-performance code, followed by an overview of what is changing in **Pandas 3.0**.\n",
    "\n",
    "### 1. Pandas Performance Best Practices\n",
    "\n",
    "* **Vectorization > Apply > Loops**:\n",
    "  * Avoid using Python `for` loops or row-wise `.apply(axis=1)` where possible. They are slow because they execute Python code for every row.\n",
    "  * Use **vectorized operations** (built-in Pandas operations or NumPy functions like `np.where()`). These execute compiled C code under the hood.\n",
    "  \n",
    "* **Method Chaining**:\n",
    "  * Instead of writing multiple lines and creating temporary variables (which bloat RAM), wrap your operations in parentheses and chain them: \n",
    "    ```python\n",
    "    df_clean = (\n",
    "        df.drop_duplicates()\n",
    "          .fillna({'age': df['age'].median()})\n",
    "          .query(\"amount > 10\")\n",
    "    )\n",
    "    ```\n",
    "    This makes code highly readable and memory efficient.\n",
    "\n",
    "* **Avoiding `SettingWithCopyWarning`**:\n",
    "  * This warning occurs when you assign values using chained indexing, e.g., `df[df['age'] > 30]['name'] = 'Adult'`.\n",
    "  * Always use **`.loc`** for assignment: `df.loc[df['age'] > 30, 'name'] = 'Adult'` to ensure you are modifying the original DataFrame in-place.\n",
    "\n",
    "* **Read/Write Efficiency (Parquet > CSV)**:\n",
    "  * CSV is a text format; loading it requires Pandas to guess data types, which is slow and memory-intensive.\n",
    "  * Use **Parquet** (`.to_parquet()`, `pd.read_parquet()`). Parquet is a binary columnar format that preserves schemas/data types, compresses data, and loads up to **10x faster** than CSVs.\n",
    "\n",
    "* **Memory Reclamation**:\n",
    "  * In RAM-constrained cloud environments (like Google Colab), explicitly delete massive intermediate DataFrames and call the Python garbage collector to reclaim memory: \n",
    "    ```python\n",
    "    import gc\n",
    "    del large_temporary_df\n",
    "    gc.collect()\n",
    "    ```\n",
    "\n",
    "---\n",
    "\n",
    "### 🚀 The Roadmap: What's New in Pandas 3.0?\n",
    "\n",
    "Pandas 3.0 represents a major architectural upgrade, resolving several historical memory and execution speed limitations:\n",
    "\n",
    "1. **PyArrow Backend by Default**:\n",
    "   * Historically, Pandas used NumPy as its backend. Strings were stored as heavy Python `object` types.\n",
    "   * Pandas 3.0 transitions string and data type storage to **PyArrow** by default. This leads to **10x faster string operations**, drastically lower memory footprints, and native support for missing values across all data types (no more converting integers to floats just because there is a missing NaN value).\n",
    "\n",
    "2. **Copy-on-Write (CoW) Active by Default**:\n",
    "   * In Pandas 3.0, Copy-on-Write is enabled by default. Under CoW, any DataFrame derived from another sharing the same memory will only copy data when a modification is actually made.\n",
    "   * This **completely eliminates the confusing `SettingWithCopyWarning`** and prevents accidental deep copies of large arrays, yielding significant speed and memory gains.\n",
    "\n",
    "3. **Improved Type Safety & Deprecations**:\n",
    "   * Pandas 3.0 deprecates many legacy keyword arguments and enforces strict schema validations to clean up the API, making it more robust and aligned with standard data engineering pipelines."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 🎉 Congratulations! \n",
    "You have completed the Advanced Pandas & Seaborn masterclass. You are now equipped with advanced data preprocessing techniques, ready to write high-performance code, and prepared for the upcoming Scikit-Learn classical Machine Learning class!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 📚 Appendix: Computer Architecture & Data Engineering 101\n",
    "\n",
    "Understanding hardware-level execution helps explain why modern libraries like Polars and DuckDB outperform legacy tools. Here is a practical reference guide covering CPU internals, data engineering paradoxes, and hardware latency.\n",
    "\n",
    "---\n",
    "\n",
    "## 🧱 Part 1: CPU Internals (Cores, Threads, Concurrency, and Parallelism)\n",
    "\n",
    "### 🧩 Cores vs. Threads: The Hardware Basics\n",
    "![CPU Threading Concepts](cpu_threading_concepts.png)\n",
    "\n",
    "* **Core**: A physical, independent hardware processor inside the CPU that executes instructions.\n",
    "* **Thread**: A virtual sequence of instructions executed by software. A single core can manage multiple threads, but multi-threading does not automatically translate to simultaneous multi-core utilization.\n",
    "\n",
    "#### 🍽️ The Restaurant Analogy\n",
    "\n",
    "| Concept | What It Is | Restaurant Analogy |\n",
    "| :--- | :--- | :--- |\n",
    "| **Core** | Physical hardware processor inside the CPU. | A physical chef working in the kitchen. |\n",
    "| **Thread** | A stream of tasks executed by software. | A recipe/order ticket that needs to be prepared. |\n",
    "| **Single-Threading** | One task is executed at a time. | One chef working on exactly one order from start to finish. |\n",
    "| **Multi-Threading** | Multiple tasks are executed concurrently. | One chef juggling multiple orders (e.g. chopping vegetables for Order B while waiting for Order A's water to boil). |\n",
    "| **Multi-Core Processing** | Multiple physical processors execute tasks at the same time. | Multiple chefs working simultaneously on different orders. |\n",
    "\n",
    "---\n",
    "\n",
    "### ⏱️ Concurrency vs. Parallelism\n",
    "* **Concurrency (Multitasking Structure)**: Managing multiple tasks by starting, running, and completing them in overlapping time frames. Concurrency is about *structure*.\n",
    "* **Parallelism (Simultaneous Execution)**: Executing multiple tasks at the exact same physical millisecond. Parallelism is about *execution*.\n",
    "\n",
    "#### How Software Meets Hardware:\n",
    "1. **Time-Slicing (Concurrency on Single Core)**: If you run a multi-threaded program on a single-core CPU, the CPU rapidly switches back and forth between threads. To a human, they appear to run simultaneously, but the core is executing only one instruction at a time.\n",
    "2. **True Parallelism (Multi-Core Execution)**: If you run a multi-threaded program on a multi-core CPU, different physical cores can execute different threads at the exact same physical millisecond.\n",
    "\n",
    "---\n",
    "\n",
    "### 🐍 Python's Execution Limit: The GIL\n",
    "Standard Python (CPython) uses a **Global Interpreter Lock (GIL)**.\n",
    "* **The Constraint**: The GIL ensures that only one thread executes Python bytecode at a time, even if your machine has 16 or 32 cores.\n",
    "* **CPU-Bound Tasks**: For data processing and numerical computations (which happen in Python space), standard Python multi-threading **will not speed up your code**.\n",
    "* **I/O-Bound Tasks**: Multi-threading is highly effective for tasks where the CPU is idle, such as downloading files, fetching web pages (scraping), or writing to disk.\n",
    "* **Hyper-Threading**: A hardware-level feature (Intel/AMD) that allows a single core to hold two threads in its pipeline at once, switching between them instantly if one thread is stalled (e.g., waiting to fetch data from RAM).\n",
    "\n",
    "---\n",
    "\n",
    "## ⚙️ Execution Models compared: Pandas, Polars & DuckDB\n",
    "\n",
    "| Framework | Threading Model | Core Utilization | Underlying Engine |\n",
    "| :--- | :--- | :--- | :--- |\n",
    "| **Pandas** | Single-threaded by default. | Utilizes 1 core (unless calling pre-compiled NumPy C-libraries). | Python/C, restricted by the Global Interpreter Lock (GIL). |\n",
    "| **Polars** | Multi-threaded by default. | Spreads workloads across all available CPU cores. | Rust-native, using Apache Arrow and the Rayon parallel framework. |\n",
    "| **DuckDB** | Multi-threaded by default. | Divides queries across all available CPU cores. | C++ vectorized, push-based execution engine. |\n",
    "\n",
    "### 1. Pandas: The Single-Threaded Traditionalist\n",
    "Pandas runs sequentially. When filtering a DataFrame, it evaluates row 1, then row 2, and so on, using a single thread on a single core. It loads the entire dataset into RAM, eagerly executing each line, which causes high memory usage and leaves the remaining CPU cores idle.\n",
    "\n",
    "### 2. Polars: The Multi-Threaded Speedster\n",
    "Polars bypasses the Python GIL because its core is written in Rust. It utilizes a thread pool to split data into chunks and processes them in parallel across all cores. By using **Lazy Evaluation**, it compiles your code into an optimized execution plan (e.g., combining filters, skipping unused columns) before processing the data.\n",
    "\n",
    "### 3. DuckDB: The Vectorized Columnar Engine\n",
    "DuckDB is optimized for analytical SQL queries. It divides tables into chunks of roughly 100,000 rows and pushes them through a concurrent execution queue, assigning one worker thread per CPU core. Its **Vectorized Engine** processes arrays of values rather than row-by-row, and it can spill intermediate results to disk if memory limits are exceeded, allowing it to query datasets larger than your physical RAM.\n",
    "\n",
    "---\n",
    "\n",
    "## 🔍 Part 2: Data Engineering Paradoxes\n",
    "\n",
    "### 📊 Paradox 1: Row-Oriented vs. Column-Oriented (CSV vs. Parquet)\n",
    "* **The Confusion**: *\"If CSV and Parquet both store the exact same table of data, why does reading a Parquet file use 100x less memory and execute 50x faster, but writing to it is slower?\"*\n",
    "* **The Hardware Reality**: Row-oriented databases store record columns adjacent to one another on disk. Columnar engines group data by column. If you only query 2 columns out of a 100-column dataset, a columnar format reads only 2% of the bytes on disk. Writing is slower because the engine has to buffer, transpose, and compress the columns in memory before flush.\n",
    "* **🍽️ The Analogy (The Phonebook)**: If you need to find the average age of everyone in a town, row-oriented is like reading every folder in city hall (finding names, addresses, phone numbers, and ages). Columnar is like being handed a single sheet containing *only* ages.\n",
    "\n",
    "### ⏳ Paradox 2: Eager vs. Lazy Evaluation (Pandas vs. Polars)\n",
    "* **The Confusion**: *\"Why does waiting until the very end of the script to execute (Lazy) run faster than executing line-by-line?\"*\n",
    "* **The Hardware Reality**: Eager execution does not know what your next line of code will be, so it must compute and allocate memory for intermediate operations immediately. Lazy execution compiles all code blocks into a logical query plan, optimizes it (e.g. projecting only required columns, pushing filter conditions to the earliest step), and compiles it into a single parallel loop.\n",
    "* **🍽️ The Analogy (The Pantry Trip)**: Eager is running to the pantry once for flour, coming back, running again for sugar, and coming back. Lazy is reading the whole recipe, realizing you need three things, and making one organized trip.\n",
    "\n",
    "### 💾 Paradox 3: The 10x RAM Rule (In-Memory vs. Disk-Spilling)\n",
    "* **The Confusion**: *\"Why does my 5 GB dataset crash my computer with an Out of Memory error when my machine has 16 GB of RAM?\"*\n",
    "* **The Hardware Reality**: Compressed datasets expand by 3-5x when decompressed into memory pointers. Furthermore, operations like Joins and Merges create temporary hash tables and copies of intermediate data in RAM, ballooning memory usage.\n",
    "* **🍽️ The Analogy (The IKEA Wardrobe)**: A flat-packed IKEA wardrobe fits in your car trunk (5 GB disk size). But once you open the box and lay all the parts on your floor to assemble it, it takes up the entire living room (15 GB memory footprint).\n",
    "\n",
    "### 🔗 Paradox 4: Views vs. Copies (Chained Indexing)\n",
    "* **The Confusion**: *\"Why does modifying a subset of my DataFrame sometimes edit my original data, but other times do absolutely nothing except print a warning?\"*\n",
    "* **The Hardware Reality**: Slicing an array can return a **View** (pointing to the original array's memory address with offsets) or a **Copy** (allocating new memory and copying the values). Chained indexing causes Pandas' compiler to lose track of whether a view or copy was returned, leading to unpredictable writes.\n",
    "* **🍽️ The Analogy (The Google Doc)**: A View is sending a collaborator a shared link to your Google Doc; edits show up immediately. A Copy is printing the document and mailing it to them; changes they make on paper do not affect your digital master.\n",
    "\n",
    "---\n",
    "\n",
    "## 📡 Part 3: Deep Hardware Architecture Concepts\n",
    "\n",
    "### ⚡ Cache Locality & SIMD (Desk Drawer vs. Library Warehouse)\n",
    "RAM is incredibly slow compared to a CPU core. To prevent the CPU from sitting idle waiting for data (memory stall), CPUs use tiny, ultra-fast **L1, L2, and L3 caches** on the chip.\n",
    "* **The Problem**: If your data is scattered all over RAM (like a standard Python list containing object pointers), the CPU must fetch each element from slow RAM. This is called a **Cache Miss**.\n",
    "* **The Columnar Solution**: If your data is stored in contiguous, aligned blocks of memory (like NumPy, Arrow, Polars, and DuckDB), the CPU fetches a whole chunk of data into L1 Cache once. It then uses **SIMD (Single Instruction, Multiple Data)** hardware registers to process multiple values (e.g. adding 8 integers) in a single CPU cycle.\n",
    "* **🍽️ The Analogy**: If you need to write 10 letters, keeping all your pens in a cup on your desk (Cache Locality) is fast. If every time you write a letter you have to walk to a warehouse in the next room to fetch a pen (Cache Miss / RAM latency), it takes all day.\n",
    "\n",
    "### ⏱️ The Latency Scale (Visualizing CPU Speeds)\n",
    "To write high-performance data pipelines, developers must appreciate the staggering scale differences between CPU speed, RAM, Disk, and Network. If **1 CPU Cycle takes 1 second** (equivalent to a human blink), the hardware operations scale as follows:\n",
    "\n",
    "| Operation | Time in CPU Scale | Human Analogy |\n",
    "| :--- | :--- | :--- |\n",
    "| **1 CPU Cycle** | 1 Second | A single blink of an eye. |\n",
    "| **L1 Cache Access** | 0.5 Seconds | Grabbing a notebook on your desk. |\n",
    "| **L3 Cache Access** | 15 Seconds | Grabbing a book from a nearby shelf. |\n",
    "| **Main Memory (RAM)** | 2 Minutes | Walking down the hall to the water cooler. |\n",
    "| **Solid State Drive (SSD)** | 2 Days | Taking a weekend trip to another city. |\n",
    "| **Mechanical Hard Drive (HDD)** | 1.5 Months | Going on a long summer vacation. |\n",
    "| **Network Call (Internet)** | 2 Years | Going to university and earning a degree. |\n",
    "\n",
    "> **Data Engineering Rule of Thumb**: Always optimize to minimize **Network Calls** and **Disk Reads** first. A single network query or reading from disk dominates your performance bottleneck, regardless of how fast your CPU code is.\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python (vm)",
   "language": "python",
   "name": "python-vm"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}