{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "72f8aa6a",
   "metadata": {},
   "source": [
    "# 13 · Capstone: Solving a Kaggle Challenge\n",
    "\n",
    "[Kaggle](https://www.kaggle.com) hosts machine-learning competitions. This capstone works through one of\n",
    "its classic *Getting Started* challenges, **House Prices: Advanced Regression Techniques**: predict the sale\n",
    "price of homes in Ames, Iowa from 79 descriptive variables. The workflow is the one every serious project\n",
    "follows: understand the data and the metric, set a baseline, engineer features, validate properly, compare\n",
    "models, and only then predict.\n",
    "\n",
    "**You will learn**\n",
    "1. Where the data comes from and whether you may use it (licensing)\n",
    "2. The competition metric, and why it is computed on log prices\n",
    "3. Exploration that changes decisions: skew, outliers, the strongest predictors\n",
    "4. Baselines, then a scikit-learn `Pipeline` with feature engineering\n",
    "5. Cross-validation, model comparison and blending\n",
    "6. What drives the predictions, and how you would submit to Kaggle"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2dd4156c",
   "metadata": {},
   "source": [
    "## 1. The data, and whether we may use it\n",
    "Kaggle datasets carry their own licences, and **competition data is governed by the competition's rules**,\n",
    "which usually allow use only for the competition and forbid redistribution. A course should therefore not\n",
    "re-host Kaggle's competition files.\n",
    "\n",
    "This competition's data is a subset of the **Ames Housing dataset**, which Dean De Cock compiled\n",
    "specifically for teaching and published with the *Journal of Statistics Education* (2011). The full dataset\n",
    "(2,930 sales from 2006 to 2010) is openly available, here through [OpenML](https://www.openml.org/d/43926)\n",
    "(licence: public). We use that, credit it, and leave the competition files on Kaggle, where you can\n",
    "download them with your own account to submit.\n",
    "\n",
    "> De Cock, D. (2011). *Ames, Iowa: Alternative to the Boston Housing Data as an End of Semester Regression\n",
    "> Project.* Journal of Statistics Education, 19(3)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "340bed08",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:29:50.399669Z",
     "iopub.status.busy": "2026-10-17T23:29:50.399284Z",
     "iopub.status.idle": "2026-10-17T23:29:53.637253Z",
     "shell.execute_reply": "2026-10-17T23:29:53.635583Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2930, 81)\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>Neighborhood</th>\n",
       "      <th>Gr_Liv_Area</th>\n",
       "      <th>Overall_Qual</th>\n",
       "      <th>Year_Built</th>\n",
       "      <th>Total_Bsmt_SF</th>\n",
       "      <th>Garage_Cars</th>\n",
       "      <th>Sale_Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>North_Ames</td>\n",
       "      <td>1656</td>\n",
       "      <td>Above_Average</td>\n",
       "      <td>1960</td>\n",
       "      <td>1080</td>\n",
       "      <td>2</td>\n",
       "      <td>215000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>North_Ames</td>\n",
       "      <td>896</td>\n",
       "      <td>Average</td>\n",
       "      <td>1961</td>\n",
       "      <td>882</td>\n",
       "      <td>1</td>\n",
       "      <td>105000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>North_Ames</td>\n",
       "      <td>1329</td>\n",
       "      <td>Above_Average</td>\n",
       "      <td>1958</td>\n",
       "      <td>1329</td>\n",
       "      <td>1</td>\n",
       "      <td>172000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>North_Ames</td>\n",
       "      <td>2110</td>\n",
       "      <td>Good</td>\n",
       "      <td>1968</td>\n",
       "      <td>2110</td>\n",
       "      <td>2</td>\n",
       "      <td>244000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Gilbert</td>\n",
       "      <td>1629</td>\n",
       "      <td>Average</td>\n",
       "      <td>1997</td>\n",
       "      <td>928</td>\n",
       "      <td>2</td>\n",
       "      <td>189900</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Neighborhood  Gr_Liv_Area   Overall_Qual  Year_Built  Total_Bsmt_SF  \\\n",
       "0   North_Ames         1656  Above_Average        1960           1080   \n",
       "1   North_Ames          896        Average        1961            882   \n",
       "2   North_Ames         1329  Above_Average        1958           1329   \n",
       "3   North_Ames         2110           Good        1968           2110   \n",
       "4      Gilbert         1629        Average        1997            928   \n",
       "\n",
       "   Garage_Cars  Sale_Price  \n",
       "0            2      215000  \n",
       "1            1      105000  \n",
       "2            1      172000  \n",
       "3            2      244000  \n",
       "4            2      189900  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np, pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.datasets import fetch_openml\n",
    "\n",
    "ames = fetch_openml(data_id=43926, as_frame=True, parser=\"auto\").frame   # downloaded once, then cached\n",
    "print(ames.shape)\n",
    "ames[[\"Neighborhood\", \"Gr_Liv_Area\", \"Overall_Qual\", \"Year_Built\", \"Total_Bsmt_SF\", \"Garage_Cars\", \"Sale_Price\"]].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0fc520d7",
   "metadata": {},
   "source": [
    "## 2. The metric\n",
    "The competition scores the **root mean squared error between the logarithms** of predicted and actual\n",
    "prices. On a log scale an error of 10% costs the same for a cheap and an expensive house, which matches\n",
    "how a buyer thinks about a bad estimate. So we model `log(price)` directly.\n",
    "\n",
    "We hold out 25% of the sales as a private \"leaderboard\" that is touched only once, at the end."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "2ea5ec5d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:29:53.642435Z",
     "iopub.status.busy": "2026-10-17T23:29:53.641890Z",
     "iopub.status.idle": "2026-10-17T23:29:54.341335Z",
     "shell.execute_reply": "2026-10-17T23:29:54.339661Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train 2197, held-out test 733\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x320 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "def rmse_log(y_true_log, y_pred_log):\n",
    "    return float(np.sqrt(np.mean((np.asarray(y_true_log) - np.asarray(y_pred_log)) ** 2)))\n",
    "\n",
    "df = ames.copy()\n",
    "df[\"log_price\"] = np.log(df.pop(\"Sale_Price\"))\n",
    "train, test = train_test_split(df, test_size=0.25, random_state=42)\n",
    "print(f\"train {len(train)}, held-out test {len(test)}\")\n",
    "\n",
    "fig, ax = plt.subplots(1, 2, figsize=(10, 3.2))\n",
    "np.exp(train[\"log_price\"]).plot.hist(bins=50, ax=ax[0], title=\"sale price ($): right-skewed\")\n",
    "train[\"log_price\"].plot.hist(bins=50, ax=ax[1], title=\"log price: close to normal\", color=\"#22c55e\")\n",
    "plt.tight_layout(); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2b57189",
   "metadata": {},
   "source": [
    "## 3. Exploration that changes decisions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "16cb3836",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:29:54.347010Z",
     "iopub.status.busy": "2026-10-17T23:29:54.346513Z",
     "iopub.status.idle": "2026-10-17T23:29:55.376959Z",
     "shell.execute_reply": "2026-10-17T23:29:55.375164Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "strongest numeric correlations with log price:\n",
      " Gr_Liv_Area       0.68\n",
      "Garage_Cars       0.66\n",
      "Garage_Area       0.63\n",
      "Total_Bsmt_SF     0.61\n",
      "Year_Built        0.60\n",
      "First_Flr_SF      0.59\n",
      "Full_Bath         0.57\n",
      "Year_Remod_Add    0.57\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "num = train.select_dtypes(\"number\")\n",
    "top = num.corr()[\"log_price\"].drop(\"log_price\").sort_values(key=abs, ascending=False).head(8)\n",
    "print(\"strongest numeric correlations with log price:\\n\", top.round(2).to_string())\n",
    "\n",
    "fig, ax = plt.subplots(1, 2, figsize=(11, 3.6))\n",
    "ax[0].scatter(train[\"Gr_Liv_Area\"], np.exp(train[\"log_price\"]), s=6, alpha=.5)\n",
    "big = train[\"Gr_Liv_Area\"] > 4000\n",
    "ax[0].scatter(train.loc[big, \"Gr_Liv_Area\"], np.exp(train.loc[big, \"log_price\"]), s=30, color=\"#ef4444\", label=\"> 4000 sq ft\")\n",
    "ax[0].set_xlabel(\"above-ground living area (sq ft)\"); ax[0].set_ylabel(\"price ($)\"); ax[0].legend()\n",
    "order = train.groupby(\"Neighborhood\", observed=True)[\"log_price\"].median().sort_values().index\n",
    "ax[1].boxplot([train.loc[train[\"Neighborhood\"] == n, \"log_price\"] for n in order], showfliers=False)\n",
    "ax[1].set_xticks(range(1, len(order) + 1)); ax[1].set_xticklabels(order, rotation=90, fontsize=6)\n",
    "ax[1].set_title(\"log price by neighbourhood (sorted by median)\")\n",
    "plt.tight_layout(); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf94c685",
   "metadata": {},
   "source": [
    "Two findings shape the model:\n",
    "- **Size and quality dominate**, and neighbourhood shifts the whole price level.\n",
    "- A few **very large houses sold at low prices** (red, partial sales of unusual properties). De Cock's own\n",
    "  documentation recommends removing houses above 4,000 sq ft for teaching models; we drop them from\n",
    "  *training* only, never from the test set."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1142b15b",
   "metadata": {},
   "source": [
    "## 4. Baselines\n",
    "Every model must beat something simple. The median price, and a one-variable regression on living area:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "70fc418b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:29:55.382054Z",
     "iopub.status.busy": "2026-10-17T23:29:55.381630Z",
     "iopub.status.idle": "2026-10-17T23:29:55.661588Z",
     "shell.execute_reply": "2026-10-17T23:29:55.660234Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "predict the median        : RMSE(log) 0.4244\n",
      "log(area) linear regression: RMSE(log) 0.2797\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "train = train[train[\"Gr_Liv_Area\"] <= 4000]\n",
    "y_train, y_test = train[\"log_price\"], test[\"log_price\"]\n",
    "\n",
    "print(f\"predict the median        : RMSE(log) {rmse_log(y_test, np.full(len(test), y_train.median())):.4f}\")\n",
    "lr = LinearRegression().fit(np.log(train[[\"Gr_Liv_Area\"]]), y_train)\n",
    "print(f\"log(area) linear regression: RMSE(log) {rmse_log(y_test, lr.predict(np.log(test[['Gr_Liv_Area']]))):.4f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1076d974",
   "metadata": {},
   "source": [
    "## 5. Feature engineering inside a Pipeline\n",
    "Feature engineering written as a function inside a scikit-learn `Pipeline` is applied identically to\n",
    "training data, validation folds and new houses, which prevents the most common competition bug: leaking\n",
    "information from the data you evaluate on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e5ecd9e3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:29:55.676272Z",
     "iopub.status.busy": "2026-10-17T23:29:55.672121Z",
     "iopub.status.idle": "2026-10-17T23:29:55.800197Z",
     "shell.execute_reply": "2026-10-17T23:29:55.797804Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "95 columns after engineering; new: ['Overall_Qual_n', 'Exter_Qual_n', 'Kitchen_Qual_n', 'Bsmt_Qual_n', 'Heating_QC_n', 'Fireplace_Qu_n', 'Garage_Qual_n', 'Total_SF', 'Total_Baths', 'Age_At_Sale', 'Years_Since_Remodel', 'Qual_x_Area', 'log_Gr_Liv_Area', 'log_Total_SF', 'log_Lot_Area']\n"
     ]
    }
   ],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.compose import ColumnTransformer, make_column_selector\n",
    "from sklearn.preprocessing import FunctionTransformer, OneHotEncoder, StandardScaler\n",
    "\n",
    "QUALITY = [\"Very_Poor\", \"Poor\", \"Fair\", \"Below_Average\", \"Average\", \"Above_Average\", \"Good\",\n",
    "           \"Very_Good\", \"Excellent\", \"Very_Excellent\"]\n",
    "GRADE = {\"No_Basement\": 0, \"No_Garage\": 0, \"No_Fireplace\": 0, \"No_Pool\": 0, \"Poor\": 1, \"Fair\": 2,\n",
    "         \"Typical\": 3, \"Good\": 4, \"Excellent\": 5}\n",
    "\n",
    "def engineer(X: pd.DataFrame) -> pd.DataFrame:\n",
    "    X = X.drop(columns=[\"log_price\"], errors=\"ignore\").copy()\n",
    "    X[\"Overall_Qual_n\"] = X[\"Overall_Qual\"].astype(str).map({q: i for i, q in enumerate(QUALITY, 1)})\n",
    "    for col in [\"Exter_Qual\", \"Kitchen_Qual\", \"Bsmt_Qual\", \"Heating_QC\", \"Fireplace_Qu\", \"Garage_Qual\"]:\n",
    "        X[col + \"_n\"] = X[col].astype(str).map(GRADE).fillna(0)\n",
    "    X[\"Total_SF\"] = X[\"Total_Bsmt_SF\"] + X[\"First_Flr_SF\"] + X[\"Second_Flr_SF\"]\n",
    "    X[\"Total_Baths\"] = X[\"Full_Bath\"] + 0.5 * X[\"Half_Bath\"] + X[\"Bsmt_Full_Bath\"] + 0.5 * X[\"Bsmt_Half_Bath\"]\n",
    "    X[\"Age_At_Sale\"] = X[\"Year_Sold\"] - X[\"Year_Built\"]\n",
    "    X[\"Years_Since_Remodel\"] = X[\"Year_Sold\"] - X[\"Year_Remod_Add\"]\n",
    "    X[\"Qual_x_Area\"] = X[\"Overall_Qual_n\"] * X[\"Gr_Liv_Area\"]\n",
    "    for col in [\"Gr_Liv_Area\", \"Total_SF\", \"Lot_Area\"]:\n",
    "        X[\"log_\" + col] = np.log1p(X[col])\n",
    "    return X\n",
    "\n",
    "fe = FunctionTransformer(engineer)\n",
    "X_tr_preview = engineer(train)\n",
    "print(f\"{X_tr_preview.shape[1]} columns after engineering; new: \"\n",
    "      f\"{[c for c in X_tr_preview.columns if c not in ames.columns]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "aad1fa7a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:29:55.807012Z",
     "iopub.status.busy": "2026-10-17T23:29:55.806471Z",
     "iopub.status.idle": "2026-10-17T23:29:56.159596Z",
     "shell.execute_reply": "2026-10-17T23:29:56.157456Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import RidgeCV\n",
    "from sklearn.ensemble import HistGradientBoostingRegressor\n",
    "\n",
    "numeric = make_column_selector(dtype_include=\"number\")\n",
    "categorical = make_column_selector(dtype_exclude=\"number\")\n",
    "\n",
    "ridge = make_pipeline(\n",
    "    fe,\n",
    "    ColumnTransformer([(\"num\", StandardScaler(), numeric),\n",
    "                       (\"cat\", OneHotEncoder(handle_unknown=\"ignore\", min_frequency=10), categorical)]),\n",
    "    RidgeCV(alphas=np.logspace(-1, 3, 30)),\n",
    ")\n",
    "\n",
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "\n",
    "# Trees take categories as integer codes. The encoder is fitted on the training data only, so a category\n",
    "# always gets the same code in training, validation and test (codes taken from each DataFrame's own\n",
    "# pandas categories would not be consistent), and an unseen category becomes \"missing\".\n",
    "cat_cols = list(X_tr_preview.select_dtypes(exclude=\"number\").columns)\n",
    "gbm = make_pipeline(\n",
    "    fe,\n",
    "    ColumnTransformer([(\"cat\", OrdinalEncoder(handle_unknown=\"use_encoded_value\", unknown_value=np.nan,\n",
    "                                              encoded_missing_value=np.nan), cat_cols)],\n",
    "                      remainder=\"passthrough\", verbose_feature_names_out=False).set_output(transform=\"pandas\"),\n",
    "    HistGradientBoostingRegressor(learning_rate=0.05, max_iter=600, max_leaf_nodes=15, min_samples_leaf=10,\n",
    "                                  l2_regularization=1.0, categorical_features=cat_cols, random_state=0),\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "447687f2",
   "metadata": {},
   "source": [
    "## 6. Cross-validation and comparison\n",
    "Five-fold cross-validation on the training set estimates how each model would score on unseen houses,\n",
    "without touching the held-out test set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bae28ed9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:29:56.165997Z",
     "iopub.status.busy": "2026-10-17T23:29:56.165422Z",
     "iopub.status.idle": "2026-10-17T23:30:21.388557Z",
     "shell.execute_reply": "2026-10-17T23:30:21.386660Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ridge (linear)       CV RMSE(log) 0.1171\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gradient boosting    CV RMSE(log) 0.1188\n",
      "best blend: 0.55 x Ridge + 0.45 x boosting -> CV RMSE(log) 0.1137\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold, cross_val_predict\n",
    "\n",
    "cv = KFold(5, shuffle=True, random_state=0)\n",
    "oof = {}\n",
    "for name, model in {\"Ridge (linear)\": ridge, \"Gradient boosting\": gbm}.items():\n",
    "    oof[name] = cross_val_predict(model, train, y_train, cv=cv)\n",
    "    print(f\"{name:<20} CV RMSE(log) {rmse_log(y_train, oof[name]):.4f}\")\n",
    "\n",
    "weights = np.linspace(0, 1, 21)\n",
    "blend_scores = [rmse_log(y_train, w * oof[\"Ridge (linear)\"] + (1 - w) * oof[\"Gradient boosting\"]) for w in weights]\n",
    "w_best = float(weights[int(np.argmin(blend_scores))])\n",
    "print(f\"best blend: {w_best:.2f} x Ridge + {1 - w_best:.2f} x boosting -> CV RMSE(log) {min(blend_scores):.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "0531d3ca",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:30:21.392136Z",
     "iopub.status.busy": "2026-10-17T23:30:21.391770Z",
     "iopub.status.idle": "2026-10-17T23:30:21.629543Z",
     "shell.execute_reply": "2026-10-17T23:30:21.627416Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x320 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6, 3.2))\n",
    "plt.plot(weights, blend_scores, \"o-\"); plt.axvline(w_best, ls=\"--\", color=\"grey\")\n",
    "plt.xlabel(\"weight on Ridge\"); plt.ylabel(\"CV RMSE (log)\"); plt.title(\"Blending two different models\")\n",
    "plt.tight_layout(); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c1167f33",
   "metadata": {},
   "source": [
    "The linear model and the tree ensemble make different kinds of mistakes, so averaging them beats either\n",
    "one alone. That is the most reliable trick in tabular competitions.\n",
    "\n",
    "## 7. The final, one-time evaluation\n",
    "Fit both models on the whole training set and score the held-out houses exactly once."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c04dbadd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:30:21.633557Z",
     "iopub.status.busy": "2026-10-17T23:30:21.633182Z",
     "iopub.status.idle": "2026-10-17T23:30:27.443418Z",
     "shell.execute_reply": "2026-10-17T23:30:27.441476Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "held-out RMSE(log), lower is better:\n",
      " Blend                   0.1101\n",
      "Gradient boosting       0.1108\n",
      "Ridge (linear)          0.1167\n",
      "log(area) regression    0.2797\n",
      "median baseline         0.4244\n",
      "\n",
      "blend: median absolute error 5.2% of the price; 80% of houses within 9.8%\n"
     ]
    }
   ],
   "source": [
    "ridge.fit(train, y_train); gbm.fit(train, y_train)\n",
    "pred = {\"Ridge (linear)\": ridge.predict(test), \"Gradient boosting\": gbm.predict(test)}\n",
    "pred[\"Blend\"] = w_best * pred[\"Ridge (linear)\"] + (1 - w_best) * pred[\"Gradient boosting\"]\n",
    "board = pd.Series({\"median baseline\": rmse_log(y_test, np.full(len(test), y_train.median())),\n",
    "                   \"log(area) regression\": rmse_log(y_test, lr.predict(np.log(test[[\"Gr_Liv_Area\"]]))),\n",
    "                   **{k: rmse_log(y_test, v) for k, v in pred.items()}}).sort_values()\n",
    "print(\"held-out RMSE(log), lower is better:\\n\", board.round(4).to_string())\n",
    "err = np.exp(pred[\"Blend\"]) / np.exp(y_test) - 1\n",
    "print(f\"\\nblend: median absolute error {np.median(np.abs(err)):.1%} of the price; \"\n",
    "      f\"80% of houses within {np.quantile(np.abs(err), .8):.1%}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "06e64fa9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:30:27.447639Z",
     "iopub.status.busy": "2026-10-17T23:30:27.447247Z",
     "iopub.status.idle": "2026-10-17T23:31:30.218172Z",
     "shell.execute_reply": "2026-10-17T23:31:30.216011Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x380 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 2, figsize=(11, 3.8))\n",
    "ax[0].scatter(np.exp(y_test), np.exp(pred[\"Blend\"]), s=8, alpha=.6)\n",
    "lims = [np.exp(y_test).min(), np.exp(y_test).max()]\n",
    "ax[0].plot(lims, lims, \"--\", color=\"grey\"); ax[0].set_xlabel(\"actual price ($)\"); ax[0].set_ylabel(\"predicted ($)\")\n",
    "ax[0].set_title(\"held-out houses\")\n",
    "\n",
    "from sklearn.inspection import permutation_importance\n",
    "X_test_only = test.drop(columns=\"log_price\")\n",
    "imp = permutation_importance(gbm, X_test_only, y_test, n_repeats=5, random_state=0, scoring=\"neg_root_mean_squared_error\")\n",
    "top = pd.Series(imp.importances_mean, index=X_test_only.columns).sort_values().tail(10)\n",
    "top.plot.barh(ax=ax[1], color=\"#4f46e5\"); ax[1].set_title(\"what the boosting model relies on (permutation)\")\n",
    "plt.tight_layout(); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1096c24f",
   "metadata": {},
   "source": [
    "## 8. Submitting to Kaggle\n",
    "A Kaggle submission is a CSV with one row per house in the competition's test file: `Id,SalePrice`. This is\n",
    "how you would produce it:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "df600d19",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-17T23:31:30.223038Z",
     "iopub.status.busy": "2026-10-17T23:31:30.222519Z",
     "iopub.status.idle": "2026-10-17T23:31:30.234023Z",
     "shell.execute_reply": "2026-10-17T23:31:30.232074Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " Id  SalePrice\n",
      "  1   163755.0\n",
      "  2   109983.0\n",
      "  3   188201.0\n",
      "  4   123558.0\n",
      "  5   106777.0\n"
     ]
    }
   ],
   "source": [
    "submission = pd.DataFrame({\"Id\": range(1, len(test) + 1), \"SalePrice\": np.exp(pred[\"Blend\"]).round(0)})\n",
    "print(submission.head().to_string(index=False))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7b7f242",
   "metadata": {},
   "source": [
    "To compete for real:\n",
    "1. Create a Kaggle account, open the *House Prices: Advanced Regression Techniques* competition and\n",
    "   **accept its rules**; then download `train.csv` and `test.csv` (or use the `kaggle` command-line tool).\n",
    "2. The competition files use different column spellings (`GrLivArea` instead of `Gr_Liv_Area`) and some\n",
    "   different category codes, and they mark \"no basement\" and similar as missing values. Rename the columns,\n",
    "   fill those values, and reuse the same pipeline.\n",
    "3. Fit on all of Kaggle's training data, predict `test.csv`, write `submission.csv`, and upload it.\n",
    "\n",
    "Scores around 0.12 RMSE(log) are typical for a solid blended model like this one; the top of the\n",
    "leaderboard is near 0.10, reached with careful feature work and many blended models.\n",
    "\n",
    "## What made the difference\n",
    "| Step | Why it mattered |\n",
    "|---|---|\n",
    "| reading the licence | legal, reusable course material |\n",
    "| modelling log price | matches the metric and tames the skew |\n",
    "| removing the documented outliers from training | a handful of huge, cheap houses distort a linear model |\n",
    "| features in a Pipeline | identical processing everywhere, no leakage |\n",
    "| cross-validation before the test set | an honest estimate, and the test set stays a true final exam |\n",
    "| blending different model families | their errors partly cancel |\n",
    "\n",
    "## Exercises\n",
    "1. Add `Neighborhood` target encoding (mean log price per neighbourhood, computed inside the cross-validation folds only) and measure the change.\n",
    "2. Tune the boosting model's `learning_rate` and `max_leaf_nodes` with `RandomizedSearchCV`.\n",
    "3. Use `Longitude` and `Latitude` (included in this dataset) to add a \"distance to the town centre\" feature."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
