{
 "cells": [
  {
   "attachments": {
    "logo.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "4ed0a233",
   "metadata": {},
   "source": [
    "![NVIDIA Structured Data Models logo](attachment:logo.png)\n",
    "\n",
    "# KumoTabular on Vehicle (OpenML 54) Walkthrough\n",
    "\n",
    "Structured Data Models (`sdm`) provides pretrained models, tensor types, and data processing for structured data.\n",
    "\n",
    "This notebook runs **Kumo Tabular** from the [Structured Data Models library](https://github.com/NVIDIA/structured-data-models) using pretrained weights from [Hugging Face](https://huggingface.co/collections/nvidia/nvidia-kumo-structured), on a **classification** task with a Vehicle dataset from **OpenML**. \n",
    "\n",
    "**A GPU is not required** to run this notebook. \n",
    "\n",
    "For tensor semantics and model APIs, refer to the [SDM documentation](https://nvidia.github.io/structured-data-models/overview.html)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "561d19ec",
   "metadata": {},
   "source": [
    "## 1. Install"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "72a2961c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-20T17:29:08.003338Z",
     "iopub.status.busy": "2026-10-20T17:29:08.003162Z",
     "iopub.status.idle": "2026-10-20T17:29:11.008662Z",
     "shell.execute_reply": "2026-10-20T17:29:11.007846Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install --quiet structured-data-models==0.2.1 pandas scikit-learn openml==0.15.1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e94f92c0",
   "metadata": {},
   "source": [
    "## 2. Load the dataset\n",
    "\n",
    "**Dataset:** [Vehicle (OpenML 54)](https://www.openml.org/d/54), using [task 53](https://www.openml.org/t/53)\n",
    "\n",
    "**Task:** Four-class classification\n",
    "\n",
    "The target `Class` identifies the vehicle pictured by a silhouette: `bus`, `Opel`, `Saab`, or `van` (846 rows and 18 numeric features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "bfdc049a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-20T17:29:11.011444Z",
     "iopub.status.busy": "2026-10-20T17:29:11.011255Z",
     "iopub.status.idle": "2026-10-20T17:29:20.115356Z",
     "shell.execute_reply": "2026-10-20T17:29:20.114354Z"
    }
   },
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "\n",
    "import numpy as np\n",
    "import openml\n",
    "import pandas as pd\n",
    "\n",
    "openml.config.set_root_cache_directory(str(Path.cwd() / \".cache\" / \"openml\"))\n",
    "task = openml.tasks.get_task(53)\n",
    "X, y, _, _ = task.get_dataset().get_data(\n",
    "    dataset_format=\"dataframe\", target=task.target_name\n",
    ")\n",
    "_, test = task.get_train_test_split_indices(fold=0)\n",
    "_, validation = task.get_train_test_split_indices(fold=1)\n",
    "train = np.setdiff1d(np.arange(len(X)), np.r_[test, validation])\n",
    "classes = sorted(y.unique())\n",
    "df = X.assign(target=pd.Categorical(y, categories=classes))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b2d96a3",
   "metadata": {},
   "source": [
    "## 3. Fit Labeled Context"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "91fccd58",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-20T17:29:20.118190Z",
     "iopub.status.busy": "2026-10-20T17:29:20.117887Z",
     "iopub.status.idle": "2026-10-20T17:29:31.623498Z",
     "shell.execute_reply": "2026-10-20T17:29:31.623072Z"
    }
   },
   "outputs": [],
   "source": [
    "import sdm\n",
    "import torch\n",
    "\n",
    "config = {\"size\": \"small\", \"num_estimators\": 8}\n",
    "# For medium or large, change config[\"size\"] to \"medium\" or \"large\".\n",
    "# config[\"num_estimators\"] controls ensemble size; reduce it to use less memory\n",
    "\n",
    "device = torch.device(\"cpu\")\n",
    "torch.set_num_threads(4)\n",
    "\n",
    "model = sdm.models.KumoTabular(\n",
    "    task=\"classification\",\n",
    "    size=config[\"size\"],\n",
    "    device=device,\n",
    ")\n",
    "table = sdm.TableTensor.from_pandas(\n",
    "    df=df, stypes={**dict.fromkeys(X.columns, \"numerical\"), \"target\": \"categorical\"},\n",
    "    device=device,\n",
    ")\n",
    "with torch.inference_mode():\n",
    "    model.fit(\n",
    "        x=table[train].drop_columns(\"target\"), y=table[train, \"target\"],\n",
    "        num_estimators=config[\"num_estimators\"],\n",
    "        generator=torch.Generator(device).manual_seed(0),\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e64dc5a",
   "metadata": {},
   "source": [
    "## 4. Read the Predictions\n",
    "\n",
    "Pass only features for the 85 held-out test rows.\n",
    "Align probability columns with the class labels before scoring."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "c46ac604",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-20T17:29:31.625025Z",
     "iopub.status.busy": "2026-10-20T17:29:31.624951Z",
     "iopub.status.idle": "2026-10-20T17:29:31.845643Z",
     "shell.execute_reply": "2026-10-20T17:29:31.845161Z"
    }
   },
   "outputs": [],
   "source": [
    "with torch.inference_mode():\n",
    "    output = model.predict(table[test].drop_columns(\"target\"))\n",
    "\n",
    "columns = output.columns[sdm.Stype.numerical]\n",
    "\n",
    "probabilities = output.numerical[:, [columns.index(c) for c in classes]].cpu().numpy()\n",
    "predictions = np.array(classes)[probabilities.argmax(axis=1)]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dc8e858",
   "metadata": {},
   "source": [
    "## 5. Evaluate predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "039b95c2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-10-20T17:29:31.846897Z",
     "iopub.status.busy": "2026-10-20T17:29:31.846839Z",
     "iopub.status.idle": "2026-10-20T17:29:32.016862Z",
     "shell.execute_reply": "2026-10-20T17:29:32.016263Z"
    }
   },
   "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>Accuracy</th>\n",
       "      <th>ROC AUC (macro OVO)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Model</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>KumoTabular small</th>\n",
       "      <td>0.9176</td>\n",
       "      <td>0.9841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Majority class</th>\n",
       "      <td>0.2588</td>\n",
       "      <td>0.5000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Accuracy  ROC AUC (macro OVO)\n",
       "Model                                           \n",
       "KumoTabular small    0.9176               0.9841\n",
       "Majority class       0.2588               0.5000"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.dummy import DummyClassifier\n",
    "from sklearn.metrics import accuracy_score, roc_auc_score\n",
    "\n",
    "baseline = DummyClassifier(strategy=\"most_frequent\").fit(X.iloc[train], y.iloc[train])\n",
    "pd.DataFrame({\n",
    "    \"Model\": [f'KumoTabular {config[\"size\"]}', \"Majority class\"],\n",
    "    \"Accuracy\": [\n",
    "        accuracy_score(y.iloc[test], predictions),\n",
    "        accuracy_score(y.iloc[test], baseline.predict(X.iloc[test])),\n",
    "    ],\n",
    "    \"ROC AUC (macro OVO)\": [\n",
    "        roc_auc_score(y.iloc[test], probabilities, labels=classes, multi_class=\"ovo\"),\n",
    "        roc_auc_score(y.iloc[test], baseline.predict_proba(X.iloc[test]),\n",
    "                      labels=baseline.classes_, multi_class=\"ovo\"),\n",
    "    ],\n",
    "}).set_index(\"Model\").round(4)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "425c9d32",
   "metadata": {},
   "source": [
    "## Next Steps\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "053b0106",
   "metadata": {},
   "source": [
    "### Other Models\n",
    "- [Kumo-Relational](https://huggingface.co/nvidia/Kumo-Relational)\n",
    "- [Kumo-Anomaly](https://huggingface.co/nvidia/Kumo-Anomaly)\n",
    "- [Kumo-Forecast](https://huggingface.co/nvidia/Kumo-Forecast)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27aae15a",
   "metadata": {},
   "source": [
    "### Task-Specific Training\n",
    "Use Kumo Engine to train a model for a prediction task on your relational data.\n",
    "- [GitHub](https://github.com/NVIDIA/kumo-engine)\n",
    "- [Documentation](https://docs.nvidia.com/sdgm/kumo-engine/get-started-with-kumo-engine/overview)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "kumotabular (3.11.15)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}