{
 "nbformat": 4,
 "nbformat_minor": 5,
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3.12.0"
  },
  "colab": {
   "provenance": []
  }
 },
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fE3_sBdoskdQ"
   },
   "source": "# 💧 QML WaterMatica — Sızıntı / Aciliyet Önceliklendirme (20 Aşama)\n\n## Senaryo\nİstanbul / Ankara / İzmir su şebekelerinde **sızıntı veya kritik olay aciliyet sınıflandırması**.\n\nHer satır bir **aday müdahale noktası** (DMA segmenti, hidrant, manhole, depo çevresi).\n\n$$\n\\hat{y}_i \\in \\{0, 1\\}\n$$\n\n| Etiket | Anlam |\n|--------|--------|\n| **1 = Yüksek aciliyet** | Önce müdahale (yüksek kayıp riski + düşük basınç + gece akış sapması + kritik tesis yakınlığı) |\n| **0 = Düşük aciliyet** | İzleme / ikinci dalga |\n\n**Sensör bağlamı (WaterMatica):** mmRadar (1 mm), basınç 0–20 bar, debimetre, SWM kalite, SMC manhole.\n\n**Kübit politikası**\n\n| Ortam | Davranış |\n|-------|----------|\n| Colab simülasyon | `SIM_MAX_QUBITS = 10` (PCA → [0,π] → ZZFeatureMap) |\n| Gerçek IBM QPU | `n_qubits = min(SIM_MAX_QUBITS, backend.num_qubits)` otomatik |\n| Yavaş simülasyon | `SIM_MAX_QUBITS = 8` yapın |\n\n**Secret:** `IBM_QUANTUM_TOKEN` (opsiyonel)  \n> Veri **sentetiktir**; gerçek şebeke kaydı değildir. Klasik-first + seçici quantum disiplini geçerlidir.\n",
   "id": "fE3_sBdoskdQ"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "k7GnciIhskdW"
   },
   "source": "## 📋 Joseph Volkan OZCAN QML Şablonu — 20 Aşama (WaterMatica uyarlaması)\n\n| # | Aşama |\n|---|--------|\n| 1 | Drive / Path Tanımlama |\n| 2 | Cihaz & Algoritma Seçim Motoru + Kübit politikası |\n| 3 | Python Kütüphane Kurulumu |\n| 4 | API Keys Yönetimi |\n| 5 | Sentetik Su Şebekesi Verisi (WaterMatica sınırları) |\n| 6 | (Opsiyonel) CSV Yükleme |\n| 7 | Keşifsel Veri Analizi (EDA) |\n| 8 | KVKK / hassas alan notu |\n| 9 | Veri Hazırlama ve Ön İşleme |\n| 10 | Klasik Makine Öğrenmesi (RBF-SVM + Decision Tree) |\n| 11 | QML Algoritma Seçimi |\n| 12 | Veri Encoding ve Qubit Maliyeti (PCA) |\n| 13 | Quantum Feature Map + Fidelity Kernel |\n| 14 | Quantum Kernel Eğitimleri (QSVC) |\n| 15 | Confusion Matrix ve ROC |\n| 16 | Gerçek Quantum Donanım (IBM — opsiyonel) |\n| 17 | QML Sonuç Kaydı |\n| 18 | Model Karşılaştırma (+ QAOA/QUBO notu) |\n| 19 | Sonuçların Yorumlanması (su şebekesi bağlamı) |\n| 20 | Rehber + Rapor + İleri Adımlar |\n\n**Kullanım:** Colab’a yükleyin → hücreleri sırayla çalıştırın.\n",
   "id": "k7GnciIhskdW"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "gsXyaqrFskdX"
   },
   "source": "## AŞAMA 1: Drive / Path Tanımlama\n\n**Amaç:** Çıktı ve proje klasörlerini sabitlemek.\n",
   "id": "gsXyaqrFskdX"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "R7kxSXGQskdX",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456178516,
     "user_tz": -180,
     "elapsed": 1243,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "ba48a44c-1096-4ef4-b20f-14b1dafda8eb"
   },
   "execution_count": 1,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Python: 3.13.15\n",
      "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n",
      "✅ Drive → /content/drive/MyDrive/QML_WaterMatica\n",
      "✅ AŞAMA 1 tamamlandı.\n"
     ]
    }
   ],
   "source": "import os, sys\nprint(f\"Python: {sys.version.split()[0]}\")\n\nDRIVE_ROOT = PROJECT_PATH = OUTPUT_PATH = None\ntry:\n    from google.colab import drive\n    drive.mount('/content/drive')\n    DRIVE_ROOT = '/content/drive/MyDrive'\n    PROJECT_PATH = os.path.join(DRIVE_ROOT, 'QML_WaterMatica')\n    OUTPUT_PATH = os.path.join(PROJECT_PATH, 'outputs')\n    os.makedirs(OUTPUT_PATH, exist_ok=True)\n    print(f\"✅ Drive → {PROJECT_PATH}\")\nexcept Exception as e:\n    print(f\"⚠️ Drive atlandı: {e}\")\n    PROJECT_PATH = os.path.abspath('./qml_watermatica')\n    OUTPUT_PATH = os.path.join(PROJECT_PATH, 'outputs')\n    os.makedirs(OUTPUT_PATH, exist_ok=True)\n    print(f\"Lokal: {PROJECT_PATH}\")\n\nREPORT_DIR = os.path.join(OUTPUT_PATH, 'report_assets')\nos.makedirs(REPORT_DIR, exist_ok=True)\nprint(\"✅ AŞAMA 1 tamamlandı.\")\n",
   "id": "R7kxSXGQskdX"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "b2IYQbU0skdZ"
   },
   "source": "## AŞAMA 2: Cihaz & Algoritma + Kübit politikası\n\n**Amaç:** Simülasyonda pratik üst sınır; gerçek cihazda otomatik tavan.\n",
   "id": "b2IYQbU0skdZ"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "-G2VSM8wskda",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456178531,
     "user_tz": -180,
     "elapsed": 11,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "83de3c99-d1cc-482b-9dd2-3812c3c1284d"
   },
   "execution_count": 2,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "SIM_MAX_QUBITS = 10\n",
      "SELECTED_ALGO  = QSVC\n",
      "✅ AŞAMA 2 tamamlandı.\n"
     ]
    }
   ],
   "source": "SIM_MAX_QUBITS = 10          # Colab için agresif; yavaşlarsa 8 yapın\nSELECTED_ALGO = \"QSVC\"       # QSVC | QAOA_demo\nRANDOM_STATE = 42\nTEST_SIZE = 0.25\n\nprint(f\"SIM_MAX_QUBITS = {SIM_MAX_QUBITS}\")\nprint(f\"SELECTED_ALGO  = {SELECTED_ALGO}\")\nprint(\"✅ AŞAMA 2 tamamlandı.\")\n",
   "id": "-G2VSM8wskda"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "12NdnEOxskdb"
   },
   "source": "## AŞAMA 3: Python Kütüphane Kurulumu\n",
   "id": "12NdnEOxskdb"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "eCxLvIXXskdb",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456233749,
     "user_tz": -180,
     "elapsed": 55221,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "90a7904b-ee6a-4dea-822f-7cbd6bc57389"
   },
   "execution_count": 3,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "✅ Kütüphaneler kuruldu / kontrol edildi.\n",
      "✅ AŞAMA 3 tamamlandı.\n"
     ]
    }
   ],
   "source": "import subprocess, sys\npkgs = [\n    \"qiskit\", \"qiskit-aer\", \"qiskit-machine-learning\",\n    \"scikit-learn\", \"pandas\", \"numpy\", \"matplotlib\", \"seaborn\",\n    \"python-docx\"\n]\nfor p in pkgs:\n    subprocess.check_call([sys.executable, \"-m\", \"pip\", \"install\", \"-q\", p])\nprint(\"✅ Kütüphaneler kuruldu / kontrol edildi.\")\nprint(\"✅ AŞAMA 3 tamamlandı.\")\n",
   "id": "eCxLvIXXskdb"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PfLwq3YMskdc"
   },
   "source": "## AŞAMA 4: API Keys Yönetimi\n\nColab Secrets veya ortam değişkeni: `IBM_QUANTUM_TOKEN`\n",
   "id": "PfLwq3YMskdc"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Icc8rpLuskdc",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456234752,
     "user_tz": -180,
     "elapsed": 997,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "c9df67a6-41a3-439a-db29-711fb672662e"
   },
   "execution_count": 4,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "✅ IBM token Colab Secrets'tan okundu.\n",
      "✅ AŞAMA 4 tamamlandı.\n"
     ]
    }
   ],
   "source": "import os\nIBM_TOKEN = None\ntry:\n    from google.colab import userdata\n    IBM_TOKEN = userdata.get('IBM_QUANTUM_TOKEN')\n    print(\"✅ IBM token Colab Secrets'tan okundu.\" if IBM_TOKEN else \"⚠️ Token yok — simülasyon ile devam.\")\nexcept Exception:\n    IBM_TOKEN = os.environ.get('IBM_QUANTUM_TOKEN')\n    print(\"✅ Token ortam değişkeninden.\" if IBM_TOKEN else \"⚠️ Token yok — sadece simülasyon.\")\nprint(\"✅ AŞAMA 4 tamamlandı.\")\n",
   "id": "Icc8rpLuskdc"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "DmZvpnU2skdd"
   },
   "source": "## AŞAMA 5: Sentetik Su Şebekesi Verisi (WaterMatica sınırları)\n\n**Özellikler (WaterMatica sensör bağlamı):**\n- `level_m` — mmRadar 0.15–15 m (1 mm hassasiyet, sınırlara yakın salınım)\n- `pressure_bar` — 0–20 bar (drift + sınır)\n- `night_flow_ratio` — MNF sapması (leak sinyali)\n- `flow_balance_residual` — giriş − sayaçlı\n- `quality_ph_delta` — sıcaklık kompanzasyonsuz sapma\n- `distance_to_critical_km` — hastane / depo / kritik tesis\n- `manhole_risk` — SMC proxy\n- `packet_loss_rate` — iletişim kalitesi\n- `elevation_m` — kot (Ankara/İstanbul)\n",
   "id": "DmZvpnU2skdd"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "OSyfXehGskdd",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456234953,
     "user_tz": -180,
     "elapsed": 56,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "25e988e1-65bc-4cca-b12e-ed7a78ed071d"
   },
   "execution_count": 5,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "urgency\n",
      "1    207\n",
      "0    193\n",
      "Name: count, dtype: int64\n",
      "       elevation_m  level_m  pressure_bar  night_flow_ratio  \\\n",
      "count       400.00   400.00        400.00            400.00   \n",
      "mean        380.28     6.78          4.41              1.00   \n",
      "std         401.08     3.39          1.95              0.33   \n",
      "min           6.10     0.20          0.30              0.30   \n",
      "25%          86.85     4.30          3.05              0.77   \n",
      "50%         182.10     6.72          4.36              0.99   \n",
      "75%         854.85     8.96          5.71              1.22   \n",
      "max        1197.20    14.70         10.60              2.10   \n",
      "\n",
      "       flow_balance_residual  quality_ph_delta  distance_to_critical_km  \\\n",
      "count                 400.00            400.00                   400.00   \n",
      "mean                    0.34              0.00                     2.45   \n",
      "std                    11.59              0.15                     2.40   \n",
      "min                   -31.25             -0.41                     0.02   \n",
      "25%                    -7.78             -0.10                     0.73   \n",
      "50%                     0.28             -0.00                     1.58   \n",
      "75%                     8.16              0.09                     3.45   \n",
      "max                    37.32              0.59                    12.26   \n",
      "\n",
      "       manhole_risk  packet_loss_rate  density_proxy  urgency  \n",
      "count        400.00            400.00         400.00   400.00  \n",
      "mean           0.28              0.16           3.91     0.52  \n",
      "std            0.16              0.11           2.84     0.50  \n",
      "min            0.00              0.00           0.09     0.00  \n",
      "25%            0.16              0.07           1.92     0.00  \n",
      "50%            0.26              0.13           3.14     1.00  \n",
      "75%            0.38              0.22           5.45     1.00  \n",
      "max            0.81              0.55          16.90     1.00  \n",
      "✅ AŞAMA 5 tamamlandı — sentetik WaterMatica verisi hazır.\n"
     ]
    }
   ],
   "source": "import numpy as np\nimport pandas as pd\n\nnp.random.seed(RANDOM_STATE)\nN = 400   # aday müdahale noktası (DMA segment / hidrant / manhole)\n\n# Coğrafi proxy (İzmir-Ankara-İstanbul karışık)\ncity = np.random.choice(['Istanbul', 'Ankara', 'Izmir'], size=N, p=[0.45, 0.30, 0.25])\nelevation = np.where(city == 'Ankara',\n                     np.random.uniform(800, 1200, N),\n                     np.random.uniform(5, 250, N))\n\nlevel_m = np.clip(np.random.normal(6.5, 3.5, N), 0.2, 14.7)\npressure_bar = np.clip(np.random.normal(4.2, 2.0, N) + 0.001 * elevation/100, 0.3, 19.5)\nnight_flow_ratio = np.clip(np.random.normal(1.0, 0.35, N), 0.3, 2.8)      # >1.3 şüpheli\nflow_balance_residual = np.random.normal(0, 12, N)                          # m³/h sapma\nquality_ph_delta = np.random.normal(0, 0.15, N)\ndistance_to_critical_km = np.random.exponential(2.5, N)\nmanhole_risk = np.random.beta(2, 5, N)\npacket_loss_rate = np.random.beta(1.5, 8, N)\ndensity_proxy = np.random.gamma(2.0, 2.0, N)\n\n# Etiket üretimi (kural tabanlı + gürültü) — yüksek aciliyet\nlogit = (\n    1.4 * (night_flow_ratio > 1.35).astype(float)\n    + 1.1 * (np.abs(flow_balance_residual) > 18).astype(float)\n    + 0.9 * (pressure_bar < 2.0).astype(float)\n    + 0.7 * (distance_to_critical_km < 1.2).astype(float)\n    + 0.5 * (manhole_risk > 0.55).astype(float)\n    + 0.4 * (packet_loss_rate > 0.15).astype(float)\n    - 0.8\n)\nprob = 1 / (1 + np.exp(-logit))\ny = (np.random.rand(N) < prob).astype(int)\n\ndf = pd.DataFrame({\n    'city': city,\n    'elevation_m': elevation.round(1),\n    'level_m': level_m.round(3),\n    'pressure_bar': pressure_bar.round(3),\n    'night_flow_ratio': night_flow_ratio.round(3),\n    'flow_balance_residual': flow_balance_residual.round(2),\n    'quality_ph_delta': quality_ph_delta.round(3),\n    'distance_to_critical_km': distance_to_critical_km.round(2),\n    'manhole_risk': manhole_risk.round(3),\n    'packet_loss_rate': packet_loss_rate.round(3),\n    'density_proxy': density_proxy.round(2),\n    'urgency': y\n})\n\nprint(df['urgency'].value_counts())\nprint(df.describe().round(2))\nprint(\"✅ AŞAMA 5 tamamlandı — sentetik WaterMatica verisi hazır.\")\n",
   "id": "OSyfXehGskdd"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZoMCFjleskde"
   },
   "source": "## AŞAMA 6: (Opsiyonel) CSV Yükleme\n\nVarsayılan: sentetik. Gerçek CSV varsa yolu verin.\n",
   "id": "ZoMCFjleskde"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "VfjQfdemskde",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456235106,
     "user_tz": -180,
     "elapsed": 150,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "04e47e90-7dd2-44bc-8865-fd259a217420"
   },
   "execution_count": 6,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Sentetik veri ile devam ediliyor.\n",
      "✅ AŞAMA 6 tamamlandı.\n"
     ]
    }
   ],
   "source": "# CSV_PATH = os.path.join(PROJECT_PATH, 'my_water_data.csv')\n# if os.path.exists(CSV_PATH):\n#     df = pd.read_csv(CSV_PATH)\n#     print(\"CSV yüklendi:\", df.shape)\nprint(\"Sentetik veri ile devam ediliyor.\")\nprint(\"✅ AŞAMA 6 tamamlandı.\")\n",
   "id": "VfjQfdemskde"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "26rqoJzNskdf"
   },
   "source": "## AŞAMA 7: Keşifsel Veri Analizi (EDA)\n",
   "id": "26rqoJzNskdf"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 942
    },
    "id": "WEU7IcJKskdf",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456238246,
     "user_tz": -180,
     "elapsed": 2996,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "7f87f130-c2b7-438c-af6d-c7fe6ffc1bbc"
   },
   "execution_count": 7,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/tmp/ipykernel_14870/89244611.py:5: FutureWarning: \n",
      "\n",
      "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n",
      "\n",
      "  sns.countplot(x='urgency', data=df, ax=axes[0,0], palette='Set2')\n"
     ]
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 1100x800 with 4 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Kayıt: /content/drive/MyDrive/QML_WaterMatica/outputs/report_assets/01_eda.png\n",
      "✅ AŞAMA 7 tamamlandı.\n"
     ]
    }
   ],
   "source": "import matplotlib.pyplot as plt\nimport seaborn as sns\n\nfig, axes = plt.subplots(2, 2, figsize=(11, 8))\nsns.countplot(x='urgency', data=df, ax=axes[0,0], palette='Set2')\naxes[0,0].set_title('Aciliyet dağılımı')\nsns.boxplot(x='urgency', y='night_flow_ratio', data=df, ax=axes[0,1])\naxes[0,1].set_title('MNF oranı vs aciliyet')\nsns.scatterplot(x='pressure_bar', y='flow_balance_residual', hue='urgency', data=df, ax=axes[1,0], alpha=0.6)\naxes[1,0].set_title('Basınç vs flow residual')\nsns.boxplot(x='city', y='level_m', data=df, ax=axes[1,1])\naxes[1,1].set_title('Şehir bazlı seviye (mmRadar)')\nplt.tight_layout()\neda_path = os.path.join(REPORT_DIR, '01_eda.png')\nplt.savefig(eda_path, dpi=120)\nplt.show()\nprint(\"Kayıt:\", eda_path)\nprint(\"✅ AŞAMA 7 tamamlandı.\")\n",
   "id": "WEU7IcJKskdf"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "91lhyDhbskdf"
   },
   "source": "## AŞAMA 8: KVKK / hassas alan notu\n\nVeri **tamamen sentetiktir**. Gerçek abone, adres veya SCADA kimliği içermez.  \nGerçek deployment’ta RLS (`app.current_tenant_id`) ve IoTservo Core tenancy kuralları geçerlidir.\n",
   "id": "91lhyDhbskdf"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "PAfFlW7Fskdg",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456238258,
     "user_tz": -180,
     "elapsed": 9,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "a926f404-8da4-4939-d959-239b177d9f4f"
   },
   "execution_count": 8,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Sentetik veri — KVKK riski yok.\n",
      "✅ AŞAMA 8 tamamlandı.\n"
     ]
    }
   ],
   "source": "print(\"Sentetik veri — KVKK riski yok.\")\nprint(\"✅ AŞAMA 8 tamamlandı.\")\n",
   "id": "PAfFlW7Fskdg"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0gpQalCyskdg"
   },
   "source": "## AŞAMA 9: Veri Hazırlama ve Ön İşleme\n",
   "id": "0gpQalCyskdg"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "306aMbEKskdh",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456238707,
     "user_tz": -180,
     "elapsed": 199,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "e6bd1ec2-83b2-4adb-cfbc-2332494a3804"
   },
   "execution_count": 9,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Train: (300, 10) | Test: (100, 10)\n",
      "✅ AŞAMA 9 tamamlandı.\n"
     ]
    }
   ],
   "source": "from sklearn.model_selection import train_test_split\nfrom sklearn.preprocessing import StandardScaler\n\nfeature_cols = [\n    'elevation_m', 'level_m', 'pressure_bar', 'night_flow_ratio',\n    'flow_balance_residual', 'quality_ph_delta', 'distance_to_critical_km',\n    'manhole_risk', 'packet_loss_rate', 'density_proxy'\n]\nX = df[feature_cols].values\ny = df['urgency'].values\n\nX_train, X_test, y_train, y_test = train_test_split(\n    X, y, test_size=TEST_SIZE, random_state=RANDOM_STATE, stratify=y\n)\nscaler = StandardScaler()\nX_train_s = scaler.fit_transform(X_train)\nX_test_s = scaler.transform(X_test)\nprint(f\"Train: {X_train.shape} | Test: {X_test.shape}\")\nprint(\"✅ AŞAMA 9 tamamlandı.\")\n",
   "id": "306aMbEKskdh"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "v8StQ9knskdh"
   },
   "source": "## AŞAMA 10: Klasik Makine Öğrenmesi (RBF-SVM + Decision Tree)\n",
   "id": "v8StQ9knskdh"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "LU9mFINQskdh",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456238907,
     "user_tz": -180,
     "elapsed": 199,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "c934f175-3866-404f-bb5b-047e74b1b4d5"
   },
   "execution_count": 10,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "=== RBF-SVM ===\n",
      "Accuracy: 0.540 | Balanced: 0.535\n",
      "=== Decision Tree ===\n",
      "Accuracy: 0.550 | Balanced: 0.560\n",
      "✅ AŞAMA 10 tamamlandı.\n"
     ]
    }
   ],
   "source": "from sklearn.svm import SVC\nfrom sklearn.tree import DecisionTreeClassifier\nfrom sklearn.metrics import accuracy_score, balanced_accuracy_score, classification_report\n\nclf_svm = SVC(kernel='rbf', C=1.0, probability=True, random_state=RANDOM_STATE)\nclf_svm.fit(X_train_s, y_train)\npred_svm = clf_svm.predict(X_test_s)\n\nclf_dt = DecisionTreeClassifier(max_depth=5, random_state=RANDOM_STATE)\nclf_dt.fit(X_train_s, y_train)\npred_dt = clf_dt.predict(X_test_s)\n\nprint(\"=== RBF-SVM ===\")\nprint(f\"Accuracy: {accuracy_score(y_test, pred_svm):.3f} | Balanced: {balanced_accuracy_score(y_test, pred_svm):.3f}\")\nprint(\"=== Decision Tree ===\")\nprint(f\"Accuracy: {accuracy_score(y_test, pred_dt):.3f} | Balanced: {balanced_accuracy_score(y_test, pred_dt):.3f}\")\nprint(\"✅ AŞAMA 10 tamamlandı.\")\n",
   "id": "LU9mFINQskdh"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vymkIqKGskdi"
   },
   "source": "## AŞAMA 11: QML Algoritma Seçimi\n\nBu çalışmada **QSVC (Quantum Support Vector Classifier)** + fidelity kernel.\n",
   "id": "vymkIqKGskdi"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4QiH_6pOskdi",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456239104,
     "user_tz": -180,
     "elapsed": 7,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "b9555852-f8f9-4c4c-ad6c-8852212d21ad"
   },
   "execution_count": 11,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Seçilen algoritma: QSVC\n",
      "Alternatif: QAOA ile küçük set-partitioning / rota demo (Aşama 18).\n",
      "✅ AŞAMA 11 tamamlandı.\n"
     ]
    }
   ],
   "source": "print(f\"Seçilen algoritma: {SELECTED_ALGO}\")\nprint(\"Alternatif: QAOA ile küçük set-partitioning / rota demo (Aşama 18).\")\nprint(\"✅ AŞAMA 11 tamamlandı.\")\n",
   "id": "4QiH_6pOskdi"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "iJEWboVNskdj"
   },
   "source": "## AŞAMA 12: Veri Encoding ve Qubit Maliyeti (PCA)\n",
   "id": "iJEWboVNskdj"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4oqO8DmLskdj",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456239248,
     "user_tz": -180,
     "elapsed": 142,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "1ecdf1e1-bfe2-4b02-b0ac-49ccb4d9126e"
   },
   "execution_count": 12,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "n_qubits = 10\n",
      "PCA explained variance: 1.000\n",
      "✅ AŞAMA 12 tamamlandı.\n"
     ]
    }
   ],
   "source": "from sklearn.decomposition import PCA\nfrom sklearn.preprocessing import MinMaxScaler\n\nn_qubits = SIM_MAX_QUBITS\npca = PCA(n_components=n_qubits, random_state=RANDOM_STATE)\nX_train_pca = pca.fit_transform(X_train_s)\nX_test_pca = pca.transform(X_test_s)\n\n# [0, π] aralığına ölçekle (ZZFeatureMap için)\nmm = MinMaxScaler(feature_range=(0, np.pi))\nX_train_enc = mm.fit_transform(X_train_pca)\nX_test_enc = mm.transform(X_test_pca)\n\nprint(f\"n_qubits = {n_qubits}\")\nprint(f\"PCA explained variance: {pca.explained_variance_ratio_.sum():.3f}\")\nprint(\"✅ AŞAMA 12 tamamlandı.\")\n",
   "id": "4oqO8DmLskdj"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "v6Z6g_f9skdj"
   },
   "source": "## AŞAMA 13: Quantum Feature Map + Fidelity Kernel\n",
   "id": "v6Z6g_f9skdj"
  },
  {
   "cell_type": "code",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ZyKA5ds-skdk",
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789456239748,
     "user_tz": -180,
     "elapsed": 497,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "63f893bc-4f6a-4efb-ee2d-8917d0772334"
   },
   "execution_count": 13,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "85\n",
      "✅ AŞAMA 13 tamamlandı.\n"
     ]
    },
    {
     "output_type": "stream",
     "name": "stderr",
     "text": [
      "/tmp/ipykernel_14870/478646144.py:5: DeprecationWarning: The class ``qiskit.circuit.library.data_preparation._zz_feature_map.ZZFeatureMap`` is deprecated as of Qiskit 2.1. It will be removed in Qiskit 3.0. Use the zz_feature_map function as a replacement. Note that this will no longer return a BlueprintCircuit, but just a plain QuantumCircuit.\n",
      "  feature_map = ZZFeatureMap(feature_dimension=n_qubits, reps=2)\n"
     ]
    }
   ],
   "source": "from qiskit.circuit.library import ZZFeatureMap\nfrom qiskit_machine_learning.kernels import FidelityQuantumKernel\nfrom qiskit.primitives import StatevectorSampler\n\nfeature_map = ZZFeatureMap(feature_dimension=n_qubits, reps=2)\nsampler = StatevectorSampler()\nquantum_kernel = FidelityQuantumKernel(feature_map=feature_map)\n\nprint(feature_map.decompose().depth() if hasattr(feature_map, 'decompose') else \"ZZFeatureMap hazır\")\nprint(\"✅ AŞAMA 13 tamamlandı.\")\n",
   "id": "ZyKA5ds-skdk"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "mIR9i21iskdk"
   },
   "source": "## AŞAMA 14: Quantum Kernel Eğitimleri (QSVC)\n",
   "id": "mIR9i21iskdk"
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "H2Og664mskdk",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789457356751,
     "user_tz": -180,
     "elapsed": 1117000,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "a3d1e0d9-3454-465f-e3da-67a2b71f924b"
   },
   "execution_count": 14,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "QSVC Accuracy: 0.475 | Balanced: 0.500\n",
      "(Eğitim alt kümesi: 80, test: 40)\n",
      "✅ AŞAMA 14 tamamlandı.\n"
     ]
    }
   ],
   "source": "from qiskit_machine_learning.algorithms import QSVC\n\n# Küçük alt küme ile hız (Colab timeout riski)\nMAX_TRAIN = min(80, len(X_train_enc))\nMAX_TEST = min(40, len(X_test_enc))\nXtr = X_train_enc[:MAX_TRAIN]\nytr = y_train[:MAX_TRAIN]\nXte = X_test_enc[:MAX_TEST]\nyte = y_test[:MAX_TEST]\n\nqsvc = QSVC(quantum_kernel=quantum_kernel)\nqsvc.fit(Xtr, ytr)\npred_qsvc = qsvc.predict(Xte)\n\nacc_q = accuracy_score(yte, pred_qsvc)\nbal_q = balanced_accuracy_score(yte, pred_qsvc)\nprint(f\"QSVC Accuracy: {acc_q:.3f} | Balanced: {bal_q:.3f}\")\nprint(f\"(Eğitim alt kümesi: {MAX_TRAIN}, test: {MAX_TEST})\")\nprint(\"✅ AŞAMA 14 tamamlandı.\")\n",
   "id": "H2Og664mskdk"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "R_M63MYBskdl"
   },
   "source": "## AŞAMA 15: Confusion Matrix ve ROC\n",
   "id": "R_M63MYBskdl"
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "vXL1dKKoskdl",
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 423
    },
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789457374650,
     "user_tz": -180,
     "elapsed": 17343,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "5c6be7e1-7991-415c-8deb-abebec50f212"
   },
   "execution_count": 15,
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ],
      "image/png": "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\n"
     },
     "metadata": {}
    },
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "✅ AŞAMA 15 tamamlandı.\n"
     ]
    }
   ],
   "source": "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay, roc_curve, auc\n\nfig, axes = plt.subplots(1, 2, figsize=(10, 4))\ncm = confusion_matrix(yte, pred_qsvc)\nConfusionMatrixDisplay(cm).plot(ax=axes[0], colorbar=False)\naxes[0].set_title('QSVC Confusion Matrix')\n\n# ROC için SVM (probability)\nif hasattr(clf_svm, 'predict_proba'):\n    proba = clf_svm.predict_proba(X_test_s[:MAX_TEST])[:, 1]\n    fpr, tpr, _ = roc_curve(yte, proba)\n    axes[1].plot(fpr, tpr, label=f'SVM AUC={auc(fpr,tpr):.2f}')\naxes[1].plot([0,1], [0,1], 'k--')\naxes[1].set_title('ROC (klasik SVM referans)')\naxes[1].legend()\nplt.tight_layout()\ncm_path = os.path.join(REPORT_DIR, '15_cm_roc.png')\nplt.savefig(cm_path, dpi=120)\nplt.show()\nprint(\"✅ AŞAMA 15 tamamlandı.\")\n",
   "id": "vXL1dKKoskdl"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-Yy1VX30skdm"
   },
   "source": "## AŞAMA 16: Gerçek Quantum Donanım (IBM — opsiyonel)\n",
   "id": "-Yy1VX30skdm"
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "UFt6alo2skdn",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789457378808,
     "user_tz": -180,
     "elapsed": 3902,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "10fbc6ce-a2dc-406c-cd83-f417e82054da"
   },
   "execution_count": 16,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "IBM bağlantı hatası: No module named 'qiskit_ibm_runtime'\n",
      "✅ AŞAMA 16 tamamlandı.\n"
     ]
    }
   ],
   "source": "real_backend_name = None\nif IBM_TOKEN:\n    try:\n        from qiskit_ibm_runtime import QiskitRuntimeService\n        service = QiskitRuntimeService(channel=\"ibm_quantum\", token=IBM_TOKEN)\n        backend = service.least_busy(operational=True, simulator=False)\n        real_backend_name = backend.name\n        n_qubits_hw = min(SIM_MAX_QUBITS, backend.num_qubits)\n        print(f\"Seçilen backend: {real_backend_name} | qubits={backend.num_qubits} → kullanılacak={n_qubits_hw}\")\n        print(\"Not: Gerçek QPU’da kernel hesabı uzun sürer; bu notebook’ta demo amaçlı atlanabilir.\")\n    except Exception as e:\n        print(f\"IBM bağlantı hatası: {e}\")\nelse:\n    print(\"Token yok — simülasyon sonuçları ile devam.\")\nprint(\"✅ AŞAMA 16 tamamlandı.\")\n",
   "id": "UFt6alo2skdn"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zdm4bcZkskdp"
   },
   "source": "## AŞAMA 17: QML Sonuç Kaydı\n",
   "id": "zdm4bcZkskdp"
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "1A1J22Uqskdq",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789457382554,
     "user_tz": -180,
     "elapsed": 3252,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "a608d031-b25a-4a25-c780-1787c2d01b5e"
   },
   "execution_count": 17,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Kayıt: /content/drive/MyDrive/QML_WaterMatica/outputs/test_predictions.csv\n",
      "✅ AŞAMA 17 tamamlandı.\n"
     ]
    }
   ],
   "source": "results_df = pd.DataFrame({\n    'y_true': yte,\n    'pred_qsvc': pred_qsvc,\n    'pred_svm_ref': pred_svm[:MAX_TEST]\n})\nout_csv = os.path.join(OUTPUT_PATH, 'test_predictions.csv')\nresults_df.to_csv(out_csv, index=False)\nprint(\"Kayıt:\", out_csv)\nprint(\"✅ AŞAMA 17 tamamlandı.\")\n",
   "id": "1A1J22Uqskdq"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "U6x0wo8fskdr"
   },
   "source": "## AŞAMA 18: Model Karşılaştırma (+ QAOA/QUBO notu)\n\nKlasik RBF-SVM güçlü baseline’dır. Küçük sentetik sette QSVC üstünlüğü **garantili değildir**.  \nSensor-placement / set-cover tipi problemlerde QAOA veya Rydberg (Pasqal/QuEra) daha doğal adaydır.\n",
   "id": "U6x0wo8fskdr"
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "8i96_RjFskdr",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789457384655,
     "user_tz": -180,
     "elapsed": 1707,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "5e946dbd-3f15-49dc-e323-8c1e87214ba6"
   },
   "execution_count": 18,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "         Model    Platform  Accuracy  Balanced\n",
      "Klasik RBF-SVM         CPU     0.540     0.535\n",
      "    QSVC (10q) statevector     0.475     0.500\n",
      "\n",
      "--- QAOA mini demo (eğitim amaçlı, 4 değişken) ---\n",
      "QAOA p=1 enerji ≈ -1.4262 | success=False\n",
      "Not: Gerçek crew/sensor-partitioning için yüzlerce değişken + hibrit gerekir.\n",
      "✅ AŞAMA 18 tamamlandı.\n"
     ]
    }
   ],
   "source": "comparison = pd.DataFrame([\n    {'Model': 'Klasik RBF-SVM', 'Platform': 'CPU',\n     'Accuracy': accuracy_score(y_test, pred_svm),\n     'Balanced': balanced_accuracy_score(y_test, pred_svm)},\n    {'Model': f'QSVC ({n_qubits}q)', 'Platform': 'statevector',\n     'Accuracy': acc_q, 'Balanced': bal_q},\n])\nprint(comparison.round(3).to_string(index=False))\ncomp_path = os.path.join(OUTPUT_PATH, 'comparison.csv')\ncomparison.to_csv(comp_path, index=False)\n\n# Mini QAOA notu (set-partitioning tarzı demo — 4 değişken)\nprint(\"\\n--- QAOA mini demo (eğitim amaçlı, 4 değişken) ---\")\ntry:\n    from qiskit.circuit.library import qaoa_ansatz\n    from qiskit.quantum_info import SparsePauliOp\n    from qiskit.primitives import StatevectorEstimator\n    from scipy.optimize import minimize\n\n    costs = [1.0, 1.2, 0.8, 1.5]\n    P = 2.0\n    ops = []\n    for i, c in enumerate(costs):\n        z = ['I']*4; z[i]='Z'\n        ops.append((''.join(z), float(c)))\n    for i in range(4):\n        for j in range(i+1,4):\n            z = ['I']*4; z[i]='Z'; z[j]='Z'\n            ops.append((''.join(z), 0.5*P))\n    cost_op = SparsePauliOp.from_list(ops)\n    ansatz = qaoa_ansatz(cost_op, reps=1)\n    estimator = StatevectorEstimator()\n    def energy(theta):\n        return float(estimator.run([(ansatz, cost_op, theta)]).result()[0].data.evs)\n    x0 = np.random.default_rng(RANDOM_STATE).uniform(0, 2*np.pi, size=ansatz.num_parameters)\n    res = minimize(energy, x0, method='COBYLA', options={'maxiter': 30})\n    print(f\"QAOA p=1 enerji ≈ {res.fun:.4f} | success={res.success}\")\n    print(\"Not: Gerçek crew/sensor-partitioning için yüzlerce değişken + hibrit gerekir.\")\nexcept Exception as e:\n    print(\"QAOA demo atlandı:\", e)\nprint(\"✅ AŞAMA 18 tamamlandı.\")\n",
   "id": "8i96_RjFskdr"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OkBPdFGHskds"
   },
   "source": "## AŞAMA 19: Sonuçların Yorumlanması (su şebekesi bağlamı)\n\n1. $\\hat{y}\\in\\{0,1\\}$ **müdahale aciliyet sınıfıdır** — “kuantum her zaman daha iyi” iddiası değildir.\n2. Klasik RBF-SVM güçlü baseline; küçük sentetik sette QSVC üstünlüğü garantili değildir.\n3. WaterMatica bağlamında asıl quantum adayı: **büyük graph sensor-placement / set-cover** (Rydberg veya QAOA).\n4. MNF oranı + flow residual + düşük basınç + kritik tesise yakınlık → yüksek aciliyet eğilimi (üretici kural).\n5. Gerçek sistemde: quality-tag, RLS, human-in-the-loop ve Core actuation disiplini zorunludur.\n",
   "id": "OkBPdFGHskds"
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "B0HKlJ7bskdt",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789457385108,
     "user_tz": -180,
     "elapsed": 1099,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "aab018db-0346-4297-ed56-932b65720418"
   },
   "execution_count": 19,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "=== YORUM ÖZETİ ===\n",
      "         Model    Platform  Accuracy  Balanced\n",
      "Klasik RBF-SVM         CPU     0.540     0.535\n",
      "    QSVC (10q) statevector     0.475     0.500\n",
      "Anlaşmazlık (QSVC vs SVM, alt küme): 32.5%\n",
      "✅ AŞAMA 19 tamamlandı.\n"
     ]
    }
   ],
   "source": "print(\"=== YORUM ÖZETİ ===\")\nprint(comparison.round(3).to_string(index=False))\nprint(f\"Anlaşmazlık (QSVC vs SVM, alt küme): {np.mean(pred_qsvc != pred_svm[:MAX_TEST]):.1%}\")\nprint(\"✅ AŞAMA 19 tamamlandı.\")\n",
   "id": "B0HKlJ7bskdt"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "OUQ6G-sLskdt"
   },
   "source": "## AŞAMA 20: Rehber + Rapor + İleri Adımlar\n",
   "id": "OUQ6G-sLskdt"
  },
  {
   "cell_type": "code",
   "metadata": {
    "id": "p8of9SK8skdt",
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "status": "ok",
     "timestamp": 1789457385509,
     "user_tz": -180,
     "elapsed": 205,
     "user": {
      "displayName": "Joseph Volkan ÖZCAN",
      "userId": "11389530967736397480"
     }
    },
    "outputId": "60f27e00-47c0-41be-b485-c695eb444f1c"
   },
   "execution_count": 20,
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "Rapor: /content/drive/MyDrive/QML_WaterMatica/outputs/qml_watermatica_rapor.docx\n",
      "\n",
      "İleri adımlar:\n",
      "- Gerçek WaterMatica MQTT / Timescale verisi\n",
      "- Quality-tag ve MNF baseline üretimi\n",
      "- Sensor-placement için Rydberg / QAOA\n",
      "- Klasik MILP ile yan yana A/B\n",
      "\n",
      "✅ AŞAMA 20 tamamlandı. Notebook sonu.\n"
     ]
    }
   ],
   "source": "from docx import Document\nfrom docx.shared import Inches, Pt\nfrom datetime import datetime\n\ndoc = Document()\ndoc.styles['Normal'].font.name = 'Calibri'\ndoc.styles['Normal'].font.size = Pt(11)\n\ndef h(t, l=1): doc.add_heading(t, level=l)\ndef p(t): doc.add_paragraph(t)\n\nh('QML WaterMatica — Sızıntı / Aciliyet Önceliklendirme', 0)\np(f'Tarih: {datetime.now().strftime(\"%d.%m.%Y %H:%M\")} | MAX_QUBITS={n_qubits} | Algo={SELECTED_ALGO}')\nh('1. Problem')\np('Her aday nokta için aciliyet y_hat in {0,1}. WaterMatica sensör proxy özellikleri → klasik SVM / QSVC.')\nh('2. Veri')\np('Sentetik: level (mmRadar), pressure, MNF ratio, flow residual, quality, manhole risk, packet loss, elevation, city.')\nh('3. Sonuçlar')\np(comparison.to_string(index=False))\nh('4. Yorum')\np('Öğretim boru hattı. Asıl quantum değeri büyük kombinatorik sensor-placement / multi-objective optimizasyonda aranmalıdır. Klasik-first disiplin korunur.')\nh('5. İleri adımlar')\np('- Gerçek DMA telemetrisi + quality-tag')\np('- Pasqal/QuEra ile büyük graph placement')\np('- IBM Runtime ile daha büyük alt küme')\np('- MILP baseline + QAOA hibrit')\np('- Control Tower entegrasyonu')\n\nout = os.path.join(OUTPUT_PATH, 'qml_watermatica_rapor.docx')\ndoc.save(out)\nprint('Rapor:', out)\nprint(\"\"\"\nİleri adımlar:\n- Gerçek WaterMatica MQTT / Timescale verisi\n- Quality-tag ve MNF baseline üretimi\n- Sensor-placement için Rydberg / QAOA\n- Klasik MILP ile yan yana A/B\n\"\"\")\nprint(\"✅ AŞAMA 20 tamamlandı. Notebook sonu.\")\n",
   "id": "p8of9SK8skdt"
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "cTxiQ0MCskdu"
   },
   "source": "---\n### Özet ve Kapanış\n\n20 aşamalı Joseph Volkan Ozcan QML şablonu **WaterMatica su şebekesi aciliyet** domain’ine uyarlandı.\n\n| Çıktı | Konum |\n|-------|--------|\n| Tahmin CSV | `outputs/test_predictions.csv` |\n| Karşılaştırma | `outputs/comparison.csv` |\n| Word rapor | `outputs/qml_watermatica_rapor.docx` |\n| Görseller | `outputs/report_assets/` |\n\n**Hesaplanan şey:** $\\hat{y}_i \\in \\{0,1\\}$ — müdahale aciliyet sınıfı.  \n**Disiplin:** Klasik-first; quantum seçici ve ölçülebilir avantaj aranan yerde.\n",
   "id": "cTxiQ0MCskdu"
  }
 ]
}