Pennylane_CVQNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Pennylane_CVQNN.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyPuAHg6K2LKyDmPFwGNJMKl",
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/BankNatchapol/CV-QNN/blob/main/Pennylane_CVQNN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "baxbWzWnWHwH"
      },
      "source": [
        "# **Installation**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2Mrw-E82mORe"
      },
      "source": [
        "!pip install pennylane-sf"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rDGOcvA-WO63"
      },
      "source": [
        "# **Import Library**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iMPWUZa1l7s4"
      },
      "source": [
        "import pennylane as qml\n",
        "from pennylane import numpy as np\n",
        "from pennylane.optimize import AdamOptimizer, QNGOptimizer\n",
        "\n",
        "import matplotlib.pyplot as plt \n",
        "import pandas as pd\n",
        "\n",
        "import tensorflow as tf\n",
        "import math"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xh2XNSSASWb7"
      },
      "source": [
        "# **Continuous Variable Quantum Neural Network : Function Fitting**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J6Dct4rIWqqq"
      },
      "source": [
        "## Load Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QuSwsasadi3h"
      },
      "source": [
        "def tanh_function(x, eps=0.0):\n",
        "    \"\"\"The function f(x)=tanh(4x)+noise\"\"\"\n",
        "    return np.tanh(4*x) + eps * np.random.normal(size=x.shape)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 292
        },
        "id": "CBjv7fZvdmjR",
        "outputId": "04fda2fa-7c2f-4130-dc22-5dc34a957c70"
      },
      "source": [
        "X = np.linspace(-1, 1, 50)\n",
        "Y = tanh_function(X, 0.1)\n",
        "\n",
        "plt.figure()\n",
        "plt.scatter(X, Y)\n",
        "plt.xlabel(\"x\", fontsize=18)\n",
        "plt.ylabel(\"f(x)\", fontsize=18)\n",
        "plt.tick_params(axis=\"both\", which=\"major\", labelsize=16)\n",
        "plt.tick_params(axis=\"both\", which=\"minor\", labelsize=16)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Zk3vx4rZWSpI"
      },
      "source": [
        "## Define Device"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c3VUz8fTW5_Q"
      },
      "source": [
        "using strawberryfields.fork device for returning continuous variable quantum fork state"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZUtj9UOUmRVV"
      },
      "source": [
        "dev = qml.device(\"strawberryfields.fock\", wires=1, cutoff_dim=10)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vhgHe1jTWWD5"
      },
      "source": [
        "## Define Continuous Variable Quantum Neural Network\n",
        "### Quantum Layer\n",
        "<img src=\"https://github.com/BankNatchapol/CV-QNN/blob/main/static/single-qubit-quantum-layer.png?raw=true\">"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8gfKU7nTmTqJ"
      },
      "source": [
        "def layer(v):\n",
        "    # Matrix multiplication of input layer\n",
        "    qml.Rotation(v[0], wires=0)\n",
        "    qml.Squeezing(v[1], 0.0, wires=0)\n",
        "    qml.Rotation(v[2], wires=0)\n",
        "\n",
        "    # Bias\n",
        "    qml.Displacement(v[3], 0.0, wires=0)\n",
        "\n",
        "    # Element-wise nonlinear transformation\n",
        "    qml.Kerr(v[4], wires=0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EiRZyt3QWgO6"
      },
      "source": [
        "### Combining Quantum Layer into Quantum Neural Network\n",
        "<img src=\"https://github.com/BankNatchapol/CV-QNN/blob/main/static/single-qubit-quantum-neural-network.png?raw=true\">"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6r-Atmr9mi9S"
      },
      "source": [
        "@qml.qnode(dev)\n",
        "def quantum_neural_net(var, x):\n",
        "    # Encode input x into quantum state\n",
        "    qml.Displacement(x, 0.0, wires=0)\n",
        "\n",
        "    # \"layer\" subcircuits\n",
        "    for v in var:\n",
        "        layer(v)\n",
        "\n",
        "    return qml.expval(qml.X(0))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XI7kohWcWmTL"
      },
      "source": [
        "### Cost Function"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "R5VZnEh5cemc"
      },
      "source": [
        "using Mean Square Error\n",
        "\n",
        "<img src=\"https://miro.medium.com/max/1400/0*oxg81Sl5QDauFysz\">"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f83ZxKyvmmKt"
      },
      "source": [
        "def square_loss(labels, predictions):\n",
        "    loss = 0\n",
        "    for l, p in zip(labels, predictions):\n",
        "        loss = loss + (l - p) ** 2\n",
        "\n",
        "    loss = loss / len(labels)\n",
        "    return loss"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5xquKmuomn7q"
      },
      "source": [
        "def cost(var, features, labels):\n",
        "    preds = [quantum_neural_net(var, x) for x in features]\n",
        "    return square_loss(labels, preds)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Q4dwPi7PWyka"
      },
      "source": [
        "## Parameters Initialization\n",
        "4x5 Matrix of 4 layers 5 gate QNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PMh6TYsOmqju",
        "outputId": "0e0ee824-fb17-4e95-84c3-6d0e126f8d39"
      },
      "source": [
        "np.random.seed(0)\n",
        "num_layers = 4\n",
        "var_init = 0.05 * np.random.randn(num_layers, 5)\n",
        "print(var_init)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[ 0.08820262  0.02000786  0.0489369   0.11204466  0.0933779 ]\n",
            " [-0.04886389  0.04750442 -0.00756786 -0.00516094  0.02052993]\n",
            " [ 0.00720218  0.07271368  0.03805189  0.00608375  0.02219316]\n",
            " [ 0.01668372  0.07470395 -0.01025791  0.01565339 -0.04270479]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TNYSKZ6Pin-_"
      },
      "source": [
        "## Training QNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uoVNvmktm63F",
        "outputId": "7de9cde3-63d5-45c8-f240-25b2411ceab3"
      },
      "source": [
        "opt = AdamOptimizer(0.01)\n",
        "\n",
        "loss_plot = []\n",
        "\n",
        "var = var_init\n",
        "for it in range(100):\n",
        "    var, _cost = opt.step_and_cost(lambda v: cost(v, X, Y), var)\n",
        "    loss_plot.append(_cost)\n",
        "    print(\"Iter: {:5d} | Cost: {:0.7f} \".format(it, _cost))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Iter:     0 | Cost: 0.1132256 \n",
            "Iter:     1 | Cost: 0.0951169 \n",
            "Iter:     2 | Cost: 0.0840686 \n",
            "Iter:     3 | Cost: 0.0764727 \n",
            "Iter:     4 | Cost: 0.0711412 \n",
            "Iter:     5 | Cost: 0.0674666 \n",
            "Iter:     6 | Cost: 0.0650490 \n",
            "Iter:     7 | Cost: 0.0635786 \n",
            "Iter:     8 | Cost: 0.0627929 \n",
            "Iter:     9 | Cost: 0.0624615 \n",
            "Iter:    10 | Cost: 0.0623837 \n",
            "Iter:    11 | Cost: 0.0623909 \n",
            "Iter:    12 | Cost: 0.0623504 \n",
            "Iter:    13 | Cost: 0.0621685 \n",
            "Iter:    14 | Cost: 0.0617908 \n",
            "Iter:    15 | Cost: 0.0611994 \n",
            "Iter:    16 | Cost: 0.0604079 \n",
            "Iter:    17 | Cost: 0.0594534 \n",
            "Iter:    18 | Cost: 0.0583885 \n",
            "Iter:    19 | Cost: 0.0572725 \n",
            "Iter:    20 | Cost: 0.0561635 \n",
            "Iter:    21 | Cost: 0.0551127 \n",
            "Iter:    22 | Cost: 0.0541601 \n",
            "Iter:    23 | Cost: 0.0533322 \n",
            "Iter:    24 | Cost: 0.0526422 \n",
            "Iter:    25 | Cost: 0.0520908 \n",
            "Iter:    26 | Cost: 0.0516681 \n",
            "Iter:    27 | Cost: 0.0513564 \n",
            "Iter:    28 | Cost: 0.0511331 \n",
            "Iter:    29 | Cost: 0.0509728 \n",
            "Iter:    30 | Cost: 0.0508499 \n",
            "Iter:    31 | Cost: 0.0507402 \n",
            "Iter:    32 | Cost: 0.0506228 \n",
            "Iter:    33 | Cost: 0.0504805 \n",
            "Iter:    34 | Cost: 0.0503009 \n",
            "Iter:    35 | Cost: 0.0500762 \n",
            "Iter:    36 | Cost: 0.0498038 \n",
            "Iter:    37 | Cost: 0.0494851 \n",
            "Iter:    38 | Cost: 0.0491256 \n",
            "Iter:    39 | Cost: 0.0487339 \n",
            "Iter:    40 | Cost: 0.0483204 \n",
            "Iter:    41 | Cost: 0.0478966 \n",
            "Iter:    42 | Cost: 0.0474735 \n",
            "Iter:    43 | Cost: 0.0470612 \n",
            "Iter:    44 | Cost: 0.0466673 \n",
            "Iter:    45 | Cost: 0.0462963 \n",
            "Iter:    46 | Cost: 0.0459498 \n",
            "Iter:    47 | Cost: 0.0456260 \n",
            "Iter:    48 | Cost: 0.0453205 \n",
            "Iter:    49 | Cost: 0.0450269 \n",
            "Iter:    50 | Cost: 0.0447381 \n",
            "Iter:    51 | Cost: 0.0444472 \n",
            "Iter:    52 | Cost: 0.0441488 \n",
            "Iter:    53 | Cost: 0.0438394 \n",
            "Iter:    54 | Cost: 0.0435177 \n",
            "Iter:    55 | Cost: 0.0431852 \n",
            "Iter:    56 | Cost: 0.0428449 \n",
            "Iter:    57 | Cost: 0.0425013 \n",
            "Iter:    58 | Cost: 0.0421592 \n",
            "Iter:    59 | Cost: 0.0418230 \n",
            "Iter:    60 | Cost: 0.0414960 \n",
            "Iter:    61 | Cost: 0.0411802 \n",
            "Iter:    62 | Cost: 0.0408760 \n",
            "Iter:    63 | Cost: 0.0405824 \n",
            "Iter:    64 | Cost: 0.0402974 \n",
            "Iter:    65 | Cost: 0.0400187 \n",
            "Iter:    66 | Cost: 0.0397439 \n",
            "Iter:    67 | Cost: 0.0394711 \n",
            "Iter:    68 | Cost: 0.0391993 \n",
            "Iter:    69 | Cost: 0.0389284 \n",
            "Iter:    70 | Cost: 0.0386590 \n",
            "Iter:    71 | Cost: 0.0383924 \n",
            "Iter:    72 | Cost: 0.0381302 \n",
            "Iter:    73 | Cost: 0.0378738 \n",
            "Iter:    74 | Cost: 0.0376244 \n",
            "Iter:    75 | Cost: 0.0373824 \n",
            "Iter:    76 | Cost: 0.0371477 \n",
            "Iter:    77 | Cost: 0.0369196 \n",
            "Iter:    78 | Cost: 0.0366972 \n",
            "Iter:    79 | Cost: 0.0364794 \n",
            "Iter:    80 | Cost: 0.0362652 \n",
            "Iter:    81 | Cost: 0.0360543 \n",
            "Iter:    82 | Cost: 0.0358464 \n",
            "Iter:    83 | Cost: 0.0356418 \n",
            "Iter:    84 | Cost: 0.0354411 \n",
            "Iter:    85 | Cost: 0.0352448 \n",
            "Iter:    86 | Cost: 0.0350536 \n",
            "Iter:    87 | Cost: 0.0348675 \n",
            "Iter:    88 | Cost: 0.0346868 \n",
            "Iter:    89 | Cost: 0.0345110 \n",
            "Iter:    90 | Cost: 0.0343399 \n",
            "Iter:    91 | Cost: 0.0341730 \n",
            "Iter:    92 | Cost: 0.0340098 \n",
            "Iter:    93 | Cost: 0.0338502 \n",
            "Iter:    94 | Cost: 0.0336940 \n",
            "Iter:    95 | Cost: 0.0335412 \n",
            "Iter:    96 | Cost: 0.0333921 \n",
            "Iter:    97 | Cost: 0.0332467 \n",
            "Iter:    98 | Cost: 0.0331051 \n",
            "Iter:    99 | Cost: 0.0329673 \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "Skaspd4nn-5F",
        "outputId": "801854d7-f6cb-4f88-a435-298e4fadf026"
      },
      "source": [
        "plt.figure()\n",
        "plt.plot(loss_plot)\n",
        "plt.xlabel(\"Iter\")\n",
        "plt.ylabel(\"Cost\")\n",
        "plt.tick_params(axis=\"both\", which=\"major\")\n",
        "plt.tick_params(axis=\"both\", which=\"minor\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "73lSS3QATafi"
      },
      "source": [
        "## Evaluate"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1S7nq2EhTgry",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e30505fe-599e-4789-e01e-449d86e08ac0"
      },
      "source": [
        "x_pred = np.linspace(-1, 1, 50)\n",
        "predictions = [quantum_neural_net(var, x_) for x_ in x_pred]\n",
        "\n",
        "print(\"Mean square error : \", cost(var, X, Y))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mean square error :  0.032832992417139745\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "au9Bf77Rm9j9",
        "outputId": "a551172e-00a4-4df5-8ee6-11a89ea0ee1c"
      },
      "source": [
        "plt.figure()\n",
        "plt.scatter(X, Y, label= \"actual\")\n",
        "plt.scatter(x_pred, predictions, color=\"green\", label = \"prediction\")\n",
        "plt.xlabel(\"x\")\n",
        "plt.ylabel(\"f(x)\")\n",
        "plt.tick_params(axis=\"both\", which=\"major\")\n",
        "plt.tick_params(axis=\"both\", which=\"minor\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OpZhLARnrjEn"
      },
      "source": [
        "# **Titanic - Machine Learning from Disaster**\n",
        "Dataset from https://www.kaggle.com/c/titanic "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "twEIwMLyT-4h"
      },
      "source": [
        "## Load Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EoLnLQfuYCQZ",
        "outputId": "8351a211-ffdf-47e2-c5db-dea39a690604"
      },
      "source": [
        "!wget https://github.com/BankNatchapol/CV-QNN/raw/main/static/train.csv"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2021-07-22 13:41:28--  https://github.com/BankNatchapol/CV-QNN/raw/main/static/train.csv\n",
            "Resolving github.com (github.com)... 52.192.72.89\n",
            "Connecting to github.com (github.com)|52.192.72.89|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/BankNatchapol/CV-QNN/main/static/train.csv [following]\n",
            "--2021-07-22 13:41:28--  https://raw.githubusercontent.com/BankNatchapol/CV-QNN/main/static/train.csv\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 60302 (59K) [text/plain]\n",
            "Saving to: ‘train.csv’\n",
            "\n",
            "train.csv           100%[===================>]  58.89K  --.-KB/s    in 0.01s   \n",
            "\n",
            "2021-07-22 13:41:28 (4.37 MB/s) - ‘train.csv’ saved [60302/60302]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 407
        },
        "id": "YxSrkU6jBKqF",
        "outputId": "2ba5b27c-b26f-4b0a-a29d-f6bb1e7022c9"
      },
      "source": [
        "train = pd.read_csv('train.csv')\n",
        "train"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Braund, Mr. Owen Harris</td>\n",
              "      <td>male</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>A/5 21171</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
              "      <td>female</td>\n",
              "      <td>38.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>PC 17599</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>C85</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>Heikkinen, Miss. Laina</td>\n",
              "      <td>female</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>STON/O2. 3101282</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
              "      <td>female</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>113803</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>C123</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Allen, Mr. William Henry</td>\n",
              "      <td>male</td>\n",
              "      <td>35.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>373450</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>886</th>\n",
              "      <td>887</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>Montvila, Rev. Juozas</td>\n",
              "      <td>male</td>\n",
              "      <td>27.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>211536</td>\n",
              "      <td>13.0000</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>887</th>\n",
              "      <td>888</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Graham, Miss. Margaret Edith</td>\n",
              "      <td>female</td>\n",
              "      <td>19.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>112053</td>\n",
              "      <td>30.0000</td>\n",
              "      <td>B42</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>888</th>\n",
              "      <td>889</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Johnston, Miss. Catherine Helen \"Carrie\"</td>\n",
              "      <td>female</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>W./C. 6607</td>\n",
              "      <td>23.4500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>889</th>\n",
              "      <td>890</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>Behr, Mr. Karl Howell</td>\n",
              "      <td>male</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>111369</td>\n",
              "      <td>30.0000</td>\n",
              "      <td>C148</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>890</th>\n",
              "      <td>891</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Dooley, Mr. Patrick</td>\n",
              "      <td>male</td>\n",
              "      <td>32.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>370376</td>\n",
              "      <td>7.7500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>Q</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>891 rows × 12 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     PassengerId  Survived  Pclass  ...     Fare Cabin  Embarked\n",
              "0              1         0       3  ...   7.2500   NaN         S\n",
              "1              2         1       1  ...  71.2833   C85         C\n",
              "2              3         1       3  ...   7.9250   NaN         S\n",
              "3              4         1       1  ...  53.1000  C123         S\n",
              "4              5         0       3  ...   8.0500   NaN         S\n",
              "..           ...       ...     ...  ...      ...   ...       ...\n",
              "886          887         0       2  ...  13.0000   NaN         S\n",
              "887          888         1       1  ...  30.0000   B42         S\n",
              "888          889         0       3  ...  23.4500   NaN         S\n",
              "889          890         1       1  ...  30.0000  C148         C\n",
              "890          891         0       3  ...   7.7500   NaN         Q\n",
              "\n",
              "[891 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AMy-6GWfUCPW"
      },
      "source": [
        "## EDA"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xrApmKdrCOL8",
        "outputId": "f85fd598-0762-4e57-94e8-eca4f11f9c70"
      },
      "source": [
        "train.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 891 entries, 0 to 890\n",
            "Data columns (total 12 columns):\n",
            " #   Column       Non-Null Count  Dtype  \n",
            "---  ------       --------------  -----  \n",
            " 0   PassengerId  891 non-null    int64  \n",
            " 1   Survived     891 non-null    int64  \n",
            " 2   Pclass       891 non-null    int64  \n",
            " 3   Name         891 non-null    object \n",
            " 4   Sex          891 non-null    object \n",
            " 5   Age          714 non-null    float64\n",
            " 6   SibSp        891 non-null    int64  \n",
            " 7   Parch        891 non-null    int64  \n",
            " 8   Ticket       891 non-null    object \n",
            " 9   Fare         891 non-null    float64\n",
            " 10  Cabin        204 non-null    object \n",
            " 11  Embarked     889 non-null    object \n",
            "dtypes: float64(2), int64(5), object(5)\n",
            "memory usage: 83.7+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "W3WNCG_QUF_O"
      },
      "source": [
        "## Feature Extraction"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y9aJBiaMB8sM"
      },
      "source": [
        "feature_cols = ['Age', 'Fare']\n",
        "label = ['Survived']\n",
        "\n",
        "train_data = train[feature_cols]\n",
        "train_label = train[label]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MJVMcGd6Ck3E",
        "outputId": "1a0971dd-7caf-47e9-bb59-bcb1db827003"
      },
      "source": [
        "train_data['Age'] = train_data['Age'].fillna(train_data['Age'].mean())\n",
        "train_data.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 891 entries, 0 to 890\n",
            "Data columns (total 2 columns):\n",
            " #   Column  Non-Null Count  Dtype  \n",
            "---  ------  --------------  -----  \n",
            " 0   Age     891 non-null    float64\n",
            " 1   Fare    891 non-null    float64\n",
            "dtypes: float64(2)\n",
            "memory usage: 14.0 KB\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 198
        },
        "id": "ojy5Laf_Cv7r",
        "outputId": "a6b0b79c-4b0e-4a8f-ef60-58093ee48d57"
      },
      "source": [
        "train_data.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Age</th>\n",
              "      <th>Fare</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>22.0</td>\n",
              "      <td>7.2500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>38.0</td>\n",
              "      <td>71.2833</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>26.0</td>\n",
              "      <td>7.9250</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>35.0</td>\n",
              "      <td>53.1000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>35.0</td>\n",
              "      <td>8.0500</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    Age     Fare\n",
              "0  22.0   7.2500\n",
              "1  38.0  71.2833\n",
              "2  26.0   7.9250\n",
              "3  35.0  53.1000\n",
              "4  35.0   8.0500"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gYTLeFEXuRZv",
        "outputId": "d73e5af6-87d1-4734-ac1c-613b0dda85d4"
      },
      "source": [
        "# Standard Scaler\n",
        "train_data[['Age']] = (train_data[['Age']] - train_data[['Age']].min())/train_data[['Age']].max()\n",
        "train_data[['Fare']] = (train_data[['Fare']] - train_data[['Fare']].min())/train_data[['Fare']].max()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/pandas/core/frame.py:3069: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame.\n",
            "Try using .loc[row_indexer,col_indexer] = value instead\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 198
        },
        "id": "jp7-6tnevIiB",
        "outputId": "d15e5768-6fda-439d-d346-f8d217ee6d72"
      },
      "source": [
        "train_data.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Age</th>\n",
              "      <th>Fare</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.26975</td>\n",
              "      <td>0.014151</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.46975</td>\n",
              "      <td>0.139136</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.31975</td>\n",
              "      <td>0.015469</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.43225</td>\n",
              "      <td>0.103644</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.43225</td>\n",
              "      <td>0.015713</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       Age      Fare\n",
              "0  0.26975  0.014151\n",
              "1  0.46975  0.139136\n",
              "2  0.31975  0.015469\n",
              "3  0.43225  0.103644\n",
              "4  0.43225  0.015713"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e2DqGmcnULv2"
      },
      "source": [
        "## Continuous Variable Quantum Neural Network"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2WgCyKooffv9"
      },
      "source": [
        "<img src=\"https://github.com/BankNatchapol/CV-QNN/blob/main/static/titanic-qnn.png?raw=true\">"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EdXwDKtSed5l"
      },
      "source": [
        "num_modes = 2"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CHzB-TimDDWT"
      },
      "source": [
        "dev = qml.device(\"strawberryfields.fock\", wires=num_modes, cutoff_dim=10)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Y-YhDIfcUUuJ"
      },
      "source": [
        "### Define Layers"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OyiFam-JC2DG"
      },
      "source": [
        "def Interferometer(theta, phi, varphi, wires, mesh='rectangular', beamsplitter='pennylane'):\n",
        "    w = wires\n",
        "    M = len(w)\n",
        "\n",
        "    if M == 1:\n",
        "        # the interferometer is a single rotation\n",
        "        qml.Rotation(varphi[0], wires=w[0])\n",
        "        return\n",
        "\n",
        "    n = 0 # keep track of free parameters\n",
        "\n",
        "    if mesh == 'rectangular':\n",
        "        # Apply the Clements beamsplitter array\n",
        "        # The array depth is N\n",
        "        for l in range(M):\n",
        "            for k, (w1, w2) in enumerate(zip(w[:-1], w[1:])):\n",
        "                #skip even or odd pairs depending on layer\n",
        "                if (l+k)%2 != 1:\n",
        "                    if beamsplitter == 'clements':\n",
        "                        qml.Rotation(phi[n], wires=[w1])\n",
        "                        qml.Beamsplitter(theta[n], 0, wires=[w1, w2])\n",
        "                    else:\n",
        "                        qml.Beamsplitter(theta[n], phi[n], wires=[w1, w2])\n",
        "                    n += 1\n",
        "\n",
        "    elif mesh == 'triangular':\n",
        "        # apply the Reck beamsplitter array\n",
        "        # The array depth is 2*N-3\n",
        "        for l in range(2*M-3):\n",
        "            for k in range(abs(l+1-(M-1)), M-1, 2):\n",
        "                if beamsplitter == 'clements':\n",
        "                    qml.Rotation(phi[n], wires=[w[k]])\n",
        "                    qml.Beamsplitter(theta[n], 0, wires=[w[k], w[k+1]])\n",
        "                else:\n",
        "                    qml.Beamsplitter(theta[n], phi[n], wires=[w[k], w[k+1]])\n",
        "                n += 1\n",
        "\n",
        "    # apply the final local phase shifts to all modes\n",
        "    for i, p in enumerate(varphi):\n",
        "        qml.Rotation(p, wires=[w[i]])\n",
        "\n",
        "def layer(v):\n",
        "    wires = list(range(num_modes))\n",
        "\n",
        "    # Matrix multiplication of input layer\n",
        "    if len(wires) == 1:\n",
        "      s = 0\n",
        "      Interferometer(None, None, v[s], wires)\n",
        "    else:\n",
        "      s = (num_list[num_modes-1] - 3*num_modes)//2\n",
        "      Interferometer(v[0:s-num_modes+1], v[s-num_modes+1:2*(s-num_modes+1)], v[2*(s-num_modes+1):s], wires)\n",
        "    for i in range(len(wires)):\n",
        "      qml.Squeezing(v[s + i], 0.0, wires= i)\n",
        "\n",
        "    n = s + len(wires)\n",
        "    if len(wires) == 1:\n",
        "      sl = 0\n",
        "      Interferometer(None, None, v[n + sl], wires)\n",
        "    else:\n",
        "      sl = n+s\n",
        "      Interferometer(v[n:n+s-num_modes+1], v[n+s-num_modes+1:n+2*(s-num_modes+1)], v[n+2*(s-num_modes+1):sl], wires)\n",
        "    \n",
        "\n",
        "    # Bias\n",
        "    for i in range(len(wires)):\n",
        "      qml.Displacement(v[sl + i], 0.0, wires= i)\n",
        "    \n",
        "\n",
        "    # Element-wise nonlinear transformation\n",
        "    ken = sl + len(wires)\n",
        "    for i in range(len(wires)):\n",
        "      qml.Kerr(v[ken + i], wires= i)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ldap90L2KYv0"
      },
      "source": [
        "@qml.qnode(dev)\n",
        "def quantum_neural_net(var, x):\n",
        "    # Encode input x into quantum state\n",
        "    for i in range(num_modes):\n",
        "      qml.Displacement(x[i], 0.0, wires=i)\n",
        "\n",
        "    # \"layer\" subcircuits\n",
        "    for v in var:\n",
        "        layer(v)\n",
        "        \n",
        "    return [qml.expval(qml.X(i)) for i in range(num_modes)]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ali4Y1i_Ua8x"
      },
      "source": [
        "### Parameters Initialization"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xWKCYpOWLwYH",
        "outputId": "88a2e381-f1ac-4882-df63-28b499235274"
      },
      "source": [
        "num_list = [5, 12, 25]\n",
        "for i in range(4, 10):\n",
        "  num_list.append(i*3 + (2*(num_list[-1] - (i-1)*3) - (num_list[-2] - (i-2)*3) + 4))\n",
        "  \n",
        "\n",
        "np.random.seed(0)\n",
        "num_layers = 2\n",
        "var_init = 0.05 * np.random.randn(num_layers, num_list[num_modes-1])\n",
        "print(var_init)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[ 0.08820262  0.02000786  0.0489369   0.11204466  0.0933779  -0.04886389\n",
            "   0.04750442 -0.00756786 -0.00516094  0.02052993  0.00720218  0.07271368]\n",
            " [ 0.03805189  0.00608375  0.02219316  0.01668372  0.07470395 -0.01025791\n",
            "   0.01565339 -0.04270479 -0.12764949  0.03268093  0.04322181 -0.03710825]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9xR_6z_dUkkr"
      },
      "source": [
        "### Cost Function"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jIqmh-Pbfyvx"
      },
      "source": [
        "using Mean Square Error\n",
        "\n",
        "<img src=\"https://miro.medium.com/max/1400/0*oxg81Sl5QDauFysz\">"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oSAWGRTlgS8U"
      },
      "source": [
        "def square_loss(labels, predictions):\n",
        "    loss = 0\n",
        "    for l, p in zip(labels, predictions):\n",
        "        loss = loss + (l - p) ** 2\n",
        "\n",
        "    loss = loss / len(labels)\n",
        "    return loss"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lUKDTiLJZF6b"
      },
      "source": [
        "def cost(var, features, labels):\n",
        "    preds = [quantum_neural_net(var, x)[0] for x in features]\n",
        "    return square_loss(labels, preds)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wWnasRreUnK6"
      },
      "source": [
        "### Training QNN"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WhRMRAkEZPTF",
        "outputId": "a42a48a7-049a-42d1-ce72-5ef9b7dd6b82"
      },
      "source": [
        "X = np.array(train_data.to_numpy())\n",
        "Y = np.array(train_label.to_numpy().T[0])\n",
        "\n",
        "opt = AdamOptimizer(0.01, beta1=0.9, beta2=0.999)\n",
        "\n",
        "loss_plot = []\n",
        "\n",
        "var = var_init\n",
        "for it in range(100):\n",
        "    var, _cost = opt.step_and_cost(lambda v: cost(v, X, Y), var)\n",
        "    loss_plot.append(_cost)\n",
        "    print(\"Iter: {:5d} | Cost: {:0.7f} \".format(it, _cost))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Iter:     0 | Cost: 0.3392135 \n",
            "Iter:     1 | Cost: 0.3373101 \n",
            "Iter:     2 | Cost: 0.3360096 \n",
            "Iter:     3 | Cost: 0.3349394 \n",
            "Iter:     4 | Cost: 0.3339723 \n",
            "Iter:     5 | Cost: 0.3330457 \n",
            "Iter:     6 | Cost: 0.3321256 \n",
            "Iter:     7 | Cost: 0.3311931 \n",
            "Iter:     8 | Cost: 0.3302394 \n",
            "Iter:     9 | Cost: 0.3292618 \n",
            "Iter:    10 | Cost: 0.3282612 \n",
            "Iter:    11 | Cost: 0.3272405 \n",
            "Iter:    12 | Cost: 0.3262029 \n",
            "Iter:    13 | Cost: 0.3251510 \n",
            "Iter:    14 | Cost: 0.3240857 \n",
            "Iter:    15 | Cost: 0.3230063 \n",
            "Iter:    16 | Cost: 0.3219098 \n",
            "Iter:    17 | Cost: 0.3207919 \n",
            "Iter:    18 | Cost: 0.3196465 \n",
            "Iter:    19 | Cost: 0.3184668 \n",
            "Iter:    20 | Cost: 0.3172457 \n",
            "Iter:    21 | Cost: 0.3159762 \n",
            "Iter:    22 | Cost: 0.3146520 \n",
            "Iter:    23 | Cost: 0.3132676 \n",
            "Iter:    24 | Cost: 0.3118188 \n",
            "Iter:    25 | Cost: 0.3103022 \n",
            "Iter:    26 | Cost: 0.3087157 \n",
            "Iter:    27 | Cost: 0.3070578 \n",
            "Iter:    28 | Cost: 0.3053276 \n",
            "Iter:    29 | Cost: 0.3035245 \n",
            "Iter:    30 | Cost: 0.3016482 \n",
            "Iter:    31 | Cost: 0.2996983 \n",
            "Iter:    32 | Cost: 0.2976742 \n",
            "Iter:    33 | Cost: 0.2955756 \n",
            "Iter:    34 | Cost: 0.2934022 \n",
            "Iter:    35 | Cost: 0.2911541 \n",
            "Iter:    36 | Cost: 0.2888317 \n",
            "Iter:    37 | Cost: 0.2864360 \n",
            "Iter:    38 | Cost: 0.2839688 \n",
            "Iter:    39 | Cost: 0.2814327 \n",
            "Iter:    40 | Cost: 0.2788309 \n",
            "Iter:    41 | Cost: 0.2761676 \n",
            "Iter:    42 | Cost: 0.2734475 \n",
            "Iter:    43 | Cost: 0.2706764 \n",
            "Iter:    44 | Cost: 0.2678608 \n",
            "Iter:    45 | Cost: 0.2650081 \n",
            "Iter:    46 | Cost: 0.2621278 \n",
            "Iter:    47 | Cost: 0.2592319 \n",
            "Iter:    48 | Cost: 0.2563367 \n",
            "Iter:    49 | Cost: 0.2534645 \n",
            "Iter:    50 | Cost: 0.2506447 \n",
            "Iter:    51 | Cost: 0.2479116 \n",
            "Iter:    52 | Cost: 0.2452986 \n",
            "Iter:    53 | Cost: 0.2428295 \n",
            "Iter:    54 | Cost: 0.2405147 \n",
            "Iter:    55 | Cost: 0.2383556 \n",
            "Iter:    56 | Cost: 0.2363511 \n",
            "Iter:    57 | Cost: 0.2345063 \n",
            "Iter:    58 | Cost: 0.2328419 \n",
            "Iter:    59 | Cost: 0.2313920 \n",
            "Iter:    60 | Cost: 0.2301714 \n",
            "Iter:    61 | Cost: 0.2291457 \n",
            "Iter:    62 | Cost: 0.2282629 \n",
            "Iter:    63 | Cost: 0.2274961 \n",
            "Iter:    64 | Cost: 0.2268419 \n",
            "Iter:    65 | Cost: 0.2262930 \n",
            "Iter:    66 | Cost: 0.2258198 \n",
            "Iter:    67 | Cost: 0.2253823 \n",
            "Iter:    68 | Cost: 0.2249555 \n",
            "Iter:    69 | Cost: 0.2245419 \n",
            "Iter:    70 | Cost: 0.2241551 \n",
            "Iter:    71 | Cost: 0.2237939 \n",
            "Iter:    72 | Cost: 0.2234378 \n",
            "Iter:    73 | Cost: 0.2230709 \n",
            "Iter:    74 | Cost: 0.2227008 \n",
            "Iter:    75 | Cost: 0.2223470 \n",
            "Iter:    76 | Cost: 0.2220129 \n",
            "Iter:    77 | Cost: 0.2216838 \n",
            "Iter:    78 | Cost: 0.2213581 \n",
            "Iter:    79 | Cost: 0.2210537 \n",
            "Iter:    80 | Cost: 0.2207765 \n",
            "Iter:    81 | Cost: 0.2205140 \n",
            "Iter:    82 | Cost: 0.2202599 \n",
            "Iter:    83 | Cost: 0.2200238 \n",
            "Iter:    84 | Cost: 0.2198116 \n",
            "Iter:    85 | Cost: 0.2196068 \n",
            "Iter:    86 | Cost: 0.2194038 \n",
            "Iter:    87 | Cost: 0.2192218 \n",
            "Iter:    88 | Cost: 0.2190481 \n",
            "Iter:    89 | Cost: 0.2188819 \n",
            "Iter:    90 | Cost: 0.2187325 \n",
            "Iter:    91 | Cost: 0.2185764 \n",
            "Iter:    92 | Cost: 0.2184349 \n",
            "Iter:    93 | Cost: 0.2182838 \n",
            "Iter:    94 | Cost: 0.2181444 \n",
            "Iter:    95 | Cost: 0.2179963 \n",
            "Iter:    96 | Cost: 0.2178599 \n",
            "Iter:    97 | Cost: 0.2177206 \n",
            "Iter:    98 | Cost: 0.2175924 \n",
            "Iter:    99 | Cost: 0.2174640 \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "woTqGc8xQsVN",
        "outputId": "a3e5e165-6741-47cb-b500-dda99289ad1d"
      },
      "source": [
        "plt.figure()\n",
        "plt.plot(loss_plot)\n",
        "plt.xlabel(\"Iter\")\n",
        "plt.ylabel(\"Cost\")\n",
        "plt.tick_params(axis=\"both\", which=\"major\")\n",
        "plt.tick_params(axis=\"both\", which=\"minor\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uxyaw3n1Usdz"
      },
      "source": [
        "### Evaluate"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9ivxBVC1nArf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "78283409-e47f-4b90-9fc3-665bad23f373"
      },
      "source": [
        "re = []\n",
        "for i in range(len(X)):\n",
        "  re.append(int(quantum_neural_net(var, X[i])[0]>0.5))\n",
        "\n",
        "su = 0\n",
        "for i in range(len(Y)):\n",
        "  su += int(Y[i]==re[i])\n",
        "print(\"Accuracy : \", su/len(Y))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy :  0.67003367003367\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SFD8GsYe3gXf"
      },
      "source": [
        "## Compare Continuous Variable Quantum Neural Network with Classical Neural Network "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "X9-HWmMUoYHw",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0c7da4d3-fd10-47ef-a80a-e56ccb049e47"
      },
      "source": [
        "model = tf.keras.models.Sequential()\n",
        "model.add(tf.keras.layers.Dense(units = 10, activation = 'relu', input_shape = (2, )))\n",
        "model.add(tf.keras.layers.Dense(units = 10, activation = 'relu'))\n",
        "model.add(tf.keras.layers.Dense(units = 1, activation = 'sigmoid'))\n",
        "\n",
        "model.compile(optimizer = 'adam', loss = 'mean_squared_error', metrics = ['accuracy'])\n",
        "\n",
        "history = model.fit(X.reshape(-1, 2), Y, epochs =100)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "28/28 [==============================] - 1s 1ms/step - loss: 0.2407 - accuracy: 0.6296\n",
            "Epoch 2/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2351 - accuracy: 0.6207\n",
            "Epoch 3/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2326 - accuracy: 0.6207\n",
            "Epoch 4/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2312 - accuracy: 0.6218\n",
            "Epoch 5/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2303 - accuracy: 0.6207\n",
            "Epoch 6/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2289 - accuracy: 0.6263\n",
            "Epoch 7/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2279 - accuracy: 0.6251\n",
            "Epoch 8/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2269 - accuracy: 0.6285\n",
            "Epoch 9/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2261 - accuracy: 0.6285\n",
            "Epoch 10/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2251 - accuracy: 0.6308\n",
            "Epoch 11/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2241 - accuracy: 0.6352\n",
            "Epoch 12/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2233 - accuracy: 0.6375\n",
            "Epoch 13/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2225 - accuracy: 0.6476\n",
            "Epoch 14/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2217 - accuracy: 0.6487\n",
            "Epoch 15/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2214 - accuracy: 0.6510\n",
            "Epoch 16/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2201 - accuracy: 0.6521\n",
            "Epoch 17/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2195 - accuracy: 0.6532\n",
            "Epoch 18/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2188 - accuracy: 0.6566\n",
            "Epoch 19/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2180 - accuracy: 0.6577\n",
            "Epoch 20/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2175 - accuracy: 0.6566\n",
            "Epoch 21/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2168 - accuracy: 0.6588\n",
            "Epoch 22/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2161 - accuracy: 0.6655\n",
            "Epoch 23/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2156 - accuracy: 0.6633\n",
            "Epoch 24/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2151 - accuracy: 0.6577\n",
            "Epoch 25/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2146 - accuracy: 0.6566\n",
            "Epoch 26/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2143 - accuracy: 0.6577\n",
            "Epoch 27/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2139 - accuracy: 0.6622\n",
            "Epoch 28/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2142 - accuracy: 0.6588\n",
            "Epoch 29/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2131 - accuracy: 0.6667\n",
            "Epoch 30/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2127 - accuracy: 0.6655\n",
            "Epoch 31/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2126 - accuracy: 0.6622\n",
            "Epoch 32/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2123 - accuracy: 0.6779\n",
            "Epoch 33/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2123 - accuracy: 0.6667\n",
            "Epoch 34/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2114 - accuracy: 0.6835\n",
            "Epoch 35/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2112 - accuracy: 0.6768\n",
            "Epoch 36/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2110 - accuracy: 0.6902\n",
            "Epoch 37/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2108 - accuracy: 0.6779\n",
            "Epoch 38/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2103 - accuracy: 0.6790\n",
            "Epoch 39/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2101 - accuracy: 0.6869\n",
            "Epoch 40/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2099 - accuracy: 0.6880\n",
            "Epoch 41/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2097 - accuracy: 0.6902\n",
            "Epoch 42/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2100 - accuracy: 0.6779\n",
            "Epoch 43/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2098 - accuracy: 0.6914\n",
            "Epoch 44/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2095 - accuracy: 0.6857\n",
            "Epoch 45/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2098 - accuracy: 0.6880\n",
            "Epoch 46/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2091 - accuracy: 0.6902\n",
            "Epoch 47/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2090 - accuracy: 0.6914\n",
            "Epoch 48/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2092 - accuracy: 0.6914\n",
            "Epoch 49/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2088 - accuracy: 0.6891\n",
            "Epoch 50/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2087 - accuracy: 0.6925\n",
            "Epoch 51/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2087 - accuracy: 0.6902\n",
            "Epoch 52/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2088 - accuracy: 0.6914\n",
            "Epoch 53/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2087 - accuracy: 0.6914\n",
            "Epoch 54/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2084 - accuracy: 0.6936\n",
            "Epoch 55/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2087 - accuracy: 0.6902\n",
            "Epoch 56/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2086 - accuracy: 0.6869\n",
            "Epoch 57/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2084 - accuracy: 0.6947\n",
            "Epoch 58/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2081 - accuracy: 0.6925\n",
            "Epoch 59/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2082 - accuracy: 0.6869\n",
            "Epoch 60/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2082 - accuracy: 0.6914\n",
            "Epoch 61/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2081 - accuracy: 0.6902\n",
            "Epoch 62/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2079 - accuracy: 0.6914\n",
            "Epoch 63/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2080 - accuracy: 0.6925\n",
            "Epoch 64/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2077 - accuracy: 0.6925\n",
            "Epoch 65/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2078 - accuracy: 0.6925\n",
            "Epoch 66/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2078 - accuracy: 0.6914\n",
            "Epoch 67/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2079 - accuracy: 0.6936\n",
            "Epoch 68/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2076 - accuracy: 0.6902\n",
            "Epoch 69/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2075 - accuracy: 0.6891\n",
            "Epoch 70/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2074 - accuracy: 0.6914\n",
            "Epoch 71/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2075 - accuracy: 0.6936\n",
            "Epoch 72/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2072 - accuracy: 0.6936\n",
            "Epoch 73/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2073 - accuracy: 0.6958\n",
            "Epoch 74/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2070 - accuracy: 0.6925\n",
            "Epoch 75/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2072 - accuracy: 0.6936\n",
            "Epoch 76/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2069 - accuracy: 0.6936\n",
            "Epoch 77/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2072 - accuracy: 0.6936\n",
            "Epoch 78/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2074 - accuracy: 0.6936\n",
            "Epoch 79/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2068 - accuracy: 0.6936\n",
            "Epoch 80/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2068 - accuracy: 0.6936\n",
            "Epoch 81/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2068 - accuracy: 0.6981\n",
            "Epoch 82/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2066 - accuracy: 0.6970\n",
            "Epoch 83/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2065 - accuracy: 0.6981\n",
            "Epoch 84/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2071 - accuracy: 0.6880\n",
            "Epoch 85/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2066 - accuracy: 0.6958\n",
            "Epoch 86/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2070 - accuracy: 0.6936\n",
            "Epoch 87/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2069 - accuracy: 0.6914\n",
            "Epoch 88/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2064 - accuracy: 0.6992\n",
            "Epoch 89/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2065 - accuracy: 0.6914\n",
            "Epoch 90/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2062 - accuracy: 0.6936\n",
            "Epoch 91/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2062 - accuracy: 0.6936\n",
            "Epoch 92/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2061 - accuracy: 0.6992\n",
            "Epoch 93/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2061 - accuracy: 0.6970\n",
            "Epoch 94/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2059 - accuracy: 0.6981\n",
            "Epoch 95/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2061 - accuracy: 0.6981\n",
            "Epoch 96/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2058 - accuracy: 0.7003\n",
            "Epoch 97/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2057 - accuracy: 0.6992\n",
            "Epoch 98/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2059 - accuracy: 0.6970\n",
            "Epoch 99/100\n",
            "28/28 [==============================] - 0s 1ms/step - loss: 0.2057 - accuracy: 0.7015\n",
            "Epoch 100/100\n",
            "28/28 [==============================] - 0s 2ms/step - loss: 0.2056 - accuracy: 0.6992\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-kmpD4G8r8YH",
        "outputId": "b80caf53-97de-4ade-f7dd-ae65a2d09eae"
      },
      "source": [
        "print(\"Classical NN Accuracy : \", history.history['accuracy'][-1])\n",
        "print(\"Quantum NN Accuracy   : \", su/len(Y))\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Classical NN Accuracy :  0.6992143392562866\n",
            "Quantum NN Accuracy   :  0.67003367003367\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "v3f5H9boVGBD",
        "outputId": "e1a92b3b-65fb-4203-f54d-dcf937d86e6f"
      },
      "source": [
        "plt.figure()\n",
        "plt.plot(history.history['loss'], label = 'ANN loss')\n",
        "plt.plot(loss_plot, label = 'QNN loss')\n",
        "plt.xlabel(\"Iter\")\n",
        "plt.legend()\n",
        "plt.tick_params(axis=\"both\", which=\"major\")\n",
        "plt.tick_params(axis=\"both\", which=\"minor\")\n",
        "plt.show()\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o9ON6QxRr4cq"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}