MNIST-classification_digit_3_6_amplitude_encoding_classification_test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "pycharm": {
     "is_executing": true
    }
   },
   "outputs": [],
   "source": [
    "import pennylane as qml\n",
    "import tensorflow as tf\n",
    "import sklearn.datasets\n",
    "from sklearn.datasets import fetch_openml\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "tf.keras.backend.set_floatx('float64')"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load and reduce the data\n",
    "\n",
    "Load the MNIST data and check number of training and test examples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of original training examples: 60000\n",
      "Number of original test examples: 10000\n"
     ]
    }
   ],
   "source": [
    "(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()\n",
    "\n",
    "# Rescale the images from [0,255] to the [0.0,1.0] range.\n",
    "x_train, x_test = x_train[..., np.newaxis]/255.0, x_test[..., np.newaxis]/255.0  # add colour dimension\n",
    "\n",
    "print(\"Number of original training examples:\", len(x_train))\n",
    "print(\"Number of original test examples:\", len(x_test))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create a function to filter out only 3 and 6"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def filter_36(x, y):\n",
    "    keep = (y == 3) | (y == 6)\n",
    "    x, y = x[keep], y[keep]\n",
    "    y = y == 3\n",
    "    return x,y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of filtered training examples: 12049\n",
      "Number of filtered test examples: 1968\n"
     ]
    }
   ],
   "source": [
    "x_train, y_train = filter_36(x_train, y_train)\n",
    "x_test, y_test = filter_36(x_test, y_test)\n",
    "\n",
    "print(\"Number of filtered training examples:\", len(x_train))\n",
    "print(\"Number of filtered test examples:\", len(x_test))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot the original images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(x_train[0, :, :, 0], cmap='Greys')\n",
    "ax = plt.gca()\n",
    "ax.set_xticks([])\n",
    "ax.set_yticks([])\n",
    "plt.title(\"28x28 image of a 0\", fontsize=20)\n",
    "# save to file as pdf\n",
    "plt.savefig('28x28_MNIST_0.pdf', bbox_inches='tight', pad_inches=0)\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Reduce images to 3x3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_train_small = tf.image.resize(x_train, (3,3)).numpy()\n",
    "# x_test_small = tf.image.resize(x_test, (3,3)).numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plt.imshow(x_train_small[0,:,:,0], vmin=0, vmax=1, cmap='Greys')\n",
    "# ax = plt.gca()\n",
    "# ax.set_xticks([])\n",
    "# ax.set_yticks([])\n",
    "# plt.title(\"9x9 image of a 0\", fontsize=20)\n",
    "# # save to file as pdf\n",
    "# plt.savefig('9x9_MNIST_0.pdf', bbox_inches='tight', pad_inches=0)\n",
    "# plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Flatten 28x28 to 784x1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# x_train_small = x_train_small[:,4,:,:].reshape((-1,9))  # take only central slices\n",
    "# x_test_small = x_test_small[:,4,:,:].reshape((-1,9))  # take only central slices\n",
    "\n",
    "x_train_small = x_train.reshape((-1,784))  \n",
    "x_test_small = x_test.reshape((-1,784))  \n",
    "\n",
    "plt.imshow(x_train_small[0,:].reshape((1,-1)), vmin=0, vmax=1, cmap='Greys')\n",
    "ax = plt.gca()\n",
    "ax.set_xticks([])\n",
    "ax.set_yticks([])\n",
    "plt.title(\"784x1 image of a 0\", fontsize=20)\n",
    "# save to file as pdf\n",
    "plt.savefig('784x1_MNIST_0.pdf', bbox_inches='tight', pad_inches=0)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(784,)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train_small[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "784\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(1968, 784)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(len(x_train_small[0]))\n",
    "x_test_small.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.02745098,\n",
       "       0.8       , 0.99215686, 0.69019608, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.02745098, 0.58823529, 0.98823529, 0.98823529,\n",
       "       0.49019608, 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.45882353,\n",
       "       0.98823529, 0.72941176, 0.21960784, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.55294118, 0.98823529, 0.4627451 ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.60392157, 0.96862745, 0.19607843, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.10196078, 0.99215686, 0.76862745,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.58823529, 0.99215686, 0.76862745, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.22352941, 0.33333333, 0.33333333, 0.14901961, 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.88235294, 0.99215686,\n",
       "       0.37647059, 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.59215686, 0.88627451, 0.95294118, 0.98823529,\n",
       "       0.98823529, 0.93333333, 0.49019608, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.03921569, 0.89803922, 0.88627451, 0.        , 0.        ,\n",
       "       0.        , 0.01568627, 0.21176471, 0.89803922, 0.99215686,\n",
       "       1.        , 0.91764706, 0.68627451, 0.88235294, 1.        ,\n",
       "       0.89411765, 0.12156863, 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.43137255, 0.98823529,\n",
       "       0.58823529, 0.        , 0.        , 0.10196078, 0.50196078,\n",
       "       0.98823529, 0.98823529, 0.89019608, 0.5254902 , 0.10980392,\n",
       "       0.        , 0.        , 0.69803922, 0.98823529, 0.21960784,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.62352941, 0.98823529, 0.44313725, 0.        ,\n",
       "       0.        , 0.58823529, 0.99215686, 0.98823529, 0.72941176,\n",
       "       0.16862745, 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.55294118, 0.98823529, 0.21960784, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.7254902 ,\n",
       "       0.98823529, 0.44313725, 0.        , 0.14901961, 0.92941176,\n",
       "       0.99215686, 0.59215686, 0.02352941, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.55294118, 0.79215686,\n",
       "       0.02352941, 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.77647059, 0.99215686, 0.44705882,\n",
       "       0.        , 0.57647059, 0.99215686, 0.63921569, 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.60392157, 0.77254902, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.77254902, 0.98823529, 0.44313725, 0.        , 0.6745098 ,\n",
       "       0.98823529, 0.7372549 , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.10196078, 0.99215686,\n",
       "       0.67058824, 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.77254902, 0.98823529,\n",
       "       0.44313725, 0.        , 0.0745098 , 0.90588235, 0.96862745,\n",
       "       0.47843137, 0.0745098 , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.78431373, 0.95686275, 0.21960784, 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.10196078, 0.87058824, 0.98823529, 0.44313725, 0.        ,\n",
       "       0.        , 0.09803922, 0.79607843, 0.98823529, 0.75686275,\n",
       "       0.05098039, 0.        , 0.29803922, 0.78431373, 0.97647059,\n",
       "       0.49019608, 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.7254902 ,\n",
       "       0.99215686, 0.70196078, 0.03921569, 0.        , 0.        ,\n",
       "       0.        , 0.29803922, 0.1372549 , 0.11372549, 0.60392157,\n",
       "       0.99215686, 0.95686275, 0.49019608, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.10980392, 0.81960784, 0.99215686,\n",
       "       0.76862745, 0.32156863, 0.22352941, 0.22352941, 0.51372549,\n",
       "       0.77254902, 0.98823529, 0.99215686, 0.83921569, 0.31764706,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.09803922, 0.84705882, 0.98823529, 0.98823529,\n",
       "       0.98823529, 0.99215686, 0.98823529, 0.98823529, 0.98823529,\n",
       "       0.61176471, 0.0745098 , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.0627451 , 0.40392157, 0.54509804, 0.94117647, 0.54901961,\n",
       "       0.54509804, 0.54509804, 0.15686275, 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "       0.        , 0.        , 0.        , 0.        ])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_test_small[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "784"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(x_train_small[0,:])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_train_small_amplitude_encoded = qml.AmplitudeEmbedding(x_train_small, wires=range(10), pad_with=0.0, normalize=True)  # for pennylane 0.11.0\n",
    "# print(x_train_small_amplitude_encoded.parameters)\n",
    "# print(len(x_train_small_amplitude_encoded.parameters[0]))\n",
    "# len(x_train_small_amplitude_encoded.parameters[0][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# x_test_small_amplitude_encoded = qml.AmplitudeEmbedding(x_test_small, wires=range(10), pad_with=0.0, normalize=True)  # for pennylane 0.11.0\n",
    "# print(x_test_small_amplitude_encoded.parameters)\n",
    "# print(len(x_test_small_amplitude_encoded.parameters[0]))\n",
    "# len(x_test_small_amplitude_encoded.parameters[0][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.max(x_train_small[0,:])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum((x_train_small[0,:]/np.sqrt(np.sum(x_train_small[0,:]**2)))**2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 784)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train_small[0,:].reshape(1,-1).shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.14901961, 0.16862745, 0.41176471, 1.        ,\n",
       "        0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.99215686,\n",
       "        0.68235294, 0.02352941, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.16862745, 0.54509804, 0.87843137,\n",
       "        0.88627451, 0.98823529, 0.99215686, 0.98823529, 0.98823529,\n",
       "        0.98823529, 0.98823529, 0.98823529, 0.98823529, 0.61960784,\n",
       "        0.05490196, 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.69803922, 0.98823529, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.99215686, 0.98823529, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.98823529, 0.98823529, 0.98823529, 0.23137255, 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.42745098, 0.98823529,\n",
       "        0.98823529, 0.90196078, 0.51764706, 0.52156863, 0.51764706,\n",
       "        0.51764706, 0.74117647, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.98823529, 0.23137255, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.01568627, 0.11372549, 0.11372549, 0.09411765,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.05490196,\n",
       "        0.88627451, 0.98823529, 0.98823529, 0.6745098 , 0.02745098,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.33333333, 0.95294118, 0.98823529,\n",
       "        0.98823529, 0.56470588, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.34509804,\n",
       "        0.74117647, 0.98823529, 0.98823529, 0.98823529, 0.05490196,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.35686275, 0.83137255, 0.96862745, 0.98823529, 0.98823529,\n",
       "        0.98823529, 0.8       , 0.03529412, 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.1254902 , 0.49019608,\n",
       "        0.75686275, 0.75686275, 0.75686275, 0.99215686, 0.98823529,\n",
       "        0.98823529, 0.98823529, 0.93333333, 0.4       , 0.10980392,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.17647059, 0.87058824, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.98823529, 0.99215686, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.69411765, 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.17647059, 0.8745098 ,\n",
       "        0.99215686, 0.99215686, 0.99215686, 0.99215686, 1.        ,\n",
       "        0.99215686, 0.99215686, 0.99215686, 0.99215686, 0.29019608,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.12156863, 0.48235294, 0.20392157,\n",
       "        0.17254902, 0.17254902, 0.17254902, 0.17254902, 0.56078431,\n",
       "        0.98823529, 0.98823529, 0.29019608, 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.05882353, 0.98823529, 0.98823529,\n",
       "        0.29019608, 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.3372549 , 0.98823529, 0.98823529, 0.29019608, 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.01960784, 0.29411765,\n",
       "        0.03529412, 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.38431373, 0.94901961, 0.98823529,\n",
       "        0.98823529, 0.29019608, 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.23921569, 0.71764706, 0.98823529, 0.11372549, 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.07058824, 0.36078431,\n",
       "        0.9372549 , 0.98823529, 0.98823529, 0.95294118, 0.25490196,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.81568627, 0.98823529,\n",
       "        0.98823529, 0.57647059, 0.5254902 , 0.5254902 , 0.5254902 ,\n",
       "        0.5254902 , 0.79607843, 0.99215686, 0.98823529, 0.98823529,\n",
       "        0.7372549 , 0.3254902 , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.81568627, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.98823529, 0.98823529, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.99215686, 0.90196078, 0.6       , 0.03137255, 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.19215686,\n",
       "        0.61568627, 0.98823529, 0.98823529, 0.98823529, 0.98823529,\n",
       "        0.98823529, 0.85098039, 0.81176471, 0.57254902, 0.17647059,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.02745098, 0.40392157,\n",
       "        0.92156863, 0.98823529, 0.6745098 , 0.40392157, 0.09411765,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        , 0.        ,\n",
       "        0.        , 0.        , 0.        , 0.        ]])"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train_small[0,:].reshape(1,-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(784,)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_train_small[0,:].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "range(0, 10)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "range(10)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Set up the Quantum Neural Network\n",
    "\n",
    "Create quantum circuit using pennylane qnode, then conver this to a Keras layer to make a Keras model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 271ms/step\n",
      "Model: \"QNN\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " inputs (InputLayer)         [(None, 784)]             0         \n",
      "                                                                 \n",
      " quantumLayer (KerasLayer)   (None, 1)                 57        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 57\n",
      "Trainable params: 57\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "n_qubits = 9\n",
    "n_layers = 1\n",
    "dev = qml.device(\"default.qubit\", wires=n_qubits)\n",
    "\n",
    "@qml.qnode(dev, diff_method=\"adjoint\")\n",
    "def qnode(inputs, weights):\n",
    "    \n",
    "    qml.AmplitudeEmbedding(features=inputs, wires=range(10),normalize=True,pad_with=0)  \n",
    "    # for ii in range(n_qubits):\n",
    "        # qml.RY(np.pi * inputs[ii], wires=ii)\n",
    "\n",
    "        \n",
    "    for jj in range(n_layers):\n",
    "        for ii in range(n_qubits-1):\n",
    "            qml.RZ(weights[jj,2*ii,0], wires=0)\n",
    "            qml.RY(weights[jj,2*ii,1], wires=0)\n",
    "            qml.RZ(weights[jj,2*ii,2], wires=0)\n",
    "\n",
    "            qml.RZ(weights[jj,2*ii+1,0], wires=ii+1)\n",
    "            qml.RY(weights[jj,2*ii+1,1], wires=ii+1)\n",
    "            qml.RZ(weights[jj,2*ii+1,2], wires=ii+1)\n",
    "\n",
    "            #qml.U3(*list(weights[jj,2*ii,:]), wires=0)\n",
    "            #qml.U3(*list(weights[jj,2*ii+1,:]), wires=ii+1)\n",
    "            qml.CNOT(wires=[ii+1,0])\n",
    "            \n",
    "        qml.RZ(weights[jj,2*(n_qubits-1),0], wires=0)\n",
    "        qml.RY(weights[jj,2*(n_qubits-1),1], wires=0)\n",
    "        qml.RZ(weights[jj,2*(n_qubits-1),2], wires=0)\n",
    "\n",
    "    return qml.expval(qml.PauliZ(0))\n",
    "\n",
    "weight_shapes = {\"weights\": (n_layers, 2*(n_qubits-1)+1, 3) }\n",
    "\n",
    "\n",
    "qlayer = qml.qnn.KerasLayer(qnode, weight_shapes, output_dim=1, name=\"quantumLayer\")\n",
    "inputs = tf.keras.Input(shape=(784,), name=\"inputs\")\n",
    "\n",
    "outputs = qlayer(inputs)\n",
    "model = tf.keras.Model(inputs=inputs, outputs=outputs, name=\"QNN\")\n",
    "\n",
    "# testImage = x_train_small[1,:]  /np.sqrt(np.sum(x_train_small[1,:]**2))\n",
    "# print(testImage)\n",
    "model.predict(x_train_small[0,:].reshape(1,-1))\n",
    "print(model.summary())"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Redefine the classification values to match the Hinge loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train_hinge = 2.0*y_train-1.0\n",
    "y_test_hinge = 2.0*y_test-1.0"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Define the hinge accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hinge_accuracy(y_true, y_pred):\n",
    "    y_true = tf.squeeze(y_true) > 0.0\n",
    "    y_pred = tf.squeeze(y_pred) > 0.0\n",
    "    result = tf.cast(y_true == y_pred, tf.float32)\n",
    "\n",
    "    return tf.reduce_mean(result)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Compile and train the model\n",
    "\n",
    "Compile the model using the hinge loss and accuracy and the Adam optimiser"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(\n",
    "    loss=tf.keras.losses.Hinge(),\n",
    "    optimizer=tf.keras.optimizers.Adam(),\n",
    "    metrics=[hinge_accuracy])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"QNN\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " inputs (InputLayer)         [(None, 784)]             0         \n",
      "                                                                 \n",
      " quantumLayer (KerasLayer)   (None, 1)                 57        \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 57\n",
      "Trainable params: 57\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(model.summary())"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Train the model over 20 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "512\n",
      "Epoch 1/20\n",
      "16/16 [==============================] - 101s 6s/step - loss: 1.0044 - hinge_accuracy: 0.4414 - val_loss: 1.0026 - val_hinge_accuracy: 0.4587\n",
      "Epoch 2/20\n",
      "16/16 [==============================] - 108s 7s/step - loss: 1.0033 - hinge_accuracy: 0.4453 - val_loss: 1.0018 - val_hinge_accuracy: 0.4738\n",
      "Epoch 3/20\n",
      "16/16 [==============================] - 100s 6s/step - loss: 1.0022 - hinge_accuracy: 0.4570 - val_loss: 1.0009 - val_hinge_accuracy: 0.4899\n",
      "Epoch 4/20\n",
      "16/16 [==============================] - 116s 8s/step - loss: 1.0011 - hinge_accuracy: 0.4746 - val_loss: 1.0000 - val_hinge_accuracy: 0.5025\n",
      "Epoch 5/20\n",
      "16/16 [==============================] - 135s 9s/step - loss: 0.9999 - hinge_accuracy: 0.4941 - val_loss: 0.9991 - val_hinge_accuracy: 0.5166\n",
      "Epoch 6/20\n",
      "16/16 [==============================] - 128s 8s/step - loss: 0.9987 - hinge_accuracy: 0.5156 - val_loss: 0.9980 - val_hinge_accuracy: 0.5292\n",
      "Epoch 7/20\n",
      "16/16 [==============================] - 131s 8s/step - loss: 0.9973 - hinge_accuracy: 0.5449 - val_loss: 0.9967 - val_hinge_accuracy: 0.5534\n",
      "Epoch 8/20\n",
      "16/16 [==============================] - 125s 8s/step - loss: 0.9956 - hinge_accuracy: 0.5684 - val_loss: 0.9952 - val_hinge_accuracy: 0.5736\n",
      "Epoch 9/20\n",
      "16/16 [==============================] - 101s 7s/step - loss: 0.9937 - hinge_accuracy: 0.6016 - val_loss: 0.9932 - val_hinge_accuracy: 0.5968\n",
      "Epoch 10/20\n",
      "16/16 [==============================] - 97s 6s/step - loss: 0.9913 - hinge_accuracy: 0.6348 - val_loss: 0.9907 - val_hinge_accuracy: 0.6270\n",
      "Epoch 11/20\n",
      "16/16 [==============================] - 100s 6s/step - loss: 0.9883 - hinge_accuracy: 0.6699 - val_loss: 0.9878 - val_hinge_accuracy: 0.6704\n",
      "Epoch 12/20\n",
      "16/16 [==============================] - 97s 6s/step - loss: 0.9848 - hinge_accuracy: 0.7148 - val_loss: 0.9841 - val_hinge_accuracy: 0.7157\n",
      "Epoch 13/20\n",
      "16/16 [==============================] - 97s 6s/step - loss: 0.9807 - hinge_accuracy: 0.7539 - val_loss: 0.9798 - val_hinge_accuracy: 0.7571\n",
      "Epoch 14/20\n",
      "16/16 [==============================] - 97s 6s/step - loss: 0.9759 - hinge_accuracy: 0.8008 - val_loss: 0.9750 - val_hinge_accuracy: 0.7959\n",
      "Epoch 15/20\n",
      "16/16 [==============================] - 98s 6s/step - loss: 0.9705 - hinge_accuracy: 0.8242 - val_loss: 0.9695 - val_hinge_accuracy: 0.8246\n",
      "Epoch 16/20\n",
      "16/16 [==============================] - 98s 6s/step - loss: 0.9646 - hinge_accuracy: 0.8496 - val_loss: 0.9636 - val_hinge_accuracy: 0.8438\n",
      "Epoch 17/20\n",
      "16/16 [==============================] - 98s 6s/step - loss: 0.9583 - hinge_accuracy: 0.8828 - val_loss: 0.9573 - val_hinge_accuracy: 0.8493\n",
      "Epoch 18/20\n",
      "16/16 [==============================] - 110s 7s/step - loss: 0.9516 - hinge_accuracy: 0.8906 - val_loss: 0.9507 - val_hinge_accuracy: 0.8644\n",
      "Epoch 19/20\n",
      "16/16 [==============================] - 97s 6s/step - loss: 0.9447 - hinge_accuracy: 0.8965 - val_loss: 0.9440 - val_hinge_accuracy: 0.8695\n",
      "Epoch 20/20\n",
      "16/16 [==============================] - 98s 6s/step - loss: 0.9378 - hinge_accuracy: 0.9043 - val_loss: 0.9373 - val_hinge_accuracy: 0.8735\n",
      "62/62 [==============================] - 75s 1s/step - loss: 0.9373 - hinge_accuracy: 0.8735\n"
     ]
    }
   ],
   "source": [
    "EPOCHS = 20\n",
    "BATCH_SIZE = 32\n",
    "\n",
    "NUM_EXAMPLES = 512 #x_train_small.shape[0]\n",
    "print(NUM_EXAMPLES)\n",
    "\n",
    "x_train_sub = x_train_small[:NUM_EXAMPLES,:]\n",
    "y_train_hinge_sub = y_train_hinge[:NUM_EXAMPLES]\n",
    "\n",
    "x_test_sub = x_test_small[:,:]\n",
    "y_test_sub = y_test[:]\n",
    "\n",
    "qnn_history_1layer = model.fit(\n",
    "      x_train_sub, y_train_hinge_sub,\n",
    "      batch_size=BATCH_SIZE,\n",
    "      epochs=EPOCHS,\n",
    "      verbose=1,\n",
    "      validation_data=(x_test_sub, y_test_sub))\n",
    "\n",
    "qnn_results_1layer = model.evaluate(x_test_sub, y_test_sub)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'loss': [1.004409705272598,\n",
       "  1.0032886756188792,\n",
       "  1.0021600469578005,\n",
       "  1.001076920294305,\n",
       "  0.9999170290270583,\n",
       "  0.9986699472477788,\n",
       "  0.9972823361987804,\n",
       "  0.9956034711325031,\n",
       "  0.9936620762124166,\n",
       "  0.9912829145356873,\n",
       "  0.9882711927589484,\n",
       "  0.9847859351434506,\n",
       "  0.9806559193061809,\n",
       "  0.9759030747458758,\n",
       "  0.9705407325567663,\n",
       "  0.9645502787457241,\n",
       "  0.9583170871585145,\n",
       "  0.9515678646171704,\n",
       "  0.9446537112724052,\n",
       "  0.937824892655233],\n",
       " 'hinge_accuracy': [0.44140625,\n",
       "  0.4453125,\n",
       "  0.45703125,\n",
       "  0.474609375,\n",
       "  0.494140625,\n",
       "  0.515625,\n",
       "  0.544921875,\n",
       "  0.568359375,\n",
       "  0.6015625,\n",
       "  0.634765625,\n",
       "  0.669921875,\n",
       "  0.71484375,\n",
       "  0.75390625,\n",
       "  0.80078125,\n",
       "  0.82421875,\n",
       "  0.849609375,\n",
       "  0.8828125,\n",
       "  0.890625,\n",
       "  0.896484375,\n",
       "  0.904296875],\n",
       " 'val_loss': [1.002605521035271,\n",
       "  1.0017558787226502,\n",
       "  1.0009206551285064,\n",
       "  1.0000341811532476,\n",
       "  0.9990987862554529,\n",
       "  0.9980154484003643,\n",
       "  0.9966983837559384,\n",
       "  0.9951534403779873,\n",
       "  0.9931841100199977,\n",
       "  0.9906973598531996,\n",
       "  0.9877893586956417,\n",
       "  0.9841470064880026,\n",
       "  0.9798490097765961,\n",
       "  0.9750222952419253,\n",
       "  0.9694985769622219,\n",
       "  0.9635821952466072,\n",
       "  0.9572713135322642,\n",
       "  0.9506545017682564,\n",
       "  0.9439900387910984,\n",
       "  0.9372605483045949],\n",
       " 'val_hinge_accuracy': [0.4586693548387097,\n",
       "  0.4737903225806452,\n",
       "  0.4899193548387097,\n",
       "  0.5025201612903226,\n",
       "  0.516633064516129,\n",
       "  0.5292338709677419,\n",
       "  0.5534274193548387,\n",
       "  0.5735887096774194,\n",
       "  0.5967741935483871,\n",
       "  0.6270161290322581,\n",
       "  0.6703629032258065,\n",
       "  0.7157258064516129,\n",
       "  0.7570564516129032,\n",
       "  0.795866935483871,\n",
       "  0.8245967741935484,\n",
       "  0.84375,\n",
       "  0.8492943548387096,\n",
       "  0.8644153225806451,\n",
       "  0.8694556451612904,\n",
       "  0.8734879032258065]}"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qnn_history_1layer.history"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create and train a 2 layer version of the QNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 97ms/step\n",
      "Model: \"QNN\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " inputs (InputLayer)         [(None, 784)]             0         \n",
      "                                                                 \n",
      " quantumLayer (KerasLayer)   (None, 1)                 114       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 114\n",
      "Trainable params: 114\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "n_qubits = 10\n",
    "n_layers = 2\n",
    "dev = qml.device(\"default.qubit\", wires=n_qubits)\n",
    "\n",
    "@qml.qnode(dev, diff_method=\"adjoint\")\n",
    "def qnode(inputs, weights):\n",
    "\n",
    "    qml.AmplitudeEmbedding(features=inputs, wires=range(10),normalize=True,pad_with=0)  \n",
    "\n",
    "    # for ii in range(n_qubits):\n",
    "    #     qml.RY(np.pi * inputs[ii], wires=ii)\n",
    "\n",
    "    for jj in range(n_layers):\n",
    "        for ii in range(n_qubits-1):\n",
    "            qml.RZ(weights[jj,2*ii,0], wires=0)\n",
    "            qml.RY(weights[jj,2*ii,1], wires=0)\n",
    "            qml.RZ(weights[jj,2*ii,2], wires=0)\n",
    "\n",
    "            qml.RZ(weights[jj,2*ii+1,0], wires=ii+1)\n",
    "            qml.RY(weights[jj,2*ii+1,1], wires=ii+1)\n",
    "            qml.RZ(weights[jj,2*ii+1,2], wires=ii+1)\n",
    "\n",
    "            #qml.U3(*list(weights[jj,2*ii,:]), wires=0)\n",
    "            #qml.U3(*list(weights[jj,2*ii+1,:]), wires=ii+1)\n",
    "            qml.CNOT(wires=[ii+1,0])\n",
    "            \n",
    "        qml.RZ(weights[jj,2*(n_qubits-1),0], wires=0)\n",
    "        qml.RY(weights[jj,2*(n_qubits-1),1], wires=0)\n",
    "        qml.RZ(weights[jj,2*(n_qubits-1),2], wires=0)\n",
    "\n",
    "    return qml.expval(qml.PauliZ(0))\n",
    "\n",
    "weight_shapes = {\"weights\": (n_layers, 2*(n_qubits-1)+1, 3) }\n",
    "\n",
    "\n",
    "qlayer = qml.qnn.KerasLayer(qnode, weight_shapes, output_dim=1, name=\"quantumLayer\")\n",
    "inputs = tf.keras.Input(shape=(784,), name=\"inputs\")\n",
    "outputs = qlayer(inputs)\n",
    "model2 = tf.keras.Model(inputs=inputs, outputs=outputs, name=\"QNN\")\n",
    "\n",
    "model2.predict(x_train_small[0,:].reshape(1,-1))\n",
    "print(model2.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "model2.compile(\n",
    "    loss=tf.keras.losses.Hinge(),\n",
    "    optimizer=tf.keras.optimizers.Adam(),\n",
    "    metrics=[hinge_accuracy])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "512\n",
      "Epoch 1/20\n",
      "16/16 [==============================] - 212s 14s/step - loss: 0.9919 - hinge_accuracy: 0.4902 - val_loss: 0.9878 - val_hinge_accuracy: 0.5116\n",
      "Epoch 2/20\n",
      "16/16 [==============================] - 205s 13s/step - loss: 0.9832 - hinge_accuracy: 0.4902 - val_loss: 0.9794 - val_hinge_accuracy: 0.5121\n",
      "Epoch 3/20\n",
      "16/16 [==============================] - 213s 14s/step - loss: 0.9732 - hinge_accuracy: 0.4922 - val_loss: 0.9708 - val_hinge_accuracy: 0.5131\n",
      "Epoch 4/20\n",
      "16/16 [==============================] - 221s 14s/step - loss: 0.9633 - hinge_accuracy: 0.4961 - val_loss: 0.9617 - val_hinge_accuracy: 0.5197\n",
      "Epoch 5/20\n",
      "16/16 [==============================] - 212s 14s/step - loss: 0.9530 - hinge_accuracy: 0.5078 - val_loss: 0.9521 - val_hinge_accuracy: 0.5307\n",
      "Epoch 6/20\n",
      "16/16 [==============================] - 217s 14s/step - loss: 0.9422 - hinge_accuracy: 0.5293 - val_loss: 0.9425 - val_hinge_accuracy: 0.5570\n",
      "Epoch 7/20\n",
      "16/16 [==============================] - 241s 16s/step - loss: 0.9315 - hinge_accuracy: 0.5781 - val_loss: 0.9329 - val_hinge_accuracy: 0.6295\n",
      "Epoch 8/20\n",
      "16/16 [==============================] - 241s 16s/step - loss: 0.9211 - hinge_accuracy: 0.6738 - val_loss: 0.9235 - val_hinge_accuracy: 0.7172\n",
      "Epoch 9/20\n",
      "16/16 [==============================] - 241s 16s/step - loss: 0.9112 - hinge_accuracy: 0.7637 - val_loss: 0.9142 - val_hinge_accuracy: 0.8231\n",
      "Epoch 10/20\n",
      "16/16 [==============================] - 242s 16s/step - loss: 0.9016 - hinge_accuracy: 0.8516 - val_loss: 0.9057 - val_hinge_accuracy: 0.8710\n",
      "Epoch 11/20\n",
      "16/16 [==============================] - 241s 16s/step - loss: 0.8928 - hinge_accuracy: 0.8926 - val_loss: 0.8975 - val_hinge_accuracy: 0.9133\n",
      "Epoch 12/20\n",
      "16/16 [==============================] - 244s 16s/step - loss: 0.8845 - hinge_accuracy: 0.9355 - val_loss: 0.8898 - val_hinge_accuracy: 0.9304\n",
      "Epoch 13/20\n",
      "16/16 [==============================] - 243s 16s/step - loss: 0.8765 - hinge_accuracy: 0.9492 - val_loss: 0.8825 - val_hinge_accuracy: 0.9400\n",
      "Epoch 14/20\n",
      "16/16 [==============================] - 239s 16s/step - loss: 0.8689 - hinge_accuracy: 0.9551 - val_loss: 0.8754 - val_hinge_accuracy: 0.9476\n",
      "Epoch 15/20\n",
      "16/16 [==============================] - 241s 16s/step - loss: 0.8616 - hinge_accuracy: 0.9570 - val_loss: 0.8685 - val_hinge_accuracy: 0.9531\n",
      "Epoch 16/20\n",
      "16/16 [==============================] - 240s 16s/step - loss: 0.8542 - hinge_accuracy: 0.9609 - val_loss: 0.8620 - val_hinge_accuracy: 0.9546\n",
      "Epoch 17/20\n",
      "16/16 [==============================] - 244s 16s/step - loss: 0.8475 - hinge_accuracy: 0.9570 - val_loss: 0.8556 - val_hinge_accuracy: 0.9516\n",
      "Epoch 18/20\n",
      "16/16 [==============================] - 244s 16s/step - loss: 0.8408 - hinge_accuracy: 0.9570 - val_loss: 0.8495 - val_hinge_accuracy: 0.9496\n",
      "Epoch 19/20\n",
      "16/16 [==============================] - 242s 16s/step - loss: 0.8346 - hinge_accuracy: 0.9551 - val_loss: 0.8438 - val_hinge_accuracy: 0.9481\n",
      "Epoch 20/20\n",
      "16/16 [==============================] - 245s 16s/step - loss: 0.8287 - hinge_accuracy: 0.9551 - val_loss: 0.8386 - val_hinge_accuracy: 0.9491\n",
      "62/62 [==============================] - 144s 2s/step - loss: 0.8386 - hinge_accuracy: 0.9491\n"
     ]
    }
   ],
   "source": [
    "EPOCHS = 20\n",
    "BATCH_SIZE = 32\n",
    "\n",
    "NUM_EXAMPLES = 512 #x_train_small.shape[0]\n",
    "print(NUM_EXAMPLES)\n",
    "\n",
    "x_train_sub = x_train_small[:NUM_EXAMPLES,:]\n",
    "y_train_hinge_sub = y_train_hinge[:NUM_EXAMPLES]\n",
    "\n",
    "x_test_sub = x_test_small[:,:]\n",
    "y_test_sub = y_test[:]\n",
    "\n",
    "qnn_history_2layer = model2.fit(\n",
    "      x_train_sub, y_train_hinge_sub,\n",
    "      batch_size=BATCH_SIZE,\n",
    "      epochs=EPOCHS,\n",
    "      verbose=1,\n",
    "      validation_data=(x_test_sub, y_test_sub))\n",
    "\n",
    "qnn_results_2layer = model2.evaluate(x_test_sub, y_test_sub)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'loss': [0.9918835658423661,\n",
       "  0.9831518684641459,\n",
       "  0.9732395253532526,\n",
       "  0.9633359900916635,\n",
       "  0.9530492973159846,\n",
       "  0.942243866326671,\n",
       "  0.9315433380557576,\n",
       "  0.9210791519461905,\n",
       "  0.9111940418979739,\n",
       "  0.9016074148175584,\n",
       "  0.8928286390999405,\n",
       "  0.8844707951543982,\n",
       "  0.8764792389835535,\n",
       "  0.868929488145401,\n",
       "  0.861610599607113,\n",
       "  0.8542287840877346,\n",
       "  0.8474743937613779,\n",
       "  0.8407677177049787,\n",
       "  0.8346174946783933,\n",
       "  0.828719607374954],\n",
       " 'hinge_accuracy': [0.490234375,\n",
       "  0.490234375,\n",
       "  0.4921875,\n",
       "  0.49609375,\n",
       "  0.5078125,\n",
       "  0.529296875,\n",
       "  0.578125,\n",
       "  0.673828125,\n",
       "  0.763671875,\n",
       "  0.8515625,\n",
       "  0.892578125,\n",
       "  0.935546875,\n",
       "  0.94921875,\n",
       "  0.955078125,\n",
       "  0.95703125,\n",
       "  0.9609375,\n",
       "  0.95703125,\n",
       "  0.95703125,\n",
       "  0.955078125,\n",
       "  0.955078125],\n",
       " 'val_loss': [0.9877831963749255,\n",
       "  0.9793629158322258,\n",
       "  0.9708110863488494,\n",
       "  0.9616685785706333,\n",
       "  0.9521454472081826,\n",
       "  0.9425446336314404,\n",
       "  0.932881105408207,\n",
       "  0.9234782224076815,\n",
       "  0.9142218276240694,\n",
       "  0.9056589503816221,\n",
       "  0.8975083405467489,\n",
       "  0.8897900934778735,\n",
       "  0.8825064733391124,\n",
       "  0.8753619430703005,\n",
       "  0.8684717618169052,\n",
       "  0.8620340761981523,\n",
       "  0.8556476913225948,\n",
       "  0.8495279118504379,\n",
       "  0.8438316721537322,\n",
       "  0.8385839047357643],\n",
       " 'val_hinge_accuracy': [0.5115927419354839,\n",
       "  0.5120967741935484,\n",
       "  0.5131048387096774,\n",
       "  0.5196572580645161,\n",
       "  0.5307459677419355,\n",
       "  0.5569556451612904,\n",
       "  0.6295362903225806,\n",
       "  0.7172379032258065,\n",
       "  0.8230846774193549,\n",
       "  0.8709677419354839,\n",
       "  0.9133064516129032,\n",
       "  0.9304435483870968,\n",
       "  0.9400201612903226,\n",
       "  0.9475806451612904,\n",
       "  0.953125,\n",
       "  0.9546370967741935,\n",
       "  0.9516129032258065,\n",
       "  0.9495967741935484,\n",
       "  0.9480846774193549,\n",
       "  0.9490927419354839]}"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qnn_history_2layer.history"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot the results!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the hinge accuracy for the qnn history for both 1 and 2 layer on the same axes\n",
    "plt.plot(np.arange(1,21), qnn_history_1layer.history[\"hinge_accuracy\"], label=\"TRAIN - 1 layer QNN\")\n",
    "plt.plot(np.arange(1,21), qnn_history_1layer.history[\"val_hinge_accuracy\"], label=\"VAL - 1 layer QNN\")\n",
    "plt.plot(np.arange(1,21), qnn_history_2layer.history[\"hinge_accuracy\"], label=\"TRAIN - 2 layer QNN\")\n",
    "plt.plot(np.arange(1,21), qnn_history_2layer.history[\"val_hinge_accuracy\"], label=\"VAL - 2 layer QNN\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Hinge Accuracy\")\n",
    "plt.legend()\n",
    "# modify x_ticks to also include 20\n",
    "plt.xticks(np.arange(0, 21, 2.5))\n",
    "# save as pdf with tight layout and no padding\n",
    "plt.savefig(\"qnn_hinge_accuracy.pdf\", bbox_inches=\"tight\", pad_inches=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the hinge accuracy for the loss for both 1 and 2 layer on the same axes\n",
    "plt.plot(np.arange(1,21), qnn_history_1layer.history[\"loss\"], label=\"TRAIN - 1 layer QNN\")\n",
    "plt.plot(np.arange(1,21), qnn_history_1layer.history[\"val_loss\"], label=\"VAL - 1 layer QNN\")\n",
    "plt.plot(np.arange(1,21), qnn_history_2layer.history[\"loss\"], label=\"TRAIN - 2 layer QNN\")\n",
    "plt.plot(np.arange(1,21), qnn_history_2layer.history[\"val_loss\"], label=\"VAL - 2 layer QNN\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.ylabel(\"Hinge Loss\")\n",
    "plt.legend()\n",
    "# modify x_ticks to also include 20\n",
    "plt.xticks(np.arange(0, 21, 2.5))\n",
    "# save as pdf with tight layout and no padding\n",
    "plt.savefig(\"qnn_hinge_loss.pdf\", bbox_inches=\"tight\", pad_inches=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "py39",
   "language": "python",
   "name": "py39"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.13"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "66a612e4ee5d9343e2fda4842cd4535190425ea2d31d59f514fa54a2d33459fb"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}