annealing_hand_written_digits.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Problem setting: Recognizing hand-written digits.\n",
    "\n",
    "Recognizing hand-written digits is a classification problem, in which the aim is to assign each input vector to one of a finite number of discrete categories, to learn observed data points from already labeled data how to predict the class of unlabeled data. In the usecase of hand-written digits dataset, the task is to predict, given an image, which digit it represents.\n",
    "\n",
    "Training model learn features of each number between 0 and 9. The model will be given samples of each of the 10 possible classes on which we fit an estimator to be able to predict the classes to which unseen samples belong.\n",
    "\n",
    "## Problem solution: Supervised learning and predicting\n",
    "\n",
    "In this simple problem, there can be no need to use *state-of-the-art* Machine Learning library and to implement deep neural network. We use a simple algorithm as the estimator for the purpose of convenience. An example of the estimator is the class [sklearn.neural_network.MLPClassifier](http://scikit-learn.org/stable/modules/generated/sklearn.neural_network.MLPClassifier.html#sklearn-neural-network-mlpclassifier) that implements Multi-layer Perceptron classifier. The constructor of an estimator takes as arguments the parameters of the model, but for the time being, we will consider the estimator as a black box."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from sklearn import datasets, metrics\n",
    "from sklearn.neural_network import MLPClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/bin/.pyenv/versions/anaconda3-5.1.0/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (100) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report for classifier MLPClassifier(activation='relu', alpha=1e-05, batch_size=100, beta_1=0.9,\n",
      "       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
      "       hidden_layer_sizes=(1000,), learning_rate='constant',\n",
      "       learning_rate_init=1e-05, max_iter=100, momentum=0.9,\n",
      "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n",
      "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n",
      "       verbose=False, warm_start=False):\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.89      0.94      0.92        88\n",
      "          1       0.88      0.81      0.85        91\n",
      "          2       0.96      0.95      0.96        86\n",
      "          3       0.83      0.87      0.85        91\n",
      "          4       0.98      0.91      0.94        92\n",
      "          5       0.80      0.95      0.87        91\n",
      "          6       0.98      0.98      0.98        91\n",
      "          7       0.89      0.96      0.92        89\n",
      "          8       0.88      0.72      0.79        88\n",
      "          9       0.80      0.78      0.79        92\n",
      "\n",
      "avg / total       0.89      0.89      0.89       899\n",
      "\n",
      "\n",
      "Confusion matrix:\n",
      "[[83  0  1  0  0  3  1  0  0  0]\n",
      " [ 0 74  1  8  0  0  0  0  3  5]\n",
      " [ 1  0 82  0  0  0  0  0  0  3]\n",
      " [ 0  3  0 79  0  0  0  6  2  1]\n",
      " [ 4  0  0  0 84  1  0  0  2  1]\n",
      " [ 0  1  0  0  0 86  1  0  0  3]\n",
      " [ 0  2  0  0  0  0 89  0  0  0]\n",
      " [ 0  0  0  0  2  2  0 85  0  0]\n",
      " [ 0  4  1  6  0  7  0  2 63  5]\n",
      " [ 5  0  0  2  0  8  0  3  2 72]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea20828f6a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.neural_network import MLPClassifier\n",
    "\n",
    "# The digits dataset\n",
    "digits = datasets.load_digits()\n",
    "\n",
    "# The data that we are interested in is made of 8x8 images of digits, let's\n",
    "# have a look at the first 4 images, stored in the `images` attribute of the\n",
    "# dataset.  If we were working from image files, we could load them using\n",
    "# matplotlib.pyplot.imread.  Note that each image must have the same size. For these\n",
    "# images, we know which digit they represent: it is given in the 'target' of\n",
    "# the dataset.\n",
    "images_and_labels = list(zip(digits.images, digits.target))\n",
    "for index, (image, label) in enumerate(images_and_labels[:4]):\n",
    "    plt.subplot(2, 4, index + 1)\n",
    "    plt.axis('off')\n",
    "    plt.imshow(image, cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "    plt.title('Training: %i' % label)\n",
    "\n",
    "# To apply a classifier on this data, we need to flatten the image, to\n",
    "# turn the data in a (samples, feature) matrix:\n",
    "n_samples = len(digits.images)\n",
    "data = digits.images.reshape((n_samples, -1))\n",
    "\n",
    "classifier = MLPClassifier(\n",
    "    # Activation function in the hidden layer.\n",
    "    activation=\"relu\",\n",
    "    # L2 penalty (regularization term) parameter.\n",
    "    alpha=0.00001,\n",
    "    # Bath size.\n",
    "    batch_size=100,\n",
    "    # The ith element represents the number of neurons in the ith hidden layer.\n",
    "    hidden_layer_sizes=(1000,), \n",
    "    # The initial learning rate used.\n",
    "    learning_rate_init=0.00001,\n",
    "    # Maximum number of iterations.\n",
    "    max_iter=100,\n",
    "    # Solver.\n",
    "    solver=\"adam\",\n",
    "    shuffle=True,\n",
    "    verbose=False\n",
    ")\n",
    "\n",
    "# We learn the digits on the first half of the digits\n",
    "classifier.fit(data[:n_samples // 2], digits.target[:n_samples // 2])\n",
    "\n",
    "# Now predict the value of the digit on the second half:\n",
    "expected = digits.target[n_samples // 2:]\n",
    "predicted = classifier.predict(data[n_samples // 2:])\n",
    "\n",
    "print(\"Classification report for classifier %s:\\n%s\\n\"\n",
    "      % (classifier, metrics.classification_report(expected, predicted)))\n",
    "print(\"Confusion matrix:\\n%s\" % metrics.confusion_matrix(expected, predicted))\n",
    "\n",
    "images_and_predictions = list(zip(digits.images[n_samples // 2:], predicted))\n",
    "for index, (image, prediction) in enumerate(images_and_predictions[:4]):\n",
    "    plt.subplot(2, 4, index + 5)\n",
    "    plt.axis('off')\n",
    "    plt.imshow(image, cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "    plt.title('Prediction: %i' % prediction)\n",
    "\n",
    "plt.show()\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Issue: Complexity of hyperparameters, or how can be hyperparameters decided? \n",
    "\n",
    "There are many hyperparameters that we have to set before the actual training process begins. Each parameter should be decided in relation to machine learning theory and it cause side effects in training model. Because of this complexity of hyperparameters, so-called the *hyperparameter tuning* must become a burden of Data scientists and R & D engineers from the perspective of not only a theoretical point of view but also implementation level. \n",
    "\n",
    "Beginners or end users who are inclined to use *state-of-the-art* Machine Learning Library as a mere blackbox and spend all one's time doing blackbox test may regard this problem setting as a kind of paradox or a bitter irony. Even if hyperparameters were chosen at random, the training for indefinitely long times makes it possible to predict with high precision to serve end user's aim. \n",
    "\n",
    "In a sense, such endless *hyperparameter tuning* implicates a behavior that supports **the infinite monkey theorem**. This theorem states that a monkey hitting keys at random on a typewriter keyboard for an infinite amount of time will almost surely type a given natural sentence, such as the complete works of William Shakespeare. Although we can believe the illimitable possibilities of human beings, but time is limited.\n",
    "\n",
    "# Problem re-setting: Combinatorial optimization problem.\n",
    "\n",
    "This issue can be considered as Combinatorial optimization problem which is an optimization problem, where an optimal solution has to be identified from a finite set of solutions. The solutions are normally discrete or can be converted into discrete. This is an important topic studied in operations research such as software engineering, artificial intelligence(AI), and machine learning. For instance, travelling sales man problem is one of the popular combinatorial optimization problem.\n",
    "\n",
    "Considering that the *hyperparameter tuning* can be formed into Combinatorial optimization problem, the optimal solution to be searched is estimation accuracy, precision, recall, or F-value and so on. Conversely, estimation error such as Mean Squared Error(MSE), Root Mean Squared Error(RMSE), or Mean Absolute Error(MAE) can be a value of penalty, which must be reduced by searching and learning.\n",
    "\n",
    "In this problem setting, combination of the hyperparameters are explanatory variables for the optimal solution. Each parameter can be generated from random sampling like Monte Carlo method."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Problem solution: Simulated Annealing.\n",
    "\n",
    "Simulated Annealing is a probabilistic single solution based search method inspired by the annealing process in metallurgy. Annealing is a physical process referred to as tempering certain alloys of metal, glass, or crystal by heating above its melting point, holding its temperature, and then cooling it very slowly until it solidifies into a perfect crystalline structure. The simulation of this process is known as simulated annealing. \n",
    "\n",
    "### Conceptual perspective.\n",
    "\n",
    "The simulated annealing is a generic algorithm. Combinatorial optimization is an one of the various functions of simulated annealing. \n",
    "\n",
    "#### Boltzmann distribution.\n",
    "\n",
    "According to statistical mechanics and thermodynamics, $P_{\\alpha}$, the probability of a physical system\n",
    "being in state $\\alpha$ with energy $E_{\\alpha}$ at absolute temperature $T$ satisfies the Boltzmann distribution as follow.\n",
    "\n",
    "$$P_{\\alpha} = \\frac{1}{Z}e^{\\frac{E_{\\alpha}}{k_{B}}}T$$\n",
    "\n",
    "where $k_{B}$ is the Boltzmann’s constant and $Z$ is the partition function, defined by\n",
    "\n",
    "$$Z = \\sum_{\\beta}^{}e^{\\frac{E_{\\beta}}{k_{B}}}T$$\n",
    "\n",
    "#### Meaning of partition function.\n",
    "\n",
    "To simplify, this partition function can be considered as the summation being taken over all states $\\alpha = \\{0, 1, ..., \\beta \\}$ with energy $E_{\\alpha}$ at temperature $T$ . At high $T$ , the Boltzmann distribution exhibits uniform preference for all the states, regardless of the energy. The system ignores small changes in\n",
    "the energy and approaches thermal equilibrium rapidly. On the other hand, when $T$ approaches zero or very low, only the states with minimum energy have non-zero probability of occurrence. The system can respond to small changes in the energy.\n",
    "\n",
    "#### To find a better minimum\n",
    "\n",
    "According to statistical mechanics, gradually decreasing temperature corresponds to gradually decreasing information entropy. So this system can perform a fine search in the neighborhood of the already determined minimum and finds a better minimum.\n",
    "\n",
    "#### Cost function\n",
    "\n",
    "The function of Combinatorial optimization algorithm is minimization of the penalty that cost function returns. Typically, the cost function is defined by\n",
    "\n",
    "$$f(p_k, y_k) = Loss_{MSE} =\\frac{1}{n}\\sum_{k=1}^n(p_k - y_k)^2$$\n",
    "\n",
    "where $p_k$ is a set of predicted labels and $y_k$ is a set of correct labels."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Specification perspective.\n",
    "\n",
    "There are various Simulated Annealing such as Boltzmann Annealing, Adaptive Simulated Annealing(SAS), and Quantum Simulated Annealing. On the premise of Combinatorial optimization problem, these annealing methods can be considered as functionally equivalent. The *Commonality/Variability* in these methods are able to keep responsibility of objects all straight as the class diagram below indicates.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/class_diagram_annealing_model.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "An abstract class `AnnealingModel` is responsible for common annealing methods. And a concrete class `SimulatedAnnealing` is-a `AnnealingModel` to behave as Simulated Annealing. `CostFunctionable` is an interface which performs as cost function. Implementation of this interface depends on user's problem settings.\n",
    "\n",
    "`AdaptiveSimulatedAnnealing` is a functionally extended version of `SimulatedAnnealing`. `QuantumMonteCarlo` is a functional equivalent of `SimulatedAnnealing`. The interface `DistanceComputable` can be distance function for `QuantumMonteCarlo`. For instance, `CostAsDistance` is-a `DistanceComputable` to obtain cost as distance by `CostFunctionable`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Implementation perspective.\n",
    "\n",
    "In order to implement `CostFunctionable` and use `AnnealingModel` in this problem setting, let us adhere fundamentally to an introduction with scikit-learn.\n",
    "\n",
    "- [An introduction to machine learning with scikit-learn — scikit-learn 0.19.1 documentation](http://scikit-learn.org/stable/tutorial/basic/tutorial.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dataset.\n",
    "\n",
    "The data that we are interested in is made of 8x8 images of digits, let's have a look at the first 4 images, stored in the `images` attribute of the dataset.  If we were working from image files, we could load them using matplotlib.pyplot.imread.  Note that each image must have the same size. For these images, we know which digit they represent: it is given in the 'target' of the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea2082da908>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# The digits dataset\n",
    "digits = datasets.load_digits()\n",
    "\n",
    "images_and_labels = list(zip(digits.images, digits.target))\n",
    "for index, (image, label) in enumerate(images_and_labels[:4]):\n",
    "    plt.subplot(2, 4, index + 1)\n",
    "    plt.axis('off')\n",
    "    plt.imshow(image, cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "    plt.title('Training: %i' % label)\n",
    "\n",
    "# To apply a classifier on this data, we need to flatten the image, to\n",
    "# turn the data in a (samples, feature) matrix:\n",
    "n_samples = len(digits.images)\n",
    "data = digits.images.reshape((n_samples, -1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Implement cost function.\n",
    "\n",
    "Let us implement `CostFunctionable` as follow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyqlearning.annealingmodel.simulated_annealing import SimulatedAnnealing\n",
    "from pyqlearning.annealingmodel.cost_functionable import CostFunctionable\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "class DigitClfCost(CostFunctionable):\n",
    "    '''\n",
    "    is-a `CostFunctionable` in relation to `SimulatedAnnealing`.\n",
    "    '''\n",
    "\n",
    "    # The list of activation function in the hidden layer.\n",
    "    __activation_list = [\"relu\", \"logistic\", \"tanh\"]\n",
    "    # The solver for weight optimization.\n",
    "    __solver_list = [\"lbfgs\", \"sgd\", \"adam\"]\n",
    "    \n",
    "    def compute(self, x):\n",
    "        '''\n",
    "        Compute cost.\n",
    "        \n",
    "        Args:\n",
    "            x:    `np.ndarray` of explanatory variables.\n",
    "        \n",
    "        Returns:\n",
    "            cost\n",
    "        '''\n",
    "        classifier = MLPClassifier(\n",
    "            # Activation function in the hidden layer.\n",
    "            activation=self.__activation_list[int(x[0])],\n",
    "            # L2 penalty (regularization term) parameter.\n",
    "            alpha=x[1],\n",
    "            # Bath size.\n",
    "            batch_size=int(x[2]),\n",
    "            # The ith element represents the number of neurons in the ith hidden layer.\n",
    "            hidden_layer_sizes=(int(x[3]),), \n",
    "            # The initial learning rate used.\n",
    "            learning_rate_init=x[4],\n",
    "            # Maximum number of iterations.\n",
    "            max_iter=int(x[5]),\n",
    "            # Solver.\n",
    "            solver=self.__solver_list[int(x[6])],\n",
    "            shuffle=True,\n",
    "            verbose=False\n",
    "        )\n",
    "\n",
    "        # We learn the digits on the first half of the digits\n",
    "        classifier.fit(data[:n_samples // 2], digits.target[:n_samples // 2])\n",
    "\n",
    "        # Now predict the value of the digit on the second half:\n",
    "        expected = digits.target[n_samples // 2:]\n",
    "        predicted = classifier.predict(data[n_samples // 2:])\n",
    "        \n",
    "        cost = mean_squared_error(expected, predicted)\n",
    "        \n",
    "        return cost\n",
    "\n",
    "    def get_activation_list(self):\n",
    "        return self.__activation_list\n",
    "\n",
    "    def get_solver_list(self):\n",
    "        return self.__solver_list\n",
    "\n",
    "    def set_readonly(self, value):\n",
    "        raise TypeError()\n",
    "\n",
    "    activation_list = property(get_activation_list, set_readonly)\n",
    "    solver_list = property(get_solver_list, set_readonly)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Sampling hyperparameters.\n",
    "\n",
    "Like Monte Carlo method, let us draw random samples from a normal (Gaussian) or unifrom distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "cost_functionable = DigitClfCost()\n",
    "\n",
    "activation_arr = np.random.randint(low=0, high=len(cost_functionable.activation_list), size=1000)\n",
    "alpha_arr = np.random.normal(loc=0.0001, scale=0.00001, size=1000)\n",
    "batch_size_arr = np.random.normal(loc=200, scale=10, size=1000).astype(int)\n",
    "hidden_arr = np.random.normal(loc=1000, scale=100, size=1000).astype(int)\n",
    "learning_rate_arr = np.random.normal(loc=0.01, scale=0.0005, size=1000)\n",
    "max_iter_arr = np.random.normal(loc=1000, scale=200, size=1000)\n",
    "solver_arr = np.random.randint(low=0, high=len(cost_functionable.solver_list), size=1000)\n",
    "\n",
    "params_arr = np.c_[\n",
    "    activation_arr,\n",
    "    alpha_arr,\n",
    "    batch_size_arr,\n",
    "    hidden_arr,\n",
    "    learning_rate_arr,\n",
    "    max_iter_arr,\n",
    "    solver_arr\n",
    "]\n",
    "\n",
    "params_arr = pd.DataFrame(params_arr, columns=[\n",
    "    \"activation\",\n",
    "    \"alpha\",\n",
    "    \"batch_size\",\n",
    "    \"hidden\",\n",
    "    \"learning_rate\",\n",
    "    \"max_iter\",\n",
    "    \"solver\"\n",
    "]).sort_values(by=[\n",
    "    \"learning_rate\",\n",
    "    \"batch_size\",\n",
    "    \"max_iter\",\n",
    "    \"hidden\",\n",
    "    \"alpha\",\n",
    "    \"activation\",\n",
    "    \"solver\"\n",
    "]).values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Use `AnnealingModel`.\n",
    "\n",
    "Instantiate `DigitClfCost` and delegate it to `SimulatedAnnealing.__init__()`. Then call the method `fit_dist_mat` to fit the combination of hyperparameters. Finally, call the method `annealing`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/media/removable/rum/accel-brain-code/Reinforcement-Learning/pyqlearning/annealing_model.py:30: FutureWarning: This property will be removed in future version. Use `var_arr`.\n",
      "  warnings.warn(\"This property will be removed in future version. Use `var_arr`.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "annealing_model = SimulatedAnnealing(\n",
    "    cost_functionable=cost_functionable,\n",
    "    cycles_num=33,\n",
    "    trials_per_cycle=3,\n",
    "    accepted_sol_num=0.0,\n",
    "    init_prob=0.7,\n",
    "    final_prob=0.001,\n",
    "    start_pos=0,\n",
    "    move_range=3\n",
    ")\n",
    "annealing_model.fit_dist_mat(params_arr)\n",
    "annealing_model.annealing()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get and plot result data.\n",
    "\n",
    "As shown by the figure below, the energy has been decreasing gradually."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/media/removable/rum/accel-brain-code/Reinforcement-Learning/pyqlearning/annealing_model.py:112: FutureWarning: This property will be removed in future version. Use `predicted_log_arr`.\n",
      "  warnings.warn(\"This property will be removed in future version. Use `predicted_log_arr`.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# Extract list: [(Cost, Delta energy, Mean of delta energy, probability, accept)]\n",
    "predicted_log_list = annealing_model.predicted_log_list\n",
    "predicted_log_arr = np.array(predicted_log_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get and plot result data.\n",
    "\n",
    "As shown by the figure below, the energy has been decreasing gradually."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea200962128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot delta_e_avg\n",
    "fig = plt.figure(figsize=(15, 12))\n",
    "plt.plot(predicted_log_arr[:, 2])\n",
    "plt.show()\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dicide optimal solution\n",
    "\n",
    "The optimal combination of hyperparameters can be confirmed as follow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Activation: logistic\n",
      "Alpha: 9.07558440806e-05\n",
      "Batch size: 193\n",
      "The number of units in hidden layer: 968\n",
      "Init of Learning rate: 0.00905655735841\n",
      "Max iter: 862\n",
      "Solver: lbfgs\n"
     ]
    }
   ],
   "source": [
    "min_e_v_arr = annealing_model.var_arr[np.argmin(predicted_log_arr[:, 0])]\n",
    "\n",
    "activation = cost_functionable.activation_list[int(min_e_v_arr[0])]\n",
    "alpha = min_e_v_arr[1]\n",
    "batch_size = int(min_e_v_arr[2])\n",
    "hidden_n = int(min_e_v_arr[3])\n",
    "learning_rate_init = min_e_v_arr[4]\n",
    "max_iter = int(min_e_v_arr[5])\n",
    "solver = cost_functionable.solver_list[int(min_e_v_arr[6])]\n",
    "\n",
    "print(\"Activation: \" + str(activation))\n",
    "print(\"Alpha: \" + str(alpha))\n",
    "print(\"Batch size: \" + str(batch_size))\n",
    "print(\"The number of units in hidden layer: \" + str(hidden_n))\n",
    "print(\"Init of Learning rate: \" + str(learning_rate_init))\n",
    "print(\"Max iter: \" + str(max_iter))\n",
    "print(\"Solver: \" + str(solver))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Training by optimal hyperparameters.\n",
    "\n",
    "Then this optimal solution makes it possible for machine learning model to learn by optimal hyperparameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report for classifier MLPClassifier(activation='logistic', alpha=9.0755844080612054e-05,\n",
      "       batch_size=193, beta_1=0.9, beta_2=0.999, early_stopping=False,\n",
      "       epsilon=1e-08, hidden_layer_sizes=(968,), learning_rate='constant',\n",
      "       learning_rate_init=0.009056557358407635, max_iter=862, momentum=0.9,\n",
      "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n",
      "       shuffle=True, solver='lbfgs', tol=0.0001, validation_fraction=0.1,\n",
      "       verbose=False, warm_start=False):\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       1.00      0.95      0.98        88\n",
      "          1       0.94      0.92      0.93        91\n",
      "          2       0.92      1.00      0.96        86\n",
      "          3       0.96      0.87      0.91        91\n",
      "          4       0.98      0.88      0.93        92\n",
      "          5       0.92      0.96      0.94        91\n",
      "          6       0.90      0.98      0.94        91\n",
      "          7       0.93      0.93      0.93        89\n",
      "          8       0.94      0.88      0.91        88\n",
      "          9       0.85      0.96      0.90        92\n",
      "\n",
      "avg / total       0.93      0.93      0.93       899\n",
      "\n",
      "\n",
      "Confusion matrix:\n",
      "[[84  0  0  0  1  1  2  0  0  0]\n",
      " [ 0 84  2  0  0  0  0  0  0  5]\n",
      " [ 0  0 86  0  0  0  0  0  0  0]\n",
      " [ 0  0  2 79  0  2  0  5  3  0]\n",
      " [ 0  1  0  0 81  0  7  0  0  3]\n",
      " [ 0  0  0  0  0 87  1  0  1  2]\n",
      " [ 0  2  0  0  0  0 89  0  0  0]\n",
      " [ 0  0  2  0  1  0  0 83  1  2]\n",
      " [ 0  2  1  2  0  3  0  0 77  3]\n",
      " [ 0  0  0  1  0  2  0  1  0 88]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea205d0f128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "classifier = MLPClassifier(\n",
    "    # Activation function in the hidden layer.\n",
    "    activation=activation,\n",
    "    # L2 penalty (regularization term) parameter.\n",
    "    alpha=alpha,\n",
    "    # Bath size.\n",
    "    batch_size=batch_size,\n",
    "    # The ith element represents the number of neurons in the ith hidden layer.\n",
    "    hidden_layer_sizes=(hidden_n,), \n",
    "    # The initial learning rate used.\n",
    "    learning_rate_init=learning_rate_init,\n",
    "    # Maximum number of iterations.\n",
    "    max_iter=max_iter,\n",
    "    # Solver.\n",
    "    solver=solver,\n",
    "    shuffle=True,\n",
    "    verbose=False\n",
    ")\n",
    "\n",
    "# We learn the digits on the first half of the digits\n",
    "classifier.fit(data[:n_samples // 2], digits.target[:n_samples // 2])\n",
    "\n",
    "# Now predict the value of the digit on the second half:\n",
    "expected = digits.target[n_samples // 2:]\n",
    "predicted = classifier.predict(data[n_samples // 2:])\n",
    "\n",
    "print(\"Classification report for classifier %s:\\n%s\\n\"\n",
    "      % (classifier, metrics.classification_report(expected, predicted)))\n",
    "print(\"Confusion matrix:\\n%s\" % metrics.confusion_matrix(expected, predicted))\n",
    "\n",
    "images_and_predictions = list(zip(digits.images[n_samples // 2:], predicted))\n",
    "for index, (image, prediction) in enumerate(images_and_predictions[:4]):\n",
    "    plt.subplot(2, 4, index + 5)\n",
    "    plt.axis('off')\n",
    "    plt.imshow(image, cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "    plt.title('Prediction: %i' % prediction)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Functional equivalent: Adaptive Simulated Annealing.\n",
    "\n",
    "Adaptive simulated annealing, also known as the very fast simulated reannealing, is a very efficient version of simulated annealing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/media/removable/rum/accel-brain-code/Reinforcement-Learning/pyqlearning/annealing_model.py:30: FutureWarning: This property will be removed in future version. Use `var_arr`.\n",
      "  warnings.warn(\"This property will be removed in future version. Use `var_arr`.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "from pyqlearning.annealingmodel.simulatedannealing.adaptive_simulated_annealing import AdaptiveSimulatedAnnealing\n",
    "\n",
    "cost_functionable = DigitClfCost()\n",
    "annealing_model = AdaptiveSimulatedAnnealing(\n",
    "    cost_functionable=cost_functionable,\n",
    "    cycles_num=33,\n",
    "    trials_per_cycle=3,\n",
    "    accepted_sol_num=0.0,\n",
    "    init_prob=0.7,\n",
    "    final_prob=0.001,\n",
    "    start_pos=0,\n",
    "    move_range=3\n",
    ")\n",
    "annealing_model.adaptive_set(\n",
    "    reannealing_per=50,\n",
    "    thermostat=0.,\n",
    "    t_min=0.001,\n",
    "    t_default=1.0\n",
    ")\n",
    "annealing_model.fit_dist_mat(params_arr)\n",
    "annealing_model.annealing()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/media/removable/rum/accel-brain-code/Reinforcement-Learning/pyqlearning/annealing_model.py:112: FutureWarning: This property will be removed in future version. Use `predicted_log_arr`.\n",
      "  warnings.warn(\"This property will be removed in future version. Use `predicted_log_arr`.\", FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "# Extract list: [(Cost, Delta energy, Mean of delta energy, probability, accept)]\n",
    "predicted_log_list = annealing_model.predicted_log_list\n",
    "predicted_log_arr = np.array(predicted_log_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea200eb4748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot delta_e_avg\n",
    "fig = plt.figure(figsize=(15, 12))\n",
    "plt.plot(predicted_log_arr[:, 2])\n",
    "plt.show()\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dicide optimal solution\n",
    "\n",
    "The optimal combination of hyperparameters can be confirmed as follow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Activation: relu\n",
      "Alpha: 0.000105971020366\n",
      "Batch size: 211\n",
      "The number of units in hidden layer: 933\n",
      "Init of Learning rate: 0.00870026005571\n",
      "Max iter: 1316\n",
      "Solver: lbfgs\n"
     ]
    }
   ],
   "source": [
    "min_e_v_arr = annealing_model.var_arr[np.argmin(predicted_log_arr[:, 0])]\n",
    "\n",
    "activation = cost_functionable.activation_list[int(min_e_v_arr[0])]\n",
    "alpha = min_e_v_arr[1]\n",
    "batch_size = int(min_e_v_arr[2])\n",
    "hidden_n = int(min_e_v_arr[3])\n",
    "learning_rate_init = min_e_v_arr[4]\n",
    "max_iter = int(min_e_v_arr[5])\n",
    "solver = cost_functionable.solver_list[int(min_e_v_arr[6])]\n",
    "\n",
    "print(\"Activation: \" + str(activation))\n",
    "print(\"Alpha: \" + str(alpha))\n",
    "print(\"Batch size: \" + str(batch_size))\n",
    "print(\"The number of units in hidden layer: \" + str(hidden_n))\n",
    "print(\"Init of Learning rate: \" + str(learning_rate_init))\n",
    "print(\"Max iter: \" + str(max_iter))\n",
    "print(\"Solver: \" + str(solver))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Training by optimal hyperparameters.\n",
    "\n",
    "Then this optimal solution makes it possible for machine learning model to learn by optimal hyperparameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report for classifier MLPClassifier(activation='relu', alpha=0.00010597102036595624, batch_size=211,\n",
      "       beta_1=0.9, beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
      "       hidden_layer_sizes=(933,), learning_rate='constant',\n",
      "       learning_rate_init=0.008700260055708401, max_iter=1316,\n",
      "       momentum=0.9, nesterovs_momentum=True, power_t=0.5,\n",
      "       random_state=None, shuffle=True, solver='lbfgs', tol=0.0001,\n",
      "       validation_fraction=0.1, verbose=False, warm_start=False):\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       1.00      0.97      0.98        88\n",
      "          1       0.93      0.89      0.91        91\n",
      "          2       0.99      1.00      0.99        86\n",
      "          3       0.98      0.87      0.92        91\n",
      "          4       0.95      0.88      0.92        92\n",
      "          5       0.88      0.93      0.90        91\n",
      "          6       0.89      0.98      0.93        91\n",
      "          7       0.98      0.94      0.96        89\n",
      "          8       0.88      0.90      0.89        88\n",
      "          9       0.85      0.93      0.89        92\n",
      "\n",
      "avg / total       0.93      0.93      0.93       899\n",
      "\n",
      "\n",
      "Confusion matrix:\n",
      "[[85  0  0  0  1  0  2  0  0  0]\n",
      " [ 0 81  0  1  1  4  1  0  1  2]\n",
      " [ 0  0 86  0  0  0  0  0  0  0]\n",
      " [ 0  0  0 79  0  3  0  0  8  1]\n",
      " [ 0  1  0  0 81  0  7  0  0  3]\n",
      " [ 0  0  1  0  0 85  1  0  0  4]\n",
      " [ 0  2  0  0  0  0 89  0  0  0]\n",
      " [ 0  0  0  0  1  0  0 84  1  3]\n",
      " [ 0  3  0  0  1  3  0  0 79  2]\n",
      " [ 0  0  0  1  0  2  0  2  1 86]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea208134ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "classifier = MLPClassifier(\n",
    "    # Activation function in the hidden layer.\n",
    "    activation=activation,\n",
    "    # L2 penalty (regularization term) parameter.\n",
    "    alpha=alpha,\n",
    "    # Bath size.\n",
    "    batch_size=batch_size,\n",
    "    # The ith element represents the number of neurons in the ith hidden layer.\n",
    "    hidden_layer_sizes=(hidden_n,), \n",
    "    # The initial learning rate used.\n",
    "    learning_rate_init=learning_rate_init,\n",
    "    # Maximum number of iterations.\n",
    "    max_iter=max_iter,\n",
    "    # Solver.\n",
    "    solver=solver,\n",
    "    shuffle=True,\n",
    "    verbose=False\n",
    ")\n",
    "\n",
    "# We learn the digits on the first half of the digits\n",
    "classifier.fit(data[:n_samples // 2], digits.target[:n_samples // 2])\n",
    "\n",
    "# Now predict the value of the digit on the second half:\n",
    "expected = digits.target[n_samples // 2:]\n",
    "predicted = classifier.predict(data[n_samples // 2:])\n",
    "\n",
    "print(\"Classification report for classifier %s:\\n%s\\n\"\n",
    "      % (classifier, metrics.classification_report(expected, predicted)))\n",
    "print(\"Confusion matrix:\\n%s\" % metrics.confusion_matrix(expected, predicted))\n",
    "\n",
    "images_and_predictions = list(zip(digits.images[n_samples // 2:], predicted))\n",
    "for index, (image, prediction) in enumerate(images_and_predictions[:4]):\n",
    "    plt.subplot(2, 4, index + 5)\n",
    "    plt.axis('off')\n",
    "    plt.imshow(image, cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "    plt.title('Prediction: %i' % prediction)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Functional equivalent: Quantum Monte Carlo Method\n",
    "\n",
    "Generally, Quantum Monte Carlo is a stochastic method to solve the Schrödinger equation. This algorithm is one of the earliest types of solution in order to simulate the Quantum Annealing in classical computer. In summary, one of the function of this algorithm is to solve the ground state search problem which is known as logically equivalent to combinatorial optimization problem.\n",
    "\n",
    "According to theory of spin glasses, the ground state search problem can be described as minimization energy determined by the hamiltonian <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/h_0.png\" /> as follow\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/hamiltonian_in_ising_model.png\" />\n",
    "\n",
    "where <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/pauli_z_i.png\" /> refers to the Pauli spin matrix below for the spin-half particle at lattice point <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/i.gif\" />. In spin glasses, random value is assigned to <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/j_i_j.png\" />. The number of combinations is enormous. If this value is <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/n.png\" />, a trial frequency is <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/2_n.png\" />. This computation complexity makes it impossible to solve the ground state search problem. Then, in theory of spin glasses, the standard hamiltonian is re-described in expanded form.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/hamiltonian_in_t_ising_model.png\" />\n",
    "\n",
    "where <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/pauli_x_i.png\" /> also refers to the Pauli spin matrix and <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/gamma.png\" /> is so-called annealing coefficient, which is hyperparameter that contains vely high value. Ising model to follow this Hamiltonian is known as the Transverse Ising model.\n",
    "\n",
    "In relation to this system, thermal equilibrium amount of a physical quantity <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/q.png?1\" /> is as follow.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/langle_q_rangle.png\" />\n",
    "\n",
    "If <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/h.png\" /> is a diagonal matrix, then also <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/e_beta_h.png\" /> is diagonal matrix. If diagonal element in <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/h.png\" /> is <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/e_i.png\" />, Each diagonal element is <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/e_beta_h_ij_e_i.png\" />. However if <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/h.png\" /> has off-diagonal elements, It is known that <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/e_beta_h_ij_e_i_neq.png\" /> since for any of the exponent <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/i.gif\" /> we must exponentiate the matrix as follow.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/e_matrix_infty.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Therefore, a path integration based on Trotter-Suzuki decomposition has been introduced in Quantum Monte Carlo Method. This path integration makes it possible to obtain the partition function <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/z.png\" />.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/z_in_t_ising_model.png\" />\n",
    "\n",
    "where if <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/m.png\" /> is large enough, relational expression below is established.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/exp_left_frac_1_m_beta_h_right.png\" /></td></tr>\n",
    "\n",
    "Then the partition function <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/z.png\" /> can be re-descibed as follow.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/z_in_t_ising_model_re_described.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "where <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/mid_sigma_k_rangle.png\" /> is <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/l.png\" /> topological products (product spaces). Because <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/h_0.png\" /> is the diagonal matrix, <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/tilde_sigma_j_z_mid_sigma.png\" />.\n",
    "\n",
    "Therefore, \n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/langle_sigma_k_mid.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The partition function <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/z.png\" /> can be re-descibed as follow.\n",
    "\n",
    "<img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/z_in_t_ising_model_re_described_last.png\" />"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "where <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/m.png\" /> is the number of trotter.\n",
    "\n",
    "This relational expression indicates that the quantum - mechanical Hamiltonian in <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/d.png\" /> dimentional Tranverse Ising model is functional equivalence to classical Hamiltonian in <img src=\"https://storage.googleapis.com/accel-brain-code/Reinforcement-Learning/img/latex/d_1.png\" /> dimentional Ising model, which means that the state of the quantum - mechanical system can be approximate by the state of classical system.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyqlearning.annealingmodel.quantum_monte_carlo import QuantumMonteCarlo\n",
    "from pyqlearning.annealingmodel.distancecomputable.cost_as_distance import CostAsDistance\n",
    "\n",
    "cost_funtionable = DigitClfCost()\n",
    "distance_computable = CostAsDistance(params_arr, cost_funtionable)\n",
    "\n",
    "annealing_model = QuantumMonteCarlo(\n",
    "    distance_computable=distance_computable,\n",
    "    cycles_num=100,\n",
    "    inverse_temperature_beta=0.1,\n",
    "    gammma=1.0,\n",
    "    fractional_reduction=0.99,\n",
    "    trotter_dimention=10,\n",
    "    mc_step=100,\n",
    "    point_num=100,\n",
    "    spin_arr=None,\n",
    "    tolerance_diff_e=0.01\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "annealing_model.annealing()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Get and plot result data.\n",
    "\n",
    "As shown by the figure below, the energy has been decreasing gradually in each trotter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea208134cc0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "fig = plt.figure(figsize=(20, 10))\n",
    "for i in range(10):\n",
    "    plt.plot(\n",
    "        annealing_model.predicted_log_arr[annealing_model.predicted_log_arr[:, 0] == i][:, 5],\n",
    "        label=\"Trotter: \" + str(i+1)\n",
    "    )\n",
    "plt.legend()\n",
    "plt.show()\n",
    "plt.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Dicide optimal solution.\n",
    "The optimal combination of hyperparameters can be confirmed as follow."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Activation: relu\n",
      "Alpha: 9.52089872787e-05\n",
      "Batch size: 203\n",
      "The number of units in hidden layer: 1027\n",
      "Init of Learning rate: 0.00933657002229\n",
      "Max iter: 800\n",
      "Solver: sgd\n"
     ]
    }
   ],
   "source": [
    "opt_arr = params_arr[int(annealing_model.predicted_log_arr[\n",
    "    annealing_model.predicted_log_arr[:, 5] == annealing_model.predicted_log_arr[:, 5].min()\n",
    "][0][4])]\n",
    "\n",
    "activation = cost_functionable.activation_list[int(opt_arr[0])]\n",
    "alpha = opt_arr[1]\n",
    "batch_size = int(opt_arr[2])\n",
    "hidden_n = int(opt_arr[3])\n",
    "learning_rate_init = opt_arr[4]\n",
    "max_iter = int(opt_arr[5])\n",
    "solver = cost_functionable.solver_list[int(opt_arr[6])]\n",
    "\n",
    "print(\"Activation: \" + str(activation))\n",
    "print(\"Alpha: \" + str(alpha))\n",
    "print(\"Batch size: \" + str(batch_size))\n",
    "print(\"The number of units in hidden layer: \" + str(hidden_n))\n",
    "print(\"Init of Learning rate: \" + str(learning_rate_init))\n",
    "print(\"Max iter: \" + str(max_iter))\n",
    "print(\"Solver: \" + str(solver))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Training by optimal hyperparameters.\n",
    "Then this optimal solution makes it possible for machine learning model to learn by optimal hyperparameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification report for classifier MLPClassifier(activation='relu', alpha=9.520898727871237e-05, batch_size=203,\n",
      "       beta_1=0.9, beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
      "       hidden_layer_sizes=(1027,), learning_rate='constant',\n",
      "       learning_rate_init=0.0093365700222927152, max_iter=800,\n",
      "       momentum=0.9, nesterovs_momentum=True, power_t=0.5,\n",
      "       random_state=None, shuffle=True, solver='sgd', tol=0.0001,\n",
      "       validation_fraction=0.1, verbose=False, warm_start=False):\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.98      0.98      0.98        88\n",
      "          1       0.94      0.91      0.93        91\n",
      "          2       1.00      0.98      0.99        86\n",
      "          3       0.94      0.87      0.90        91\n",
      "          4       0.98      0.91      0.94        92\n",
      "          5       0.91      0.95      0.92        91\n",
      "          6       0.95      0.99      0.97        91\n",
      "          7       0.94      0.99      0.96        89\n",
      "          8       0.94      0.91      0.92        88\n",
      "          9       0.86      0.93      0.90        92\n",
      "\n",
      "avg / total       0.94      0.94      0.94       899\n",
      "\n",
      "\n",
      "Confusion matrix:\n",
      "[[86  0  0  0  1  0  1  0  0  0]\n",
      " [ 0 83  0  1  0  1  0  0  0  6]\n",
      " [ 0  0 84  2  0  0  0  0  0  0]\n",
      " [ 0  0  0 79  0  3  0  4  5  0]\n",
      " [ 1  1  0  0 84  0  3  0  0  3]\n",
      " [ 0  0  0  0  0 86  1  0  0  4]\n",
      " [ 0  1  0  0  0  0 90  0  0  0]\n",
      " [ 0  0  0  0  0  0  0 88  0  1]\n",
      " [ 0  3  0  1  1  3  0  0 80  0]\n",
      " [ 1  0  0  1  0  2  0  2  0 86]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ea2009dfbe0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "classifier = MLPClassifier(\n",
    "    # Activation function in the hidden layer.\n",
    "    activation=activation,\n",
    "    # L2 penalty (regularization term) parameter.\n",
    "    alpha=alpha,\n",
    "    # Bath size.\n",
    "    batch_size=batch_size,\n",
    "    # The ith element represents the number of neurons in the ith hidden layer.\n",
    "    hidden_layer_sizes=(hidden_n,), \n",
    "    # The initial learning rate used.\n",
    "    learning_rate_init=learning_rate_init,\n",
    "    # Maximum number of iterations.\n",
    "    max_iter=max_iter,\n",
    "    # Solver.\n",
    "    solver=solver,\n",
    "    shuffle=True,\n",
    "    verbose=False\n",
    ")\n",
    "\n",
    "# We learn the digits on the first half of the digits\n",
    "classifier.fit(data[:n_samples // 2], digits.target[:n_samples // 2])\n",
    "\n",
    "# Now predict the value of the digit on the second half:\n",
    "expected = digits.target[n_samples // 2:]\n",
    "predicted = classifier.predict(data[n_samples // 2:])\n",
    "\n",
    "print(\"Classification report for classifier %s:\\n%s\\n\"\n",
    "      % (classifier, metrics.classification_report(expected, predicted)))\n",
    "print(\"Confusion matrix:\\n%s\" % metrics.confusion_matrix(expected, predicted))\n",
    "\n",
    "images_and_predictions = list(zip(digits.images[n_samples // 2:], predicted))\n",
    "for index, (image, prediction) in enumerate(images_and_predictions[:4]):\n",
    "    plt.subplot(2, 4, index + 5)\n",
    "    plt.axis('off')\n",
    "    plt.imshow(image, cmap=plt.cm.gray_r, interpolation='nearest')\n",
    "    plt.title('Prediction: %i' % prediction)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparison of accuracy.\n",
    "\n",
    "As described above, it is backed by data such as precision, recall, f1-score, and support that training models in the case of using simulated annealing, adaptive simulated annealing, and Quantum Monte Carlo have a higher score than the model based on default hyperparameters.\n",
    "\n",
    "#### In case of default hyperparameters.\n",
    "```\n",
    "precision    recall  f1-score   support\n",
    "\n",
    "          0       0.89      0.94      0.92        88\n",
    "          1       0.88      0.81      0.85        91\n",
    "          2       0.96      0.95      0.96        86\n",
    "          3       0.83      0.87      0.85        91\n",
    "          4       0.98      0.91      0.94        92\n",
    "          5       0.80      0.95      0.87        91\n",
    "          6       0.98      0.98      0.98        91\n",
    "          7       0.89      0.96      0.92        89\n",
    "          8       0.88      0.72      0.79        88\n",
    "          9       0.80      0.78      0.79        92\n",
    "\n",
    "avg / total       0.89      0.89      0.89       899\n",
    "```\n",
    "\n",
    "#### In case of using Simulated Annealing.\n",
    "\n",
    "```\n",
    "precision    recall  f1-score   support\n",
    "\n",
    "          0       1.00      0.95      0.98        88\n",
    "          1       0.94      0.92      0.93        91\n",
    "          2       0.92      1.00      0.96        86\n",
    "          3       0.96      0.87      0.91        91\n",
    "          4       0.98      0.88      0.93        92\n",
    "          5       0.92      0.96      0.94        91\n",
    "          6       0.90      0.98      0.94        91\n",
    "          7       0.93      0.93      0.93        89\n",
    "          8       0.94      0.88      0.91        88\n",
    "          9       0.85      0.96      0.90        92\n",
    "\n",
    "avg / total       0.93      0.93      0.93       899\n",
    "```\n",
    "\n",
    "#### In case of using Adaptive Simulated Annealing.\n",
    "\n",
    "```\n",
    "precision    recall  f1-score   support\n",
    "\n",
    "          0       1.00      0.97      0.98        88\n",
    "          1       0.93      0.89      0.91        91\n",
    "          2       0.99      1.00      0.99        86\n",
    "          3       0.98      0.87      0.92        91\n",
    "          4       0.95      0.88      0.92        92\n",
    "          5       0.88      0.93      0.90        91\n",
    "          6       0.89      0.98      0.93        91\n",
    "          7       0.98      0.94      0.96        89\n",
    "          8       0.88      0.90      0.89        88\n",
    "          9       0.85      0.93      0.89        92\n",
    "\n",
    "avg / total       0.93      0.93      0.93       899```\n",
    "\n",
    "#### In case of using Quantum Monte Carlo\n",
    "\n",
    "```\n",
    "precision    recall  f1-score   support\n",
    "\n",
    "          0       0.98      0.98      0.98        88\n",
    "          1       0.94      0.91      0.93        91\n",
    "          2       1.00      0.98      0.99        86\n",
    "          3       0.94      0.87      0.90        91\n",
    "          4       0.98      0.91      0.94        92\n",
    "          5       0.91      0.95      0.92        91\n",
    "          6       0.95      0.99      0.97        91\n",
    "          7       0.94      0.99      0.96        89\n",
    "          8       0.94      0.91      0.92        88\n",
    "          9       0.86      0.93      0.90        92\n",
    "\n",
    "avg / total       0.94      0.94      0.94       899```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Issue: Hyperparameters in order to tune hyperparameters.\n",
    "\n",
    "The function of `AnnealingModel` is to tune hyperparameters. But this model itself requires to be tuned hyperparameters. Then how can be hyperparameters decided? Can `AnnealingModel` decide hyperparameters of `AnnealingModel` ? So the problem solution as described above can raise a paradox: **hyperparameters in order to tune hyperparameters in order to tune hyperparameters in order to tune hyperparameters...**.\n",
    "\n",
    "It is not important for a mere end-user in the machine learning library to think that how can be this paradox *unfolded*. For instance, a kind of verification procedure such as so-called appeal from authority, in any case, makes it possible to *de-paradoxize* this issue. \n",
    "\n",
    "In spite of the fact, it is not necessary for us to follow expert advice and to trust primer's intuition about hyperparameter tuning blindly. According to modal logic, the relationship between algorithms and data structures in *your* problem setting can raise contingent result. Without customization, the machine learning library may well be of no practical use.\n",
    "\n",
    "This issue is not finished yet.  True is not revealed.  And it circulates through developments──It is an infinity loop !\n",
    "\n",
    "## Problem re-setting: Self-referential machine learning, or Autologie.\n",
    "\n",
    "It is important for us to produce new PoC in relation to theory of *self-referential systems* so as to detoxify the paradox. The systems are a kind of machine learning models to interconnect, tuning their hyperparameters mutually. With the completion of those models, the machine learning model will be called Artificial intelligence(AI)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# References\n",
    "\n",
    "- Bektas, T. (2006). The multiple traveling salesman problem: an overview of formulations and solution procedures. Omega, 34(3), 209-219.\n",
    "- Bertsimas, D., & Tsitsiklis, J. (1993). Simulated annealing. Statistical science, 8(1), 10-15.\n",
    "- Das, A., & Chakrabarti, B. K. (Eds.). (2005). Quantum annealing and related optimization methods (Vol. 679). Springer Science & Business Media.\n",
    "- Du, K. L., & Swamy, M. N. S. (2016). Search and optimization by metaheuristics. New York City: Springer.\n",
    "- Edwards, S. F., & Anderson, P. W. (1975). Theory of spin glasses. Journal of Physics F: Metal Physics, 5(5), 965.\n",
    "- Facchi, P., & Pascazio, S. (2008). Quantum Zeno dynamics: mathematical and physical aspects. Journal of Physics A: Mathematical and Theoretical, 41(49), 493001.\n",
    "- Heim, B., Rønnow, T. F., Isakov, S. V., & Troyer, M. (2015). Quantum versus classical annealing of Ising spin glasses. Science, 348(6231), 215-217.\n",
    "- Heisenberg, W. (1925) Über quantentheoretische Umdeutung kinematischer und mechanischer Beziehungen. Z. Phys. 33, pp.879—893.\n",
    "- Heisenberg, W. (1927). Über den anschaulichen Inhalt der quantentheoretischen Kinematik und Mechanik. Zeitschrift fur Physik, 43, 172-198.\n",
    "- Heisenberg, W. (1984). The development of quantum mechanics. In Scientific Review Papers, Talks, and Books -Wissenschaftliche Übersichtsartikel, Vorträge und Bücher (pp. 226-237). Springer Berlin Heidelberg.\n",
    "Hilgevoord, Jan and Uffink, Jos, \"The Uncertainty Principle\", The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), Edward N. Zalta (ed.), URL = ＜https://plato.stanford.edu/archives/win2016/entries/qt-uncertainty/＞.\n",
    "- Jarzynski, C. (1997). Nonequilibrium equality for free energy differences. Physical Review Letters, 78(14), 2690.\n",
    "- Messiah, A. (1966). Quantum mechanics. 2 (1966). North-Holland Publishing Company.\n",
    "- Mezard, M., & Montanari, A. (2009). Information, physics, and computation. Oxford University Press.\n",
    "- Nallusamy, R., Duraiswamy, K., Dhanalaksmi, R., & Parthiban, P. (2009). Optimization of non-linear multiple traveling salesman problem using k-means clustering, shrink wrap algorithm and meta-heuristics. International Journal of Nonlinear Science, 8(4), 480-487.\n",
    "- Schrödinger, E. (1926). Quantisierung als eigenwertproblem. Annalen der physik, 385(13), S.437-490.\n",
    "- Somma, R. D., Batista, C. D., & Ortiz, G. (2007). Quantum approach to classical statistical mechanics. Physical review letters, 99(3), 030603.\n",
    "- 鈴木正. (2008). 「組み合わせ最適化問題と量子アニーリング: 量子断熱発展の理論と性能評価」.,『物性研究』, 90(4): pp598-676. 参照箇所はpp619-624.\n",
    "- 西森秀稔、大関真之(2018) 『量子アニーリングの基礎』須藤 彰三、岡 真 監修、共立出版、参照箇所はpp9-46.\n",
    "\n",
    "## My PoC\n",
    "\n",
    "- [量子力学、統計力学、熱力学における天才物理学者たちの神学的な形象について](https://accel-brain.com/das-theologische-bild-genialer-physiker-in-der-quantenmechanik-und-der-statistischen-mechanik-und-thermodynamik/) (Japanese)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}