FYS-STK4155/doc/Programs/IsingModel/logisticIsing.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predicting phase transition of the two-dimensional Ising model\n",
    "\n",
    "In this notebook we will be doing a binary classification of the two-dimensional Ising model where we try to predict if a given state is `ordered` or `disordered`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The Ising model\n",
    "\n",
    "The Hamiltonian of the two-dimensional Ising model without an external field for a constant coupling constant $J$ is given by\n",
    "\n",
    "\\begin{align}\n",
    "    H = -J \\sum_{\\langle ij\\rangle} S_i S_j,\n",
    "\\end{align}\n",
    "\n",
    "where $S_i \\in \\{-1, 1\\}$ and $\\langle ij \\rangle$ signifies that we only iterate over the nearest neighbors in the lattice. We will be looking at a system of $L = 40$ spins in each dimension, i.e., $L^2 = 1600$ spins in total. Opposed to the one-dimensional Ising model we will get a phase transition from an _ordered_ phase to a _disordered_ phase at the critical temperature\n",
    "\n",
    "\\begin{align}\n",
    "    \\frac{T_c}{J} = \\frac{2}{\\log\\left(1 + \\sqrt{2}\\right)} \\approx 2.26,\n",
    "\\end{align}\n",
    "\n",
    "as shown by Lars Onsager."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### The goal\n",
    "\n",
    "We wish in this notebook to use _logistic regression_ to predict when a phase transition occurs. The data we will look at is a set of spin configurations, i.e., individual lattices with spins, labeled _ordered_ `1` or _disordered_ `0`. Our job is to build a model which will take in a spin configuration and predict whether or not the spin configuration constitutes an ordered or a disordered phase. To achieve this we will represent the lattices as flattened arrays with $1600$ elements instead of a matrix of $40 \\times 40$ elements. As an extra test of the performance of the algorithms we will divide the dataset into three pieces. We will do a conventional train-test-split on a combination of totally ordered and totally disordered phases. The remaining \"critical-like\" states will be used as test data which we hope the model will be able to make good extrapolated predictions on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "import os\n",
    "import glob\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import sklearn.model_selection as skms\n",
    "import sklearn.linear_model as skl\n",
    "import sklearn.metrics as skm\n",
    "import tqdm\n",
    "import copy\n",
    "import time\n",
    "from IPython.display import display\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "sns.set(color_codes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "// Fill entire cell with output\n",
       "IPython.OutputArea.prototype._should_scroll = function(lines) {\n",
       "    return false;\n",
       "}"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%javascript\n",
    "// Fill entire cell with output\n",
    "IPython.OutputArea.prototype._should_scroll = function(lines) {\n",
    "    return false;\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reading in the data\n",
    "\n",
    "Using the data from [here](https://physics.bu.edu/~pankajm/ML-Review-Datasets/isingMC/) (specifically the two datasets named `Ising2DFM_reSample_L40_T=All.pkl` and `Ising2DFM_reSample_L40_T=All_labels.pkl`) we have to unpack the data into numpy arrays."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "filenames = glob.glob(os.path.join(\"..\", \"dat\", \"*\"))\n",
    "label_filename = list(filter(lambda x: \"label\" in x, filenames))[0]\n",
    "dat_filename = list(filter(lambda x: \"label\" not in x, filenames))[0]\n",
    "\n",
    "# Read in the labels\n",
    "with open(label_filename, \"rb\") as f:\n",
    "    labels = pickle.load(f)\n",
    "\n",
    "# Read in the corresponding configurations\n",
    "with open(dat_filename, \"rb\") as f:\n",
    "    data = np.unpackbits(pickle.load(f)).reshape(-1, 1600).astype(\"int\")\n",
    "\n",
    "# Set spin-down to -1\n",
    "data[data == 0] = -1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This dataset consists of $10000$ samples, i.e., $10000$ spin configurations with $40 \\times 40$ spins each, for $16$ temperatures between $0.25$ to $4.0$. Next we create a train/test-split and keep the data in the critical phase as a separate dataset for extrapolation-testing."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Set up slices of the dataset\n",
    "ordered = slice(0, 70000)\n",
    "critical = slice(70000, 100000)\n",
    "disordered = slice(100000, 160000)\n",
    "\n",
    "X_train, X_test, y_train, y_test = skms.train_test_split(\n",
    "    np.concatenate((data[ordered], data[disordered])),\n",
    "    np.concatenate((labels[ordered], labels[disordered])),\n",
    "    test_size=0.95\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__Note:__ Using a small training set yields a better accuracy. This will be discussed in the end."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Logistic regression\n",
    "\n",
    "Logistic regression is a linear model for classification. Recalling the cost function for ordinary least squares with both L2 (ridge) and L1 (LASSO) penalties we will see that the logistic cost function is very similar. In OLS we wish to predict a continuous variable $\\hat{y}$ using\n",
    "\n",
    "\\begin{align}\n",
    "    \\hat{y} = X\\omega,\n",
    "\\end{align}\n",
    "\n",
    "where $X \\in \\mathbb{R}^{n \\times p}$ is the input data and $\\omega^{p \\times d}$ are the weights of the regression. In a classification setting (binary classification in our situation) we are interested in a positive or negative answer. We can thus define either answer to be above or below some threshold. But, in order to limit the size of the answer and also to get a probability interpretation on how sure we are for either answer we can compute the sigmoid function of OLS. That is,\n",
    "\n",
    "\\begin{align}\n",
    "    f(X\\omega) = \\frac{1}{1 + \\exp(-X\\omega)}.\n",
    "\\end{align}\n",
    "\n",
    "We are thus interested in minimizing the following cost function\n",
    "\n",
    "\\begin{align}\n",
    "    C(X, \\omega) = \\sum_{i = 1}^n \\left\\{\n",
    "        - y_i\\log\\left( f(x_i^T\\omega) \\right)\n",
    "        - (1 - y_i)\\log\\left[1 - f(x_i^T\\omega)\\right]\n",
    "    \\right\\},\n",
    "\\end{align}\n",
    "\n",
    "where we will restrict ourselves to a value for $f(z)$ as the sigmoid described above. We can also tack on a L2 (Ridge) or L1 (LASSO) penalization to this cost function in the same manner we did for linear regression."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Exploring the logistic regression\n",
    "\n",
    "The penalization factor $\\lambda$ is inverted in the case of the logistic regression model we use. We will explore several values of $\\lambda$ using both L1 and L2 penalization. We do this using a grid search over different parameters and run a 3-fold cross validation for each configuration. In other words, we fit a model 3 times for each configuration of the hyper parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "lambdas = np.logspace(-7, -1, 7)\n",
    "\n",
    "param_grid = {\n",
    "    \"C\": list(1.0/lambdas),\n",
    "    \"penalty\": [\"l1\", \"l2\"]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "clf = skms.GridSearchCV(\n",
    "    skl.LogisticRegression(),\n",
    "    param_grid=param_grid,\n",
    "    n_jobs=-1,\n",
    "    return_train_score=True\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Time spent fitting GridSearchCV(LogisticRegression): 949.417 sec\n"
     ]
    }
   ],
   "source": [
    "t0 = time.time()\n",
    "clf.fit(X_train, y_train)\n",
    "t1 = time.time()\n",
    "\n",
    "print (\n",
    "    \"Time spent fitting GridSearchCV(LogisticRegression): {0:.3f} sec\".format(\n",
    "        t1 - t0\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that logistic regression is quite slow and using the grid search and cross validation results in quite a heavy computation. Below we show the results of the different configurations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean_fit_time</th>\n",
       "      <th>std_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>param_C</th>\n",
       "      <th>param_penalty</th>\n",
       "      <th>params</th>\n",
       "      <th>split0_test_score</th>\n",
       "      <th>split1_test_score</th>\n",
       "      <th>split2_test_score</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>rank_test_score</th>\n",
       "      <th>split0_train_score</th>\n",
       "      <th>split1_train_score</th>\n",
       "      <th>split2_train_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>133.985549</td>\n",
       "      <td>43.129200</td>\n",
       "      <td>0.043782</td>\n",
       "      <td>0.002575</td>\n",
       "      <td>1e+07</td>\n",
       "      <td>l1</td>\n",
       "      <td>{'C': 10000000.0, 'penalty': 'l1'}</td>\n",
       "      <td>0.694047</td>\n",
       "      <td>0.709737</td>\n",
       "      <td>0.700831</td>\n",
       "      <td>0.701538</td>\n",
       "      <td>0.006425</td>\n",
       "      <td>12</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>51.797149</td>\n",
       "      <td>7.266426</td>\n",
       "      <td>0.058810</td>\n",
       "      <td>0.008064</td>\n",
       "      <td>1e+07</td>\n",
       "      <td>l2</td>\n",
       "      <td>{'C': 10000000.0, 'penalty': 'l2'}</td>\n",
       "      <td>0.694509</td>\n",
       "      <td>0.711583</td>\n",
       "      <td>0.702216</td>\n",
       "      <td>0.702769</td>\n",
       "      <td>0.006982</td>\n",
       "      <td>8</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>125.188562</td>\n",
       "      <td>46.865003</td>\n",
       "      <td>0.028435</td>\n",
       "      <td>0.009611</td>\n",
       "      <td>1e+06</td>\n",
       "      <td>l1</td>\n",
       "      <td>{'C': 1000000.0, 'penalty': 'l1'}</td>\n",
       "      <td>0.693586</td>\n",
       "      <td>0.714352</td>\n",
       "      <td>0.700831</td>\n",
       "      <td>0.702923</td>\n",
       "      <td>0.008606</td>\n",
       "      <td>6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>46.434756</td>\n",
       "      <td>6.818460</td>\n",
       "      <td>0.047470</td>\n",
       "      <td>0.021621</td>\n",
       "      <td>1e+06</td>\n",
       "      <td>l2</td>\n",
       "      <td>{'C': 1000000.0, 'penalty': 'l2'}</td>\n",
       "      <td>0.694509</td>\n",
       "      <td>0.711121</td>\n",
       "      <td>0.702216</td>\n",
       "      <td>0.702615</td>\n",
       "      <td>0.006789</td>\n",
       "      <td>9</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>131.480284</td>\n",
       "      <td>49.637990</td>\n",
       "      <td>0.025367</td>\n",
       "      <td>0.001607</td>\n",
       "      <td>100000</td>\n",
       "      <td>l1</td>\n",
       "      <td>{'C': 99999.99999999999, 'penalty': 'l1'}</td>\n",
       "      <td>0.693586</td>\n",
       "      <td>0.711583</td>\n",
       "      <td>0.700831</td>\n",
       "      <td>0.702000</td>\n",
       "      <td>0.007394</td>\n",
       "      <td>11</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>48.442208</td>\n",
       "      <td>4.821450</td>\n",
       "      <td>0.025852</td>\n",
       "      <td>0.000660</td>\n",
       "      <td>100000</td>\n",
       "      <td>l2</td>\n",
       "      <td>{'C': 99999.99999999999, 'penalty': 'l2'}</td>\n",
       "      <td>0.694509</td>\n",
       "      <td>0.712506</td>\n",
       "      <td>0.703139</td>\n",
       "      <td>0.703385</td>\n",
       "      <td>0.007350</td>\n",
       "      <td>3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>136.005517</td>\n",
       "      <td>63.967902</td>\n",
       "      <td>0.035429</td>\n",
       "      <td>0.009921</td>\n",
       "      <td>10000</td>\n",
       "      <td>l1</td>\n",
       "      <td>{'C': 10000.0, 'penalty': 'l1'}</td>\n",
       "      <td>0.693586</td>\n",
       "      <td>0.712967</td>\n",
       "      <td>0.700831</td>\n",
       "      <td>0.702462</td>\n",
       "      <td>0.007997</td>\n",
       "      <td>10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>45.492904</td>\n",
       "      <td>4.962405</td>\n",
       "      <td>0.024428</td>\n",
       "      <td>0.003608</td>\n",
       "      <td>10000</td>\n",
       "      <td>l2</td>\n",
       "      <td>{'C': 10000.0, 'penalty': 'l2'}</td>\n",
       "      <td>0.693586</td>\n",
       "      <td>0.712506</td>\n",
       "      <td>0.703601</td>\n",
       "      <td>0.703231</td>\n",
       "      <td>0.007729</td>\n",
       "      <td>5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>109.751415</td>\n",
       "      <td>29.097978</td>\n",
       "      <td>0.036205</td>\n",
       "      <td>0.012117</td>\n",
       "      <td>1000</td>\n",
       "      <td>l1</td>\n",
       "      <td>{'C': 1000.0, 'penalty': 'l1'}</td>\n",
       "      <td>0.693586</td>\n",
       "      <td>0.712044</td>\n",
       "      <td>0.703139</td>\n",
       "      <td>0.702923</td>\n",
       "      <td>0.007538</td>\n",
       "      <td>6</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>40.780339</td>\n",
       "      <td>5.498792</td>\n",
       "      <td>0.022993</td>\n",
       "      <td>0.002149</td>\n",
       "      <td>1000</td>\n",
       "      <td>l2</td>\n",
       "      <td>{'C': 1000.0, 'penalty': 'l2'}</td>\n",
       "      <td>0.693124</td>\n",
       "      <td>0.711583</td>\n",
       "      <td>0.705448</td>\n",
       "      <td>0.703385</td>\n",
       "      <td>0.007676</td>\n",
       "      <td>3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>171.943614</td>\n",
       "      <td>10.621141</td>\n",
       "      <td>0.046334</td>\n",
       "      <td>0.018095</td>\n",
       "      <td>100</td>\n",
       "      <td>l1</td>\n",
       "      <td>{'C': 100.0, 'penalty': 'l1'}</td>\n",
       "      <td>0.684818</td>\n",
       "      <td>0.708814</td>\n",
       "      <td>0.699908</td>\n",
       "      <td>0.697846</td>\n",
       "      <td>0.009905</td>\n",
       "      <td>14</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>35.711317</td>\n",
       "      <td>5.045432</td>\n",
       "      <td>0.021141</td>\n",
       "      <td>0.000833</td>\n",
       "      <td>100</td>\n",
       "      <td>l2</td>\n",
       "      <td>{'C': 100.0, 'penalty': 'l2'}</td>\n",
       "      <td>0.694509</td>\n",
       "      <td>0.712506</td>\n",
       "      <td>0.705448</td>\n",
       "      <td>0.704154</td>\n",
       "      <td>0.007405</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>111.015806</td>\n",
       "      <td>21.466425</td>\n",
       "      <td>0.031371</td>\n",
       "      <td>0.011843</td>\n",
       "      <td>10</td>\n",
       "      <td>l1</td>\n",
       "      <td>{'C': 10.0, 'penalty': 'l1'}</td>\n",
       "      <td>0.685741</td>\n",
       "      <td>0.716198</td>\n",
       "      <td>0.698523</td>\n",
       "      <td>0.700154</td>\n",
       "      <td>0.012488</td>\n",
       "      <td>13</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>30.197917</td>\n",
       "      <td>4.607725</td>\n",
       "      <td>0.021942</td>\n",
       "      <td>0.004698</td>\n",
       "      <td>10</td>\n",
       "      <td>l2</td>\n",
       "      <td>{'C': 10.0, 'penalty': 'l2'}</td>\n",
       "      <td>0.694047</td>\n",
       "      <td>0.714352</td>\n",
       "      <td>0.702216</td>\n",
       "      <td>0.703538</td>\n",
       "      <td>0.008342</td>\n",
       "      <td>2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999769</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.999923</td>\n",
       "      <td>0.000109</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    mean_fit_time  std_fit_time  mean_score_time  std_score_time param_C  \\\n",
       "0      133.985549     43.129200         0.043782        0.002575   1e+07   \n",
       "1       51.797149      7.266426         0.058810        0.008064   1e+07   \n",
       "2      125.188562     46.865003         0.028435        0.009611   1e+06   \n",
       "3       46.434756      6.818460         0.047470        0.021621   1e+06   \n",
       "4      131.480284     49.637990         0.025367        0.001607  100000   \n",
       "5       48.442208      4.821450         0.025852        0.000660  100000   \n",
       "6      136.005517     63.967902         0.035429        0.009921   10000   \n",
       "7       45.492904      4.962405         0.024428        0.003608   10000   \n",
       "8      109.751415     29.097978         0.036205        0.012117    1000   \n",
       "9       40.780339      5.498792         0.022993        0.002149    1000   \n",
       "10     171.943614     10.621141         0.046334        0.018095     100   \n",
       "11      35.711317      5.045432         0.021141        0.000833     100   \n",
       "12     111.015806     21.466425         0.031371        0.011843      10   \n",
       "13      30.197917      4.607725         0.021942        0.004698      10   \n",
       "\n",
       "   param_penalty                                     params  \\\n",
       "0             l1         {'C': 10000000.0, 'penalty': 'l1'}   \n",
       "1             l2         {'C': 10000000.0, 'penalty': 'l2'}   \n",
       "2             l1          {'C': 1000000.0, 'penalty': 'l1'}   \n",
       "3             l2          {'C': 1000000.0, 'penalty': 'l2'}   \n",
       "4             l1  {'C': 99999.99999999999, 'penalty': 'l1'}   \n",
       "5             l2  {'C': 99999.99999999999, 'penalty': 'l2'}   \n",
       "6             l1            {'C': 10000.0, 'penalty': 'l1'}   \n",
       "7             l2            {'C': 10000.0, 'penalty': 'l2'}   \n",
       "8             l1             {'C': 1000.0, 'penalty': 'l1'}   \n",
       "9             l2             {'C': 1000.0, 'penalty': 'l2'}   \n",
       "10            l1              {'C': 100.0, 'penalty': 'l1'}   \n",
       "11            l2              {'C': 100.0, 'penalty': 'l2'}   \n",
       "12            l1               {'C': 10.0, 'penalty': 'l1'}   \n",
       "13            l2               {'C': 10.0, 'penalty': 'l2'}   \n",
       "\n",
       "    split0_test_score  split1_test_score  split2_test_score  mean_test_score  \\\n",
       "0            0.694047           0.709737           0.700831         0.701538   \n",
       "1            0.694509           0.711583           0.702216         0.702769   \n",
       "2            0.693586           0.714352           0.700831         0.702923   \n",
       "3            0.694509           0.711121           0.702216         0.702615   \n",
       "4            0.693586           0.711583           0.700831         0.702000   \n",
       "5            0.694509           0.712506           0.703139         0.703385   \n",
       "6            0.693586           0.712967           0.700831         0.702462   \n",
       "7            0.693586           0.712506           0.703601         0.703231   \n",
       "8            0.693586           0.712044           0.703139         0.702923   \n",
       "9            0.693124           0.711583           0.705448         0.703385   \n",
       "10           0.684818           0.708814           0.699908         0.697846   \n",
       "11           0.694509           0.712506           0.705448         0.704154   \n",
       "12           0.685741           0.716198           0.698523         0.700154   \n",
       "13           0.694047           0.714352           0.702216         0.703538   \n",
       "\n",
       "    std_test_score  rank_test_score  split0_train_score  split1_train_score  \\\n",
       "0         0.006425               12                 1.0            1.000000   \n",
       "1         0.006982                8                 1.0            1.000000   \n",
       "2         0.008606                6                 1.0            1.000000   \n",
       "3         0.006789                9                 1.0            1.000000   \n",
       "4         0.007394               11                 1.0            1.000000   \n",
       "5         0.007350                3                 1.0            1.000000   \n",
       "6         0.007997               10                 1.0            1.000000   \n",
       "7         0.007729                5                 1.0            1.000000   \n",
       "8         0.007538                6                 1.0            1.000000   \n",
       "9         0.007676                3                 1.0            1.000000   \n",
       "10        0.009905               14                 1.0            1.000000   \n",
       "11        0.007405                1                 1.0            1.000000   \n",
       "12        0.012488               13                 1.0            1.000000   \n",
       "13        0.008342                2                 1.0            0.999769   \n",
       "\n",
       "    split2_train_score  mean_train_score  std_train_score  \n",
       "0                  1.0          1.000000         0.000000  \n",
       "1                  1.0          1.000000         0.000000  \n",
       "2                  1.0          1.000000         0.000000  \n",
       "3                  1.0          1.000000         0.000000  \n",
       "4                  1.0          1.000000         0.000000  \n",
       "5                  1.0          1.000000         0.000000  \n",
       "6                  1.0          1.000000         0.000000  \n",
       "7                  1.0          1.000000         0.000000  \n",
       "8                  1.0          1.000000         0.000000  \n",
       "9                  1.0          1.000000         0.000000  \n",
       "10                 1.0          1.000000         0.000000  \n",
       "11                 1.0          1.000000         0.000000  \n",
       "12                 1.0          1.000000         0.000000  \n",
       "13                 1.0          0.999923         0.000109  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "logreg_df = pd.DataFrame(clf.cv_results_)\n",
    "\n",
    "display(logreg_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Accuracy of a classification model\n",
    "\n",
    "To determine how well a classification model is performing we count the number of correctly labeled classes and divide by the number of classes in total. The accuracy is thus given by\n",
    "\n",
    "\\begin{align}\n",
    "    a(y, \\hat{y}) = \\frac{1}{n}\\sum_{i = 1}^{n} I(y_i = \\hat{y}_i),\n",
    "\\end{align}\n",
    "\n",
    "where $I(y_i = \\hat{y}_i)$ is the indicator function given by\n",
    "\n",
    "\\begin{align}\n",
    "    I(x = y) = \\begin{cases}\n",
    "    1 & x = y, \\\\\n",
    "    0 & x \\neq y.\n",
    "    \\end{cases}\n",
    "\\end{align}\n",
    "\n",
    "This is the accuracy provided by Scikit-learn when using `sklearn.metrics.accuracy_score`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below we compute the accuracy of the best fit model on the training data (which should give a good accuracy), the test data (which has not been shown to the model) and the critical data (completely new data that needs to be extrapolated)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on train data: 0.876\n",
      "Accuracy on test data: 0.6896275303643725\n",
      "Accuracy on critical data: 0.5445\n"
     ]
    }
   ],
   "source": [
    "train_accuracy = skm.accuracy_score(y_train, clf.predict(X_train))\n",
    "test_accuracy = skm.accuracy_score(y_test, clf.predict(X_test))\n",
    "critical_accuracy = skm.accuracy_score(labels[critical], clf.predict(data[critical]))\n",
    "\n",
    "print (\"Accuracy on train data: {0}\".format(train_accuracy))\n",
    "print (\"Accuracy on test data: {0}\".format(test_accuracy))\n",
    "print (\"Accuracy on critical data: {0}\".format(critical_accuracy))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that we get quite good accuracy on the training data, but gradually worsening accuracy on the test and critical data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Receiver operating characteristic\n",
    "\n",
    "Below we show a different metric for determining the quality of our model, namely the _reciever operating characteristic_ (ROC). The ROC curve tells us how well the model correctly classifies the different labels. We plot the _true positive rate_ (the rate of predicted positive classes that are positive) versus the _false positive rate_ (the rate of predicted positive classes that are negative). The ROC curve is built by computing the true positive rate and the false positive rate for varying _thresholds_, i.e, which probability we should acredit a certain class.\n",
    "\n",
    "By computing the _area under the curve_ (AUC) of the ROC curve we get an estimate of how well our model is performing. Pure guessing will get an AUC of $0.5$. A perfect score will get an AUC of $1.0$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression AUC (Train): 0.927466526230532\n",
      "LogisticRegression AUC (Test): 0.4975344101021899\n",
      "LogisticRegression AUC (Critical): 0.500251495\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKEAAANKCAYAAABS413oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3f8DHL4eSfRrbQkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1098a2208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(20, 14))\n",
    "\n",
    "for (_X, _y), label in zip(\n",
    "    [\n",
    "        (X_train, y_train),\n",
    "        (X_test, y_test),\n",
    "        (data[critical], labels[critical])\n",
    "    ],\n",
    "    [\"Train\", \"Test\", \"Critical\"]\n",
    "):\n",
    "    proba = clf.predict_proba(_X)\n",
    "    fpr, tpr, _ = skm.roc_curve(_y, proba[:, 1])\n",
    "    roc_auc = skm.auc(fpr, tpr)\n",
    "\n",
    "    print (\"LogisticRegression AUC ({0}): {1}\".format(label, roc_auc))\n",
    "\n",
    "    plt.plot(fpr, tpr, label=\"{0} (AUC = {1})\".format(label, roc_auc), linewidth=4.0)\n",
    "\n",
    "plt.plot([0, 1], [0, 1], \"--\", label=\"Guessing (AUC = 0.5)\", linewidth=4.0)\n",
    "\n",
    "plt.title(r\"The ROC curve for LogisticRegression\", fontsize=18)\n",
    "plt.xlabel(r\"False positive rate\", fontsize=18)\n",
    "plt.ylabel(r\"True positive rate\", fontsize=18)\n",
    "plt.axis([-0.01, 1.01, -0.01, 1.01])\n",
    "plt.xticks(fontsize=18)\n",
    "plt.yticks(fontsize=18)\n",
    "plt.legend(loc=\"best\", fontsize=18)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that this plot of the ROC looks very strange. This tells us that logistic regression is quite inept at predicting the Ising model transition and is therefore highly non-linear. The ROC curve for the training data looks quite good, but as the testing data is so far off we see that we are dealing with an overfit model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A previous run with $50\\%$ of the data used for training yielded a worse performance than using a smaller training set. This again gives confidence to the fact that logistic regression is not able to correctly fit the Ising model as it is not a linear model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![](large_training_data.jpg)"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}