vbm_hpc_prediction_multi.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/angela/anaconda3/envs/vcog_paper_py35/lib/python3.5/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n",
      "/home/angela/anaconda3/envs/vcog_paper_py35/lib/python3.5/site-packages/statsmodels/compat/pandas.py:56: FutureWarning: The pandas.core.datetools module is deprecated and will be removed in a future version. Please use the pandas.tseries module instead.\n",
      "  from pandas.core import datetools\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import scipy.io\n",
    "from os import listdir\n",
    "from sklearn.utils import shuffle\n",
    "from sklearn import preprocessing\n",
    "#from nilearn import plotting\n",
    "#from proteus.io import util\n",
    "from proteus.visu import sbp_visu\n",
    "\n",
    "import glob,os\n",
    "#import nibabel as nib\n",
    "\n",
    "\n",
    "import pickle\n",
    "from proteus.predic import high_confidence_at\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from proteus.predic import prediction\n",
    "from sklearn import preprocessing\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "\n",
    "from sklearn import metrics\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import cm\n",
    "import seaborn as sns\n",
    "\n",
    "import statsmodels.formula.api as smf\n",
    "import statsmodels.api as sm\n",
    "from statsmodels.stats.multicomp import pairwise_tukeyhsd\n",
    "\n",
    "from copy import deepcopy\n",
    "\n",
    "from itertools import cycle\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import label_binarize\n",
    "from sklearn.multiclass import OneVsRestClassifier\n",
    "from scipy import interp\n",
    "\n",
    "from sklearn.metrics import average_precision_score\n",
    "from sklearn.metrics import precision_recall_curve"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def stats_mask(y_true, y_pred, mask_selected=None):\n",
    "    if mask_selected is None:\n",
    "        mask_selected = np.ones(y_pred.shape).astype(bool)\n",
    "    print('------------------------')\n",
    "    print('Ratio:', y_true[mask_selected].sum()/y_true.sum()) \n",
    "    print('#    : ', y_true[mask_selected].sum()) \n",
    "    print('# true values: ',mask_selected.sum())\n",
    "    print('ACC  : ', np.mean((y_true == y_pred)[mask_selected])) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predic_stats(y_, y_pred, lr_decision):\n",
    "    # number of AD subjects\n",
    "    n_ad = sum(y_)\n",
    "    print('Total number of TARGET subjects: ', n_ad)\n",
    "\n",
    "    # number of CN subjects\n",
    "    n_cn = len(y_) - sum(y_)\n",
    "    print('Total number of NON-TARGET subjects: ', n_cn)\n",
    "    \n",
    "    # number of subjects predicted as AD at stage 1\n",
    "    n_pos = sum(y_pred)\n",
    "    print('Stage 1 number of hits (true and false positives): ', n_pos)\n",
    "    \n",
    "    # true positives at stage 1\n",
    "    n_pos_ad = sum(y_pred[y_.astype(bool)])\n",
    "    print('Stage 1 TRUE positives: ', n_pos_ad)\n",
    "    \n",
    "    # false positives at stage 1\n",
    "    n_pos_cn = n_pos - n_pos_ad\n",
    "    print('Stage 1 FALSE positives: ', n_pos_cn)\n",
    "    \n",
    "    # number of CN subjects not identified as positive (true negatives)\n",
    "    n_neg1_cn = n_cn - n_pos_cn\n",
    "    print('Stage 1 TRUE negatives: ', n_neg1_cn)\n",
    "\n",
    "    # number of all flagged HPC-AD subjects\n",
    "    n_flag = sum(y_pred[lr_decision>0])\n",
    "    print('Total number of flagged HPC-AD subjects: ', n_flag)\n",
    "\n",
    "    # number of flagged HPC-AD subjects who are actually AD (true positives)\n",
    "    y_pred_true = y_ + y_pred\n",
    "    y_pred_true = y_pred_true==2\n",
    "    n_flag_ad = sum(y_pred_true[lr_decision>0])\n",
    "    print('Number of flagged HPC-AD subjects that are TRUE positives: ', n_flag_ad)\n",
    "\n",
    "    # number of flagged HPC-AD subjects that are actually CN (false positives)\n",
    "    n_flag_cn = n_flag - n_flag_ad\n",
    "    print('Number of flagged HPC-AD subjects that are FALSE positives: ', n_flag_cn)\n",
    "\n",
    "    # number of CN subjects that were not flagged (true negatives)\n",
    "    n_neg_cn = n_cn - n_flag_cn\n",
    "    print('Number of true negatives: ', n_neg_cn)\n",
    "    \n",
    "    print('#############################')\n",
    "    print('Stage 1 stats for TARGET vs NON-TARGET')\n",
    "    print('Precision for AD: ', n_pos_ad/(n_pos_ad + n_pos_cn))\n",
    "    prec = n_pos_ad/(n_pos_ad + n_pos_cn)\n",
    "    print('Recall (or sensitivity)  for AD: ', n_pos_ad/n_ad)\n",
    "    sens = n_pos_ad/n_ad\n",
    "    print('Specificity: ', n_neg1_cn/n_cn)\n",
    "    spec = n_neg1_cn/n_cn\n",
    "    fp = (1-spec)*664\n",
    "    tp = sens*336\n",
    "    adj_prec = tp/(tp+fp)\n",
    "    print('Adjusted precision for 33.6% baseline rate: ', adj_prec)\n",
    "    print('Accuracy: ', (n_pos_ad + n_neg1_cn)/(n_ad + n_cn))\n",
    "    acc = (n_pos_ad + n_neg1_cn)/(n_ad + n_cn)\n",
    "\n",
    "    print('#############################')\n",
    "    print('Stage 2 stats for TARGET vs NON-TARGET')\n",
    "    print('Precision for HPC-AD: ', n_flag_ad/n_flag)\n",
    "    prec_2 = n_flag_ad/n_flag\n",
    "    print('Recall (or sensitivity) for HPC-AD: ', n_flag_ad/n_ad)\n",
    "    sens_2 = n_flag_ad/n_ad\n",
    "    print('Specificity: ', n_neg_cn/n_cn)\n",
    "    spec_2 = n_neg_cn/n_cn\n",
    "    fp_2 = (1-spec_2)*664\n",
    "    tp_2 = sens_2*336\n",
    "    adj_prec_2 = tp_2/(tp_2 + fp_2)\n",
    "    print('Adjusted precision for 33.6% baseline rate: ', adj_prec_2)\n",
    "    print('Accuracy: ', (n_flag_ad + n_neg_cn)/(n_ad + n_cn))\n",
    "    acc_2 = (n_flag_ad + n_neg_cn)/(n_ad + n_cn)\n",
    "    \n",
    "    return sens, spec, prec, acc, sens_2, spec_2, prec_2, acc_2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set(font_scale=2)\n",
    "sns.set_style(\"white\")\n",
    "#sns.set_context(\"paper\")\n",
    "#sns.set_palette(\"colorblind\")\n",
    "#sns.set_palette(\"GnBu_d\")\n",
    "\n",
    "#sns.set_palette(sns.cubehelix_palette(n_colors=8))\n",
    "#sns.set_palette(sns.color_palette(\"BrBG\", 6))\n",
    "\n",
    "cpal = [\"#F0DFB2\", \"#CFA255\", \"#995D12\", \"#B3E2DB\", \"#58B0A6\", \"#0D7068\"]\n",
    "sns.set_palette(cpal)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "path_results = '/home/angela/Desktop/vcog_paper/gigascience/third_submission/roc/vbm/'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load longitudinal data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.seed(1)\n",
    "#np.random.RandomState(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/angela/anaconda3/envs/vcog_paper_py35/lib/python3.5/site-packages/IPython/core/interactiveshell.py:2785: DtypeWarning: Columns (101) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n"
     ]
    }
   ],
   "source": [
    "long_data = pd.read_csv('/home/angela/Documents/adni_csv/adnimerge_upenn_unw_av45_neurobat.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load data for the training set (ADNI1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('/home/angela/Desktop/vcog_paper/adni1_vbm_adcn_subtypes_20171209/7clus/adni1_model_weights.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.dropna(subset=['sub1','age_scan','gender','mean_gm','tiv'],\n",
    "            inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Organize the AD & CN data for classification task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Mask of the CN and AD subjects only\n",
    "mask_cnad = data.loc[:,['CN','AD']].values.sum(1).astype(bool)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((377, 11), (377,), (377, 4))"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#reload(high_confidence)\n",
    "scalerX = preprocessing.StandardScaler()\n",
    "scaler = preprocessing.StandardScaler()\n",
    "\n",
    "x_ = data.iloc[mask_cnad,data.columns.get_loc(\"sub1\"):data.columns.get_loc(\"sub7\")+1].values\n",
    "#x_ = scalerX.fit_transform(x_)\n",
    "y_ = data[['AD']].values.ravel()[mask_cnad]\n",
    "\n",
    "confounds = data[['gender','age_scan','mean_gm','tiv']].values[mask_cnad,:]\n",
    "#confounds = data[['sex','age_r']].values[mask_cnad,:]\n",
    "#confounds[:, 1:] = scaler.fit_transform(confounds[:, 1:])\n",
    "#confounds[:, 0] = preprocessing.binarize(confounds[:, 0].reshape(-1, 1), threshold=1)[:, 0]\n",
    "\n",
    "#crm = prediction.ConfoundsRm(confounds, x_)\n",
    "#x_ = crm.transform(confounds, x_)\n",
    "\n",
    "x_ = scaler.fit_transform(np.hstack((x_,confounds)))\n",
    "#x_ = np.hstack((x_,confounds))\n",
    "\n",
    "\n",
    "x_.shape, y_.shape, confounds.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Cross-validation HPC in ADNI1 AD & CN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stage 1\n",
      "Proba:\n",
      "[0.9957265  1.         1.         1.         1.         0.65254237\n",
      " 0.98380567 0.50746269 1.         1.         0.56818182 1.\n",
      " 1.         1.         0.93589744 1.         0.99574468 0.71814672\n",
      " 1.         1.         0.04918033 0.98333333 1.         0.88416988\n",
      " 0.608      0.92828685 1.         1.         1.         1.\n",
      " 0.996      0.99607843 1.         1.         1.         0.996139\n",
      " 0.85470085 0.31746032 0.76377953 1.         1.         0.69635628\n",
      " 1.         1.         1.         0.99607843 0.98297872 1.\n",
      " 1.         1.         1.         0.56097561 0.         0.61111111\n",
      " 1.         0.94466403 1.         0.99239544 0.98       1.\n",
      " 0.99596774 1.         0.93495935 1.         1.         0.99193548\n",
      " 1.         1.         0.96186441 0.85098039 1.         1.\n",
      " 1.         0.63559322 1.         0.99618321 1.         1.\n",
      " 1.         1.         0.94023904 1.         1.         0.89105058\n",
      " 0.99595142 1.         0.         1.         1.         0.77205882\n",
      " 1.         1.         0.01321586 1.         1.         0.96875\n",
      " 0.48015873 0.8828125  0.         1.         0.9057377  1.\n",
      " 1.         1.         0.57377049 1.         1.         0.53648069\n",
      " 1.         0.         0.         1.         0.78731343 0.98431373\n",
      " 0.12809917 0.98181818 0.         1.         0.97131148 0.99256506\n",
      " 0.73151751 1.         0.12195122 1.         0.71369295 1.\n",
      " 0.95217391 1.         1.         0.99603175 0.84063745 0.96590909\n",
      " 1.         1.         0.65560166 1.         0.94163424 1.\n",
      " 1.         1.         0.07364341 1.         1.         0.14007782\n",
      " 1.         0.84680851 1.         0.         0.         1.\n",
      " 0.92369478 1.         0.94676806 0.60869565 0.99616858 1.\n",
      " 0.94402985 1.         1.         1.         1.         0.72\n",
      " 0.83613445 0.         1.         1.         0.06273063 0.34920635\n",
      " 0.33962264 0.02380952 1.         1.         0.         1.\n",
      " 1.         1.         1.         1.         1.         0.95983936\n",
      " 0.33807829 1.         1.         0.99107143 0.         1.\n",
      " 1.         1.         1.         0.95850622 1.         0.\n",
      " 0.98333333 0.75502008 1.         1.         0.34412955 0.996\n",
      " 0.49795918 1.         1.         0.99206349 1.         0.\n",
      " 1.         0.95081967 0.99078341 1.         1.         0.53781513\n",
      " 0.30125523 1.         0.82       1.         1.         0.99598394\n",
      " 0.         1.         1.         0.36821705 1.         0.24701195\n",
      " 0.64453125 1.         0.00406504 1.         0.72177419 0.01581028\n",
      " 0.02996255 0.49797571 1.         0.03614458 1.         1.\n",
      " 0.6872428  0.08527132 1.         1.         1.         1.\n",
      " 1.         1.         0.91774892 0.05952381 0.97942387 1.\n",
      " 0.99156118 1.         0.06866953 0.01153846]\n",
      "Average hm score 0.516\n",
      "Stage 2\n",
      "Adjusted gamma:  1.0\n",
      "Adjusted gamma:  1.0\n",
      "Classifying AD vs CN...\n",
      "[0. 0. 1. 0. 1. 0. 0. 0. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0.\n",
      " 0. 0. 1. 0. 1. 0. 0. 1. 1. 1. 1. 0. 0. 0. 1. 1. 1. 0. 0. 0. 1. 1. 0. 0.\n",
      " 1. 0. 1. 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 1. 0. 0. 1. 1.\n",
      " 0. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1.\n",
      " 0. 0. 0. 1. 1. 1. 1. 0. 1. 0. 0. 1. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.\n",
      " 1. 1. 1. 1. 1. 1. 1.]\n",
      "Total number of TARGET subjects:  58.0\n",
      "Total number of NON-TARGET subjects:  69.0\n",
      "Stage 1 number of hits (true and false positives):  54.0\n",
      "Stage 1 TRUE positives:  49.0\n",
      "Stage 1 FALSE positives:  5.0\n",
      "Stage 1 TRUE negatives:  64.0\n",
      "Total number of flagged HPC-AD subjects:  35.0\n",
      "Number of flagged HPC-AD subjects that are TRUE positives:  33\n",
      "Number of flagged HPC-AD subjects that are FALSE positives:  2.0\n",
      "Number of true negatives:  67.0\n",
      "#############################\n",
      "Stage 1 stats for TARGET vs NON-TARGET\n",
      "Precision for AD:  0.9074074074074074\n",
      "Recall (or sensitivity)  for AD:  0.8448275862068966\n",
      "Specificity:  0.927536231884058\n",
      "Adjusted precision for 33.6% baseline rate:  0.8550628642998218\n",
      "Accuracy:  0.889763779527559\n",
      "#############################\n",
      "Stage 2 stats for TARGET vs NON-TARGET\n",
      "Precision for HPC-AD:  0.9428571428571428\n",
      "Recall (or sensitivity) for HPC-AD:  0.5689655172413793\n",
      "Specificity:  0.9710144927536232\n",
      "Adjusted precision for 33.6% baseline rate:  0.9085329938629324\n",
      "Accuracy:  0.7874015748031497\n",
      "Stage 1\n",
      "Proba:\n",
      "[0.89344262 1.         1.         1.         0.97692308 1.\n",
      " 1.         1.         0.         1.         0.16595745 0.9958159\n",
      " 0.97131148 0.99619772 1.         0.99206349 1.         1.\n",
      " 1.         0.96047431 1.         1.         0.99193548 1.\n",
      " 1.         1.         0.         1.         1.         0.\n",
      " 1.         0.47881356 1.         0.57429719 1.         1.\n",
      " 1.         0.13168724 1.         1.         0.30327869 1.\n",
      " 0.99621212 0.97142857 1.         1.         0.79475983 1.\n",
      " 0.         1.         1.         1.         1.         1.\n",
      " 0.00396825 1.         0.         0.83461538 0.4488189  0.36051502\n",
      " 1.         0.99619772 1.         1.         0.83529412 1.\n",
      " 1.         0.97285068 1.         1.         0.64092664 0.62645914\n",
      " 0.89772727 1.         0.5        0.81974249 1.         1.\n",
      " 1.         1.         0.98876404 1.         1.         1.\n",
      " 0.92622951 1.         0.04615385 0.66666667 1.         1.\n",
      " 1.         0.69402985 0.98380567 0.86343612 0.99212598 1.\n",
      " 1.         0.         1.         0.68060837 1.         1.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 0.         1.         1.         1.         1.         1.\n",
      " 1.         0.98832685 1.         1.         1.         0.99622642\n",
      " 1.         0.96414343 1.         1.         1.         0.94444444\n",
      " 1.         1.         0.06589147 0.99586777 0.96282528 1.\n",
      " 1.         1.         1.         1.         0.93927126 0.16393443\n",
      " 1.         0.95149254 0.99598394 0.94067797 1.         1.\n",
      " 0.98406375 1.         1.         1.         0.94537815 1.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 0.23420074 1.         0.87763713 1.         1.         0.88211382\n",
      " 0.93076923 1.         1.         1.         1.         0.54471545\n",
      " 0.01612903 0.32283465 0.31983806 0.00904977 1.         1.\n",
      " 0.         1.         0.99609375 1.         1.         1.\n",
      " 1.         0.83805668 0.1124498  1.         1.         1.\n",
      " 0.         1.         1.         1.         1.         0.97008547\n",
      " 1.         0.         1.         0.84814815 1.         1.\n",
      " 0.31538462 0.99618321 0.55947137 1.         1.         0.95884774\n",
      " 1.         0.00398406 1.         0.81960784 1.         1.\n",
      " 1.         0.28740157 0.79919679 1.         0.61445783 1.\n",
      " 1.         1.         0.         1.         1.         0.15352697\n",
      " 1.         0.79051383 0.66932271 1.         0.00403226 1.\n",
      " 0.67460317 0.00803213 0.08695652 0.46946565 1.         0.01893939\n",
      " 1.         0.99209486 0.42696629 0.1547619  1.         1.\n",
      " 1.         0.99588477 1.         1.         0.67068273 0.00403226\n",
      " 1.         1.         0.890625   1.         0.02553191 0.07083333]\n",
      "Average hm score 0.5912698412698413\n",
      "Stage 2\n",
      "Adjusted gamma:  1.0\n",
      "Adjusted gamma:  1.0\n",
      "Classifying AD vs CN...\n",
      "[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.\n",
      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 1. 0. 0. 1. 0. 1. 0. 0.\n",
      " 0. 1. 0. 0. 1. 1. 1. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 0. 0. 1. 1.\n",
      " 0. 1. 1. 0. 1. 0. 1. 1. 0. 1. 0. 1. 0. 1. 1. 0. 1. 1. 1. 1. 0. 1. 0. 1.\n",
      " 1. 1. 0. 1. 1. 0. 0. 1. 1. 0. 0. 0. 1. 1. 1. 1. 1. 0. 0. 0. 1. 1. 1. 1.\n",
      " 1. 1. 0. 1. 1.]\n",
      "Total number of TARGET subjects:  57.0\n",
      "Total number of NON-TARGET subjects:  68.0\n",
      "Stage 1 number of hits (true and false positives):  49.0\n",
      "Stage 1 TRUE positives:  44.0\n",
      "Stage 1 FALSE positives:  5.0\n",
      "Stage 1 TRUE negatives:  63.0\n",
      "Total number of flagged HPC-AD subjects:  32.0\n",
      "Number of flagged HPC-AD subjects that are TRUE positives:  30\n",
      "Number of flagged HPC-AD subjects that are FALSE positives:  2.0\n",
      "Number of true negatives:  66.0\n",
      "#############################\n",
      "Stage 1 stats for TARGET vs NON-TARGET\n",
      "Precision for AD:  0.8979591836734694\n",
      "Recall (or sensitivity)  for AD:  0.7719298245614035\n",
      "Specificity:  0.9264705882352942\n",
      "Adjusted precision for 33.6% baseline rate:  0.8415807767263376\n",
      "Accuracy:  0.856\n",
      "#############################\n",
      "Stage 2 stats for TARGET vs NON-TARGET\n",
      "Precision for HPC-AD:  0.9375\n",
      "Recall (or sensitivity) for HPC-AD:  0.5263157894736842\n",
      "Specificity:  0.9705882352941176\n",
      "Adjusted precision for 33.6% baseline rate:  0.9005486535914737\n",
      "Accuracy:  0.768\n",
      "Stage 1\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proba:\n",
      "[0.28688525 0.99603175 1.         1.         0.91153846 1.\n",
      " 0.99579832 0.86521739 0.0112782  1.         0.28085106 0.9874477\n",
      " 0.99590164 0.90114068 1.         0.8968254  1.         1.\n",
      " 0.98846154 1.         1.         0.94117647 0.93951613 1.\n",
      " 0.97482014 1.         0.         0.92857143 1.         0.\n",
      " 1.         0.8559322  1.         0.21285141 1.         1.\n",
      " 0.99206349 0.22222222 1.         1.         0.13114754 1.\n",
      " 1.         0.9755102  1.         1.         0.99563319 1.\n",
      " 0.         1.         1.         1.         0.98367347 1.\n",
      " 0.05555556 0.94849785 0.01298701 0.99615385 0.14173228 0.3776824\n",
      " 1.         0.40304183 1.         1.         0.90588235 1.\n",
      " 1.         1.         1.         1.         0.96911197 0.91828794\n",
      " 0.67424242 1.         0.64344262 0.89699571 1.         1.\n",
      " 1.         1.         0.99625468 1.         1.         1.\n",
      " 0.97540984 1.         0.26538462 0.58634538 1.         1.\n",
      " 1.         0.76865672 0.65991903 0.66079295 1.         0.98319328\n",
      " 1.         0.         1.         0.9391635  1.         1.\n",
      " 1.         1.         1.         0.88715953 1.         1.\n",
      " 1.         1.         0.         1.         1.         1.\n",
      " 1.         1.         1.         1.         1.         0.82462687\n",
      " 1.         0.75769231 1.         1.         0.96108949 1.\n",
      " 0.6147541  1.         0.99626866 1.         1.         0.96610169\n",
      " 0.99621212 1.         0.99595142 1.         0.86055777 1.\n",
      " 0.99622642 1.         0.03658537 1.         1.         0.97609562\n",
      " 1.         0.97254902 0.77642276 0.98387097 1.         1.\n",
      " 1.         1.         1.         0.8697318  0.99148936 1.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 0.83817427 1.         0.10638298 0.7944664  1.         0.99596774\n",
      " 0.91056911 1.         1.         0.98867925 1.         0.90566038\n",
      " 1.         1.         1.         1.         0.72373541 0.\n",
      " 0.75510204 1.         0.93133047 1.         0.99601594 0.99236641\n",
      " 0.99559471 1.         1.         0.99176955 1.         1.\n",
      " 0.97286822 1.         1.         0.91566265 0.92083333 1.\n",
      " 1.         1.         0.87916667 1.         0.92105263 1.\n",
      " 1.         1.         1.         0.93145161 1.         1.\n",
      " 0.85355649 0.99642857 1.         0.         1.         1.\n",
      " 0.94615385 0.94583333 1.         0.04780876 0.98431373 0.\n",
      " 1.         0.87692308 1.         0.25403226 0.16141732 1.\n",
      " 0.         0.         1.         1.         0.         0.56818182\n",
      " 0.80566802 1.         0.93632959 0.76190476 1.         1.\n",
      " 0.03149606 0.         0.00408163 1.         0.91967871 0.94354839\n",
      " 1.         1.         0.9140625  0.         1.         1.        ]\n",
      "Average hm score 0.5436507936507936\n",
      "Stage 2\n",
      "Adjusted gamma:  1.0\n",
      "Adjusted gamma:  1.0\n",
      "Classifying AD vs CN...\n",
      "[0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0.\n",
      " 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 1. 1. 1. 0.\n",
      " 0. 1. 0. 1. 0. 0. 1. 1. 1. 1. 0. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 0. 1.\n",
      " 0. 1. 0. 1. 0. 0. 0. 1. 1. 0. 0. 1. 0. 1. 0. 1. 1. 0. 1. 1. 1. 1. 0. 0.\n",
      " 0. 1. 1. 0. 0. 1. 0. 1. 1. 0. 0. 1. 1. 0. 0. 1. 1. 1. 1. 1. 1. 1. 0. 1.\n",
      " 1. 1. 1. 0. 0.]\n",
      "Total number of TARGET subjects:  57.0\n",
      "Total number of NON-TARGET subjects:  68.0\n",
      "Stage 1 number of hits (true and false positives):  56.0\n",
      "Stage 1 TRUE positives:  44.0\n",
      "Stage 1 FALSE positives:  12.0\n",
      "Stage 1 TRUE negatives:  56.0\n",
      "Total number of flagged HPC-AD subjects:  35.0\n",
      "Number of flagged HPC-AD subjects that are TRUE positives:  30\n",
      "Number of flagged HPC-AD subjects that are FALSE positives:  5.0\n",
      "Number of true negatives:  63.0\n",
      "#############################\n",
      "Stage 1 stats for TARGET vs NON-TARGET\n",
      "Precision for AD:  0.7857142857142857\n",
      "Recall (or sensitivity)  for AD:  0.7719298245614035\n",
      "Specificity:  0.8235294117647058\n",
      "Adjusted precision for 33.6% baseline rate:  0.6888114187989213\n",
      "Accuracy:  0.8\n",
      "#############################\n",
      "Stage 2 stats for TARGET vs NON-TARGET\n",
      "Precision for HPC-AD:  0.8571428571428571\n",
      "Recall (or sensitivity) for HPC-AD:  0.5263157894736842\n",
      "Specificity:  0.9264705882352942\n",
      "Adjusted precision for 33.6% baseline rate:  0.7836465907531899\n",
      "Accuracy:  0.744\n"
     ]
    }
   ],
   "source": [
    "scores_ad_cn=[]\n",
    "scores_s2 = []\n",
    "ad_precision = []\n",
    "cn_precision = []\n",
    "ad_recall = []\n",
    "cn_recall = []\n",
    "ad_f1_score = []\n",
    "cn_f1_score = []\n",
    "\n",
    "s1_spec = []\n",
    "s1_sens = []\n",
    "s1_prec = []\n",
    "s1_acc = []\n",
    "\n",
    "s2_spec = []\n",
    "s2_sens = []\n",
    "s2_prec = []\n",
    "s2_acc = []\n",
    "\n",
    "\n",
    "skf = StratifiedKFold(n_splits=3)\n",
    "for train_index, val_index in skf.split(x_,y_):\n",
    "    X_training, X_val = x_[train_index], x_[val_index]\n",
    "    y_training, y_val = y_[train_index], y_[val_index]\n",
    "    \n",
    "    hpc = high_confidence_at.TwoStagesPrediction(\n",
    "        n_iter=500,\n",
    "        shuffle_test_split=0.5,\n",
    "        min_gamma=.99,\n",
    "        thresh_ratio=0.1)\n",
    "    \n",
    "    hpc.fit(X_training, X_training, y_training)\n",
    "    \n",
    "    _, dic_results = hpc.predict(X_val, X_val)\n",
    "    \n",
    "    # test in validation sample\n",
    "    acc = metrics.accuracy_score(y_val, (dic_results['s1df'][:,0]>0).astype(float))\n",
    "    tmp_mask = (dic_results['s2df'][:,1]>0)\n",
    "    acc_s2 = metrics.accuracy_score(y_val[tmp_mask], (dic_results['s1df'][:,0]>0).astype(float)[tmp_mask])\n",
    "    scores_ad_cn.append(acc)\n",
    "    scores_s2.append(acc_s2)\n",
    "    print('Classifying AD vs CN...')\n",
    "    print((dic_results['s1df'][:,0]>0).astype(float))\n",
    "    \n",
    "    y_pred = (dic_results['s1df'][:,0]>0).astype(float)\n",
    "    lr_decision = dic_results['s2df'][:,1]\n",
    "    \n",
    "    # BASE SVM PERFORMANCE\n",
    "    ad_p = metrics.precision_score(y_val, y_pred)\n",
    "    ad_precision.append(ad_p)\n",
    "    cn_p = metrics.precision_score(y_val, y_pred, pos_label=0)\n",
    "    cn_precision.append(cn_p)\n",
    "    ad_r = metrics.recall_score(y_val, y_pred)\n",
    "    ad_recall.append(ad_r)\n",
    "    cn_r = metrics.recall_score(y_val, y_pred, pos_label=0)\n",
    "    cn_recall.append(cn_r)\n",
    "    ad_f1 = metrics.f1_score(y_val, y_pred)\n",
    "    ad_f1_score.append(ad_f1)\n",
    "    cn_f1 = metrics.f1_score(y_val, y_pred, pos_label=0)\n",
    "    cn_f1_score.append(cn_f1)\n",
    "    \n",
    "    \n",
    "    sens, spec, prec, acc, sens_2, spec_2, prec_2, acc_2 = predic_stats(y_val, y_pred, lr_decision)\n",
    "    s1_spec.append(spec)\n",
    "    s1_sens.append(sens)\n",
    "    s1_prec.append(prec)\n",
    "    s1_acc.append(acc)\n",
    "    s2_spec.append(spec_2)\n",
    "    s2_sens.append(sens_2)\n",
    "    s2_prec.append(prec_2)\n",
    "    s2_acc.append(acc_2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stage 1\n",
      "Mean sensitivity:  0.7962290784432344\n",
      "Mean specificity:  0.8925120772946858\n",
      "Mean precision:  0.8636936255983875\n",
      "Mean accuracy:  0.8485879265091864\n",
      "##########\n",
      "Stage 2\n",
      "Mean sensitivity:  0.5405323653962492\n",
      "Mean specificity:  0.9560244387610117\n",
      "Mean precision:  0.9125\n",
      "Mean accuracy:  0.7664671916010498\n"
     ]
    }
   ],
   "source": [
    "print('Stage 1')\n",
    "print('Mean sensitivity: ', np.mean(s1_sens))\n",
    "print('Mean specificity: ', np.mean(s1_spec))\n",
    "print('Mean precision: ', np.mean(s1_prec))\n",
    "print('Mean accuracy: ', np.mean(s1_acc))\n",
    "print('#'*10)\n",
    "print('Stage 2')\n",
    "print('Mean sensitivity: ', np.mean(s2_sens))\n",
    "print('Mean specificity: ', np.mean(s2_spec))\n",
    "print('Mean precision: ', np.mean(s2_prec))\n",
    "print('Mean accuracy: ', np.mean(s2_acc))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train HPC (on whole training set of ADNI1 AD & CN)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stage 1\n",
      "Proba:\n",
      "[0.648      1.         1.         1.         0.99212598 1.\n",
      " 1.         1.         0.         1.         0.26217228 1.\n",
      " 1.         0.99588477 1.         1.         1.         1.\n",
      " 1.         0.99628253 1.         1.         1.         1.\n",
      " 1.         1.         0.         1.         1.         0.\n",
      " 1.         0.7375     1.         0.54585153 1.         1.\n",
      " 1.         0.1259542  1.         1.         0.16862745 1.\n",
      " 1.         1.         1.         1.         0.97154472 1.\n",
      " 0.         1.         1.         1.         1.         1.\n",
      " 0.         1.         0.         0.99596774 0.29296875 0.3372093\n",
      " 1.         0.97021277 1.         1.         0.9280303  1.\n",
      " 1.         1.         1.         1.         0.87951807 0.856\n",
      " 0.89112903 1.         0.67901235 0.93233083 1.         1.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 0.9921875  1.         0.09745763 0.72962963 1.         1.\n",
      " 1.         0.79400749 0.95850622 0.91304348 1.         1.\n",
      " 1.         0.         1.         0.81818182 1.         1.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 1.         1.         0.         1.         1.         1.\n",
      " 1.         1.         1.         1.         1.         0.74678112\n",
      " 1.         0.79607843 1.         1.         0.99593496 1.\n",
      " 0.43534483 1.         1.         1.         1.         0.96603774\n",
      " 1.         1.         1.         1.         0.85483871 1.\n",
      " 1.         1.         0.01960784 1.         1.         0.99603175\n",
      " 1.         0.87209302 0.75510204 0.97744361 1.         1.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 0.84552846 1.         0.20564516 0.65587045 1.         1.\n",
      " 0.83461538 1.         1.         1.         1.         1.\n",
      " 1.         1.         1.         1.         0.50413223 0.\n",
      " 0.63424125 1.         0.98867925 1.         1.         0.99134199\n",
      " 1.         1.         1.         0.99607843 1.         1.\n",
      " 1.         1.         1.         0.96414343 0.98818898 1.\n",
      " 1.         1.         0.51405622 1.         0.99583333 1.\n",
      " 1.         1.         1.         0.96934866 1.         1.\n",
      " 0.83984375 1.         1.         0.         1.         1.\n",
      " 0.976      1.         1.         0.00826446 1.         1.\n",
      " 0.98425197 0.3619403  0.85258964 0.         1.         0.93145161\n",
      " 1.         1.         1.         0.3385214  1.         1.\n",
      " 0.51818182 1.         0.         0.         1.         0.96832579\n",
      " 1.         0.04958678 1.         0.         1.         1.\n",
      " 1.         0.86746988 1.         0.09311741 1.         0.49382716\n",
      " 1.         0.99212598 1.         1.         0.99578059 0.78688525\n",
      " 0.99245283 1.         1.         0.8245614  0.99606299 0.99264706\n",
      " 1.         1.         1.         0.         1.         1.\n",
      " 0.27799228 1.         0.96666667 1.         0.         0.\n",
      " 1.         0.98418972 1.         0.99616858 0.6798419  1.\n",
      " 1.         0.96484375 1.         1.         1.         1.\n",
      " 0.54661017 0.88256228 0.         1.         1.         0.\n",
      " 0.29296875 0.33744856 0.         1.         1.         0.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 0.93442623 0.22131148 1.         1.         1.         0.\n",
      " 1.         1.         1.         1.         1.         1.\n",
      " 0.         0.99224806 0.82113821 1.         1.         0.11788618\n",
      " 0.99607843 0.59848485 1.         1.         0.99606299 1.\n",
      " 0.         1.         0.95419847 1.         1.         1.\n",
      " 0.67636364 0.70916335 1.         0.85714286 1.         1.\n",
      " 1.         0.         1.         1.         0.42857143 1.\n",
      " 0.59656652 0.90039841 1.         0.         1.         0.87795276\n",
      " 0.00727273 0.01132075 0.45864662 1.         0.01239669 1.\n",
      " 1.         0.40377358 0.03292181 1.         1.         1.\n",
      " 1.         1.         1.         0.89767442 0.         1.\n",
      " 1.         0.9921875  1.         0.0334728  0.0125    ]\n",
      "Average hm score 0.6392572944297082\n",
      "Stage 2\n",
      "Adjusted gamma:  1.0\n",
      "Adjusted gamma:  1.0\n"
     ]
    }
   ],
   "source": [
    "#reload(high_confidence)\n",
    "hpc = high_confidence_at.TwoStagesPrediction(\n",
    "    n_iter=500,\n",
    "    shuffle_test_split=0.5,\n",
    "    min_gamma=.99,\n",
    "    thresh_ratio=0.1)\n",
    "\n",
    "hpc.fit(x_, x_, y_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Level 1\n",
      "------------------------\n",
      "Ratio: 1.0\n",
      "#    :  172.0\n",
      "# true values:  377\n",
      "ACC  :  0.8647214854111406\n",
      "Level 2\n",
      "------------------------\n",
      "Ratio: 0.622093023255814\n",
      "#    :  107.0\n",
      "# true values:  114\n",
      "ACC  :  0.9385964912280702\n"
     ]
    }
   ],
   "source": [
    "array_results, dic_results = hpc.predict(x_, x_)\n",
    "\n",
    "# Level 1\n",
    "print('Level 1')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float))\n",
    "\n",
    "print('Level 2')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float), dic_results['s2df'][:,1]>0)\n",
    "#stats_mask(dic_results['s2df'][:,2]>0)\n",
    "#stats_mask(dic_results['s2df'][:,3]>0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = (dic_results['s1df'][:,0]>0).astype(float)\n",
    "lr_decision = dic_results['s2df'][:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stage 1 stats for AD vs CN\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "        0.0       0.85      0.91      0.88       205\n",
      "        1.0       0.88      0.81      0.85       172\n",
      "\n",
      "avg / total       0.87      0.86      0.86       377\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print('Stage 1 stats for AD vs CN')\n",
    "print(metrics.classification_report(y_, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of TARGET subjects:  172.0\n",
      "Total number of NON-TARGET subjects:  205.0\n",
      "Stage 1 number of hits (true and false positives):  159.0\n",
      "Stage 1 TRUE positives:  140.0\n",
      "Stage 1 FALSE positives:  19.0\n",
      "Stage 1 TRUE negatives:  186.0\n",
      "Total number of flagged HPC-AD subjects:  114.0\n",
      "Number of flagged HPC-AD subjects that are TRUE positives:  107\n",
      "Number of flagged HPC-AD subjects that are FALSE positives:  7.0\n",
      "Number of true negatives:  198.0\n",
      "#############################\n",
      "Stage 1 stats for TARGET vs NON-TARGET\n",
      "Precision for AD:  0.8805031446540881\n",
      "Recall (or sensitivity)  for AD:  0.813953488372093\n",
      "Specificity:  0.9073170731707317\n",
      "Adjusted precision for 33.6% baseline rate:  0.8163104986713116\n",
      "Accuracy:  0.8647214854111406\n",
      "#############################\n",
      "Stage 2 stats for TARGET vs NON-TARGET\n",
      "Precision for HPC-AD:  0.9385964912280702\n",
      "Recall (or sensitivity) for HPC-AD:  0.622093023255814\n",
      "Specificity:  0.9658536585365853\n",
      "Adjusted precision for 33.6% baseline rate:  0.9021427690114199\n",
      "Accuracy:  0.8090185676392573\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.813953488372093,\n",
       " 0.9073170731707317,\n",
       " 0.8805031446540881,\n",
       " 0.8647214854111406,\n",
       " 0.622093023255814,\n",
       " 0.9658536585365853,\n",
       " 0.9385964912280702,\n",
       " 0.8090185676392573)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predic_stats(y_, y_pred, lr_decision)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# compare with other methods "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ROC curve"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### base"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(377,)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base = high_confidence_at.BaseSvc()\n",
    "base.fit(x_, y_)\n",
    "y_predicted = base.predict(x_)\n",
    "y_score = base.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_b = dict()\n",
    "tpr_b = dict()\n",
    "roc_auc_b = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_b[i], tpr_b[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_b[i] = auc(fpr_b[i], tpr_b[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_b[\"micro\"], tpr_b[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_b[\"micro\"] = auc(fpr_b[\"micro\"], tpr_b[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision-recall score: 0.91\n"
     ]
    }
   ],
   "source": [
    "average_precision_b = average_precision_score(y_true, y_score)\n",
    "\n",
    "print('Average precision-recall score: {0:0.2f}'.format(\n",
    "      average_precision_b))\n",
    "\n",
    "precision_b, recall_b, _ = precision_recall_curve(y_true, y_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### HPS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(lr_decision, (y_score.shape[0],1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_h = dict()\n",
    "tpr_h = dict()\n",
    "roc_auc_h = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_h[i], tpr_h[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_h[i] = auc(fpr_h[i], tpr_h[i])\n",
    "\n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_h[\"micro\"], tpr_h[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_h[\"micro\"] = auc(fpr_h[\"micro\"], tpr_h[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision-recall score: 0.91\n"
     ]
    }
   ],
   "source": [
    "average_precision_h = average_precision_score(y_true, y_score)\n",
    "\n",
    "print('Average precision-recall score: {0:0.2f}'.format(\n",
    "      average_precision_h))\n",
    "\n",
    "precision_h, recall_h, _ = precision_recall_curve(y_true, y_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### rbf kernel svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(377,)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svm_param_grid=dict(C=(np.logspace(-2, 1, 15)))\n",
    "\n",
    "clf_svm = SVC(kernel='rbf', class_weight='balanced', decision_function_shape='ovr', random_state=1)\n",
    "grclf_svm = GridSearchCV(clf_svm, param_grid=svm_param_grid, \n",
    "                         cv=StratifiedShuffleSplit(n_splits=50, test_size=.2, random_state=1))               \n",
    "\n",
    "grclf_svm.fit(x_, y_)\n",
    "y_predicted = grclf_svm.predict(x_)\n",
    "y_score = grclf_svm.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_svcrbf = dict()\n",
    "tpr_svcrbf = dict()\n",
    "roc_auc_svcrbf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_svcrbf[i], tpr_svcrbf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_svcrbf[i] = auc(fpr_svcrbf[i], tpr_svcrbf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_svcrbf[\"micro\"] = auc(fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### k nearest neighbours"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(377,)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "k_range = list(range(3,7))\n",
    "weight_opt = [\"uniform\", \"distance\"]\n",
    "knn_param_grid = dict(n_neighbors = k_range, weights = weight_opt)\n",
    "\n",
    "clf_knn = KNeighborsClassifier(algorithm='auto')\n",
    "grclf_knn = GridSearchCV(clf_knn, param_grid=knn_param_grid, \n",
    "                         cv=StratifiedShuffleSplit(n_splits=50, test_size=.2, random_state=1))\n",
    "grclf_knn.fit(x_, y_)\n",
    "y_predicted = grclf_knn.predict(x_)\n",
    "#y_score = clf.decision_function(x_)\n",
    "y_score = grclf_knn.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_knn = dict()\n",
    "tpr_knn = dict()\n",
    "roc_auc_knn = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_knn[i], tpr_knn[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_knn[i] = auc(fpr_knn[i], tpr_knn[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_knn[\"micro\"], tpr_knn[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_knn[\"micro\"] = auc(fpr_knn[\"micro\"], tpr_knn[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### random forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "n_features = x_.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(377,)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_param_grid = [\n",
    "{'n_estimators': [10, 25], 'max_features': [5, n_features], \n",
    " 'max_depth': [10, 50, None], 'bootstrap': [True, False]}\n",
    "]\n",
    "\n",
    "clf_rf = RandomForestClassifier(random_state=1)\n",
    "grclf_rf = GridSearchCV(clf_rf, param_grid=rf_param_grid, \n",
    "                       cv=StratifiedShuffleSplit(n_splits=50, test_size=.2, random_state=1))\n",
    "grclf_rf.fit(x_, y_)\n",
    "y_predicted = grclf_rf.predict(x_)\n",
    "#y_score = clf.decision_function(x_)\n",
    "y_score = grclf_rf.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_rf = dict()\n",
    "tpr_rf = dict()\n",
    "roc_auc_rf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_rf[i], tpr_rf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_rf[i] = auc(fpr_rf[i], tpr_rf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_rf[\"micro\"], tpr_rf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_rf[\"micro\"] = auc(fpr_rf[\"micro\"], tpr_rf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### gaussian naive bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(377,)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_gnb = GaussianNB()\n",
    "clf_gnb.fit(x_, y_)\n",
    "y_predicted = clf_gnb.predict(x_)\n",
    "#y_score = clf.decision_function(x_)\n",
    "y_score = clf_gnb.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_nb = dict()\n",
    "tpr_nb = dict()\n",
    "roc_auc_nb = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_nb[i], tpr_nb[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_nb[i] = auc(fpr_nb[i], tpr_nb[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_nb[\"micro\"], tpr_nb[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_nb[\"micro\"] = auc(fpr_nb[\"micro\"], tpr_nb[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plt.figure()\n",
    "fig, ax = plt.subplots()\n",
    "fig.set_size_inches(10,8)\n",
    "lw = 4\n",
    "plt.rc('xtick', labelsize=40)\n",
    "plt.rc('ytick', labelsize=40)\n",
    "plt.plot(fpr_svcrbf[0], tpr_svcrbf[0], color='green',\n",
    "         lw=lw, label='RBF SVM (AUC=%0.3f)' % roc_auc_svcrbf[0])\n",
    "plt.plot(fpr_knn[0], tpr_knn[0], color='pink',\n",
    "         lw=lw, label='KNN (AUC=%0.3f)' % roc_auc_knn[0])\n",
    "plt.plot(fpr_rf[0], tpr_rf[0], color='brown',\n",
    "         lw=lw, label='RF (AUC=%0.3f)' % roc_auc_rf[0])\n",
    "plt.plot(fpr_nb[0], tpr_nb[0], color='orange',\n",
    "         lw=lw, label='GNB (AUC=%0.3f)' % roc_auc_nb[0])\n",
    "plt.plot(fpr_b[0], tpr_b[0], color='blue',\n",
    "         lw=lw, label='Base (AUC=%0.3f)' % roc_auc_b[0])\n",
    "plt.plot(fpr_h[0], tpr_h[0], color='red',\n",
    "         lw=lw, label='HPS (AUC=%0.3f)' % roc_auc_h[0])\n",
    "plt.plot([0, 1], [0, 1], color='grey', lw=lw, linestyle='--')\n",
    "plt.xlim([-0.05, 1.00])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('FPR', fontdict={'size': 40})\n",
    "plt.ylabel('TPR', fontdict={'size': 40})\n",
    "plt.title('ADNI1 AD vs CN', fontdict={'size': 40})\n",
    "plt.legend(loc=\"lower right\", prop={'size': 25})\n",
    "plt.show()\n",
    "fig.savefig(path_results + 'adni1_ad_roc_multi.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Test on ADNI1 MCI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load the data\n",
    "adni1_mci = pd.read_csv('/home/angela/Desktop/vcog_paper/adni1_vbm_adcn_subtypes_20171209/7clus/adni1_mci_bl_demog_weights.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "mask_mci = adni1_mci.loc[:,'MCI'].values.astype(bool)\n",
    "adni1_mci = adni1_mci.iloc[mask_mci]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "adni1_mci.dropna(subset=['sub1','age_scan','gender','mean_gm','tiv'],inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "238"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(adni1_mci)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((238, 11), (238,), (238, 4))"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_ = adni1_mci.iloc[:,adni1_mci.columns.get_loc(\"sub1\"):adni1_mci.columns.get_loc(\"sub7\")+1].values\n",
    "#x_ = scalerX.transform(x_)\n",
    "y_ = adni1_mci['conv_2_ad'].values.ravel()\n",
    "\n",
    "\n",
    "confounds = adni1_mci[['gender','age_scan','mean_gm','tiv']].values\n",
    "#confounds = data[['sex','age_r']].values[mask_mci,:]\n",
    "#confounds[:, 1:] = scaler.transform(confounds[:, 1:])\n",
    "#confounds[:, 0] = preprocessing.binarize(confounds[:, 0].reshape(-1, 1), threshold=1)[:, 0]\n",
    "#confounds = scaler.transform(confounds)\n",
    "#x_ = crm.transform(confounds, x_)\n",
    "\n",
    "x_ = scaler.transform(np.hstack((x_,confounds)))\n",
    "\n",
    "x_.shape, y_.shape, confounds.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Level 1\n",
      "------------------------\n",
      "Ratio: 1.0\n",
      "#    :  149.0\n",
      "# true values:  238\n",
      "ACC  :  0.6890756302521008\n",
      "Level 2\n",
      "------------------------\n",
      "Ratio: 0.42953020134228187\n",
      "#    :  64.0\n",
      "# true values:  73\n",
      "ACC  :  0.8767123287671232\n"
     ]
    }
   ],
   "source": [
    "array_results, dic_results = hpc.predict(x_, x_)\n",
    "\n",
    "# Level 1\n",
    "print('Level 1')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float))\n",
    "\n",
    "print('Level 2')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float), dic_results['s2df'][:,1]>0)\n",
    "#stats_mask(dic_results['s2df'][:,2]>0)\n",
    "#stats_mask(dic_results['s2df'][:,3]>0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = (dic_results['s1df'][:,0]>0).astype(float)\n",
    "lr_decision = dic_results['s2df'][:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of TARGET subjects:  149.0\n",
      "Total number of NON-TARGET subjects:  89.0\n",
      "Stage 1 number of hits (true and false positives):  123.0\n",
      "Stage 1 TRUE positives:  99.0\n",
      "Stage 1 FALSE positives:  24.0\n",
      "Stage 1 TRUE negatives:  65.0\n",
      "Total number of flagged HPC-AD subjects:  73.0\n",
      "Number of flagged HPC-AD subjects that are TRUE positives:  64\n",
      "Number of flagged HPC-AD subjects that are FALSE positives:  9.0\n",
      "Number of true negatives:  80.0\n",
      "#############################\n",
      "Stage 1 stats for TARGET vs NON-TARGET\n",
      "Precision for AD:  0.8048780487804879\n",
      "Recall (or sensitivity)  for AD:  0.6644295302013423\n",
      "Specificity:  0.7303370786516854\n",
      "Adjusted precision for 33.6% baseline rate:  0.5549237482567817\n",
      "Accuracy:  0.6890756302521008\n",
      "#############################\n",
      "Stage 2 stats for TARGET vs NON-TARGET\n",
      "Precision for HPC-AD:  0.8767123287671232\n",
      "Recall (or sensitivity) for HPC-AD:  0.42953020134228187\n",
      "Specificity:  0.898876404494382\n",
      "Adjusted precision for 33.6% baseline rate:  0.6824767854850442\n",
      "Accuracy:  0.6050420168067226\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.6644295302013423,\n",
       " 0.7303370786516854,\n",
       " 0.8048780487804879,\n",
       " 0.6890756302521008,\n",
       " 0.42953020134228187,\n",
       " 0.898876404494382,\n",
       " 0.8767123287671232,\n",
       " 0.6050420168067226)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predic_stats(y_, y_pred, lr_decision)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ROC curve "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### base"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(238,)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#base = high_confidence_at.BaseSvc()\n",
    "#base.fit(x_, y_)\n",
    "y_predicted = base.predict(x_)\n",
    "y_score = base.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_b = dict()\n",
    "tpr_b = dict()\n",
    "roc_auc_b = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_b[i], tpr_b[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_b[i] = auc(fpr_b[i], tpr_b[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_b[\"micro\"], tpr_b[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_b[\"micro\"] = auc(fpr_b[\"micro\"], tpr_b[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision-recall score: 0.84\n"
     ]
    }
   ],
   "source": [
    "average_precision_b = average_precision_score(y_true, y_score)\n",
    "\n",
    "print('Average precision-recall score: {0:0.2f}'.format(\n",
    "      average_precision_b))\n",
    "\n",
    "precision_b, recall_b, _ = precision_recall_curve(y_true, y_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### HPS "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(lr_decision, (y_score.shape[0],1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_h = dict()\n",
    "tpr_h = dict()\n",
    "roc_auc_h = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_h[i], tpr_h[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_h[i] = auc(fpr_h[i], tpr_h[i])\n",
    "\n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_h[\"micro\"], tpr_h[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_h[\"micro\"] = auc(fpr_h[\"micro\"], tpr_h[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### rbf kernel svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(238,)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_svm.predict(x_)\n",
    "y_score = grclf_svm.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_svcrbf = dict()\n",
    "tpr_svcrbf = dict()\n",
    "roc_auc_svcrbf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_svcrbf[i], tpr_svcrbf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_svcrbf[i] = auc(fpr_svcrbf[i], tpr_svcrbf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_svcrbf[\"micro\"] = auc(fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### k nearest neighbours"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(238,)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_knn.predict(x_)\n",
    "y_score = grclf_knn.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_knn = dict()\n",
    "tpr_knn = dict()\n",
    "roc_auc_knn = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_knn[i], tpr_knn[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_knn[i] = auc(fpr_knn[i], tpr_knn[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_knn[\"micro\"], tpr_knn[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_knn[\"micro\"] = auc(fpr_knn[\"micro\"], tpr_knn[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### random forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(238,)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_rf.predict(x_)\n",
    "y_score = grclf_rf.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_rf = dict()\n",
    "tpr_rf = dict()\n",
    "roc_auc_rf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_rf[i], tpr_rf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_rf[i] = auc(fpr_rf[i], tpr_rf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_rf[\"micro\"], tpr_rf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_rf[\"micro\"] = auc(fpr_rf[\"micro\"], tpr_rf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### gaussian naive bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(238,)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = clf_gnb.predict(x_)\n",
    "y_score = clf_gnb.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_nb = dict()\n",
    "tpr_nb = dict()\n",
    "roc_auc_nb = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_nb[i], tpr_nb[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_nb[i] = auc(fpr_nb[i], tpr_nb[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_nb[\"micro\"], tpr_nb[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_nb[\"micro\"] = auc(fpr_nb[\"micro\"], tpr_nb[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plt.figure()\n",
    "fig, ax = plt.subplots()\n",
    "fig.set_size_inches(10,8)\n",
    "lw = 4\n",
    "plt.rc('xtick', labelsize=40)\n",
    "plt.rc('ytick', labelsize=40)\n",
    "plt.plot(fpr_svcrbf[0], tpr_svcrbf[0], color='green',\n",
    "         lw=lw, label='RBF SVM (AUC=%0.3f)' % roc_auc_svcrbf[0])\n",
    "plt.plot(fpr_knn[0], tpr_knn[0], color='pink',\n",
    "         lw=lw, label='KNN (AUC=%0.3f)' % roc_auc_knn[0])\n",
    "plt.plot(fpr_rf[0], tpr_rf[0], color='brown',\n",
    "         lw=lw, label='RF (AUC=%0.3f)' % roc_auc_rf[0])\n",
    "plt.plot(fpr_nb[0], tpr_nb[0], color='orange',\n",
    "         lw=lw, label='GNB (AUC=%0.3f)' % roc_auc_nb[0])\n",
    "plt.plot(fpr_b[0], tpr_b[0], color='blue',\n",
    "         lw=lw, label='Base (AUC=%0.3f)' % roc_auc_b[0])\n",
    "plt.plot(fpr_h[0], tpr_h[0], color='red',\n",
    "         lw=lw, label='HPS (AUC=%0.3f)' % roc_auc_h[0])\n",
    "plt.plot([0, 1], [0, 1], color='grey', lw=lw, linestyle='--')\n",
    "plt.xlim([-0.05, 1.00])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('FPR', fontdict={'size': 40})\n",
    "plt.ylabel('TPR', fontdict={'size': 40})\n",
    "plt.title('ADNI1 pMCI vs sMCI', fontdict={'size': 40})\n",
    "plt.legend(loc=\"lower right\", prop={'size': 25})\n",
    "#ax.legend(loc='center left', bbox_to_anchor=(1,0.5), prop={'size': 30})\n",
    "plt.show()\n",
    "fig.savefig(path_results + 'adni1_mci_roc_multi.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# REPLICATION IN ADNI2 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "adni2_df = pd.read_csv('/home/angela/Desktop/vcog_paper/adni1_vbm_adcn_subtypes_20171209/7clus/adni2_model_weights.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "adni2_df.drop(adni2_df[adni2_df.RID < 2000].index,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get rid of NaNs\n",
    "adni2_df.dropna(axis=0,how='any',subset=['sub1','gender','age_scan','mean_gm','tiv','conv_2_ad'],inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## AD vs CN "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((278, 11), (278,), (278, 4))"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Mask of the AD and CN subjects only\n",
    "mask_cnad = adni2_df.loc[:,['CN','AD']].values.sum(1).astype(bool)\n",
    "\n",
    "x_ = adni2_df.iloc[mask_cnad,adni2_df.columns.get_loc(\"sub1\"):adni2_df.columns.get_loc(\"sub7\")+1].values\n",
    "#x_ = scalerX.transform(x_)\n",
    "y_ = adni2_df[['AD']].values.ravel()[mask_cnad]\n",
    "\n",
    "\n",
    "confounds = adni2_df[['gender','age_scan','mean_gm','tiv']].values[mask_cnad,:]\n",
    "#confounds = data[['sex','age_r']].values[mask_mci,:]\n",
    "#confounds[:, 1:] = scaler.transform(confounds[:, 1:])\n",
    "#confounds[:, 0] = preprocessing.binarize(confounds[:, 0].reshape(-1, 1), threshold=1)[:, 0]\n",
    "#confounds = scaler.transform(confounds)\n",
    "#x_ = crm.transform(confounds, x_)\n",
    "\n",
    "x_ = scaler.transform(np.hstack((x_,confounds)))\n",
    "\n",
    "x_.shape, y_.shape, confounds.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Level 1\n",
      "------------------------\n",
      "Ratio: 1.0\n",
      "#    :  90.0\n",
      "# true values:  278\n",
      "ACC  :  0.7230215827338129\n",
      "Level 2\n",
      "------------------------\n",
      "Ratio: 0.3333333333333333\n",
      "#    :  30.0\n",
      "# true values:  46\n",
      "ACC  :  0.6521739130434783\n"
     ]
    }
   ],
   "source": [
    "array_results, dic_results = hpc.predict(x_, x_)\n",
    "\n",
    "# Level 1\n",
    "print('Level 1')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float))\n",
    "\n",
    "print('Level 2')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float), dic_results['s2df'][:,1]>0)\n",
    "#stats_mask(dic_results['s2df'][:,2]>0)\n",
    "#stats_mask(dic_results['s2df'][:,3]>0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = (dic_results['s1df'][:,0]>0).astype(float)\n",
    "lr_decision = dic_results['s2df'][:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of TARGET subjects:  90.0\n",
      "Total number of NON-TARGET subjects:  188.0\n",
      "Stage 1 number of hits (true and false positives):  89.0\n",
      "Stage 1 TRUE positives:  51.0\n",
      "Stage 1 FALSE positives:  38.0\n",
      "Stage 1 TRUE negatives:  150.0\n",
      "Total number of flagged HPC-AD subjects:  46.0\n",
      "Number of flagged HPC-AD subjects that are TRUE positives:  30\n",
      "Number of flagged HPC-AD subjects that are FALSE positives:  16.0\n",
      "Number of true negatives:  172.0\n",
      "#############################\n",
      "Stage 1 stats for TARGET vs NON-TARGET\n",
      "Precision for AD:  0.5730337078651685\n",
      "Recall (or sensitivity)  for AD:  0.5666666666666667\n",
      "Specificity:  0.7978723404255319\n",
      "Adjusted precision for 33.6% baseline rate:  0.5865450159928688\n",
      "Accuracy:  0.7230215827338129\n",
      "#############################\n",
      "Stage 2 stats for TARGET vs NON-TARGET\n",
      "Precision for HPC-AD:  0.6521739130434783\n",
      "Recall (or sensitivity) for HPC-AD:  0.3333333333333333\n",
      "Specificity:  0.9148936170212766\n",
      "Adjusted precision for 33.6% baseline rate:  0.6646464646464647\n",
      "Accuracy:  0.7266187050359713\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.5666666666666667,\n",
       " 0.7978723404255319,\n",
       " 0.5730337078651685,\n",
       " 0.7230215827338129,\n",
       " 0.3333333333333333,\n",
       " 0.9148936170212766,\n",
       " 0.6521739130434783,\n",
       " 0.7266187050359713)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predic_stats(y_, y_pred, lr_decision)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ROC curve"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### base"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(278,)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = base.predict(x_)\n",
    "y_score = base.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_b = dict()\n",
    "tpr_b = dict()\n",
    "roc_auc_b = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_b[i], tpr_b[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_b[i] = auc(fpr_b[i], tpr_b[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_b[\"micro\"], tpr_b[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_b[\"micro\"] = auc(fpr_b[\"micro\"], tpr_b[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision-recall score: 0.56\n"
     ]
    }
   ],
   "source": [
    "average_precision_b = average_precision_score(y_true, y_score)\n",
    "\n",
    "print('Average precision-recall score: {0:0.2f}'.format(\n",
    "      average_precision_b))\n",
    "\n",
    "precision_b, recall_b, _ = precision_recall_curve(y_true, y_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### HPS "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(lr_decision, (y_score.shape[0],1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_h = dict()\n",
    "tpr_h = dict()\n",
    "roc_auc_h = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_h[i], tpr_h[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_h[i] = auc(fpr_h[i], tpr_h[i])\n",
    "\n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_h[\"micro\"], tpr_h[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_h[\"micro\"] = auc(fpr_h[\"micro\"], tpr_h[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision-recall score: 0.52\n"
     ]
    }
   ],
   "source": [
    "average_precision_h = average_precision_score(y_true, y_score)\n",
    "\n",
    "print('Average precision-recall score: {0:0.2f}'.format(\n",
    "      average_precision_h))\n",
    "\n",
    "precision_h, recall_h, _ = precision_recall_curve(y_true, y_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### rbf kernel svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(278,)"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_svm.predict(x_)\n",
    "y_score = grclf_svm.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_svcrbf = dict()\n",
    "tpr_svcrbf = dict()\n",
    "roc_auc_svcrbf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_svcrbf[i], tpr_svcrbf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_svcrbf[i] = auc(fpr_svcrbf[i], tpr_svcrbf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_svcrbf[\"micro\"] = auc(fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### k nearest neighbours"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(278,)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_knn.predict(x_)\n",
    "y_score = grclf_knn.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_knn = dict()\n",
    "tpr_knn = dict()\n",
    "roc_auc_knn = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_knn[i], tpr_knn[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_knn[i] = auc(fpr_knn[i], tpr_knn[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_knn[\"micro\"], tpr_knn[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_knn[\"micro\"] = auc(fpr_knn[\"micro\"], tpr_knn[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### random forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(278,)"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_rf.predict(x_)\n",
    "y_score = grclf_rf.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_rf = dict()\n",
    "tpr_rf = dict()\n",
    "roc_auc_rf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_rf[i], tpr_rf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_rf[i] = auc(fpr_rf[i], tpr_rf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_rf[\"micro\"], tpr_rf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_rf[\"micro\"] = auc(fpr_rf[\"micro\"], tpr_rf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### gaussian naive bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(278,)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = clf_gnb.predict(x_)\n",
    "y_score = clf_gnb.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_nb = dict()\n",
    "tpr_nb = dict()\n",
    "roc_auc_nb = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_nb[i], tpr_nb[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_nb[i] = auc(fpr_nb[i], tpr_nb[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_nb[\"micro\"], tpr_nb[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_nb[\"micro\"] = auc(fpr_nb[\"micro\"], tpr_nb[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plt.figure()\n",
    "fig, ax = plt.subplots()\n",
    "fig.set_size_inches(10,8)\n",
    "lw = 4\n",
    "plt.rc('xtick', labelsize=40)\n",
    "plt.rc('ytick', labelsize=40)\n",
    "plt.plot(fpr_svcrbf[0], tpr_svcrbf[0], color='green',\n",
    "         lw=lw, label='RBF SVM (AUC=%0.3f)' % roc_auc_svcrbf[0])\n",
    "plt.plot(fpr_knn[0], tpr_knn[0], color='pink',\n",
    "         lw=lw, label='KNN (AUC=%0.3f)' % roc_auc_knn[0])\n",
    "plt.plot(fpr_rf[0], tpr_rf[0], color='brown',\n",
    "         lw=lw, label='RF (AUC=%0.3f)' % roc_auc_rf[0])\n",
    "plt.plot(fpr_nb[0], tpr_nb[0], color='orange',\n",
    "         lw=lw, label='GNB (AUC=%0.3f)' % roc_auc_nb[0])\n",
    "plt.plot(fpr_b[0], tpr_b[0], color='blue',\n",
    "         lw=lw, label='Base (AUC=%0.3f)' % roc_auc_b[0])\n",
    "plt.plot(fpr_h[0], tpr_h[0], color='red',\n",
    "         lw=lw, label='HPS (AUC=%0.3f)' % roc_auc_h[0])\n",
    "plt.plot([0, 1], [0, 1], color='grey', lw=lw, linestyle='--')\n",
    "plt.xlim([-0.05, 1.00])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('FPR', fontdict={'size': 40})\n",
    "plt.ylabel('TPR', fontdict={'size': 40})\n",
    "plt.title('ADNI2 AD vs CN', fontdict={'size': 40})\n",
    "plt.legend(loc=\"lower right\", prop={'size': 25})\n",
    "#ax.legend(loc='center left', bbox_to_anchor=(1,0.5), prop={'size': 30})\n",
    "plt.show()\n",
    "fig.savefig(path_results + 'adni2_ad_roc_multi.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MCI stable vs converters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "adni2_mci = pd.read_csv('/home/angela/Desktop/vcog_paper/adni1_vbm_adcn_subtypes_20171209/7clus/adni2_mci_bl_demog_weights.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "mask_mci = adni2_mci.loc[:,'MCI'].values.astype(bool)\n",
    "adni2_mci = adni2_mci.iloc[mask_mci]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "adni2_mci.dropna(subset=['sub1','age_scan','gender','mean_gm','tiv'],inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((237, 11), (237,), (237, 4))"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x_ = adni2_mci.iloc[:, adni2_mci.columns.get_loc(\"sub1\"):adni2_mci.columns.get_loc(\"sub7\")+1].values\n",
    "#x_ = scalerX.transform(x_)\n",
    "y_ = adni2_mci[['conv_2_ad']].values.ravel()\n",
    "\n",
    "\n",
    "confounds = adni2_mci[['gender','age_scan','mean_gm','tiv']].values\n",
    "#confounds = data[['sex','age_r']].values[mask_mci,:]\n",
    "#confounds[:, 1:] = scaler.transform(confounds[:, 1:])\n",
    "#confounds[:, 0] = preprocessing.binarize(confounds[:, 0].reshape(-1, 1), threshold=1)[:, 0]\n",
    "#confounds = scaler.transform(confounds)\n",
    "#x_ = crm.transform(confounds, x_)\n",
    "\n",
    "x_ = scaler.transform(np.hstack((x_,confounds)))\n",
    "\n",
    "x_.shape, y_.shape, confounds.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Level 1\n",
      "------------------------\n",
      "Ratio: 1.0\n",
      "#    :  55.0\n",
      "# true values:  237\n",
      "ACC  :  0.759493670886076\n",
      "Level 2\n",
      "------------------------\n",
      "Ratio: 0.4727272727272727\n",
      "#    :  26.0\n",
      "# true values:  44\n",
      "ACC  :  0.5909090909090909\n"
     ]
    }
   ],
   "source": [
    "array_results, dic_results = hpc.predict(x_, x_)\n",
    "\n",
    "# Level 1\n",
    "print('Level 1')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float))\n",
    "\n",
    "print('Level 2')\n",
    "stats_mask(y_, (dic_results['s1df'][:,0]>0).astype(float), dic_results['s2df'][:,1]>0)\n",
    "#stats_mask(dic_results['s2df'][:,2]>0)\n",
    "#stats_mask(dic_results['s2df'][:,3]>0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = (dic_results['s1df'][:,0]>0).astype(float)\n",
    "lr_decision = dic_results['s2df'][:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of TARGET subjects:  55.0\n",
      "Total number of NON-TARGET subjects:  182.0\n",
      "Stage 1 number of hits (true and false positives):  76.0\n",
      "Stage 1 TRUE positives:  37.0\n",
      "Stage 1 FALSE positives:  39.0\n",
      "Stage 1 TRUE negatives:  143.0\n",
      "Total number of flagged HPC-AD subjects:  44.0\n",
      "Number of flagged HPC-AD subjects that are TRUE positives:  26\n",
      "Number of flagged HPC-AD subjects that are FALSE positives:  18.0\n",
      "Number of true negatives:  164.0\n",
      "#############################\n",
      "Stage 1 stats for TARGET vs NON-TARGET\n",
      "Precision for AD:  0.4868421052631579\n",
      "Recall (or sensitivity)  for AD:  0.6727272727272727\n",
      "Specificity:  0.7857142857142857\n",
      "Adjusted precision for 33.6% baseline rate:  0.6136921384446137\n",
      "Accuracy:  0.759493670886076\n",
      "#############################\n",
      "Stage 2 stats for TARGET vs NON-TARGET\n",
      "Precision for HPC-AD:  0.5909090909090909\n",
      "Recall (or sensitivity) for HPC-AD:  0.4727272727272727\n",
      "Specificity:  0.9010989010989011\n",
      "Adjusted precision for 33.6% baseline rate:  0.7074905487088575\n",
      "Accuracy:  0.8016877637130801\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.6727272727272727,\n",
       " 0.7857142857142857,\n",
       " 0.4868421052631579,\n",
       " 0.759493670886076,\n",
       " 0.4727272727272727,\n",
       " 0.9010989010989011,\n",
       " 0.5909090909090909,\n",
       " 0.8016877637130801)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predic_stats(y_, y_pred, lr_decision)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ROC curve"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### base"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(237,)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = base.predict(x_)\n",
    "y_score = base.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_b = dict()\n",
    "tpr_b = dict()\n",
    "roc_auc_b = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_b[i], tpr_b[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_b[i] = auc(fpr_b[i], tpr_b[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_b[\"micro\"], tpr_b[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_b[\"micro\"] = auc(fpr_b[\"micro\"], tpr_b[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision-recall score: 0.54\n"
     ]
    }
   ],
   "source": [
    "average_precision_b = average_precision_score(y_true, y_score)\n",
    "\n",
    "print('Average precision-recall score: {0:0.2f}'.format(\n",
    "      average_precision_b))\n",
    "\n",
    "precision_b, recall_b, _ = precision_recall_curve(y_true, y_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### HPS "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(lr_decision, (y_score.shape[0],1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_h = dict()\n",
    "tpr_h = dict()\n",
    "roc_auc_h = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_h[i], tpr_h[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_h[i] = auc(fpr_h[i], tpr_h[i])\n",
    "\n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_h[\"micro\"], tpr_h[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_h[\"micro\"] = auc(fpr_h[\"micro\"], tpr_h[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average precision-recall score: 0.56\n"
     ]
    }
   ],
   "source": [
    "average_precision_h = average_precision_score(y_true, y_score)\n",
    "\n",
    "print('Average precision-recall score: {0:0.2f}'.format(\n",
    "      average_precision_h))\n",
    "\n",
    "precision_h, recall_h, _ = precision_recall_curve(y_true, y_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### rbf kernel svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(237,)"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_svm.predict(x_)\n",
    "y_score = grclf_svm.decision_function(x_)\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_svcrbf = dict()\n",
    "tpr_svcrbf = dict()\n",
    "roc_auc_svcrbf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_svcrbf[i], tpr_svcrbf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_svcrbf[i] = auc(fpr_svcrbf[i], tpr_svcrbf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_svcrbf[\"micro\"] = auc(fpr_svcrbf[\"micro\"], tpr_svcrbf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### k nearest neighbours"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(237,)"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_knn.predict(x_)\n",
    "y_score = grclf_knn.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_knn = dict()\n",
    "tpr_knn = dict()\n",
    "roc_auc_knn = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_knn[i], tpr_knn[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_knn[i] = auc(fpr_knn[i], tpr_knn[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_knn[\"micro\"], tpr_knn[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_knn[\"micro\"] = auc(fpr_knn[\"micro\"], tpr_knn[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### random forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(237,)"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = grclf_rf.predict(x_)\n",
    "y_score = grclf_rf.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_rf = dict()\n",
    "tpr_rf = dict()\n",
    "roc_auc_rf = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_rf[i], tpr_rf[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_rf[i] = auc(fpr_rf[i], tpr_rf[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_rf[\"micro\"], tpr_rf[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_rf[\"micro\"] = auc(fpr_rf[\"micro\"], tpr_rf[\"micro\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### gaussian naive bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(237,)"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_predicted = clf_gnb.predict(x_)\n",
    "y_score = clf_gnb.predict_proba(x_)\n",
    "y_score = y_score[:,1] # take positive class\n",
    "y_score.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_true = y_.astype(int)\n",
    "y_true = label_binarize(y_true, classes=[0, 1])\n",
    "n_classes = y_true.shape[1]\n",
    "y_score = np.reshape(y_score, (y_score.shape[0],1))\n",
    "\n",
    "# Compute ROC curve and ROC area for each class\n",
    "fpr_nb = dict()\n",
    "tpr_nb = dict()\n",
    "roc_auc_nb = dict()\n",
    "for i in range(n_classes):\n",
    "    fpr_nb[i], tpr_nb[i], _ = roc_curve(y_true[:, i], y_score[:, i])\n",
    "    roc_auc_nb[i] = auc(fpr_nb[i], tpr_nb[i])\n",
    "    \n",
    "# Compute micro-average ROC curve and ROC area\n",
    "fpr_nb[\"micro\"], tpr_nb[\"micro\"], _ = roc_curve(y_true.ravel(), y_score.ravel())\n",
    "roc_auc_nb[\"micro\"] = auc(fpr_nb[\"micro\"], tpr_nb[\"micro\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plt.figure()\n",
    "fig, ax = plt.subplots()\n",
    "fig.set_size_inches(10,8)\n",
    "lw = 4\n",
    "plt.rc('xtick', labelsize=40)\n",
    "plt.rc('ytick', labelsize=40)\n",
    "plt.plot(fpr_svcrbf[0], tpr_svcrbf[0], color='green',\n",
    "         lw=lw, label='RBF SVM (AUC=%0.3f)' % roc_auc_svcrbf[0])\n",
    "plt.plot(fpr_knn[0], tpr_knn[0], color='pink',\n",
    "         lw=lw, label='KNN (AUC=%0.3f)' % roc_auc_knn[0])\n",
    "plt.plot(fpr_rf[0], tpr_rf[0], color='brown',\n",
    "         lw=lw, label='RF (AUC=%0.3f)' % roc_auc_rf[0])\n",
    "plt.plot(fpr_nb[0], tpr_nb[0], color='orange',\n",
    "         lw=lw, label='GNB (AUC=%0.3f)' % roc_auc_nb[0])\n",
    "plt.plot(fpr_b[0], tpr_b[0], color='blue',\n",
    "         lw=lw, label='Base (AUC=%0.3f)' % roc_auc_b[0])\n",
    "plt.plot(fpr_h[0], tpr_h[0], color='red',\n",
    "         lw=lw, label='HPS (AUC=%0.3f)' % roc_auc_h[0])\n",
    "plt.plot([0, 1], [0, 1], color='grey', lw=lw, linestyle='--')\n",
    "plt.xlim([-0.05, 1.00])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('FPR', fontdict={'size': 40})\n",
    "plt.ylabel('TPR', fontdict={'size': 40})\n",
    "plt.title('ADNI2 pMCI vs sMCI', fontdict={'size': 40})\n",
    "plt.legend(loc=\"lower right\", prop={'size': 25})\n",
    "#ax.legend(loc='center left', bbox_to_anchor=(1,0.5), prop={'size': 30})\n",
    "plt.show()\n",
    "fig.savefig(path_results + 'adni2_mci_roc_multi.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [default]",
   "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": "3.5.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}