Notebooks/CSP_features_Classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#import scipy.sparse as sps\n",
    "from sklearn.linear_model import LogisticRegression\t\n",
    "#from keras.models import Sequential\n",
    "#from keras.layers import Dense\n",
    "import scipy.io\n",
    "import numpy as np\n",
    "from scipy import linalg as LA\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "#from keras import optimizers\n",
    "import xgboost as xgb\n",
    "#from sklearn.decomposition import PCA\n",
    "import matplotlib.pyplot as plt\n",
    "#from mlxtend.plotting import plot_decision_regions\n",
    "#from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "#from sklearn.cross_validation import train_test_split\n",
    "#from keras.models import Sequential\n",
    "#from keras import regularizers\n",
    "#from keras import losses\n",
    "#from keras import optimizers\n",
    "#from keras.layers.normalization import BatchNormalization\n",
    "#from keras.layers.core import Dense,Dropout,Activation,Lambda\n",
    "##from pyglmnet import GLM #change pwd to E:\\Brain Machine\\Thesis\\Potential datasets\\BCI Competition 2008 – Graz data set B\\pyglmnet-master\n",
    "#import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature extraction using CSP\n",
    "\n",
    "Sources:\n",
    "\n",
    "1.*Common Spatial Pattern revisited by Riemannian geometry*\n",
    "\n",
    "2.*Model based generalization analysis of common spatial pattern in brain computer interfaces*\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def covariance(X):\n",
    "    return np.dot(X,X.T)/np.trace(np.dot(X,X.T))\n",
    "\n",
    "def get_feat(data,sf):\n",
    "    return np.log(np.var(np.dot(sf.T,data.T),axis=1)) #check axis\n",
    "\n",
    "def get_spatial(sum_left,sum_right,J):\n",
    "    C=sum_right+sum_left\n",
    "    eigvals,eigvecs=LA.eig(C)\n",
    "#    sort_eigvals=np.sort(eigvals)[::-1]\n",
    "    diag_inv=np.zeros((C.shape[1],C.shape[1]))\n",
    "    for i in range(eigvals.shape[0]):\n",
    "        diag_inv[i,i]=(1/np.abs(eigvals[i].real)) #considering absolute value of the real parts! need to verify if approach is correct  \n",
    "\n",
    "    P=np.sqrt(diag_inv)*eigvecs.T\n",
    "    S_l=P*sum_left*P.T\n",
    "    S_r=P*sum_right*P.T   \n",
    "\n",
    "    E1,U1=LA.eig(S_l,S_r)\n",
    "    ord1 = np.argsort(E1)\n",
    "    ord1 = ord1[::-1]\n",
    "    E1 = E1[ord1]\n",
    "    U1 = U1[:,ord1]\n",
    "    W=np.dot(U1,P)#projection matrix\n",
    "#consider the first 10 columns of W as the required feautres \n",
    "#ideally you want to pick the first three features and the last three\n",
    "    W_select=np.zeros([np.shape(W)[0],2*J])\n",
    "    W_select[:,0:J]=W[:,0:J]\n",
    "    W_select[:,J::]=W[:,np.shape(W)[1]-J:np.shape(W)[1]]\n",
    "\n",
    "    return W_select"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def lda(feat_train,labels_train,feat_test):\n",
    "#     print(\"\\nLinear Discriminant Analysis\")\n",
    "    clf=LinearDiscriminantAnalysis()\n",
    "    clf.fit(feat_train,labels_train)\n",
    "    pred_train=clf.predict(feat_train)\n",
    "    pred_test=clf.predict(feat_test)\n",
    "    return pred_train,pred_test    \n",
    "\n",
    "def neural_net(feat_train,labels_train,feat_test,neuron):\n",
    "#     print(\"\\nSingle layered Feedforward neural network with %d neurons\"%neuron)\n",
    "    model=Sequential()\n",
    "    model.add(Dense(neuron,input_dim=np.shape(feat_train)[1],\n",
    "                activation='sigmoid',kernel_initializer='uniform'))\n",
    "    #model.add(Dropout(0.05))\n",
    "    #model.add(BatchNormalization())\n",
    "    #model.add(Dense(100,        \n",
    "    #                activation='sigmoid',kernel_initializer='uniform'))\n",
    "    ##model.add(BatchNormalization())\n",
    "    #model.add(Dense(90,activation='sigmoid',kernel_initializer='uniform'))\n",
    "#    model.add(Dense(20,activation='sigmoid',kernel_initializer='uniform'))\n",
    "    model.add(Dense(1,activation='sigmoid',kernel_initializer='uniform'))\n",
    "    #optim=optimizers.SGD(lr=0.0001)\n",
    "    model.compile(loss='binary_crossentropy',optimizer=optimizers.adam(lr=0.0001),metrics=['accuracy'])\n",
    "    model.fit(feat_train,labels_train , epochs=500,verbose=0)\n",
    "\n",
    "    pred_train=np.round(model.predict(feat_train))\n",
    "    \n",
    "    pred_test=np.round(model.predict(feat_test))\n",
    "    return pred_train,pred_test\n",
    "\n",
    "\n",
    "#train_error=np.mean(labels_train!=np.array(pred_train))\n",
    "#print(\"train error:\",train_error)\n",
    "\n",
    "def KNN(feat_train,y_train,feat_test,neighbor):\n",
    "#     print(\"\\n%d-Nearest neighbor\" % neighbor)\n",
    "    neigh = KNeighborsClassifier(n_neighbors=neighbor)\n",
    "    neigh.fit(feat_train, y_train)\n",
    "    pred_train=neigh.predict(feat_train)\n",
    "    pred_test=neigh.predict(feat_test)\n",
    "    return pred_train,pred_test\n",
    "\n",
    "def log_reg(feat_train,y_train,feat_test):\n",
    "#     print(\"\\nLogistic regression\")\n",
    "    clf=LogisticRegression(random_state=0,solver='liblinear',multi_class='ovr').fit(feat_train,y_train)\n",
    "    pred_train=clf.predict(feat_train)  \n",
    "    pred_test=clf.predict(feat_test)    \n",
    "    return pred_train,pred_test\n",
    "\n",
    "def boost(feat_train,y_train,feat_test,depth):\n",
    "#     print(\"\\nXGBOOST max_depth:\",depth)\n",
    "#    clf=xgb.XGBClassifier(max_depth,min_child_weight=1,n_estimators=1000)\n",
    "#    dtrain=xgb.DMatrix(features_train,labels_train)\n",
    "    clf=xgb.XGBClassifier(depth=depth,\n",
    "                seed= 0, #for reproducibility\n",
    "                silent= 1,\n",
    "                learning_rate= 0.05,\n",
    "                n_estimators= 500)\n",
    "    clf.fit(feat_train,y_train,verbose=False)\n",
    "    pred_train=clf.predict(feat_train)\n",
    "    pred_test=clf.predict(feat_test)\n",
    "    return pred_train,pred_test\n",
    "\n",
    "def random_forest(feat_train,labels_train,feat_test,depth):\n",
    "#     print(\"\\nRandom forest with max depth:\",depth)\n",
    "\n",
    "    model=RandomForestClassifier(\n",
    "          max_depth=depth,random_state=0,n_estimators=1500)\n",
    "    #      min_samples_leaf=10,\n",
    "    #      min_weight_fraction_leaf= 0.4,n_estimators= 5000)\n",
    "    model.fit(feat_train,labels_train)\n",
    "    pred_train=model.predict(feat_train)\n",
    "    pred_test=model.predict(feat_test)\n",
    "    return pred_train,pred_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training Features Computed\n",
      " Computing Testing features...\n",
      "Features for training and Testing computed\n"
     ]
    }
   ],
   "source": [
    "num_sub=14\n",
    "features_test=dict()\n",
    "features_train=dict()\n",
    "labels_train=dict()\n",
    "labels_test=dict()\n",
    "num_channels=15\n",
    "sum_hand=np.zeros((num_channels,num_channels))\n",
    "sum_leg=np.zeros((num_channels,num_channels))\n",
    "labels=[]\n",
    "spatial_filt=dict()\n",
    "for sub in range(num_sub):\n",
    "    features_train[sub]=[]\n",
    "    labels_train[sub]=[]\n",
    "    if (sub+1)<10:    \n",
    "        file_train='E:\\\\Brain Machine\\\\Thesis\\\\Potential datasets\\\\New folder\\\\S0%dT.mat'% (sub+1)\n",
    "        mat_train=scipy.io.loadmat(file_train)\n",
    "    else:\n",
    "        file_train='E:\\\\Brain Machine\\\\Thesis\\\\Potential datasets\\\\New folder\\\\S%dT.mat'% (sub+1)\n",
    "        mat_train=scipy.io.loadmat(file_train)\n",
    "    data_train=mat_train['data']\n",
    "    for k in range(5):\n",
    "        cell_train=data_train[0][k]\n",
    "        X_train=cell_train[0][0][0]\n",
    "        time_train=cell_train[0][0][1]\n",
    "        labels_tmp_train=cell_train[0][0][2]\n",
    "        labels_train[sub].extend(labels_tmp_train[0])\n",
    "        var=0\n",
    "        for l_tmp in range(len(labels_tmp_train[0])):\n",
    "            if labels_tmp_train[0][l_tmp]==1:\n",
    "#                train_leg[sub].append\n",
    "                sum_hand+=covariance(X_train[var:time_train[0][l_tmp],:].T) #transpose because we need num_channel vs num_channel\n",
    "            else:\n",
    "                sum_leg+=covariance(X_train[var:time_train[0][l_tmp],:].T)\n",
    "            var=time_train[0][l_tmp]\n",
    "        mean_hand=sum_hand/num_sub\n",
    "        mean_leg=sum_leg/num_sub\n",
    "        num_feat=3 #times 2\n",
    "        spatial_filt[sub]=get_spatial(mean_hand,mean_leg,num_feat)\n",
    "        \n",
    "        #Computing Spatial features for training\n",
    "        var=0\n",
    "        for count in range(len(time_train[0])):\n",
    "            features_train[sub].append(get_feat(X_train[var:time_train[0][count],:],spatial_filt[sub]))\n",
    "            var=time_train[0][count]    \n",
    "    features_train[sub]=np.array(features_train[sub]) # to introduce randomness shuffle all the features\n",
    "    np.random.shuffle(features_train[sub])\n",
    "print(\"Training Features Computed\\n Computing Testing features...\")\n",
    "for sub in range(num_sub): \n",
    "    features_test[sub]=[]\n",
    "    labels_test[sub]=[]\n",
    "    if (sub+1)<10:\n",
    "        file_test='E:\\\\Brain Machine\\\\Thesis\\\\Potential datasets\\\\New folder\\\\S0%dE.mat'% (sub+1)\n",
    "        mat_test=scipy.io.loadmat(file_test)\n",
    "    else:\n",
    "        file_test='E:\\\\Brain Machine\\\\Thesis\\\\Potential datasets\\\\New folder\\\\S%dE.mat'% (sub+1)\n",
    "        mat_test=scipy.io.loadmat(file_test)\n",
    "#    print(\"Mat file read\")\n",
    "    data_test=mat_test['data']\n",
    "    for k in range(3):    \n",
    "        cell_test=data_test[0][k]\n",
    "        X_test=cell_test[0][0][0]\n",
    "        time_test=cell_test[0][0][1]\n",
    "        labels_tmp_test=cell_test[0][0][2]\n",
    "        labels_test[sub].extend(labels_tmp_test[0])    \n",
    "        #Computing Spatial features for testing\n",
    "        var=0\n",
    "        for count in range(len(time_test[0])):\n",
    "            features_test[sub].append(get_feat(X_test[var:time_test[0][count],:],spatial_filt[sub]))\n",
    "            var=time_test[0][count]   \n",
    "    features_test[sub]=np.array(features_test[sub])\n",
    "    np.random.shuffle(features_test[sub])\n",
    "print(\"Features for training and Testing computed\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Classifiers Assemble! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Subject 1 classified\n",
      "Subject 2 classified\n",
      "Subject 3 classified\n",
      "Subject 4 classified\n",
      "Subject 5 classified\n",
      "Subject 6 classified\n",
      "Subject 7 classified\n",
      "Subject 8 classified\n",
      "Subject 9 classified\n",
      "Subject 10 classified\n",
      "Subject 11 classified\n",
      "Subject 12 classified\n",
      "Subject 13 classified\n",
      "Subject 14 classified\n"
     ]
    }
   ],
   "source": [
    "''' Important to remember that each classifier is trained on the individual subjects separately.\n",
    "This means that we are training 14(i.e. number of subjects) models\n",
    "'''\n",
    "#\n",
    "rando_train_err=[]\n",
    "rando_test_err=[]\n",
    "knn_train_err=[]\n",
    "knn_test_err=[]\n",
    "reg_train_err=[]\n",
    "reg_test_err=[]\n",
    "xgb_train_err=[]\n",
    "xgb_test_err=[]\n",
    "for sub in range(num_sub):\n",
    "    xtrain=features_train[sub]\n",
    "    ytrain=labels_train[sub]\n",
    "    xtest=features_test[sub]\n",
    "    ytest=labels_test[sub]\n",
    "    \n",
    "    xgb_tr,xgb_tst=boost(xtrain,ytrain,xtest,depth=2)\n",
    "    xgb_train_err.append(np.mean(ytrain!=xgb_tr))\n",
    "    xgb_test_err.append(np.mean(ytest!=xgb_tst))\n",
    "    \n",
    "    rando_tr,rando_tst=random_forest(xtrain,ytrain,xtest,depth=2)\n",
    "    rando_train_err.append(np.mean(ytrain!=rando_tr))\n",
    "    rando_test_err.append(np.mean(ytest!=rando_tst))\n",
    "    \n",
    "    knn_tr,knn_tst=KNN(xtrain,ytrain,xtest,neighbor=4)\n",
    "    knn_train_err.append(np.mean(ytrain!=knn_tr))\n",
    "    knn_test_err.append(np.mean(ytest!=knn_tst))\n",
    "    \n",
    "    reg_tr,reg_tst=log_reg(xtrain,ytrain,xtest)\n",
    "    reg_train_err.append(np.mean(ytrain!=reg_tr))\n",
    "    reg_test_err.append(np.mean(ytest!=reg_tst))\n",
    "    \n",
    "    print(\"Subject %d classified\"%(sub+1))\n",
    " "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEICAYAAACzliQjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEICAYAAACzliQjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nroB40XfOuQLiRd855wqIF33nnCsg/wWDIopdoe4FfAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEICAYAAACktLTqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCFJyFxGJICV3EZEI+v93yYSxV1wbeAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(rando_train_err,label=\"Train\")\n",
    "plt.plot(rando_test_err,label=\"Test\")\n",
    "plt.title(\"Random Forest\")\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.plot(knn_train_err,label=\"Train\")\n",
    "plt.plot(knn_test_err,label=\"Test\")\n",
    "plt.title(\"4 nearest neighbors\")\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.plot(reg_train_err,label=\"Train\")\n",
    "plt.plot(reg_test_err,label=\"Test\")\n",
    "plt.title(\"Logistic regression\")\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.plot(xgb_train_err,label=\"Train\")\n",
    "plt.plot(xgb_test_err,label=\"Test\")\n",
    "plt.title(\"XGBoost\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}