ML4DM Lab8 - (LDA, Decision Tree, Random Forest).ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"ML4DM Lab7 - (LDA, Decision Tree, Random Forest).ipynb","provenance":[],"collapsed_sections":["EVAMj5D_zcI9","HxXOtiOrzfs3"]},"kernelspec":{"name":"python3","display_name":"Python 3"}},"cells":[{"cell_type":"markdown","metadata":{"id":"EVAMj5D_zcI9","colab_type":"text"},"source":["#Importing dataset"]},{"cell_type":"code","metadata":{"id":"UxR-zTV-QSWG","colab_type":"code","outputId":"3a557570-1497-43c8-baaf-3378517e4e29","executionInfo":{"status":"ok","timestamp":1573647146292,"user_tz":-330,"elapsed":4327,"user":{"displayName":"201601422 Daiict","photoUrl":"","userId":"06333073350363496926"}},"colab":{"base_uri":"https://localhost:8080/","height":35}},"source":["from google.colab import drive\n","drive.mount('/content/drive')\n","\n","import scipy.io\n","mat = scipy.io.loadmat('/content/drive/My Drive/Colab Notebooks/Machine Learning For Data Mining/EEG Data/S3.mat')\n"],"execution_count":0,"outputs":[{"output_type":"stream","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"],"name":"stdout"}]},{"cell_type":"markdown","metadata":{"id":"HxXOtiOrzfs3","colab_type":"text"},"source":["# Data Preprocessing"]},{"cell_type":"code","metadata":{"id":"XHzWSR2XSuMj","colab_type":"code","colab":{}},"source":["print(len(mat))\n","print(mat)"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"nuMhO6vPQzrg","colab_type":"code","colab":{}},"source":["import numpy as np \n","import pandas as pd \n","import numpy as np\n","import scipy.io\n","from sklearn.decomposition import PCA\n","from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n","from sklearn.tree import DecisionTreeClassifier\n","from sklearn.model_selection import train_test_split\n","from sklearn.ensemble import RandomForestClassifier\n","import matplotlib.pyplot as plt\n","import os\n","from sklearn.preprocessing import StandardScaler"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"Oq-h1_nSUY_g","colab_type":"code","colab":{}},"source":["df=pd.DataFrame.from_dict(mat,orient='index')\n","\n","print(df.head())\n","\n","data=df[0].X_3D\n","print(\"------------------\")\n","print(data)\n","print(\"------------------\")\n","\n","data=data.transpose(1,2,0)\n","\n","print(\"------------------\")\n","print(data[0])\n","print(\"------------------\")\n","\n","label=df[0].categoryLabels\n","\n","y=np.transpose(label)\n","\n","y=y-1\n","\n","print(\"------------------\")\n","print(\"y: \",y.shape)\n","print(y[0])\n","print(\"------------------\")\n","\n","print(len(data))"],"execution_count":0,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"HQf62mCFzn5n","colab_type":"text"},"source":["# Classify using LDA, RF, DT "]},{"cell_type":"code","metadata":{"id":"S3ma56viVdHa","colab_type":"code","colab":{}},"source":["def classify(input_data,i):\n","  #print(\"-----------------------------------------------------------------------------------------------------------------------\")\n","  #print(\"TIME SAMPLE \",i)\n","  #print()\n","  #print(\"initial shape of data: \",input_data.shape)\n","\n","  x = StandardScaler().fit_transform(input_data)\n","\n","  pca = PCA(n_components=0.95)\n","  principalComponents = pca.fit_transform(x)\n","\n","  #print(\"Shape of data after PCA: \",principalComponents.shape)\n","  '''\n","  print(\"PCA explained varience ratio: \",pca.explained_variance_ratio_)\n","  plt.plot(pca.explained_variance_ratio_)\n","  plt.title('PCA Explained Variance Ratio')\n","  plt.grid(True)\n","  plt.show()\n","  '''\n","\n","  X_train, X_test, y_train, y_test = train_test_split(principalComponents, y, test_size=0.10)\n","  X_train=( X_train - (X_train.mean()) ) / X_train.std()\n","  X_test=( X_test - (X_test.mean()) ) / X_test.std()\n","\n","\n","  lda = LinearDiscriminantAnalysis()\n","  lda_components = lda.fit_transform(X_train,y_train.ravel())\n","\n","  '''\n","  print(\"Shape of data after LDA: \",lda_components.shape)\n","  print(\"LDA explained variance ratio: \",lda.explained_variance_ratio_)\n","  plt.plot(pca.explained_variance_ratio_)\n","  plt.title('LDA Explained Variance Ratio')\n","  plt.grid(True)\n","  plt.show()\n","\n","  plt.scatter(lda_components[:,0],lda_components[:,1])\n","  plt.title('Scatter Plot of 1st and 2nd LDA Components')\n","  plt.grid(True)\n","  plt.show()\n","  '''\n","\n","  lda_mean_accuracy = lda.score(X_test,y_test)\n","  #print(\"Mean Accuracy of LDA: \",lda_mean_accuracy)\n","\n","  dtc = DecisionTreeClassifier(random_state=0)\n","  dtc_components = dtc.fit(X_train,y_train.ravel())\n","\n","  '''\n","  print(\"DTC feature importances: \",dtc.feature_importances_)\n","  plt.plot(dtc.feature_importances_)\n","  plt.title('DTC Feature Importance')\n","  plt.grid(True)\n","  plt.show()\n","  '''\n","  dtc_mean_accuracy = dtc.score(X_test,y_test)\n","  #print(\"Mean Accuracy of DTC: \", dtc_mean_accuracy)\n","\n","  rfc = RandomForestClassifier(n_estimators=150, max_depth=15, random_state=0)\n","  rfc_components = rfc.fit(X_train,y_train.ravel())\n","\n","  '''\n","  print(\"RFC feature importances: \",rfc.feature_importances_)\n","  plt.plot(rfc.feature_importances_)\n","  plt.title('RFC Feature Importance')\n","  plt.grid(True)\n","  plt.show()\n","  '''\n","\n","  rfc_mean_accuracy = rfc.score(X_test,y_test)\n","  #print(\"Mean Accuracy of RFC: \", rfc_mean_accuracy)\n","  \n","  print(\"-----------------------------------------------------------------------------------------------------------------------\")\n","  print(\"Time \",i,\": \",\"LDA: \",lda_mean_accuracy,\"  \",\"DTC: \",dtc_mean_accuracy,\"  \",\"RFC: \",rfc_mean_accuracy)\n","\n","  return lda_mean_accuracy, dtc_mean_accuracy, rfc_mean_accuracy"],"execution_count":0,"outputs":[]},{"cell_type":"code","metadata":{"id":"XOHdVN6iwzrC","colab_type":"code","outputId":"35fabd62-5888-483c-a3c5-b6c84264e3eb","executionInfo":{"status":"ok","timestamp":1573649324799,"user_tz":-330,"elapsed":173961,"user":{"displayName":"201601422 Daiict","photoUrl":"","userId":"06333073350363496926"}},"colab":{"base_uri":"https://localhost:8080/","height":1000}},"source":["lda=np.zeros([32])\n","dt=np.zeros([32])\n","rf=np.zeros([32])\n","\n","print(\"Table of Mean Accuracy\")\n","for i in range(0,32):\n","  lda[i],dt[i],rf[i] = classify(data[i],i)"],"execution_count":0,"outputs":[{"output_type":"stream","text":["Table of Mean Accuracy\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  0 :  LDA:  0.17533718689788053    DTC:  0.1464354527938343    RFC:  0.1811175337186898\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  1 :  LDA:  0.16377649325626203    DTC:  0.17533718689788053    RFC:  0.1676300578034682\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  2 :  LDA:  0.1579961464354528    DTC:  0.13872832369942195    RFC:  0.17148362235067438\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  3 :  LDA:  0.15606936416184972    DTC:  0.16377649325626203    RFC:  0.18497109826589594\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  4 :  LDA:  0.16955684007707128    DTC:  0.17341040462427745    RFC:  0.17148362235067438\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  5 :  LDA:  0.1599229287090559    DTC:  0.15028901734104047    RFC:  0.1791907514450867\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  6 :  LDA:  0.18882466281310212    DTC:  0.14836223506743737    RFC:  0.1676300578034682\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  7 :  LDA:  0.18497109826589594    DTC:  0.1579961464354528    RFC:  0.18689788053949905\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  8 :  LDA:  0.21579961464354527    DTC:  0.1811175337186898    RFC:  0.19653179190751446\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  9 :  LDA:  0.21965317919075145    DTC:  0.17148362235067438    RFC:  0.20616570327552985\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  10 :  LDA:  0.279383429672447    DTC:  0.19653179190751446    RFC:  0.27167630057803466\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  11 :  LDA:  0.28901734104046245    DTC:  0.21001926782273603    RFC:  0.26396917148362237\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  12 :  LDA:  0.2774566473988439    DTC:  0.17341040462427745    RFC:  0.23699421965317918\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  13 :  LDA:  0.279383429672447    DTC:  0.17533718689788053    RFC:  0.2466281310211946\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  14 :  LDA:  0.2543352601156069    DTC:  0.19460500963391136    RFC:  0.2504816955684008\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  15 :  LDA:  0.2832369942196532    DTC:  0.2003853564547206    RFC:  0.26396917148362237\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  16 :  LDA:  0.2524084778420039    DTC:  0.1791907514450867    RFC:  0.2138728323699422\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  17 :  LDA:  0.29865125240847784    DTC:  0.18304431599229287    RFC:  0.25626204238921\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  18 :  LDA:  0.2774566473988439    DTC:  0.18497109826589594    RFC:  0.2138728323699422\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  19 :  LDA:  0.21579961464354527    DTC:  0.14065510597302505    RFC:  0.19845857418111754\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  20 :  LDA:  0.26011560693641617    DTC:  0.14836223506743737    RFC:  0.2138728323699422\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  21 :  LDA:  0.27552986512524086    DTC:  0.18882466281310212    RFC:  0.21772639691714837\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  22 :  LDA:  0.27360308285163776    DTC:  0.19845857418111754    RFC:  0.2138728323699422\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  23 :  LDA:  0.2466281310211946    DTC:  0.16955684007707128    RFC:  0.22928709055876687\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  24 :  LDA:  0.2543352601156069    DTC:  0.20809248554913296    RFC:  0.23699421965317918\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  25 :  LDA:  0.23892100192678228    DTC:  0.17726396917148363    RFC:  0.18689788053949905\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  26 :  LDA:  0.2350674373795761    DTC:  0.15606936416184972    RFC:  0.2350674373795761\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  27 :  LDA:  0.27167630057803466    DTC:  0.18497109826589594    RFC:  0.20616570327552985\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  28 :  LDA:  0.24855491329479767    DTC:  0.16570327552986513    RFC:  0.2504816955684008\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  29 :  LDA:  0.2543352601156069    DTC:  0.17341040462427745    RFC:  0.20616570327552985\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  30 :  LDA:  0.25626204238921    DTC:  0.19460500963391136    RFC:  0.2138728323699422\n","-----------------------------------------------------------------------------------------------------------------------\n","Time  31 :  LDA:  0.2504816955684008    DTC:  0.18304431599229287    RFC:  0.2023121387283237\n"],"name":"stdout"}]},{"cell_type":"code","metadata":{"id":"F9GDLycI0KDC","colab_type":"code","outputId":"18cfadea-d2f7-4754-9d27-04288f272073","executionInfo":{"status":"ok","timestamp":1573649435122,"user_tz":-330,"elapsed":1341,"user":{"displayName":"201601422 Daiict","photoUrl":"","userId":"06333073350363496926"}},"colab":{"base_uri":"https://localhost:8080/","height":265}},"source":["plt.figure()\n","plt.plot(lda,label='lda')\n","plt.plot(dt,label='dt')\n","plt.plot(rf,label='rf')\n","plt.legend()\n","plt.show()"],"execution_count":0,"outputs":[{"output_type":"display_data","data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAXoAAAD4CAYAAADiry33AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nh/P/AblUIvtcNlqkAAAAAElFTkSuQmCC\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"tags":[]}}]}]}