initial work

Session-4/digit_svm.ipynb

+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "SVC(C=0.1, cache_size=200, class_weight=None, coef0=0.0,\n",
+       "  decision_function_shape='ovr', degree=3, gamma=0.001, kernel='rbf',\n",
+       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
+       "  tol=0.001, verbose=False)"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Standard scientific Python imports\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "# Import datasets, classifiers and performance metrics\n",
+    "from sklearn import datasets, svm, metrics\n",
+    "from PIL import Image\n",
+    "import numpy as np\n",
+    "import math\n",
+    "\n",
+    "#เตรียมข้อมูลโดยการ load ข้อมูลจากไฟล์ภาพ ในโฟลเดอร์ images/numbers/\n",
+    "digit_data = list()\n",
+    "digit_label = list()\n",
+    "for eachNum in range(0,10):\n",
+    "    #print eachNum\n",
+    "    for furtherNum in range(1,10):\n",
+    "        #print(str(eachNum))\n",
+    "        imgFilePath = 'images/numbers/'+str(eachNum)+'.'+str(furtherNum)+'.png'\n",
+    "        ei = Image.open(imgFilePath)\n",
+    "        eiar = np.array(ei)\n",
+    "        eiarl = str(eiar.tolist())\n",
+    "\n",
+    "        data = np.zeros(64)\n",
+    "        len(eiar[0])\n",
+    "        count = 0\n",
+    "        for i in range(len(eiar)):\n",
+    "            for j in range(len(eiar[i])):\n",
+    "                #ทำ Normalize ด้วยการ (R+G+B) หาร 765 มาจาก (255+255+255)\n",
+    "                data[count] = math.floor(((np.sum(eiar[i][j][:3])/765)))\n",
+    "                count+=1\n",
+    "        digit_data.append( data )\n",
+    "        digit_label.append( eachNum )\n",
+    "#ข้อมูลสำหรับนำไปฝึก\n",
+    "digit_data = np.array(digit_data)  #ข้อมูล\n",
+    "digit_label = np.array(digit_label) # label ของข้อมูล\n",
+    "\n",
+    "# Create a classifier: a support vector classifier\n",
+    "classifier = svm.SVC(gamma=0.001, C=.1)\n",
+    "\n",
+    "# We learn the digits on the first half of the digits\n",
+    "classifier.fit(digit_data, digit_label)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2,\n",
+       "       2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 4, 5,\n",
+       "       5, 5, 5, 5, 5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7, 7, 7,\n",
+       "       7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9, 9])"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "digit_label"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "#การทดสอบ\n",
+    "i = Image.open('images/t5.png')\n",
+    "i = Image.open('images/t7.png')\n",
+    "iar = np.array(i)\n",
+    "\n",
+    "test_data = np.zeros(64)\n",
+    "len(iar[0])\n",
+    "count = 0\n",
+    "for i in range(len(iar)):\n",
+    "    for j in range(len(iar[i])):\n",
+    "        test_data[count] = math.floor(((np.sum(iar[i][j][:3])/765)))\n",
+    "        #print(test_data[count])\n",
+    "        count+=1\n",
+    "    #print('----')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "array([7])"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "predicted = classifier.predict([test_data])\n",
+    "predicted"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Classification report for classifier SVC(C=0.1, cache_size=200, class_weight=None, coef0=0.0,\n",
+      "  decision_function_shape='ovr', degree=3, gamma=0.001, kernel='rbf',\n",
+      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
+      "  tol=0.001, verbose=False):\n",
+      "             precision    recall  f1-score   support\n",
+      "\n",
+      "          0       1.00      0.89      0.94         9\n",
+      "          1       0.82      1.00      0.90         9\n",
+      "          2       1.00      0.89      0.94         9\n",
+      "          3       0.69      1.00      0.82         9\n",
+      "          4       1.00      0.78      0.88         9\n",
+      "          5       1.00      0.78      0.88         9\n",
+      "          6       1.00      0.78      0.88         9\n",
+      "          7       1.00      0.89      0.94         9\n",
+      "          8       0.73      0.89      0.80         9\n",
+      "          9       0.70      0.78      0.74         9\n",
+      "\n",
+      "avg / total       0.89      0.87      0.87        90\n",
+      "\n",
+      "\n",
+      "Confusion matrix:\n",
+      "[[8 0 0 0 0 0 0 0 0 1]\n",
+      " [0 9 0 0 0 0 0 0 0 0]\n",
+      " [0 0 8 1 0 0 0 0 0 0]\n",
+      " [0 0 0 9 0 0 0 0 0 0]\n",
+      " [0 0 0 0 7 0 0 0 0 2]\n",
+      " [0 0 0 2 0 7 0 0 0 0]\n",
+      " [0 0 0 0 0 0 7 0 2 0]\n",
+      " [0 1 0 0 0 0 0 8 0 0]\n",
+      " [0 0 0 1 0 0 0 0 8 0]\n",
+      " [0 1 0 0 0 0 0 0 1 7]]\n",
+      "\n",
+      "accuracy =  0.866666666667\n"
+     ]
+    }
+   ],
+   "source": [
+    "#วัดประสิทธิภาพ\n",
+    "from sklearn.metrics import accuracy_score\n",
+    "expected = digit_label\n",
+    "predicted = classifier.predict(digit_data)\n",
+    "print(\"Classification report for classifier %s:\\n%s\\n\"\n",
+    "      % (classifier, metrics.classification_report(expected, predicted)))\n",
+    "print(\"Confusion matrix:\\n%s\" % metrics.confusion_matrix(expected, predicted))\n",
+    "print(\"\\naccuracy = \",accuracy_score(digit_label, predicted))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Fitting the classifier to the training set\n",
+      "done in 0.436s\n",
+      "Best estimator found by grid search:\n",
+      "SVC(C=1000.0, cache_size=200, class_weight='balanced', coef0=0.0,\n",
+      "  decision_function_shape='ovr', degree=3, gamma=0.0005, kernel='rbf',\n",
+      "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
+      "  tol=0.001, verbose=False)\n"
+     ]
+    }
+   ],
+   "source": [
+    "#optimize parameter\n",
+    "from time import time\n",
+    "from sklearn.model_selection import GridSearchCV\n",
+    "from sklearn.svm import SVC\n",
+    "# #############################################################################\n",
+    "# Train a SVM classification model\n",
+    "\n",
+    "print(\"Fitting the classifier to the training set\")\n",
+    "t0 = time()\n",
+    "param_grid = {'C': [1e3, 5e3, 1e4, 5e4, 1e5],\n",
+    "              'gamma': [0.0001, 0.0005, 0.001, 0.005, 0.01, 0.1], }\n",
+    "clf = GridSearchCV(SVC(kernel='rbf', class_weight='balanced'), param_grid)\n",
+    "clf = clf.fit(digit_data, digit_label)\n",
+    "print(\"done in %0.3fs\" % (time() - t0))\n",
+    "print(\"Best estimator found by grid search:\")\n",
+    "print(clf.best_estimator_)\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": "3.5.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}