SVM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/darshanakg/.conda/envs/tf/lib/python3.7/site-packages/sklearn/externals/joblib/__init__.py:15: DeprecationWarning: sklearn.externals.joblib is deprecated in 0.21 and will be removed in 0.23. Please import this functionality directly from joblib, which can be installed with: pip install joblib. If this warning is raised when loading pickled models, you may need to re-serialize those models with scikit-learn 0.21+.\n",
      "  warnings.warn(msg, category=DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn import svm\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.metrics import classification_report, accuracy_score, precision_score, recall_score\n",
    "import numpy as np\n",
    "from sklearn.externals import joblib\n",
    "import matplotlib.pyplot as plt\n",
    "import sys\n",
    "from pickle import dump, load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((314, 270), (314,))"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_x = np.loadtxt('gesture_input_x_v12.csv', dtype=float)\n",
    "data_y = np.loadtxt('gesture_input_y_v12.csv', dtype=float)\n",
    "\n",
    "data_x.shape, data_y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(251, 63)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# split into train and test sets\n",
    "train_size = int(len(data_x) * 0.8)\n",
    "test_size = len(data_x) - train_size\n",
    "train_size, test_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_x, test_x = data_x[0:train_size,:], data_x[train_size:,:]\n",
    "train_y, test_y = data_y[0:train_size], data_y[train_size:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2., 0., 1., 2., 0., 0., 1., 0., 1., 2., 3., 1., 3., 4., 0., 3., 0.,\n",
       "       2., 1., 3., 0., 4., 2., 3., 2., 3., 3., 3., 4., 3., 1., 0., 3., 4.,\n",
       "       2., 3., 1., 3., 2., 0., 3., 2., 4., 0., 0., 0., 3., 1., 3., 0., 3.,\n",
       "       0., 1., 0., 2., 1., 1., 1., 1., 1., 2., 1., 0.])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "    decision_function_shape='ovr', degree=3, gamma='scale', kernel='poly',\n",
       "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "    tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classifier = svm.SVC(kernel='poly', gamma='scale')\n",
    "classifier.fit(train_x, train_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "dump(classifier, open(\"svm.pkl\", \"wb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/darshanakg/.conda/envs/tf/lib/python3.7/site-packages/sklearn/model_selection/_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([0.58333333, 0.64      , 0.85333333])"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cross_val_score(classifier, test_x, test_y, scoring='recall_macro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       1.00      0.94      0.97        49\n",
      "         1.0       0.85      1.00      0.92        52\n",
      "         2.0       0.93      0.91      0.92        56\n",
      "         3.0       0.90      0.88      0.89        51\n",
      "         4.0       0.85      0.77      0.80        43\n",
      "\n",
      "    accuracy                           0.90       251\n",
      "   macro avg       0.91      0.90      0.90       251\n",
      "weighted avg       0.91      0.90      0.90       251\n",
      "\n"
     ]
    }
   ],
   "source": [
    "predictions = classifier.predict(train_x)\n",
    "print(classification_report(train_y, predictions))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9043824701195219\n",
      "Precision: 0.9051771179640031\n",
      "Recall: 0.899856919511991\n"
     ]
    }
   ],
   "source": [
    "print(\"Accuracy:\", accuracy_score(train_y, predictions))\n",
    "print(\"Precision:\", precision_score(train_y, predictions, average='macro'))\n",
    "print(\"Recall:\", recall_score(train_y, predictions, average='macro'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7142857142857143\n",
      "Precision: 0.747008547008547\n",
      "Recall: 0.7029545454545455\n"
     ]
    }
   ],
   "source": [
    "predictions_test = classifier.predict(test_x)\n",
    "print(\"Accuracy:\", accuracy_score(test_y, predictions_test))\n",
    "print(\"Precision:\", precision_score(test_y, predictions_test, average='macro'))\n",
    "print(\"Recall:\", recall_score(test_y, predictions_test, average='macro'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import learning_curve\n",
    "\n",
    "#Create CV training and test scores for various training set sizes\n",
    "train_sizes, train_scores, test_scores = learning_curve(svm.SVC(kernel='linear', gamma='scale'), \n",
    "                                                        train_x, \n",
    "                                                        train_y,\n",
    "                                                        # Number of folds in cross-validation\n",
    "                                                        cv=2,\n",
    "                                                        # Evaluation metric\n",
    "                                                        scoring='accuracy',\n",
    "                                                        # Use all computer cores\n",
    "                                                        n_jobs=-1)\n",
    "\n",
    "# # Create means and standard deviations of training set scores\n",
    "# train_mean = np.mean(train_scores, axis=1)\n",
    "# train_std = np.std(train_scores, axis=1)\n",
    "\n",
    "# # Create means and standard deviations of test set scores\n",
    "# test_mean = np.mean(test_scores, axis=1)\n",
    "# test_std = np.std(test_scores, axis=1)\n",
    "\n",
    "# Draw lines\n",
    "plt.plot(train_sizes, train_mean, '--', color=\"#111111\",  label=\"Training score\")\n",
    "plt.plot(train_sizes, test_mean, color=\"#111111\", label=\"Cross-validation score\")\n",
    "\n",
    "# # Draw bands\n",
    "# plt.fill_between(train_sizes, train_mean - train_std, train_mean + train_std, color=\"#DDDDDD\")\n",
    "# plt.fill_between(train_sizes, test_mean - test_std, test_mean + test_std, color=\"#DDDDDD\")\n",
    "\n",
    "# Create plot\n",
    "plt.title(\"Learning Curve\")\n",
    "plt.xlabel(\"Training Set Size\"), plt.ylabel(\"Accuracy Score\"), plt.legend(loc=\"best\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = load(open(\"svm.pkl\", \"rb\" ) )"
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}