Compare with different Classification Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Compare with different classification model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compare the training time and performance between five classification models:\n",
    "\n",
    "1. KNN\n",
    "2. Logistic Regression\n",
    "3. Decision Tree\n",
    "4. Neural Network\n",
    "5. SVM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I use the 3 sample datasets and make average to compare the differenct models:\n",
    "\n",
    "1. Iris (The most simplist dataset to train)\n",
    "2. Wine (A little bit complex... The feature is more than Iris dataset)\n",
    "3. Digit (Image dataset)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion\n",
    "\n",
    "The average training time: Neural Network > Logistic Regression > KNN > SVM > Decision Tree  \n",
    "The average f1-score: Logistic Regression > SVM > Decision Tree > KNN > Neural Network\n",
    "\n",
    "So we could see that the neural need more resourse to training the data set.  \n",
    "Although the nerual network have the good performance in classification, it need more time and complicate tuning to training the data.\n",
    "\n",
    "If we want to try to see the performance very quickly, we could try the decision tree or SVM first.\n",
    "These two is more faster than other model and could get a good performance.\n",
    "\n",
    "In structure data, we also could try the logistic regression first.\n",
    "The performance of logistic regression is worth to try on the dataset.\n",
    "\n",
    "Neural Network is a very powerful to classify the data, but it need more time to training and tuning the model like add more layer or neuron.\n",
    "\n",
    "KNN get a average performance between these model, but it also is a very useful tool been used in feature engineering (predict the null data) or recommendation system field.\n",
    "\n",
    "I think the better way is sampling the data into small dataset, and use the all model to train and get a baseline.\n",
    "This baseline is very useful to let you decide which model you could pay more time to dig in.\n",
    "\n",
    "Hope everyone get a nice performance in all dataset!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Start Comparing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sn\n",
    "import numpy as np\n",
    "\n",
    "from sklearn.datasets import load_iris, load_wine, load_digits\n",
    "\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.svm import LinearSVC\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "\n",
    "from sklearn.cross_validation import train_test_split as cv\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "import time\n",
    "\n",
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Define dataset loading function\n",
    "def load_data(dataset_fn):\n",
    "    data = dataset_fn\n",
    "    return data.data, data.target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define all classfier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Define the classifier\n",
    "knn = KNeighborsClassifier()\n",
    "svm = LinearSVC()\n",
    "lr = LogisticRegression()\n",
    "tree = DecisionTreeClassifier()\n",
    "nn = MLPClassifier()\n",
    "\n",
    "classifier_lst = [knn, lr, svm, tree, nn]\n",
    "classifier_lst_name = ['KNN', 'Logistic_Regression', 'SVM', 'DecisionTree', 'NeuralNetwork']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define classifier training function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def make_baseline(dataset_fn, classifier):\n",
    "    \n",
    "    start_time = time.time()\n",
    "    \n",
    "    feature, target = load_data(dataset_fn)\n",
    "    x_train, x_test, y_train, y_test = cv(feature, target, random_state = 42)\n",
    "    classifier.fit(x_train, y_train)\n",
    "    \n",
    "    training_score = classifier.score(x_train, y_train)\n",
    "    testing_score = classifier.score(x_test, y_test)\n",
    "    score = f1_score(y_test, classifier.predict(x_test), average='weighted')\n",
    "    \n",
    "    print(classifier)\n",
    "    print('Training Score: {:.4f}'.format(training_score))\n",
    "    print('Testing Score: {:.4f}'.format(testing_score))\n",
    "    print('The f1 score is: {}'.format(score))\n",
    "\n",
    "    end_time = time.time()\n",
    "    training_time = end_time-start_time\n",
    "    \n",
    "    print('The training time is: {:.4f}\\n'.format((training_time)))\n",
    "    \n",
    "    return score, training_time"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Start to training differenct dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "           metric_params=None, n_jobs=1, n_neighbors=5, p=2,\n",
      "           weights='uniform')\n",
      "Training Score: 0.9643\n",
      "Testing Score: 1.0000\n",
      "The f1 score is: 1.0\n",
      "The training time is: 0.0039\n",
      "\n",
      "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
      "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
      "          penalty='l2', random_state=None, solver='liblinear', tol=0.0001,\n",
      "          verbose=0, warm_start=False)\n",
      "Training Score: 0.9643\n",
      "Testing Score: 1.0000\n",
      "The f1 score is: 1.0\n",
      "The training time is: 0.1681\n",
      "\n",
      "LinearSVC(C=1.0, class_weight=None, dual=True, fit_intercept=True,\n",
      "     intercept_scaling=1, loss='squared_hinge', max_iter=1000,\n",
      "     multi_class='ovr', penalty='l2', random_state=None, tol=0.0001,\n",
      "     verbose=0)\n",
      "Training Score: 0.9732\n",
      "Testing Score: 1.0000\n",
      "The f1 score is: 1.0\n",
      "The training time is: 0.0072\n",
      "\n",
      "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n",
      "            max_features=None, max_leaf_nodes=None,\n",
      "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "            min_samples_leaf=1, min_samples_split=2,\n",
      "            min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n",
      "            splitter='best')\n",
      "Training Score: 1.0000\n",
      "Testing Score: 1.0000\n",
      "The f1 score is: 1.0\n",
      "The training time is: 0.0032\n",
      "\n",
      "MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n",
      "       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
      "       hidden_layer_sizes=(100,), learning_rate='constant',\n",
      "       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n",
      "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n",
      "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n",
      "       verbose=False, warm_start=False)\n",
      "Training Score: 0.9732\n",
      "Testing Score: 0.9737\n",
      "The f1 score is: 0.9735839598997494\n",
      "The training time is: 0.1033\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:564: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n",
      "  % self.max_iter, ConvergenceWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'F1-Score')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x864 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Iris dataset\n",
    "iris = load_iris()\n",
    "iris_score_lst = []\n",
    "iris_time_lst = []\n",
    "\n",
    "for i in classifier_lst:\n",
    "    score, training_time = make_baseline(iris,i)\n",
    "    iris_score_lst.append(score)\n",
    "    iris_time_lst.append(training_time)\n",
    "\n",
    "plt.figure(figsize=(16,12))\n",
    "subplot(2,2,1)\n",
    "sn.barplot(x=classifier_lst_name, y=iris_time_lst, palette=sn.color_palette('hls'))\n",
    "plt.xticks(rotation=30)\n",
    "plt.title('Training Time')\n",
    "\n",
    "subplot(2,2,2)\n",
    "sn.barplot(x=classifier_lst_name, y=iris_score_lst, palette=sn.color_palette('hls'))\n",
    "plt.xticks(rotation=30)\n",
    "plt.title('F1-Score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "           metric_params=None, n_jobs=1, n_neighbors=5, p=2,\n",
      "           weights='uniform')\n",
      "Training Score: 0.7669\n",
      "Testing Score: 0.7111\n",
      "The f1 score is: 0.7111111111111111\n",
      "The training time is: 0.0035\n",
      "\n",
      "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
      "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
      "          penalty='l2', random_state=None, solver='liblinear', tol=0.0001,\n",
      "          verbose=0, warm_start=False)\n",
      "Training Score: 0.9774\n",
      "Testing Score: 0.9778\n",
      "The f1 score is: 0.9776949363156261\n",
      "The training time is: 0.0033\n",
      "\n",
      "LinearSVC(C=1.0, class_weight=None, dual=True, fit_intercept=True,\n",
      "     intercept_scaling=1, loss='squared_hinge', max_iter=1000,\n",
      "     multi_class='ovr', penalty='l2', random_state=None, tol=0.0001,\n",
      "     verbose=0)\n",
      "Training Score: 0.8722\n",
      "Testing Score: 0.9111\n",
      "The f1 score is: 0.9104105090311987\n",
      "The training time is: 0.0177\n",
      "\n",
      "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n",
      "            max_features=None, max_leaf_nodes=None,\n",
      "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "            min_samples_leaf=1, min_samples_split=2,\n",
      "            min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n",
      "            splitter='best')\n",
      "Training Score: 1.0000\n",
      "Testing Score: 0.9556\n",
      "The f1 score is: 0.9558589126489389\n",
      "The training time is: 0.0031\n",
      "\n",
      "MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n",
      "       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
      "       hidden_layer_sizes=(100,), learning_rate='constant',\n",
      "       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n",
      "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n",
      "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n",
      "       verbose=False, warm_start=False)\n",
      "Training Score: 0.5940\n",
      "Testing Score: 0.5111\n",
      "The f1 score is: 0.4216666666666667\n",
      "The training time is: 0.0097\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/anaconda3/lib/python3.6/site-packages/sklearn/metrics/classification.py:1135: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'F1-Score')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Wine dataset\n",
    "wine = load_wine()\n",
    "wine_score_lst = []\n",
    "wine_time_lst = []\n",
    "\n",
    "for i in classifier_lst:\n",
    "    score, training_time = make_baseline(wine,i)\n",
    "    wine_score_lst.append(score)\n",
    "    wine_time_lst.append(training_time)\n",
    "\n",
    "plt.figure(figsize=(16,4))\n",
    "subplot(1,2,1)\n",
    "sn.barplot(x=classifier_lst_name, y=wine_time_lst, palette=sn.color_palette('hls'))\n",
    "plt.xticks(rotation=30)\n",
    "plt.title('Training Time')\n",
    "subplot(1,2,2)\n",
    "sn.barplot(x=classifier_lst_name, y=wine_score_lst, palette=sn.color_palette('hls'))\n",
    "plt.xticks(rotation=30)\n",
    "plt.title('F1-Score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
      "           metric_params=None, n_jobs=1, n_neighbors=5, p=2,\n",
      "           weights='uniform')\n",
      "Training Score: 0.9889\n",
      "Testing Score: 0.9933\n",
      "The f1 score is: 0.9933082556240451\n",
      "The training time is: 0.2941\n",
      "\n",
      "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
      "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
      "          penalty='l2', random_state=None, solver='liblinear', tol=0.0001,\n",
      "          verbose=0, warm_start=False)\n",
      "Training Score: 0.9948\n",
      "Testing Score: 0.9667\n",
      "The f1 score is: 0.9670228496107158\n",
      "The training time is: 0.1705\n",
      "\n",
      "LinearSVC(C=1.0, class_weight=None, dual=True, fit_intercept=True,\n",
      "     intercept_scaling=1, loss='squared_hinge', max_iter=1000,\n",
      "     multi_class='ovr', penalty='l2', random_state=None, tol=0.0001,\n",
      "     verbose=0)\n",
      "Training Score: 0.9926\n",
      "Testing Score: 0.9600\n",
      "The f1 score is: 0.9605515894145703\n",
      "The training time is: 0.1141\n",
      "\n",
      "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=None,\n",
      "            max_features=None, max_leaf_nodes=None,\n",
      "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
      "            min_samples_leaf=1, min_samples_split=2,\n",
      "            min_weight_fraction_leaf=0.0, presort=False, random_state=None,\n",
      "            splitter='best')\n",
      "Training Score: 1.0000\n",
      "Testing Score: 0.8778\n",
      "The f1 score is: 0.8769737594870076\n",
      "The training time is: 0.0210\n",
      "\n",
      "MLPClassifier(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n",
      "       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
      "       hidden_layer_sizes=(100,), learning_rate='constant',\n",
      "       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n",
      "       nesterovs_momentum=True, power_t=0.5, random_state=None,\n",
      "       shuffle=True, solver='adam', tol=0.0001, validation_fraction=0.1,\n",
      "       verbose=False, warm_start=False)\n",
      "Training Score: 1.0000\n",
      "Testing Score: 0.9822\n",
      "The f1 score is: 0.982199076170979\n",
      "The training time is: 0.9443\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5,1,'F1-Score')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Digits Dataset\n",
    "digits = load_digits()\n",
    "digits_score_lst = []\n",
    "digits_time_lst = []\n",
    "\n",
    "for i in classifier_lst:\n",
    "    score, training_time = make_baseline(digits,i)\n",
    "    digits_score_lst.append(score)\n",
    "    digits_time_lst.append(training_time)\n",
    "\n",
    "plt.figure(figsize=(16,4))\n",
    "subplot(1,2,1)\n",
    "sn.barplot(x=classifier_lst_name, y=digits_time_lst, palette=sn.color_palette('hls'))\n",
    "plt.xticks(rotation=30)\n",
    "plt.title('Training Time')\n",
    "subplot(1,2,2)\n",
    "sn.barplot(x=classifier_lst_name, y=digits_score_lst, palette=sn.color_palette('hls'))\n",
    "plt.xticks(rotation=30)\n",
    "plt.title('F1-Score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The classifier is: KNN\n",
      "Average training time is: 0.10048906008402507\n",
      "Average training score is 0.9014731222450522\n",
      "\n",
      "The classifier is: Logistic_Regression\n",
      "Average training time is: 0.11396439870198567\n",
      "Average training score is 0.9815725953087805\n",
      "\n",
      "The classifier is: SVM\n",
      "Average training time is: 0.04633522033691406\n",
      "Average training score is 0.9569873661485896\n",
      "\n",
      "The classifier is: DecisionTree\n",
      "Average training time is: 0.009070316950480143\n",
      "Average training score is 0.9442775573786487\n",
      "\n",
      "The classifier is: NeuralNetwork\n",
      "Average training time is: 0.35243646303812665\n",
      "Average training score is 0.7924832342457985\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for idx, name in enumerate(classifier_lst_name):\n",
    "    print('The classifier is: {}'.format(name))\n",
    "    avg_time = np.mean([iris_time_lst[idx], wine_time_lst[idx], digits_time_lst[idx]])\n",
    "    avg_score = np.mean([iris_score_lst[idx], wine_score_lst[idx], digits_score_lst[idx]])\n",
    "    print('Average training time is: {}'.format(avg_time))\n",
    "    print('Average training score is {}\\n'.format(avg_score))"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}