Jean-N'dah_perceptron.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import random as rd\n",
    "import numpy as np\n",
    "import seaborn as sn\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "mu1 = -1\n",
    "mu2 = 3\n",
    "sig1 = 0.5\n",
    "sig2 = 1\n",
    "N = 100\n",
    "np.random.seed(10)\n",
    "x11=np.random.randn(N,1)*sig1 + mu1\n",
    "x12=np.random.randn(N,1)*sig1 + mu1+3\n",
    "x21=np.random.randn(N,1)*sig2 + mu2\n",
    "x22=np.random.randn(N,1)*sig2 + mu2+3\n",
    "c = np.vstack((-np.ones((N,1)), np.ones((N,1))))\n",
    "x1 = np.hstack((x11,x12))\n",
    "x2 = np.hstack((x21,x22))\n",
    "\n",
    "X = np.hstack( (np.vstack( (x1,x2) ),c) )\n",
    "np.random.shuffle(X)\n",
    "dataset = pd.DataFrame(data=X, columns=['x','y','c'])\n",
    "\n",
    "\n",
    "## Pseudo code used: pseudo code in class \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>c</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.334207</td>\n",
       "      <td>2.058738</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.264395</td>\n",
       "      <td>5.636814</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.669884</td>\n",
       "      <td>1.856185</td>\n",
       "      <td>-1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.995284</td>\n",
       "      <td>4.484924</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.588041</td>\n",
       "      <td>4.739172</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          x         y    c\n",
       "0 -0.334207  2.058738 -1.0\n",
       "1  3.264395  5.636814  1.0\n",
       "2 -0.669884  1.856185 -1.0\n",
       "3  1.995284  4.484924  1.0\n",
       "4  3.588041  4.739172  1.0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "Dataset=dataset.drop('c', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "Dataset.insert(0,'const',np.ones_like(Dataset['x']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>const</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.334207</td>\n",
       "      <td>2.058738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.264395</td>\n",
       "      <td>5.636814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.669884</td>\n",
       "      <td>1.856185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.995284</td>\n",
       "      <td>4.484924</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>3.588041</td>\n",
       "      <td>4.739172</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   const         x         y\n",
       "0    1.0 -0.334207  2.058738\n",
       "1    1.0  3.264395  5.636814\n",
       "2    1.0 -0.669884  1.856185\n",
       "3    1.0  1.995284  4.484924\n",
       "4    1.0  3.588041  4.739172"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "class Perceptron(object):\n",
    "    \n",
    "    def __init__(self, data, target, w, tolerance, alpha=0.01, Max_iter=1000):\n",
    "        self.data=data\n",
    "        self.target=target\n",
    "        self.w=w\n",
    "        self.tolerance=tolerance\n",
    "        self.alpha=alpha\n",
    "        self.Max_iter=Max_iter\n",
    "        \n",
    "    def get_gradient(self, vectors, coeffs):\n",
    "        '''This function computes the gradient of the loss function in the predictions'''\n",
    "        return -sum([coeffs[i]*vectors[i] for i in range(len(coeffs))]) #This is the formula for the gredient of \n",
    "#the loss function\n",
    "\n",
    "    def fit(self):\n",
    "        '''This function uses the gradient given by the above get_gredient function\n",
    "        to update the parameter w until we reach the desired level of accuracy in our\n",
    "        predictions'''\n",
    "        Norm=np.linalg.norm\n",
    "        old_w=self.w+1\n",
    "        num_iter=0\n",
    "        while(Norm(self.w-old_w)>self.tolerance and num_iter<self.Max_iter):\n",
    "            old_w=self.w\n",
    "            y_hat=[np.sign(self.w.dot(np.array(self.data.iloc[i,:]))) for i in range(self.data.shape[0])]\n",
    "            Misclassified_indices=[i for i in range(len(y_hat)) if y_hat[i]!=self.target[i]] # finding the indices of the misclassified\n",
    "            if len(Misclassified_indices)>0:#If there are misclassified points, then we compute the gradient and update w\n",
    "                gradient=self.get_gradient(np.array([self.data.iloc[j,:] for j in Misclassified_indices]), np.array([self.target[j] for j in Misclassified_indices]))\n",
    "            else:\n",
    "                gradient=np.zeros_like(self.data.iloc[0,:])\n",
    "            self.w=self.w-self.alpha*gradient \n",
    "            num_iter += 1\n",
    "            print(\"gradient: {} and number of misclassified: {}\".format(gradient,len(Misclassified_indices))) #We print this \n",
    "        # to track the progression of the classifications. The desired stopping point is when there are no misclassified\n",
    "        #points but it can come that we stop the process when the maximal number of iterations we fixed is exceeded\n",
    "        return self.w"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The following plot is for a good guess of w, and requires 0 iteration."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gradient: [0. 0. 0.] and number of misclassified: 0\n",
      "Final w: [-3.5  2.   0.5]\n"
     ]
    }
   ],
   "source": [
    "w=np.array([-3.5,2,0.5]) # Guess of w, a good one actually\n",
    "perceptron=Perceptron(Dataset, dataset['c'], w, tolerance=1e-6)\n",
    "w=perceptron.fit()\n",
    "print(\"Final w: {}\".format(w)) # Diplaying w"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_the_line(w,x):\n",
    "    return (-w[0]/w[2])-(w[1]/w[2])*x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "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": [
    "x=np.linspace(-3,6)\n",
    "sn.scatterplot(Dataset['x'],Dataset['y'], s=40)\n",
    "plt.plot(x,plot_the_line(w,x), color='red')\n",
    "plt.xlim(-2.5,5.5)\n",
    "plt.ylim(-0,11)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Here we give an example where the classification process takes time, that is when our guess of w is not good enough"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gradient: [-100.         -297.60944397 -607.95909262] and number of misclassified: 100\n",
      "gradient: [ 50.         -31.49600162 111.65158597] and number of misclassified: 50\n",
      "gradient: [15.         -3.3060193  33.15182277] and number of misclassified: 15\n",
      "gradient: [10.         -1.04373005 21.77613877] and number of misclassified: 10\n",
      "gradient: [ 8.         -0.30457386 17.00111752] and number of misclassified: 8\n",
      "gradient: [ 6.          0.3249215  11.87189999] and number of misclassified: 6\n",
      "gradient: [4.         0.61620332 8.03825682] and number of misclassified: 4\n",
      "gradient: [4.         0.61620332 8.03825682] and number of misclassified: 4\n",
      "gradient: [4.         0.61620332 8.03825682] and number of misclassified: 4\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [3.         0.62366103 6.74869708] and number of misclassified: 3\n",
      "gradient: [2.         0.43117736 3.51603454] and number of misclassified: 2\n",
      "gradient: [2.         0.43117736 3.51603454] and number of misclassified: 2\n",
      "gradient: [2.         0.43117736 3.51603454] and number of misclassified: 2\n",
      "gradient: [2.         0.43117736 3.51603454] and number of misclassified: 2\n",
      "gradient: [1.         0.23382553 1.86957026] and number of misclassified: 1\n",
      "gradient: [1.         0.23382553 1.86957026] and number of misclassified: 1\n",
      "gradient: [1.         0.23382553 1.86957026] and number of misclassified: 1\n",
      "gradient: [0. 0. 0.] and number of misclassified: 0\n",
      "Final w: [-0.86        6.44170763 -0.35270668]\n"
     ]
    }
   ],
   "source": [
    "w=np.array([-0.5,3.2,-3.5]) # Guess of w, a good one actually\n",
    "perceptron=Perceptron(Dataset, dataset['c'], w, tolerance=1e-10)\n",
    "w=perceptron.fit()\n",
    "print(\"Final w: {}\".format(w)) # Diplaying w"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "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": [
    "x=np.linspace(-3,6)\n",
    "sn.scatterplot(Dataset['x'],Dataset['y'], s=40)\n",
    "plt.plot(x,plot_the_line(w,x), color='red')\n",
    "plt.xlim(-2.5,5.5)\n",
    "plt.ylim(-0,11)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "def quick_sort_helper(arr,pivot):\n",
    "    for i in range(len(arr)):\n",
    "        Restart=False\n",
    "        if arr[i] <= arr[pivot]:\n",
    "            pass\n",
    "        else:\n",
    "            temp = arr[i]\n",
    "            arr[i] = arr[pivot-1]\n",
    "            arr[pivot -1] = arr[pivot]\n",
    "            arr[pivot] = temp\n",
    "            Restart=True\n",
    "            break\n",
    "    if Restart:\n",
    "        arr = quick_sort_helper(arr,pivot)\n",
    "    return arr\n",
    "        \n",
    "            \n",
    "def very_very_quick_sort(arr):\n",
    "    l = len(arr)\n",
    "    pivot = len(arr)-1\n",
    "    for i in range(l):\n",
    "        if i >= 1:\n",
    "            arr[:-i] = quick_sort_helper(arr[:-i],pivot-i)\n",
    "        else:\n",
    "            arr = quick_sort_helper(arr,pivot-i)\n",
    "    return arr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "a= [110,4,2,7,9,2,12,-56]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[-56, 2, 2, 4, 7, 9, 12, 110]"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quick_sort(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "locations = {'North America': {'USA': ['Mountain View']}, 'Asia': {'India': ['Bangalore']}, 'North America': {'USA': ['Atlanta']}, 'Africa': {'Egypt': 'Cairo'}, 'Asia': {'China': ['Shanghai']}}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "ar=np.array([12.3,4,3.9])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([12.3,  4. ,  3.9])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ar"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[12.3,  4. ,  3.9]])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ar[None,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.00345667, -0.37448701, -1.33237747,  0.04661349,  1.26947331])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.randn(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Note that here again, the classification process ended well even though with our eyes we are  not convinced."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "array_split() missing 2 required positional arguments: 'ary' and 'indices_or_sections'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-17-68a941c26909>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0marray_split\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m: array_split() missing 2 required positional arguments: 'ary' and 'indices_or_sections'"
     ]
    }
   ],
   "source": [
    "np.array_split()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}