Perceptron_Intuicao.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__Objetivos__: \n",
    "- entender como o perceptron funciona intuitivamente, tanto em regressão quanto em classificação."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sumário"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Regressão](#Regressão)\n",
    "\n",
    "[Classificação](#Classificação)\n",
    "- [Porta AND](#Porta-AND)\n",
    "- [Porta OR](#Porta-OR)\n",
    "- [Porta XOR](#Porta-XOR)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:19:56.559341Z",
     "start_time": "2017-09-12T12:19:55.344478Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import ipywidgets as wg\n",
    "from ipywidgets import interactive, fixed\n",
    "\n",
    "%matplotlib inline\n",
    "\n",
    "# jupyter nbextension enable --py widgetsnbextension --sys-prefix\n",
    "# restart jupyter notebook"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Regressão "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:19:58.026525Z",
     "start_time": "2017-09-12T12:19:57.992499Z"
    }
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/medidas.csv')\n",
    "print(df.shape)\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:20:01.271158Z",
     "start_time": "2017-09-12T12:20:00.922905Z"
    }
   },
   "outputs": [],
   "source": [
    "x = df.Altura\n",
    "y = df.Peso\n",
    "\n",
    "plt.figure()\n",
    "plt.scatter(x, y)\n",
    "plt.xlabel('Altura')\n",
    "plt.ylabel('Peso')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:20:55.215232Z",
     "start_time": "2017-09-12T12:20:55.145178Z"
    }
   },
   "outputs": [],
   "source": [
    "def plot_line(w, b):\n",
    "    plt.figure(0, figsize=(20,4))\n",
    "    plt.subplot(1,3,3)\n",
    "    plt.scatter(x, y)\n",
    "    y_pred = x*w + b\n",
    "    plt.plot(x, y_pred, c='red')\n",
    "    plt.xlim(140, 210)\n",
    "    plt.ylim(40, 120)\n",
    "    \n",
    "    plt.subplot(1,3,2)\n",
    "    x_ = np.array([0, x.max()])\n",
    "    y_ = x_*w + b\n",
    "    plt.scatter(x, y)\n",
    "    plt.plot(x_, y_, c='red')\n",
    "    plt.xlim(0, 210)\n",
    "    plt.ylim(-160, 120)\n",
    "    \n",
    "    plt.subplot(1,3,1)\n",
    "    mse = np.mean((y - y_pred)**2)\n",
    "    loss.append(mse)\n",
    "    plt.plot(loss)\n",
    "    plt.title('Loss')\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:20:56.782344Z",
     "start_time": "2017-09-12T12:20:55.705578Z"
    }
   },
   "outputs": [],
   "source": [
    "loss = []\n",
    "\n",
    "interactive_plot = interactive(plot_line, w=(1, 1.5, 0.01), b=(-200, 0, 1))\n",
    "output = interactive_plot.children[-1]\n",
    "output.layout_height = '350px'\n",
    "interactive_plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:22:03.437838Z",
     "start_time": "2017-09-12T12:21:50.268473Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "reg = LinearRegression()\n",
    "reg.fit(x.values.reshape(-1,1), y)\n",
    "print(\"w: {:.2f} \\nb: {:.2f}\".format(reg.coef_[0], reg.intercept_))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Classificação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:57:48.364600Z",
     "start_time": "2017-09-12T12:57:48.232507Z"
    }
   },
   "outputs": [],
   "source": [
    "def plot_line(w1, w2, b):\n",
    "    x1, x2 = np.meshgrid(np.linspace(0,1,100), np.linspace(0,1,100))\n",
    "    x_mesh = np.array([x1.ravel(), x2.ravel()]).T\n",
    "    \n",
    "    plt.figure(0, figsize=(10,4))\n",
    "    plt.subplot(1,2,2)\n",
    "    plt.scatter(x[:,0], x[:,1], c=y.ravel(), s=100, cmap='bwr')\n",
    "    \n",
    "    y_mesh = np.dot(x_mesh, np.array([w1, w2]).T) + b\n",
    "    y_mesh = np.where(y_mesh <= 0, 0, 1)\n",
    "\n",
    "    plt.contourf(x1, x2, y_mesh.reshape(x1.shape), cmap='bwr')\n",
    "    \n",
    "    y_pred = np.dot(x, np.array([w1, w2]).T) + b\n",
    "    y_bin = np.where(y_pred <= 0, 0, 1)\n",
    "    print('{0} => {1}'.format(y_pred, y_bin))\n",
    "    \n",
    "    plt.subplot(1,2,1)\n",
    "    mse = np.mean((y.ravel() - y_bin)**2)\n",
    "    loss.append(mse)\n",
    "    plt.plot(loss)\n",
    "    plt.title('Loss')\n",
    "    \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Porta AND"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:22:17.785319Z",
     "start_time": "2017-09-12T12:22:17.770308Z"
    }
   },
   "outputs": [],
   "source": [
    "x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])\n",
    "y = np.array([[0, 0, 0, 1]]).T\n",
    "\n",
    "print(x, y, sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:22:20.966580Z",
     "start_time": "2017-09-12T12:22:20.670370Z"
    }
   },
   "outputs": [],
   "source": [
    "plt.scatter(x[:,0], x[:,1], c=y.ravel(), s=50, cmap='bwr')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-09-12T12:57:49.470386Z",
     "start_time": "2017-09-12T12:57:48.796908Z"
    }
   },
   "outputs": [],
   "source": [
    "loss = []\n",
    "\n",
    "interactive_plot = interactive(plot_line, w1=(-1,1,0.01), w2=(-1,1,0.01), b=(-1.5, 1.5, 0.01))\n",
    "interactive_plot"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Porta OR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])\n",
    "y = np.array([[0, 1, 1, 1]]).T\n",
    "\n",
    "print(x, y, sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(x[:,0], x[:,1], c=y.ravel(), s=50, cmap='bwr')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss = []\n",
    "\n",
    "interactive_plot = interactive(plot_line, w1=(-1,1,0.01), w2=(-1,1,0.01), b=(-1.5, 1.5, 0.01))\n",
    "interactive_plot"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Porta XOR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])\n",
    "y = np.array([[0, 1, 1, 0]]).T\n",
    "\n",
    "print(x, y, sep='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.scatter(x[:,0], x[:,1], c=y.ravel(), s=50, cmap='bwr')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss = []\n",
    "\n",
    "interactive_plot = interactive(plot_line, w1=(-1,1,0.01), w2=(-1,1,0.01), b=(-1.5, 1.5, 0.01))\n",
    "interactive_plot"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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"
  },
  "widgets": {
   "state": {
    "ad801d6429424cc183bd50005dba0cc5": {
     "views": [
      {
       "cell_index": 9
      }
     ]
    },
    "d48f520c0c534642ba50d4079c3ceead": {
     "views": [
      {
       "cell_index": 21
      }
     ]
    },
    "f4b33603eeec488391bdb5d73d9c16a3": {
     "views": [
      {
       "cell_index": 16
      }
     ]
    }
   },
   "version": "1.2.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}