{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "c = Counter((w for l in open(\"train.txt\") for w in l.strip().split() if w != \"<unk>\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "ls = np.array([n for _, n in c.most_common(1000)])\n",
    "words = [w for w, _ in c.most_common(1000)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.4/dist-packages/matplotlib/__init__.py:892: UserWarning: axes.color_cycle is deprecated and replaced with axes.prop_cycle; please use the latter.\n",
      "  warnings.warn(self.msg_depr % (key, alt_key))\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApAAAAGvCAYAAADsceW1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRERERHpEAVJEREREekQBUkRERER65P8BQERdGvm5ebQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f3497817be0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plt.plot(ls[:200])\n",
    "sns.set_context(\"talk\", rc={\"lines.linewidth\": 2.5})\n",
    "sns.set_style(\"darkgrid\")\n",
    "f, ax = plt.subplots(figsize=(10, 7))\n",
    "\n",
    "#plt.plot(figsize=(10,10))\n",
    "sns.barplot(x=\"total\", y=\"abbrev\", data={\"total\": ls[:200], \"abbrev\": words[:200]},\n",
    "            label=\"Total\")\n",
    "_ = ax.set_yticklabels([ w if i % 10 ==0 else \"\" for i, w in enumerate(words[:200]) ])\n",
    "plt.savefig(\"zipf.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -1.        , -1.        , -1.        , -1.        , -1.        ,\n",
       "       -0.99999997, -0.99999856, -0.99991793, -0.99534489, -0.76580379,\n",
       "        0.76580379,  0.99534489,  0.99991793,  0.99999856,  0.99999997,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ,\n",
       "        1.        ,  1.        ,  1.        ,  1.        ,  1.        ])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAekAAAFVCAYAAADLxheZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x56f7c90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = np.linspace(-5, 5, 1000)\n",
    "y = np.tanh(x)\n",
    "plt.plot(x, y)\n",
    "#sns.regplot(y=\"y\", x=\"x\", data={\"x\":x, \"y\":y},  scatter=False, order=5)\n",
    "plt.savefig(\"tanh.png\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdsAAAFVCAYAAACnwEWcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xb105f50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = np.linspace(-5, 5, 1000)\n",
    "\n",
    "y = np.array([max(a, 0) for a in list(x)])\n",
    "plt.plot(x, y)\n",
    "#sns.regplot(y=\"y\", x=\"x\", data={\"x\":x, \"y\":y},  scatter=False, order=5)\n",
    "plt.savefig(\"Relu.png\")\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}