{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Validation curves\n",
    "\n",
    "See discussion at: http://scikit-learn.org/stable/modules/learning_curve.html\n",
    "\n",
    "Try out the example (almost as given in the documentation):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.model_selection import validation_curve\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.linear_model import Ridge\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "np.random.seed(0)\n",
    "iris = load_iris()\n",
    "X, y = iris.data, iris.target\n",
    "indices = np.arange(y.shape[0])\n",
    "np.random.shuffle(indices)\n",
    "X, y = X[indices], y[indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "param_range = np.logspace(-7, 3, 3)\n",
    "train_scores, test_scores = validation_curve(Ridge(), X, y, \"alpha\", param_range)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_scores_mean = np.mean(train_scores, axis=1)\n",
    "train_scores_std = np.std(train_scores, axis=1)\n",
    "test_scores_mean = np.mean(test_scores, axis=1)\n",
    "test_scores_std = np.std(test_scores, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEaCAYAAAD+E0veAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11ac4f2d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title(\"Validation Curve with Ridge regression\")\n",
    "plt.xlabel(\"alpha\")\n",
    "plt.ylabel(\"Score\")\n",
    "plt.ylim(0.0, 1.1)\n",
    "plt.semilogx(param_range, train_scores_mean, label=\"Training score\", color=\"r\")\n",
    "plt.fill_between(param_range, train_scores_mean - train_scores_std,\n",
    "                 train_scores_mean + train_scores_std, alpha=0.2, color=\"r\")\n",
    "plt.semilogx(param_range, test_scores_mean, label=\"Cross-validation score\",\n",
    "             color=\"g\")\n",
    "plt.fill_between(param_range, test_scores_mean - test_scores_std,\n",
    "                 test_scores_mean + test_scores_std, alpha=0.2, color=\"g\")\n",
    "plt.legend(loc=\"best\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "See another example at http://scikit-learn.org/stable/auto_examples/model_selection/plot_validation_curve.html#sphx-glr-auto-examples-model-selection-plot-validation-curve-py"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "** TODO: ** select one problem from the examples above and use k-NN to plot the \"validation curve\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Learning curves\n",
    "Execute the example at:\n",
    "\n",
    "http://scikit-learn.org/stable/auto_examples/model_selection/plot_learning_curve.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}