{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Przykład losowego podziału danych na zbiór uczący oraz testujący:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.cross_validation import train_test_split\n",
    "data = np.arange(30).reshape(-1,3)\n",
    "labels = np.arange(10)\n",
    "data_train, data_test, labels_train, labels_test = train_test_split(data, labels, test_size=0.25, random_state=1234)\n",
    "print(data)\n",
    "print(labels)\n",
    "print(data_train)\n",
    "print(data_test)\n",
    "print(labels_train)\n",
    "print(labels_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Przykład zastosowania k-krotnej walidacji:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "iris = datasets.load_iris()\n",
    "knn = KNeighborsClassifier()\n",
    "scores = cross_val_score(knn, iris.data, iris.target, cv=5)\n",
    "print(scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Badanie wpływu parametru klasyfikatora na wyniki:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "scores = []\n",
    "n = []\n",
    "for i in range(1,30):\n",
    "    knn.n_neighbors = i\n",
    "    n.append(i)\n",
    "    scores.append(cross_val_score(knn, iris.data, iris.target, cv=5).mean())\n",
    "    print(n[-1], scores[-1])\n",
    "plt.plot(n, scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Szukanie najlepszego zestawu parametrów z użyciem k-krotnej walidacji:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import svm, grid_search\n",
    "parameters = {'n_neighbors': list(range(1, 30)), 'p':[1,2,3,4] }\n",
    "clf = grid_search.GridSearchCV(knn, parameters, cv=5)\n",
    "clf.fit(iris.data, iris.target)\n",
    "print(clf.best_estimator_)\n",
    "print(clf.best_params_)\n",
    "print(clf.best_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": "3.4.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}