MultiModelSearch_test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import datasets\n",
    "from sklearn.model_selection import MultiModelSearch\n",
    "\n",
    "#iris = datasets.load_iris()\n",
    "#X = iris.data\n",
    "#y = iris.target\n",
    "X, y = datasets.make_moons(n_samples=1000, noise=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import (ExtraTreesClassifier, RandomForestClassifier, \n",
    "                              AdaBoostClassifier, GradientBoostingClassifier)\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "models = { \n",
    "    'ExtraTreesClassifier': ExtraTreesClassifier(),\n",
    "    'RandomForestClassifier': RandomForestClassifier(),\n",
    "    'AdaBoostClassifier': AdaBoostClassifier(),\n",
    "    'GradientBoostingClassifier': GradientBoostingClassifier(),\n",
    "    'SVC': SVC(),\n",
    "}\n",
    "\n",
    "param_sets = { \n",
    "    'ExtraTreesClassifier': { 'n_estimators': [16, 32] },\n",
    "    'RandomForestClassifier': { 'n_estimators': [16, 32] },\n",
    "    'AdaBoostClassifier':  { 'n_estimators': [16, 32] },\n",
    "    'GradientBoostingClassifier': { 'n_estimators': [16, 32], 'learning_rate': [0.8, 1.0] },\n",
    "    'SVC': [\n",
    "        {'kernel': ['linear'], 'C': [1, 10]},\n",
    "        {'kernel': ['rbf'], 'C': [1, 10], 'gamma': [0.001, 0.0001]},\n",
    "    ]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AdaBoostClassifier(algorithm='SAMME.R', base_estimator=None,\n",
      "          learning_rate=1.0, n_estimators=16, random_state=None)\n"
     ]
    }
   ],
   "source": [
    "multi_grid = MultiModelSearch(models, param_sets, \n",
    "                              scoring='accuracy', n_jobs=-1,\n",
    "                             )\n",
    "multi_grid.fit(X, y)\n",
    "\n",
    "print(multi_grid.best_estimator_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'AdaBoostClassifier': {'param_n_estimators': masked_array(data = [16 32],\n",
       "               mask = [False False],\n",
       "         fill_value = ?),\n",
       "  'params': ({'n_estimators': 16}, {'n_estimators': 32}),\n",
       "  'test_mean_score': array([ 0.803,  0.796]),\n",
       "  'test_rank_score': array([1, 2], dtype=int32),\n",
       "  'test_split0_score': array([ 0.81736527,  0.79341317]),\n",
       "  'test_split1_score': array([ 0.79640719,  0.7994012 ]),\n",
       "  'test_split2_score': array([ 0.79518072,  0.79518072]),\n",
       "  'test_std_score': array([ 0.01018532,  0.00251428])},\n",
       " 'ExtraTreesClassifier': {'param_n_estimators': masked_array(data = [16 32],\n",
       "               mask = [False False],\n",
       "         fill_value = ?),\n",
       "  'params': ({'n_estimators': 16}, {'n_estimators': 32}),\n",
       "  'test_mean_score': array([ 0.779,  0.772]),\n",
       "  'test_rank_score': array([1, 2], dtype=int32),\n",
       "  'test_split0_score': array([ 0.79041916,  0.79041916]),\n",
       "  'test_split1_score': array([ 0.76946108,  0.75449102]),\n",
       "  'test_split2_score': array([ 0.77710843,  0.77108434]),\n",
       "  'test_std_score': array([ 0.00866785,  0.01469645])},\n",
       " 'GradientBoostingClassifier': {'param_learning_rate': masked_array(data = [0.8 0.8 1.0 1.0],\n",
       "               mask = [False False False False],\n",
       "         fill_value = ?),\n",
       "  'param_n_estimators': masked_array(data = [16 32 16 32],\n",
       "               mask = [False False False False],\n",
       "         fill_value = ?),\n",
       "  'params': ({'learning_rate': 0.8, 'n_estimators': 16},\n",
       "   {'learning_rate': 0.8, 'n_estimators': 32},\n",
       "   {'learning_rate': 1.0, 'n_estimators': 16},\n",
       "   {'learning_rate': 1.0, 'n_estimators': 32}),\n",
       "  'test_mean_score': array([ 0.791,  0.766,  0.775,  0.759]),\n",
       "  'test_rank_score': array([1, 3, 2, 4], dtype=int32),\n",
       "  'test_split0_score': array([ 0.77844311,  0.77844311,  0.7754491 ,  0.76047904]),\n",
       "  'test_split1_score': array([ 0.7754491 ,  0.75449102,  0.77844311,  0.75449102]),\n",
       "  'test_split2_score': array([ 0.81927711,  0.76506024,  0.77108434,  0.76204819]),\n",
       "  'test_std_score': array([ 0.01997249,  0.00981057,  0.00301949,  0.00325668])},\n",
       " 'RandomForestClassifier': {'param_n_estimators': masked_array(data = [16 32],\n",
       "               mask = [False False],\n",
       "         fill_value = ?),\n",
       "  'params': ({'n_estimators': 16}, {'n_estimators': 32}),\n",
       "  'test_mean_score': array([ 0.779,  0.783]),\n",
       "  'test_rank_score': array([2, 1], dtype=int32),\n",
       "  'test_split0_score': array([ 0.79341317,  0.79341317]),\n",
       "  'test_split1_score': array([ 0.76347305,  0.76946108]),\n",
       "  'test_split2_score': array([ 0.78012048,  0.78614458]),\n",
       "  'test_std_score': array([ 0.01226069,  0.01003608])},\n",
       " 'SVC': {'param_C': masked_array(data = [1 10 1 1 10 10],\n",
       "               mask = [False False False False False False],\n",
       "         fill_value = ?),\n",
       "  'param_gamma': masked_array(data = [-- -- 0.001 0.0001 0.001 0.0001],\n",
       "               mask = [ True  True False False False False],\n",
       "         fill_value = ?),\n",
       "  'param_kernel': masked_array(data = ['linear' 'linear' 'rbf' 'rbf' 'rbf' 'rbf'],\n",
       "               mask = [False False False False False False],\n",
       "         fill_value = ?),\n",
       "  'params': ({'C': 1, 'kernel': 'linear'},\n",
       "   {'C': 10, 'kernel': 'linear'},\n",
       "   {'C': 1, 'gamma': 0.001, 'kernel': 'rbf'},\n",
       "   {'C': 1, 'gamma': 0.0001, 'kernel': 'rbf'},\n",
       "   {'C': 10, 'gamma': 0.001, 'kernel': 'rbf'},\n",
       "   {'C': 10, 'gamma': 0.0001, 'kernel': 'rbf'}),\n",
       "  'test_mean_score': array([ 0.799,  0.799,  0.769,  0.773,  0.789,  0.77 ]),\n",
       "  'test_rank_score': array([1, 1, 6, 4, 3, 5], dtype=int32),\n",
       "  'test_split0_score': array([ 0.79041916,  0.79041916,  0.74850299,  0.78143713,  0.77844311,\n",
       "          0.74850299]),\n",
       "  'test_split1_score': array([ 0.80239521,  0.80239521,  0.77844311,  0.75748503,  0.79640719,\n",
       "          0.78143713]),\n",
       "  'test_split2_score': array([ 0.80421687,  0.80421687,  0.78012048,  0.78012048,  0.79216867,\n",
       "          0.78012048]),\n",
       "  'test_std_score': array([ 0.00612197,  0.00612197,  0.01453145,  0.01100033,  0.00767349,\n",
       "          0.01523296])}}"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "multi_grid._results"
   ]
  }
 ],
 "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": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}