Parallel GridSearchCV with Custom Transformer.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[mean: 0.95960, std: 0.03299, params: {'rf__n_estimators': 10},\n",
       " mean: 0.95973, std: 0.04439, params: {'rf__n_estimators': 100}]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "iris = load_iris()\n",
    "\n",
    "X = iris[\"data\"]\n",
    "y = iris[\"target\"]\n",
    "\n",
    "class DummyTransformer(BaseEstimator, TransformerMixin):\n",
    "    def fit(self, X, y=None):\n",
    "        return self\n",
    "    def transform(self, X):\n",
    "        return X\n",
    "\n",
    "cv = GridSearchCV(estimator=Pipeline(steps=[('dummy', DummyTransformer()),\n",
    "                                            ('rf', RandomForestClassifier())]),\n",
    "                  param_grid={\"rf__n_estimators\": [10, 100]},\n",
    "                  scoring=\"f1_weighted\",\n",
    "                  cv=10,\n",
    "                  n_jobs=2) # n_jobs = 1 works fine, but setting n_jobs = 2 makes the script run forever... :-(\n",
    "cv.fit(X, y)\n",
    "\n",
    "cv.grid_scores_"
   ]
  }
 ],
 "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
}