dask-sklearn-gridsearch.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"http://dask.readthedocs.io/en/latest/_images/dask_horizontal.svg\"\n",
    "     align=\"right\"\n",
    "     width=\"30%\"\n",
    "     alt=\"Dask logo\">\n",
    "\n",
    "\n",
    "Model Parallelism with SKLearn and Dask\n",
    "=======================\n",
    "\n",
    "\n",
    "<img src=\"https://avatars2.githubusercontent.com/u/365630?v=3&s=400\"\n",
    "     align=\"right\"\n",
    "     width=\"25%\"\n",
    "     alt=\"SKLearn logo\">\n",
    "\n",
    "\n",
    "*How do we choose the right parameters for a machine learning pipeline?*\n",
    "\n",
    "This notebook takes a standard example from the Scikit-Learn documententation and parallelizes it using a Dask-powered `GridSearchCV` function, which is a drop in replacement.  We achieve significant speedup on a cluster on an important problem just by changing an import.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### SKLearn example\n",
    "\n",
    "Taken from: http://scikit-learn.org/stable/auto_examples/plot_digits_pipe.html#sphx-glr-auto-examples-plot-digits-pipe-py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Code source: Gaël Varoquaux\n",
    "# Modified for documentation by Jaques Grobler\n",
    "# License: BSD 3 clause\n",
    "\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn import linear_model, decomposition, datasets\n",
    "from sklearn.pipeline import Pipeline\n",
    "from dklearn.model_selection import DaskGridSearchCV as GridSearchCV\n",
    "\n",
    "logistic = linear_model.LogisticRegression()\n",
    "\n",
    "pca = decomposition.PCA()\n",
    "pipe = Pipeline(steps=[('pca', pca), ('logistic', logistic)])\n",
    "\n",
    "digits = datasets.load_digits()\n",
    "X_digits = digits.data\n",
    "y_digits = digits.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "%%time\n",
    "\n",
    "n_components = [10, 20, 30, 40, 50, 64]\n",
    "Cs = np.logspace(-4, 4, 30)\n",
    "\n",
    "#Parameters of pipelines can be set using ‘__’ separated parameter names:\n",
    "\n",
    "estimator = GridSearchCV(pipe,\n",
    "                         dict(pca__n_components=n_components,\n",
    "                              logistic__C=Cs))\n",
    "estimator.fit(X_digits, y_digits)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "estimator.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from dask.distributed import Client\n",
    "c = Client('localhost:8786')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "c"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}