GridSearchCV memory usage

gistfile1.txt

{
 "metadata": {
  "name": "",
  "signature": "sha256:dfb7fdbd72babce26c85e0d1420b1e20600c1dd13eb385470e9a3bc420f2bf64"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from sklearn import grid_search, linear_model\n",
      "import numpy as np\n",
      "%load_ext memory_profiler"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "X = np.random.randn(24000, 2000)\n",
      "w = np.random.randn(2000)\n",
      "y = np.dot(X, w)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "heading",
     "level": 1,
     "metadata": {},
     "source": [
      "lsqr"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "rid = linear_model.Ridge(solver='lsqr')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit rid.fit(X, y)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 800.43 MiB, increment: 367.09 MiB\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit [rid.fit(X, y) for _ in range(4)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 1899.48 MiB, increment: 1465.39 MiB\n"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit [rid.fit(X, y) for _ in range(8)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 2726.49 MiB, increment: 2291.86 MiB\n"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "sparse_cg"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "rid = linear_model.Ridge(solver='sparse_cg')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 8
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit rid.fit(X, y)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 789.27 MiB, increment: 366.42 MiB\n"
       ]
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit [rid.fit(X, y) for _ in range(4)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 1888.10 MiB, increment: 1464.92 MiB\n"
       ]
      }
     ],
     "prompt_number": 10
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit [rid.fit(X, y) for _ in range(8)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 3352.96 MiB, increment: 2929.69 MiB\n"
       ]
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "heading",
     "level": 2,
     "metadata": {},
     "source": [
      "Cholesky"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "rid = linear_model.Ridge(solver='cholesky')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit rid.fit(X, y)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 859.59 MiB, increment: 436.30 MiB\n"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit [rid.fit(X, y) for _ in range(4)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 859.64 MiB, increment: 366.24 MiB\n"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "%memit [rid.fit(X, y) for _ in range(8)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "peak memory: 859.64 MiB, increment: 366.21 MiB\n"
       ]
      }
     ],
     "prompt_number": 15
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}