{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# PyOpenCL Parallel Patterns: Reduction"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Setup Code"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "import pyopencl as cl\n",
        "import pyopencl.array\n",
        "import pyopencl.clrandom\n",
        "import numpy as np"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "ctx = cl.create_some_context()\n",
        "queue = cl.CommandQueue(ctx)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "n = 10**7\n",
        "x = cl.clrandom.rand(queue, n, np.float64)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Setting up the kernel: Computing a sum of squares\n",
        "\n",
        "Want to compute the sum of the squares of all entries in `x`.\n",
        "\n",
        "First, using `numpy`, as `result1` (watch out: `.get()`)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "result1 = np.sum(x.get()**2)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Then, using PyOpenCL:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "from pyopencl.reduction import ReductionKernel"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Syntax:\n",
        "\n",
        "ReductionKernel(context, dtype, netural, reduce_expr, map_expr, arguments)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "rknl = ReductionKernel(ctx, np.float64,\n",
        "        neutral=\"0\",\n",
        "        reduce_expr=\"a+b\", map_expr=\"x[i]*x[i]\",\n",
        "        arguments=\"double *x\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## Testing the result"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "collapsed": false
      },
      "outputs": [],
      "source": [
        "result2 = rknl(x)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "collapsed": false
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "pyopencl.array.Array"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "type(result2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "collapsed": false
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "()"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "result2.shape"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "Now check the result:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "collapsed": false
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "9.31322574615e-10\n"
          ]
        }
      ],
      "source": [
        "print(result2.get()-result1)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "* Change this to find maximum.\n",
        "* Works on structured types, too.\n",
        "* What if you wanted to find maximum *and* location?"
      ]
    },
    {
      "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": "3.5.1+"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}