eventrelated_and_fxmapper.json

{
 "metadata": {
  "name": "TestAnalysis"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Some setup..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from parseBehav import parseBhLog\n",
      "from mvpa2.suite import *\n",
      "\n",
      "PROCS_OF_INTEREST = ['MixedBivalent1', 'MixedBivalent2', 'MixedBivalent3',\n",
      "                     'MixedUnivalent1', 'MixedUnivalent2', 'MixedUnivalent3']\n",
      "\n",
      "FILES = [\"testbold%02d.nii\" % n for n in xrange(1,11)]\n",
      "\n",
      "sas, events = parseBhLog('./testbehav.txt')\n",
      "events = [ev for ev in events if ev['ev_block_proc'] in PROCS_OF_INTEREST]\n",
      "\n",
      "ds = fmri_dataset(FILES, mask='testroi.nii')\n",
      "\n",
      "for key in sas.keys(): ds.sa[key] = sas[key]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Vanilla event related dataset..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "dse = eventrelated_dataset(ds,events=events)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print dse.nsamples\n",
      "print dse.nfeatures\n",
      "vi = dse.fa['voxel_indices']\n",
      "print vi[0]\n",
      "# all voxel indices repeated twice?\n",
      "nFeats = int(floor(dse.nfeatures/2))\n",
      "print all([vi[i] == vi[nFeats+i] for i in xrange(nFeats)])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "192\n",
        "1140\n",
        "[23 31  6]\n",
        "True\n"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This all looks good."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now let's use FxMapper to do a little temporal compression..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fx = FxMapper(axis='features', fx=np.mean)\n",
      "dse2 = eventrelated_dataset(ds,events=events,event_mapper=fx)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print dse2.nsamples\n",
      "print dse2.nfeatures\n",
      "# all voxel indices equal to mean of same voxels from\n",
      "# the two volumes?\n",
      "nSamps =  dse2.nsamples\n",
      "s1 = dse.samples\n",
      "s2 = dse2.samples\n",
      "print all([[s2[n,i] == np.mean([s1[n,i], s1[n,nFeats+i]]) for i in xrange(nFeats)] for n in xrange(nSamps)])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "192\n",
        "570\n",
        "True"
       ]
      },
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "\n"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This looks good too. Everything is averaged correctly."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's check to see if the voxel indices are intact..."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print dse.fa['voxel_indices'][0]\n",
      "print dse2.fa['voxel_indices'][0]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "[23 31  6]\n",
        "6+7+8+9+10+11+12+13+14+15+16+17+18+19+20+21+22+23+24+25+26+27+28+29+30+31+32+33+34+35+36+37+38+39+40+41\n"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Okay, lost me there. Where does this value come from?"
     ]
    }
   ],
   "metadata": {}
  }
 ]
}