PyFITS_Intro.ipynb

{
 "metadata": {
  "name": ""
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "This notebook will be a basic tutorial on how FITS (Flexible Image Transport System) format images can be manipulated. The first step is to load the required Python package."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# This imports the pyfits package.\n",
      "\n",
      "import pyfits "
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 20
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "The next step is to open a FITS file. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fitsfile = pyfits.open(\"PSF.fits\")"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 21
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "We have opened the FITS file, but what precisely is it? Printing its type can give as an idea."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print type(fitsfile)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "<class 'pyfits.hdu.hdulist.HDUList'>\n"
       ]
      }
     ],
     "prompt_number": 22
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "The above tells us that the contents of the fitsfile variable is an HDU (Header Data Unit) list object. This is simpler than it sounds. It is basically a list of HDUs. An HDU usually consists of a header and some data. A header, like a dictionary, contains keys and assosciated values which are relevant to the data. Before trying to access either the header or the data, it is a good idea to know a little more about the HDU list."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# This prints some useful information about our HDU list.\n",
      "\n",
      "fitsfile.info()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "Filename: PSF.fits\n",
        "No.    Name         Type      Cards   Dimensions   Format\n",
        "0    PRIMARY     PrimaryHDU      71   (2048, 2048, 1, 1)   float32   \n"
       ]
      }
     ],
     "prompt_number": 23
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "We can now see that our HDU list only has one element, which is assigned the index 0. Note that in Python most things are indexed from 0. The cards value indicated the number of entries in the header. This can be thought of as the number of keys in a dictionary. The dimensions value gives the dimensions of the data assosicated with the HDU. Note, that this value is displayed in reverse order here, and the first two values refer to columns and rows respectively.\n",
      "\n",
      "The next step is to actually separate the header and the data and load them into variables."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# We know which element in the HDU list we want, so we access it explicitly.\n",
      "\n",
      "fitsheader = fitsfile[0].header\n",
      "fitsdata   = fitsfile[0].data"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 24
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "Let's verify that we got what we wanted."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print type(fitsheader)\n",
      "print type(fitsdata)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "<class 'pyfits.header.Header'>\n",
        "<type 'numpy.ndarray'>\n"
       ]
      }
     ],
     "prompt_number": 25
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "A quick thing to note - by opening a FITS file as shown above and assigning its contents to variables, the data we need is now in memory. It is not necessary to keep the file open once we have the desired contents."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# This is a good habit ot get into.\n",
      "\n",
      "fitsfile.close()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 26
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "Now we have the pieces of out FITS file in a form which is easier to use. Lets start by looking at our data. It is simply a numpy array, and thus all the usual numpy syntax applies. To use it, however, we want to import numpy. This should be done at the beginning, but we can get away with importing it here."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# I alias numpy as np for ease of use.\n",
      "\n",
      "import numpy as np\n",
      "\n",
      "# Verify our data shape.\n",
      "\n",
      "print fitsdata.shape"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "(1, 1, 2048, 2048)\n"
       ]
      }
     ],
     "prompt_number": 27
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "This agrees with the shape we saw in the fitsfile.info() output we saw earlier. Note again how the order is different here.\n",
      "\n",
      "This introduction will not explore numpy in detail, but slicing the data is useful for plotting. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# We only want the last two dimensions - the rows and columns which make up a single image.\n",
      "\n",
      "dataslice = fitsdata[0,0,:,:]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 28
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "Plotting the data is always useful. I tend to import pylab for this purpose. Again, this should be done at the beginning, but we can get away with it here."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# I alias pylab as pb for ease of use. The second command simply enables inline plotting for the purposes of this notebook.\n",
      "\n",
      "import pylab as pb\n",
      "%matplotlib inline\n",
      "\n",
      "# Simple colour-map plot commands.\n",
      "\n",
      "pb.imshow(dataslice)\n",
      "pb.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAARAAAAEACAYAAAB73CmOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f2f655ba690>"
       ]
      }
     ],
     "prompt_number": 29
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# We can even slice on the fly, effectively creating a zoomed image.\n",
      "\n",
      "pb.imshow(dataslice[768:1280,768:1280])\n",
      "pb.show()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAQcAAAEACAYAAAC+rrMfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "text": [
        "<matplotlib.figure.Figure at 0x7f2f63344410>"
       ]
      }
     ],
     "prompt_number": 30
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "There is not much more to be said about the data from a technical perspective. A few final point will be touched on when we discuss saving a FITS file.\n",
      "\n",
      "We return to our header. Now that we know the basics, it is possible to elaborate a little on the structure of the header. It is not quite a dictionary, but can be thought of in a similar fashion. A header is made up of cards. Each card contains a keyword, a value, and a comment. The keyword and value are identical to the more familar Python dictionary. The comment is an extra field which allows desriptive text strings to be added to the header. Let start by printing the header."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# The repr command just makes the output more readable.\n",
      "\n",
      "print repr(fitsheader)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "SIMPLE  =                    T /Standard FITS                                   \n",
        "BITPIX  =                  -32 /Floating point (32 bit)                         \n",
        "NAXIS   =                    4                                                  \n",
        "NAXIS1  =                 2048                                                  \n",
        "NAXIS2  =                 2048                                                  \n",
        "NAXIS3  =                    1                                                  \n",
        "NAXIS4  =                    1                                                  \n",
        "BSCALE  =   1.000000000000E+00 /PHYSICAL = PIXEL*BSCALE + BZERO                 \n",
        "BZERO   =   0.000000000000E+00                                                  \n",
        "BTYPE   = 'Intensity'                                                           \n",
        "OBJECT  = 'BEAM_0  '                                                            \n",
        "                                                                                \n",
        "BUNIT   = 'JY/BEAM '           /Brightness (pixel) unit                         \n",
        "EQUINOX =   2.000000000000E+03                                                  \n",
        "LONPOLE =   1.800000000000E+02                                                  \n",
        "LATPOLE =  -4.000000000000E+01                                                  \n",
        "PC001001=   1.000000000000E+00                                                  \n",
        "PC002001=   0.000000000000E+00                                                  \n",
        "PC003001=   0.000000000000E+00                                                  \n",
        "PC004001=   0.000000000000E+00                                                  \n",
        "PC001002=   0.000000000000E+00                                                  \n",
        "PC002002=   1.000000000000E+00                                                  \n",
        "PC003002=   0.000000000000E+00                                                  \n",
        "PC004002=   0.000000000000E+00                                                  \n",
        "PC001003=   0.000000000000E+00                                                  \n",
        "PC002003=   0.000000000000E+00                                                  \n",
        "PC003003=   1.000000000000E+00                                                  \n",
        "PC004003=   0.000000000000E+00                                                  \n",
        "PC001004=   0.000000000000E+00                                                  \n",
        "PC002004=   0.000000000000E+00                                                  \n",
        "PC003004=   0.000000000000E+00                                                  \n",
        "PC004004=   1.000000000000E+00                                                  \n",
        "CTYPE1  = 'RA---SIN'                                                            \n",
        "CRVAL1  =   0.000000000000E+00                                                  \n",
        "CDELT1  =  -2.777777777778E-04                                                  \n",
        "CRPIX1  =   1.025000000000E+03                                                  \n",
        "CUNIT1  = 'deg     '                                                            \n",
        "CTYPE2  = 'DEC--SIN'                                                            \n",
        "CRVAL2  =  -4.000000000000E+01                                                  \n",
        "CDELT2  =   2.777777777778E-04                                                  \n",
        "CRPIX2  =   1.025000000000E+03                                                  \n",
        "CUNIT2  = 'deg     '                                                            \n",
        "CTYPE3  = 'STOKES  '                                                            \n",
        "CRVAL3  =   1.000000000000E+00                                                  \n",
        "CDELT3  =   1.000000000000E+00                                                  \n",
        "CRPIX3  =   1.000000000000E+00                                                  \n",
        "CUNIT3  = '        '                                                            \n",
        "CTYPE4  = 'FREQ    '                                                            \n",
        "CRVAL4  =   1.014581950000E+09                                                  \n",
        "CDELT4  =   3.900000000000E+03                                                  \n",
        "CRPIX4  =   1.000000000000E+00                                                  \n",
        "CUNIT4  = 'HZ      '                                                            \n",
        "PV2_1   =   0.000000000000E+00                                                  \n",
        "PV2_2   =   0.000000000000E+00                                                  \n",
        "RESTFREQ=   1.014581950000E+09 /Rest Frequency (Hz)                             \n",
        "ALTRVAL =  -0.000000000000E+00 /Alternate frequency reference value             \n",
        "ALTRPIX =   1.000000000000E+00 /Alternate frequency reference pixel             \n",
        "VELREF  =                    3 /1 LSR, 2 HEL, 3 OBS, +256 Radio                 \n",
        "COMMENT casacore non-standard usage: 4 LSD, 5 GEO, 6 SOU, 7 GAL                 \n",
        "TELESCOP= 'LOFAR   '                                                            \n",
        "OBSERVER= 'MSCreate'                                                            \n",
        "DATE-OBS= '2011-11-16T15:00:00.000000'                                          \n",
        "TIMESYS = 'UTC     '                                                            \n",
        "OBSRA   =   0.000000000000E+00                                                  \n",
        "OBSDEC  =  -4.000000000000E+01                                                  \n",
        "OBJECT  = 'BEAM_0  '                                                            \n",
        "TELESCOP= 'LOFAR   '                                                            \n",
        "INSTRUME= 'LOFAR   '                                                            \n",
        "DISTANCE=   0.000000000000E+00                                                  \n",
        "DATE    = '2014-02-02T20:41:19.161000' /Date FITS file was written              \n",
        "ORIGIN  = 'CASA casacore alma-evla '                                            \n"
       ]
      }
     ],
     "prompt_number": 31
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "We can now see the whole header. This is useful for when the contents of the header are unknown. The format of each card is also clear in the above output: keyword = value /comment\n",
      "\n",
      "However, this output can also be a bit overwhelming, so sometimes you only want the keywords."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# This will print only the keywords.\n",
      "\n",
      "print fitsheader.keys()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "['SIMPLE', 'BITPIX', 'NAXIS', 'NAXIS1', 'NAXIS2', 'NAXIS3', 'NAXIS4', 'BSCALE', 'BZERO', 'BTYPE', 'OBJECT', '', 'BUNIT', 'EQUINOX', 'LONPOLE', 'LATPOLE', 'PC001001', 'PC002001', 'PC003001', 'PC004001', 'PC001002', 'PC002002', 'PC003002', 'PC004002', 'PC001003', 'PC002003', 'PC003003', 'PC004003', 'PC001004', 'PC002004', 'PC003004', 'PC004004', 'CTYPE1', 'CRVAL1', 'CDELT1', 'CRPIX1', 'CUNIT1', 'CTYPE2', 'CRVAL2', 'CDELT2', 'CRPIX2', 'CUNIT2', 'CTYPE3', 'CRVAL3', 'CDELT3', 'CRPIX3', 'CUNIT3', 'CTYPE4', 'CRVAL4', 'CDELT4', 'CRPIX4', 'CUNIT4', 'PV2_1', 'PV2_2', 'RESTFREQ', 'ALTRVAL', 'ALTRPIX', 'VELREF', 'COMMENT', 'TELESCOP', 'OBSERVER', 'DATE-OBS', 'TIMESYS', 'OBSRA', 'OBSDEC', 'OBJECT', 'TELESCOP', 'INSTRUME', 'DISTANCE', 'DATE', 'ORIGIN']\n"
       ]
      }
     ],
     "prompt_number": 32
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "We often only want to access a single value."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# To access a value, the syntax is the same as for a Python dictionary.\n",
      "\n",
      "print fitsheader[\"BUNIT\"]\n",
      "\n",
      "# If we want to access a comment, the syntax is a little different.\n",
      "\n",
      "print fitsheader.comments[\"BUNIT\"]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "JY/BEAM\n",
        "Brightness (pixel) unit\n"
       ]
      }
     ],
     "prompt_number": 33
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "It is very easy to update the value of a card."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Print the current value of the TELESCOP card.\n",
      "\n",
      "print fitsheader[\"TELESCOP\"]\n",
      "\n",
      "# Change its value.\n",
      "\n",
      "fitsheader[\"TELESCOP\"] = \"ATCA\"\n",
      "\n",
      "# Verify that it has been updated.\n",
      "\n",
      "print fitsheader[\"TELESCOP\"]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "LOFAR\n",
        "ATCA\n"
       ]
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "It is also very easy to add new cards."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Adding a card is a simple as specifying new key and assigning it a value.\n",
      "# header[key] = (value,comment) allows for comments.\n",
      "# header[key] = value is also correct when no comment is required. \n",
      "\n",
      "fitsheader[\"NEWFIELD\"] = (\"value\",\"comment\")\n",
      "\n",
      "# Verify that the new card has been added.\n",
      "\n",
      "print fitsheader[\"NEWFIELD\"]\n",
      "print fitsheader.comments[\"NEWFIELD\"]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "value\n",
        "comment\n"
       ]
      }
     ],
     "prompt_number": 35
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "The final step when manipulating a FITS file is the ability to save it. The syntax for this is quite simple. All that is required is a header and some data which is to be stored in the new FITS file. We are going to make use of our existing header. Remember how we changed the \"TELESCOP\" card? We are going to save that change. Additionally, we are going to save only the data which appeared in our zoomed image. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# The first step is to create a new primary HDU object which contains the header and data we want to save.\n",
      "\n",
      "newfits = pyfits.PrimaryHDU(dataslice[768:1280,768:1280], fitsheader)\n",
      "\n",
      "# Now we save this new object with a name of our choosing. Remember the extension. \n",
      "# clobber=True just allows this to overwrite existing files with the same name.\n",
      "\n",
      "newfits.writeto(\"MODPSF.fits\", clobber=True)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 36
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "We are almost at the end of the basics. The following is just a quick test to show that our save was successful, and that the header amended itself to reflect the new data size automatically."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# This is an example of the quick and dirty way of quickly seeing the contents of a header.\n",
      "# Do not use this approach in application code.\n",
      "\n",
      "print repr(pyfits.open(\"MODPSF.fits\")[0].header)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "SIMPLE  =                    T / conforms to FITS standard                      \n",
        "BITPIX  =                  -32 / array data type                                \n",
        "NAXIS   =                    2 / number of array dimensions                     \n",
        "NAXIS1  =                  512                                                  \n",
        "NAXIS2  =                  512                                                  \n",
        "BTYPE   = 'Intensity'                                                           \n",
        "OBJECT  = 'BEAM_0  '                                                            \n",
        "                                                                                \n",
        "BUNIT   = 'JY/BEAM '           /Brightness (pixel) unit                         \n",
        "EQUINOX =   2.000000000000E+03                                                  \n",
        "LONPOLE =   1.800000000000E+02                                                  \n",
        "LATPOLE =  -4.000000000000E+01                                                  \n",
        "PC001001=   1.000000000000E+00                                                  \n",
        "PC002001=   0.000000000000E+00                                                  \n",
        "PC003001=   0.000000000000E+00                                                  \n",
        "PC004001=   0.000000000000E+00                                                  \n",
        "PC001002=   0.000000000000E+00                                                  \n",
        "PC002002=   1.000000000000E+00                                                  \n",
        "PC003002=   0.000000000000E+00                                                  \n",
        "PC004002=   0.000000000000E+00                                                  \n",
        "PC001003=   0.000000000000E+00                                                  \n",
        "PC002003=   0.000000000000E+00                                                  \n",
        "PC003003=   1.000000000000E+00                                                  \n",
        "PC004003=   0.000000000000E+00                                                  \n",
        "PC001004=   0.000000000000E+00                                                  \n",
        "PC002004=   0.000000000000E+00                                                  \n",
        "PC003004=   0.000000000000E+00                                                  \n",
        "PC004004=   1.000000000000E+00                                                  \n",
        "CTYPE1  = 'RA---SIN'                                                            \n",
        "CRVAL1  =   0.000000000000E+00                                                  \n",
        "CDELT1  =  -2.777777777778E-04                                                  \n",
        "CRPIX1  =   1.025000000000E+03                                                  \n",
        "CUNIT1  = 'deg     '                                                            \n",
        "CTYPE2  = 'DEC--SIN'                                                            \n",
        "CRVAL2  =  -4.000000000000E+01                                                  \n",
        "CDELT2  =   2.777777777778E-04                                                  \n",
        "CRPIX2  =   1.025000000000E+03                                                  \n",
        "CUNIT2  = 'deg     '                                                            \n",
        "CTYPE3  = 'STOKES  '                                                            \n",
        "CRVAL3  =   1.000000000000E+00                                                  \n",
        "CDELT3  =   1.000000000000E+00                                                  \n",
        "CRPIX3  =   1.000000000000E+00                                                  \n",
        "CUNIT3  = '        '                                                            \n",
        "CTYPE4  = 'FREQ    '                                                            \n",
        "CRVAL4  =   1.014581950000E+09                                                  \n",
        "CDELT4  =   3.900000000000E+03                                                  \n",
        "CRPIX4  =   1.000000000000E+00                                                  \n",
        "CUNIT4  = 'HZ      '                                                            \n",
        "PV2_1   =   0.000000000000E+00                                                  \n",
        "PV2_2   =   0.000000000000E+00                                                  \n",
        "RESTFREQ=   1.014581950000E+09 /Rest Frequency (Hz)                             \n",
        "ALTRVAL =  -0.000000000000E+00 /Alternate frequency reference value             \n",
        "ALTRPIX =   1.000000000000E+00 /Alternate frequency reference pixel             \n",
        "VELREF  =                    3 /1 LSR, 2 HEL, 3 OBS, +256 Radio                 \n",
        "COMMENT casacore non-standard usage: 4 LSD, 5 GEO, 6 SOU, 7 GAL                 \n",
        "TELESCOP= 'ATCA    '                                                            \n",
        "OBSERVER= 'MSCreate'                                                            \n",
        "DATE-OBS= '2011-11-16T15:00:00.000000'                                          \n",
        "TIMESYS = 'UTC     '                                                            \n",
        "OBSRA   =   0.000000000000E+00                                                  \n",
        "OBSDEC  =  -4.000000000000E+01                                                  \n",
        "OBJECT  = 'BEAM_0  '                                                            \n",
        "TELESCOP= 'LOFAR   '                                                            \n",
        "INSTRUME= 'LOFAR   '                                                            \n",
        "DISTANCE=   0.000000000000E+00                                                  \n",
        "DATE    = '2014-02-02T20:41:19.161000' /Date FITS file was written              \n",
        "ORIGIN  = 'CASA casacore alma-evla '                                            \n",
        "NEWFIELD= 'value   '           / comment                                        \n"
       ]
      }
     ],
     "prompt_number": 37
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "All the changes we made have been saved. It is also possible to create a FITS file from scratch, though we do this less frequently as we often need all the cards to be filled in."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# We create some data.\n",
      "\n",
      "testdata = np.zeros([512,512])\n",
      "\n",
      "# We put the data in an HDU.\n",
      "\n",
      "testhdu = pyfits.PrimaryHDU(testdata)\n",
      "\n",
      "# And then we save it.\n",
      "\n",
      "testhdu.writeto(\"EMPTY.fits\", clobber=True)\n",
      "\n",
      "# And check what we have created.\n",
      "\n",
      "print repr(pyfits.open(\"EMPTY.fits\")[0].header)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "SIMPLE  =                    T / conforms to FITS standard                      \n",
        "BITPIX  =                  -64 / array data type                                \n",
        "NAXIS   =                    2 / number of array dimensions                     \n",
        "NAXIS1  =                  512                                                  \n",
        "NAXIS2  =                  512                                                  \n",
        "EXTEND  =                    T                                                  \n"
       ]
      }
     ],
     "prompt_number": 38
    },
    {
     "cell_type": "raw",
     "metadata": {},
     "source": [
      "As you can see, there isn't much in the FITS file. But we know how to add cards, and if you wanted to, you could add everything you wanted to.\n",
      "\n",
      "That concludes the basics of pyfits."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": []
    }
   ],
   "metadata": {}
  }
 ]
}