15_Spectroscopy_in_Python.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this lecture, I will demonstrate how to do the spectroscopic data reduction using python. Since the concepts of the tasks, i.e., the procedures what you should do are clear, you must be able to do it via any language you can use.\n",
    "\n",
    "## Literatures\n",
    "\n",
    "Some useful materials to understand spectroscopic data reduction:\n",
    "\n",
    "- [Wagner's writing in 1992](http://adsabs.harvard.edu/abs/1992ASPC...23..160W)\n",
    "  - As far as I know, this is the most comprehensive writing about spectroscopic data reduction, including theoretical background and practical usage (using IRAF).\n",
    "- [FOCAS cookbook](https://www.subarutelescope.org/Observing/DataReduction/Cookbooks/FOCAS_cookbook_2010jan05.pdf)\n",
    "  - This is written specifically for FOCAS, and is few of available English manual for that specific instrument. It gives you idea about how to reduce data using IRAF.\n",
    "  - Even though you are not using FOCAS, Sections 5, 6, and appendix B will generally be useful. Especially Section 5.5 (and 5.7) are very valuable.\n",
    "- Among [IRAF archive](http://iraf.noao.edu/iraf-ftp.html), the writing by [Massey at al in 1992](ftp://iraf.noao.edu/ftp/docs/spect.ps.Z) is THE official manual to do spectroscopic data reduction in IRAF.\n",
    "\n",
    "As you can see, all the important materials are made for IRAF, not Python. I could only find few informative packages, including [PyDIS](https://github.com/jradavenport/pydis) by Dr. James Davenport. There is a currently ongoing discussion about making spectroscopy package in astropy, called [specutils](https://github.com/astropy/specutils). The discussion about specutils are now going on as of November 2017 ([APE PR #28](https://github.com/astropy/astropy-APEs/pull/28)) in [\"Astropy Proposals for Enhancement\"](https://github.com/astropy/astropy-APEs). \n",
    "\n",
    "In short, there is currently no perfect python tools or packages to satisfactorily do the spectroscopic data reduction in python. So my basic philosophy in producing this material is\n",
    "\n",
    "1. Make explicit codes step by step so that every process is visible.\n",
    "   - This means I intentionally do not make package for all the processes. Only simple code snippets will be shown here.\n",
    "   - As I am not an expert in programming (and I believe a scientist sometimes *should not* consider too much about this technical issues), the codes are never carefully optimized. But they work at least.\n",
    "2. Make the code similar to IRAF.\n",
    "   - This is basically because IRAF is still the *de facto* standard for spectroscopic data reduction. If I try to make some different approach than IRAF, it may require significant effort from my side (which is impossible due to the lack of my ability), and I do believe that IRAF is very carefully designed.\n",
    "   \n",
    "In the following, I show my own code, and **it can be extremely slow compared to optimized codes**. I didn't care much about technical details (faster calculation, etc), basically due to the lack of my ability plus time limits.\n",
    "\n",
    "> The data which I used throughout this tutorial can be downloaded from my google drive link: [here](https://drive.google.com/open?id=0B-MLFRYnMxUvVS1jSjBLV1lOLWs)\n",
    ">\n",
    "> I tried to compare the step-by-step results with IRAF. The cl script used are only the followings. Save it as ``test.cl`` and type ``--> cl < test.cl`` on IRAF or PyRAF. Since it's tedious job to reproduce all my jobs, I uploaded database file which contains [identify](https://drive.google.com/open?id=0B-MLFRYnMxUvUTdJdG80V1VVYzg) and [aperture trace](https://drive.google.com/open?id=0B-MLFRYnMxUvWkxDa244X2E5MVU).\n",
    ">\n",
    ">    ```\n",
    ">    noao\n",
    ">    onedspec\n",
    ">    twodspec\n",
    ">    longslit\n",
    ">    apextract\n",
    ">    \n",
    ">    identify pcomp_medcomb section=\"middle line\" coordli='' funct=cheby order=4\n",
    ">    reidentify pcomp_medcomb pcomp_medcomb section=\"middle line\" override- trace- newaps- refit+ step=15 nsum=15 cradius=5 verbose+ \n",
    ">    apall pobj_mls170213_0025 find- resize- nsum=10 b_func=cheb b_order=3 b_nav=1 b_nit=5 t_nsum=10 t_step=10 t_func=cheb t_ord=3 t_nav=1 t_nit=5 background=fit skybox=1 pfit=fit1d\n",
    ">    fitcoords pcomp_medcomb interac+\n",
    ">    refspec pobj_mls170213_0025.ms reference=pcomp_medcomb sort='' group='' answer='yes'\n",
    ">    dispcor pobj_mls170213_0025.ms[*,1,1] final_0025.fits\n",
    ">    splot final_0025.fits\n",
    ">    ```\n",
    ">    **NOTE**: Be careful about re-doing any of the process! You MUST delete the pre-existing database files. Otherwise, IRAF will not re-write, but adds many lines to the pre-existing files...\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing\n",
    "\n",
    "Since this is the second semester for AO takers of SNU, I didn't spend time for preprocessing. Simply follow this:\n",
    "\n",
    "1. Download git to your computer from [here](https://git-scm.com/downloads)\n",
    "2. Clone [this repo](https://github.com/ysBach/AO2_tutorial) by, e.g., ``git clone https://github.com/ysBach/AO2_tutorial``\n",
    "3. Install by ``python3 setup.py install`` at the downloaded directory.\n",
    "4. There will never be an effort  to support python 2, so please use python 3.\n",
    "\n",
    "Then run the following code. All the parts should be very clear to you, if you finished AO1.\n",
    "\n",
    "```python\n",
    "import os\n",
    "import glob\n",
    "from astropy.io import fits\n",
    "from astropy.table import Table, Column\n",
    "import numpy as np\n",
    "from AO2tutorial.util import preproc, filemgmt\n",
    "\n",
    "#%%\n",
    "allfits = glob.glob(os.path.join('data', '*.fits'), recursive=True)\n",
    "\n",
    "# save one example header file\n",
    "os.chmod(allfits[0], 777)\n",
    "ex_hdr = fits.getheader(allfits[0])\n",
    "ex_hdr.totextfile('example_header.txt', overwrite=True)\n",
    "\n",
    "# Which cards to save? What are their corresponding data types?\n",
    "cards = ['DATE-OBS', 'NAXIS1', 'NAXIS2',\n",
    "         'XBINNING', 'YBINNING', 'EXPTIME', 'AIRMASS',\n",
    "         'OBJECT', 'GRATING', 'SLIT-WID', 'ORDERCUT']\n",
    "\n",
    "dtypes = ['U24', int, int,\n",
    "          int, int, float, float,\n",
    "          'U16', int, float, 'U16']\n",
    "\n",
    "# Initialize the table and fill it, save it.\n",
    "summarytab = Table(names=cards, dtype=dtypes)\n",
    "fnames = []\n",
    "\n",
    "for fitsfile in allfits:\n",
    "    os.chmod(fitsfile, 777)\n",
    "    fnames.append(fitsfile)    \n",
    "    hdr = fits.getheader(fitsfile)\n",
    "    row = []\n",
    "    for card in cards:\n",
    "        row.append(hdr[card])\n",
    "    summarytab.add_row(row)\n",
    "\n",
    "fnames = Column(data=fnames, name='file')\n",
    "summarytab.add_column(fnames, index=0)\n",
    "summarytab.sort('file')\n",
    "summarytab.write('summary.csv', format='ascii.csv', overwrite=True)\n",
    "\n",
    "'''\n",
    "# You may want to use\n",
    "from ccdproc import ImageFileCollection\n",
    "# set the keywords we are interested in.\n",
    "# Also, you don't have to use upper case but you **MUST be consistent** in the\n",
    "# later codes.\n",
    "cards = ['DATE-OBS', 'NAXIS1', 'NAXIS2',\n",
    "         'XBINNING', 'YBINNING', 'EXPTIME', 'AIRMASS',\n",
    "         'OBJECT', 'COMPLAMP', 'FLATLAMP', 'LAMP_SW']\n",
    "\n",
    "IC = ImageFileCollection(os.getcwd(),            # in current working directory\n",
    "#                         glob_include='FCSA*',   # files starts with 'FCSA'\n",
    "                         keywords=cards)         # only the predefined keys\n",
    "# the ``glob_include`` will be available from ``ccdproc`` version 1.3.\n",
    "# TODO: Change this part after ccdproc 1.3 release\n",
    "\n",
    "# Make a summary file.\n",
    "fits_tab = IC.summary\n",
    "fits_tab.write('summary.csv', format='ascii.csv', overwrite=True)\n",
    "'''\n",
    "#EXPTIME =                   5. / duration of exposure in seconds                \n",
    "#XBINNING=                    1 / binning factor used on X axis                  \n",
    "#YBINNING=                    1 / binning factor used on Y axis                  \n",
    "#DATE-OBS= '2017-02-13T21:59:43' / date of observation (UTC)                     \n",
    "#OBJECT  = 'DARK    '           / Target description                             \n",
    "#AIRMASS =                1.000 / Typical air mass during exposure               \n",
    "#GRATING =                  150 / gratOBJ_GRATING = 150 # lines per mming(l/mm)                                  \n",
    "#GRT_ANG =     29.7995882352941 / Grating angle(deg)                             \n",
    "#ORDERCUT= 'WG320   '           / Order cut filter Name                          \n",
    "#SLIT-WID=                  1.6 / Slit width(arcsec)                             \n",
    "#SLIT-LEN=                  300 / Slit length(arcsec)                            \n",
    "#COMPLAMP= '        '           / Comparison lamp(ON/OFF)                        \n",
    "#FLATLAMP= 'OFF     '           / Flat lamp(ON/OFF)                              \n",
    "#LAMP_SW = 'INSTFLAT'           / Selected lamp(COMPARISON/INSTFLAT)             \n",
    "\n",
    "#%%\n",
    "OBJ_SLITWID = 1.6 # arcsecond\n",
    "OBJ_ORDERCUT= 'WG320' # Order cut filter name\n",
    "OBJ_GRATING = 150 # lines per mm\n",
    "\n",
    "# http://www.nhao.jp/~malls/malls_wiki/index.php?MALLS\n",
    "# 150 lines/mm (3700A--9500A, wavelengh range = 5700A), R~600 (550nm, 1.2\")\n",
    "\n",
    "biastab = summarytab[((summarytab['OBJECT'] == 'DARK')\n",
    "                     & (summarytab['EXPTIME'] == 0.5 )) ]\n",
    "\n",
    "darktab = summarytab[((summarytab['OBJECT'] == 'DARK')\n",
    "                     & (summarytab['EXPTIME'] > 100 )) ]\n",
    "\n",
    "flattab = summarytab[((summarytab['OBJECT'] == 'flat')\n",
    "                     & (summarytab['GRATING'] == OBJ_GRATING)\n",
    "                     & (summarytab['SLIT-WID'] == OBJ_SLITWID)\n",
    "                     & (summarytab['ORDERCUT'] == OBJ_ORDERCUT))]\n",
    "\n",
    "bias_fname = 'bias.fits'\n",
    "dark_fname = 'dark1s.fits'\n",
    "flat_fname = 'flat.fits'\n",
    "\n",
    "if os.path.exists(bias_fname):\n",
    "    mbias = fits.getdata(bias_fname)\n",
    "else:\n",
    "    mbias = preproc.make_master_bias(biastab, sigma=3, iters=5, min_value=0,\n",
    "                                     output = bias_fname)\n",
    "\n",
    "if os.path.exists(dark_fname):\n",
    "    mdark = fits.getdata(dark_fname)    \n",
    "else:\n",
    "    mdark = preproc.make_master_dark(darktab, mbias=mbias, sigma=3, iters=5,\n",
    "                                     min_value=0, \n",
    "                                     output = dark_fname)\n",
    "if os.path.exists(flat_fname):\n",
    "    mflat = fits.getdata(flat_fname)\n",
    "else:\n",
    "    mflat = preproc.make_master_flat(flattab, mbias=mbias, sigma=3, iters=5,\n",
    "                                     min_value=5000, \n",
    "                                     output = flat_fname)\n",
    "\n",
    "#%%\n",
    "objtab = summarytab[( ((summarytab['OBJECT'] == '45P')\n",
    "                       |(summarytab['OBJECT'] == 'HD129184') )\n",
    "                     & (summarytab['AIRMASS'] < 1.5)) ]\n",
    "    \n",
    "comptab = summarytab[( (summarytab['OBJECT'] == 'comparison')\n",
    "                      & (summarytab['GRATING'] == OBJ_GRATING)\n",
    "                      &(summarytab['SLIT-WID'] == OBJ_SLITWID)\n",
    "                      &(summarytab['ORDERCUT'] == OBJ_ORDERCUT)) ]\n",
    "\n",
    "# Preprocess\n",
    "GAIN = 2.20 # electrons/ADU \n",
    "RONOISE = 12.3 # electrons\n",
    "# above from http://www.nhao.jp/~malls/malls_wiki/index.php?%BB%C5%CD%CD\n",
    "\n",
    "# I will trim the data with [350:1250,50:-50], where +- 50 is due to the overscan.\n",
    "TRIM = [350, 1250, 50, 2098] # y_lower, y_upper, x_lower, x_upper\n",
    "\n",
    "for fname in objtab['file']:\n",
    "    outputname = 'pobj_' + os.path.split(fname)[-1]\n",
    "    preproc.bdfgt_process(fname=fname,\n",
    "                          mbiasname=bias_fname,\n",
    "                          mdarkname=dark_fname,\n",
    "                          mflatname=flat_fname,\n",
    "                          mdark_seconds=1.0,\n",
    "                          trim=TRIM,\n",
    "                          gain=GAIN,\n",
    "                          exposure_key='EXPTIME',\n",
    "                          dtype=np.float32,\n",
    "                          output=outputname)\n",
    "    \n",
    "for fname in comptab['file']:\n",
    "    outputname = 'pcomp_' + os.path.split(fname)[-1]\n",
    "    preproc.bdfgt_process(fname=fname,\n",
    "                          mbiasname=bias_fname,\n",
    "                          mdarkname=dark_fname,\n",
    "                          mflatname=flat_fname,\n",
    "                          mdark_seconds=1.0,\n",
    "                          trim=TRIM,\n",
    "                          gain=GAIN,\n",
    "                          exposure_key='EXPTIME',\n",
    "                          dtype=np.float32,\n",
    "                          output=outputname)\n",
    "    \n",
    "#%%\n",
    "comp_orig = np.empty((TRIM[1] - TRIM[0], TRIM[3] - TRIM[2], 3), dtype=np.float32)\n",
    "header0 = fits.getheader(comptab['file'][0])\n",
    "\n",
    "for i in range(len(comptab['file'])):\n",
    "    pfname = 'pcomp_' + os.path.split(comptab['file'][i])[-1]\n",
    "    comp_orig[:, :, i] = fits.getdata(pfname).astype(np.float32)\n",
    "\n",
    "comp_comb = preproc.clip_median_combine(comp_orig, iters=0)\n",
    "fits.writeto('pcomp_medcomb.fits', data=comp_comb.data, header=header0, \n",
    "             output_verify='fix', overwrite=True)\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Postprocessing\n",
    "\n",
    "This is the tricky part. There is no single best way, so we need a lot of interaction with computers. That includes\n",
    "\n",
    "1. Wavelength identification (ID): First, you cut a small fraction of the comparison lamp image, and let the computer to calculate the wavelength $\\lambda = \\lambda(x)$ for given pixel value ``x``.\n",
    "2. Reidentification (REID): Then you must check this function makes sense throughout the whole image. REID checks it.\n",
    "3. Aperture trace (aptrace): The object spectrum cannot be perfectly parallel to the dispersion direction. Aptrace (in IRAF, ``APALL``) is to find how the source moves along the dispersion direction.\n",
    "4. Sky subtraction: The sky should be subtracted *at each wavelength*.\n",
    "5. Aperture sum: We then sum the pixel values within the selected aperture. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 0. Preparation\n",
    "\n",
    "We first load the data, and set some variables to be used:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "setting done!\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "\n",
    "from numpy.polynomial.chebyshev import chebfit, chebval\n",
    "\n",
    "%matplotlib notebook \n",
    "# comment above out when you copy this code to your spyder, etc.\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "from matplotlib import gridspec, rcParams, rc\n",
    "from matplotlib.widgets import Cursor\n",
    "\n",
    "from astropy.table import Table, Column\n",
    "from astropy.io import fits\n",
    "from astropy.stats import sigma_clip, gaussian_fwhm_to_sigma\n",
    "from astropy.modeling.models import Gaussian1D, Chebyshev2D\n",
    "from astropy.modeling.fitting import LevMarLSQFitter\n",
    "\n",
    "from skimage.feature import peak_local_max\n",
    "\n",
    "def disable_mplkeymaps():\n",
    "    rc('keymap', \n",
    "       fullscreen='',\n",
    "       home='',\n",
    "       back='',\n",
    "       forward='',\n",
    "       pan='',\n",
    "       zoom='',\n",
    "       save='',\n",
    "       quit='',\n",
    "       grid='',\n",
    "       yscale='',\n",
    "       xscale='',\n",
    "       all_axes=''\n",
    "       )\n",
    "#%%\n",
    "DISPAXIS = 1 # 1 = line = python_axis_1 // 2 = column = python_axis_0\n",
    "FONTSIZE = 12 # Change it on your computer if you wish.\n",
    "rcParams.update({'font.size': FONTSIZE})\n",
    "COMPIMAGE = os.path.join('45P', 'pcomp_medcomb.fits') # Change directory if needed!\n",
    "OBJIMAGE  = os.path.join('45P', 'pobj_mls170213_0025.fits')\n",
    "LINE_FITTER = LevMarLSQFitter()\n",
    "\n",
    "# Parameters for IDENTIFY\n",
    "FITTING_MODEL_ID = 'Chebyshev'\n",
    "ORDER_ID = 4 \n",
    "NSUM_ID = 10\n",
    "FWHM_ID = 4 # rough guess of FWHM of lines in IDENTIFY (pixels)\n",
    "\n",
    "# Parameters for REIDENTIFY\n",
    "FITTING_MODEL_REID = 'Chebyshev' # 2-D fitting function\n",
    "ORDER_SPATIAL_REID = 6\n",
    "ORDER_WAVELEN_REID = 6\n",
    "STEP_REID = 15  # Reidentification step size in pixels (spatial direction)\n",
    "NSUM_REID = 10\n",
    "TOL_REID = 5 # tolerence to lose a line in pixels\n",
    "\n",
    "# Parameters for APALL (sky fitting and aperture extract after sky subtraction)\n",
    "## parameters for finding aperture\n",
    "NSUM_AP = 10\n",
    "FWHM_AP = 10\n",
    "STEP_AP = 10  # Recentering step size in pixels (dispersion direction)\n",
    "## parameters for sky fitting\n",
    "FITTING_MODEL_APSKY = 'Chebyshev'\n",
    "ORDER_APSKY = 3\n",
    "SIGMA_APSKY = 3\n",
    "ITERS_APSKY = 5\n",
    "## parameters for aperture tracing\n",
    "FITTING_MODEL_APTRACE = 'Chebyshev'\n",
    "ORDER_APTRACE = 3\n",
    "SIGMA_APTRACE = 3\n",
    "ITERS_APTRACE = 5 \n",
    "# The fitting is done by SIGMA_APTRACE-sigma ITERS_APTRACE-iters clipped on the\n",
    "# residual of data. \n",
    "\n",
    "#%%\n",
    "lamphdu = fits.open(COMPIMAGE)\n",
    "objhdu = fits.open(OBJIMAGE)\n",
    "lampimage = lamphdu[0].data\n",
    "objimage  = objhdu[0].data\n",
    "\n",
    "if lampimage.shape != objimage.shape:\n",
    "    raise ValueError('lamp and obj images should have same sizes!')\n",
    "\n",
    "if DISPAXIS == 2:\n",
    "    lampimage = lampimage.T\n",
    "    objimage = objimage.T\n",
    "elif DISPAXIS != 1:\n",
    "    raise ValueError('DISPAXIS must be 1 or 2 (it is now {:d})'.format(DISPAXIS))\n",
    "\n",
    "EXPTIME = objhdu[0].header['EXPTIME']\n",
    "OBJNAME = objhdu[0].header['OBJECT']\n",
    "# Now python axis 0 (Y-direction) is the spatial axis \n",
    "# and 1 (X-direciton) is the wavelength (dispersion) axis.\n",
    "N_SPATIAL, N_WAVELEN = np.shape(lampimage)\n",
    "N_REID = N_SPATIAL//STEP_REID # No. of reidentification\n",
    "N_AP = N_WAVELEN//STEP_AP # No. of aperture finding\n",
    "\n",
    "# ``peak_local_max`` calculates the peak location using maximum filter:\n",
    "#   med1d_max = scipy.ndimage.maximum_filter(med1d, size=10, mode='constant')\n",
    "# I will use this to show peaks in a primitive manner.\n",
    "MINSEP_PK = 5   # minimum separation of peaks\n",
    "MINAMP_PK = 0.01 # fraction of minimum amplitude (wrt maximum) to regard as peak\n",
    "NMAX_PK = 50\n",
    "print(\"setting done!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the above code, I defined some parameters that will be used for the future codes. The variables with underbar is in the notation that ``parameter_task``, e.g., ``ORDER_ID`` = the order of fitting function for the task ID (identify). \n",
    "\n",
    "The most important thing I did is the part of ``DSIPAXIS``. I forced that ANY 2-D input data with proper ``DISPAXIS`` value to have the dispersion along x-axis (in python, ``axis=0`` is the Y-axis and ``axis=1`` is the X-axis, as many numerical languages: see [google search](https://www.google.co.kr/search?q=numpy+axis&newwindow=1&client=ubuntu&hs=KNf&channel=fs&dcr=0&source=lnms&tbm=isch&sa=X&ved=0ahUKEwj55aSH-53XAhXJUZQKHZ-IBNMQ_AUICigB&biw=1863&bih=909)). \n",
    "\n",
    "The meaning of other parameters will be clear in the following."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Identify\n",
    "\n",
    "We now do the line identification. That is, we find \"peaks\" from our preprocessed data of comparison lamps (``pcomp_medcomb``). First cut it along dispersion axis so that we can see many peaks:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"640\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# Identify (1): plot for manual input\n",
    "# =============================================================================\n",
    "# mimics IRAF IDENTIFY\n",
    "#   IDENTIIFY image.fits section='middle line' nsum=NSUM_ID\n",
    "lowercut_ID = N_SPATIAL//2 - NSUM_ID//2 \n",
    "uppercut_ID = N_SPATIAL//2 + NSUM_ID//2\n",
    "identify_1 = np.median(lampimage[lowercut_ID:uppercut_ID, :], axis=0)\n",
    "\n",
    "# For plot and visualization\n",
    "max_intens = np.max(identify_1)\n",
    "\n",
    "peak_pix = peak_local_max(identify_1, indices=True, num_peaks=NMAX_PK,\n",
    "                          min_distance=MINSEP_PK,\n",
    "                          threshold_abs=max_intens * MINAMP_PK)\n",
    "# ``peak_pix`` corresponds to the x value, since x = pixels starting from 0.\n",
    "\n",
    "disable_mplkeymaps()\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "title_str = r'Peak ($A \\geq {:.2f} A_\\mathrm{{max}}$, $\\Delta x \\geq {:.0f}$)'\n",
    "# Plot original spectrum + found peak locations\n",
    "x_identify = np.arange(0, len(identify_1))\n",
    "ax.plot(x_identify, identify_1, lw=1)\n",
    "\n",
    "for i in peak_pix:\n",
    "    ax.plot((i, i), \n",
    "            (identify_1[i]+0.01*max_intens, identify_1[i]+0.05*max_intens),\n",
    "            color='r', ls='-', lw=1)\n",
    "    ax.annotate(i[0], (i, 0.9),\n",
    "                xycoords = ('data', 'axes fraction'),\n",
    "                fontsize='xx-small', rotation=70)\n",
    "ax.grid(ls=':')\n",
    "ax.set_xlabel('Pixel number')\n",
    "ax.set_ylabel('Pixel value sum')\n",
    "ax.set_xlim(len(identify_1), 0) # to invert x-axis\n",
    "ax.set_title(title_str.format(MINAMP_PK, MINSEP_PK))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the plot, red vertical ticks are \"local maxima\" found by [``scikit.feature.peak_local_max``](http://scikit-image.org/docs/dev/api/skimage.feature.html#peak-local-max). On top, you can see the local peak in pixel numbers. They overlap a lot, so zoom in by yourself. \n",
    "\n",
    "* **QUESTION**: I intentionally inverted x-axis. Why do you think so?\n",
    "\n",
    "* **TIP**: I also used ``NSUM_ID=10``, i.e., I summed 10 pixels around the middle line of the comparison lamp image. Why not just one-pixel-width line, or taking median? It has been a customary to use sum rather than median in spectroscopy. It is not clear for this comparison lamp image, but is visible for the object image, which I will show you later. For the lamp image, there is practically no difference as you can see from this code:\n",
    "```python\n",
    "fig, ax = plt.subplots(1,1)\n",
    "test_sum = np.average(lampimage[lowercut_ID:uppercut_ID, :], axis=0)\n",
    "test_med = np.median(lampimage[lowercut_ID:uppercut_ID, :], axis=0)\n",
    "test_1px = lampimage[N_SPATIAL//2, :]\n",
    "ax.plot(test_sum, lw=1, label='sum')\n",
    "ax.plot(test_med, lw=1, label='med')\n",
    "ax.plot(test_1px, lw=1, label='1pix')\n",
    "ax.set_xlim(700, 1000)\n",
    "ax.legend()\n",
    "plt.show()\n",
    "```\n",
    "\n",
    "We are using NHAO MALLS comparison lamp, and the desired peaks with wavelength is given by the observatory ([here](http://www.nhao.jp/~malls/malls_wiki/index.php?MALLS%2F150%CB%DC%A5%B0%A5%EC%A1%BC%A5%C6%A5%A3%A5%F3%A5%B0%CD%D1%B5%B1%C0%FEID%CD%D1%A5%DE%A5%C3%A5%D7)). Comparing with this IRAF figure and the above plot, I could obtain the following table:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "line_ID = Table(np.array([[1696, 4764.8650],\n",
    "                          [1556, 5167.4870],\n",
    "                          [1521, 5269.5370],\n",
    "                          [1318, 5852.4878],\n",
    "                          [1233, 6096.1630],\n",
    "                          [1216, 6143.0627],\n",
    "                          [1126, 6402.2480],\n",
    "                          [1029, 6678.2766],\n",
    "                          [929 , 6965.4310],\n",
    "                          [893 , 7067.2180],\n",
    "                          [822 , 7272.9360],\n",
    "                          [783 , 7383.9800],\n",
    "                          [695 , 7635.1050],\n",
    "                          [665 , 7723.7610],\n",
    "                          [588 , 7948.1750],\n",
    "                          [477 , 8264.5210],\n",
    "                          [388 , 8521.4410],\n",
    "                          [339 , 8667.9440],\n",
    "                          [182 , 9122.9670],\n",
    "                          [148 , 9224.4990]\n",
    "                          ]),\n",
    "                names=['pixel_init', 'wavelength'],\n",
    "                dtype=[int, float])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you prefer, you can do ``Table.read(filename, names=[~~], dtype=[~~])`` after making the pixel-to-angstrom table by a spread sheet app. I just did it in python to visualize within the code. Now, the first column of ``line_ID`` is the initial guess, and the second column is the \"true\" wavelength values. The process to obtain the function to transform pixel to wavelength is two-fold:\n",
    "1. Find the Gaussian-fitted peak of lines (from the initial guess ``line_ID[\"pixel_init\"]``).\n",
    "2. Fit that fitted peak (in pixel unit) and wavelengths (``line_ID[\"wavelength\"]``).\n",
    "3. The function to fit is Chebyshev (``FITTING_MODEL_ID``) of order 4(``ORDER_ID``), in this tutorial."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " pixel_gauss  pixel_init   pixel_shift    wavelength\n",
      "------------- ---------- ---------------- ----------\n",
      "1695.87861951       1696  -0.121380485194   4764.865\n",
      "1556.16710632       1556   0.167106317136   5167.487\n",
      "1520.66430661       1521  -0.335693394075   5269.537\n",
      "1317.69003808       1318  -0.309961924933  5852.4878\n",
      "  1232.665724       1233  -0.334276002561   6096.163\n",
      "1216.26143145       1216   0.261431454362  6143.0627\n",
      "1125.83657181       1126  -0.163428186823   6402.248\n",
      "1029.45999883       1029   0.459998826328  6678.2766\n",
      "929.176832386        929   0.176832385543   6965.431\n",
      "893.644010277        893   0.644010276588   7067.218\n",
      " 821.97873593        822 -0.0212640704833   7272.936\n",
      "783.284202812        783   0.284202811702    7383.98\n",
      "695.777139763        695   0.777139762781   7635.105\n",
      "664.954803691        665 -0.0451963090643   7723.761\n",
      "587.118439143        588  -0.881560857384   7948.175\n",
      " 477.48385855        477   0.483858550447   8264.521\n",
      "388.642095367        388   0.642095366565   8521.441\n",
      "338.293494195        339  -0.706505805002   8667.944\n",
      "181.874203964        182  -0.125796036479   9122.967\n",
      "147.334194579        148  -0.665805421457   9224.499\n"
     ]
    }
   ],
   "source": [
    "peak_gauss = []\n",
    "fitter = LevMarLSQFitter()\n",
    "\n",
    "for peak_pix in line_ID['pixel_init']:\n",
    "    #TODO: put something like \"lost_factor\" to multiply to FWHM_ID in the bounds.\n",
    "    g_init = Gaussian1D(amplitude = identify_1[peak_pix], \n",
    "                       mean = peak_pix, \n",
    "                       stddev = FWHM_ID * gaussian_fwhm_to_sigma,\n",
    "                       bounds={'amplitude': (0, 2*identify_1[peak_pix]),\n",
    "                               'mean':(peak_pix-FWHM_ID, peak_pix+FWHM_ID),\n",
    "                               'stddev':(0, FWHM_ID)})\n",
    "    fitted = LINE_FITTER(g_init, x_identify, identify_1)\n",
    "    peak_gauss.append(fitted.mean.value)\n",
    "    \n",
    "peak_gauss = Column(data=peak_gauss, name='pixel_gauss', dtype=float)\n",
    "peak_shift = Column(data=peak_gauss - line_ID['pixel_init'],\n",
    "                    name='pixel_shift', dtype=float)\n",
    "line_ID.add_column(peak_gauss, index=0)\n",
    "line_ID.add_column(peak_shift, index=2)\n",
    "line_ID.sort('wavelength')\n",
    "line_ID.pprint()\n",
    "# If you want to save:\n",
    "#line_ID.write('user_input_identify.csv', \n",
    "#              format='ascii.csv', overwrite=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The line findings are done clearly! The column means the central pixel after gaussian fitting, initial guess in pixel, the shift (difference between the two), and the true wavelength we had to manually put in.\n",
    "\n",
    "Now, how can we fit? Simply using ``numpy``'s [Chebyshev module](https://www.google.co.kr/search?client=ubuntu&channel=fs&q=numpy+chebyshev&ie=utf-8&oe=utf-8&gfe_rd=cr&dcr=0&ei=LzjuWcWjMdCA0AT2zIboCA):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"1000\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "if FITTING_MODEL_ID.lower() == 'chebyshev':\n",
    "    coeff_ID, fitfull = chebfit(line_ID['pixel_gauss'], \n",
    "                                line_ID['wavelength'], \n",
    "                                deg=ORDER_ID,\n",
    "                                full=True)\n",
    "    fitRMS = np.sqrt(fitfull[0][0]/len(line_ID))\n",
    "    rough_error = ( np.ptp(line_ID['wavelength']) \n",
    "                   / np.ptp(line_ID['pixel_gauss']) ) / 2\n",
    "    # rough_error = (wave_max - wave_min) / (spatial_max - spatial_min)\n",
    "    residual = ( line_ID['wavelength'] \n",
    "                - chebval(line_ID['pixel_gauss'], coeff_ID))\n",
    "    res_range = np.max(np.abs(residual))\n",
    "else:\n",
    "    raise ValueError('Function {:s} is not implemented.'.format(FITTING_MODEL_REID))\n",
    "\n",
    "fig = plt.figure(figsize=(10, 5))\n",
    "gs = gridspec.GridSpec(3, 1)\n",
    "ax1 = plt.subplot(gs[0:2])\n",
    "ax2 = plt.subplot(gs[2])\n",
    "plt.setp(ax1.get_xticklabels(), visible=False)\n",
    "ax1.plot(line_ID['pixel_gauss'], line_ID['wavelength'],\n",
    "         ls=':', color='k', ms=10, marker='+')\n",
    "ax2.plot(line_ID['pixel_gauss'], residual, \n",
    "         ls='', color='k', ms=10, marker='+')\n",
    "ax2.axhline(y=0, color='b', ls=':')\n",
    "# rough error ~ +- wavelength resolution/2\n",
    "ax2.axhline(y=-rough_error/2, color='r', ls=':')\n",
    "ax2.axhline(y=+rough_error/2, color='r', ls=':')\n",
    "ax2.set_ylim(min(-rough_error/2 * 1.1, -1.1*res_range), \n",
    "             max(+rough_error/2 * 1.1, +1.1*res_range))\n",
    "ax1.set_ylabel(r'Wavelength ($\\AA$)')\n",
    "ax2.set_ylabel(r'Residual ($\\AA$)')\n",
    "ax2.set_xlabel('Pixel along dispersion axis')\n",
    "ax1.grid(ls=':')\n",
    "ax2.grid(ls=':')\n",
    "plt.suptitle(('First Identify (Chebyshev order {:d})\\n'.format(ORDER_ID) \n",
    "              + r'RMSE = {:.2f} $\\AA$'.format(fitRMS)))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As can be seen, the wavelngth decreases as increasing x-axis. The pixel coordinate can be transformed to wavelength with 0.14 Angstrom of RMS error. \n",
    "\n",
    "For comparison, I turned on the IRAF and obtained the following plots:\n",
    "\n",
    "```\n",
    "--> identify pcomp_medcomb section=\"middle line\" coordli='' funct=cheby order=4\n",
    "```\n",
    "(IRAF plot has x-axis in wavelength unit, so it is inverted with respect to the above plot)\n",
    "![](images/identify_iraf.png)\n",
    "\n",
    "* **TIP**: You MUST specify ``coordli=''`` in IRAF. If not, IRAF seems to internally change the user input wavelength. It compares the wavelength you put in with the internally stored line list, and changes your input to the closest wavelength. This will give unsatisfactorily large residuals (e.g., I got ~ 2 Angstroms). For cross check, I uploaded the ``idpcomp_medcomb`` file [here](https://drive.google.com/open?id=0B-MLFRYnMxUvUTdJdG80V1VVYzg). Delete ``_BACKUP`` from the filename and save it to the ``database/`` as IRAF default and run ``IDENTIFY`` for check."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Re-identify\n",
    "\n",
    "From identify process (ID), we used only the middle row of the image, not the whole 2-D image. If it were a perfect spectrograph, we expect the wavelength must not vary along the spatial (slit) direction. But in reality, it is possible that ``(x, y) = (1000, 100)`` can have different wavelength value compared to ``(x, y) = (1000, 300)``. Re-identification (REID) deals with such 2-D wavelength variation. \n",
    "\n",
    "1. We will find the same lines as we did from identify (ID) for many different rows. For each row, there will be fitted line centers (pixel) and the true wavelengths.\n",
    "2. For 2-D image, the locations of the line peaks are now specified, and thus we have to find $\\lambda = \\lambda(x, y)$. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reidentify each section by Chebyshev (order 4)\n",
      "section      |  found  |  RMS\n",
      "[0000:0015]\t17/20\t0.842\n",
      "[0015:0030]\t17/20\t0.995\n",
      "[0030:0045]\t19/20\t0.987\n",
      "[0045:0060]\t20/20\t0.951\n",
      "[0060:0075]\t20/20\t0.909\n",
      "[0075:0090]\t20/20\t0.901\n",
      "[0090:0105]\t20/20\t0.919\n",
      "[0105:0120]\t20/20\t0.652\n",
      "[0120:0135]\t20/20\t0.762\n",
      "[0135:0150]\t20/20\t0.095\n",
      "[0150:0165]\t20/20\t0.117\n",
      "[0165:0180]\t20/20\t0.119\n",
      "[0180:0195]\t20/20\t0.093\n",
      "[0195:0210]\t20/20\t0.126\n",
      "[0210:0225]\t20/20\t0.127\n",
      "[0225:0240]\t20/20\t0.139\n",
      "[0240:0255]\t20/20\t0.127\n",
      "[0255:0270]\t20/20\t0.134\n",
      "[0270:0285]\t20/20\t0.122\n",
      "[0285:0300]\t20/20\t0.115\n",
      "[0300:0315]\t20/20\t0.117\n",
      "[0315:0330]\t20/20\t0.109\n",
      "[0330:0345]\t20/20\t0.105\n",
      "[0345:0360]\t20/20\t0.160\n",
      "[0360:0375]\t20/20\t0.150\n",
      "[0375:0390]\t20/20\t0.178\n",
      "[0390:0405]\t20/20\t0.142\n",
      "[0405:0420]\t20/20\t0.152\n",
      "[0420:0435]\t20/20\t0.148\n",
      "[0435:0450]\t20/20\t0.123\n",
      "[0450:0465]\t20/20\t0.118\n",
      "[0465:0480]\t20/20\t0.111\n",
      "[0480:0495]\t20/20\t0.152\n",
      "[0495:0510]\t20/20\t0.117\n",
      "[0510:0525]\t20/20\t0.094\n",
      "[0525:0540]\t20/20\t0.092\n",
      "[0540:0555]\t20/20\t0.105\n",
      "[0555:0570]\t20/20\t0.090\n",
      "[0570:0585]\t20/20\t0.078\n",
      "[0585:0600]\t20/20\t0.074\n",
      "[0600:0615]\t20/20\t0.080\n",
      "[0615:0630]\t20/20\t0.095\n",
      "[0630:0645]\t20/20\t0.095\n",
      "[0645:0660]\t20/20\t0.104\n",
      "[0660:0675]\t20/20\t0.128\n",
      "[0675:0690]\t20/20\t0.133\n",
      "[0690:0705]\t20/20\t0.164\n",
      "[0705:0720]\t20/20\t0.169\n",
      "[0720:0735]\t20/20\t0.174\n",
      "[0735:0750]\t20/20\t0.162\n",
      "[0750:0765]\t20/20\t0.175\n",
      "[0765:0780]\t20/20\t0.181\n",
      "[0780:0795]\t20/20\t0.181\n",
      "[0795:0810]\t20/20\t0.165\n",
      "[0810:0825]\t20/20\t0.162\n",
      "[0825:0840]\t20/20\t0.150\n",
      "[0840:0855]\t20/20\t0.112\n",
      "[0855:0870]\t20/20\t0.111\n",
      "[0870:0885]\t20/20\t0.137\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# reidentify\n",
    "# =============================================================================\n",
    "# For REIDENTIFY, I used astropy.modeling.models.Chebyshev2D\n",
    "\n",
    "line_REID = np.zeros((len(line_ID), N_REID-1))\n",
    "spatialcoord = np.arange(0, (N_REID - 1) * STEP_REID, STEP_REID) + STEP_REID / 2\n",
    "\n",
    "print('Reidentify each section by Chebyshev (order {:d})'.format(ORDER_ID))\n",
    "print('section      |  found  |  RMS')\n",
    "\n",
    "for i in range(0, N_REID-1):\n",
    "    lower_cut, upper_cut = i*STEP_REID, (i+1)*STEP_REID\n",
    "    reidentify_i = np.sum(lampimage[lower_cut:upper_cut, :], \n",
    "                          axis=0)\n",
    "    peak_gauss_REID = []\n",
    "    \n",
    "    for peak_pix_init in line_ID['pixel_gauss']:\n",
    "        # TODO: Will a simple astropy fitting work well for general cases?\n",
    "        #TODO: put something like \"lost_factor\" to multiply to FWHM_ID in the bounds.\n",
    "        search_min = int(np.around(peak_pix_init - TOL_REID))\n",
    "        search_max = int(np.around(peak_pix_init + TOL_REID))\n",
    "        cropped = reidentify_i[search_min:search_max]\n",
    "        x_cropped = np.arange(len(cropped)) + search_min\n",
    "        \n",
    "        #TODO: put something like \"lost_factor\" to multiply to FWHM_ID in the bounds.\n",
    "        A_init = np.max(cropped)\n",
    "        mean_init = peak_pix_init\n",
    "        stddev_init = FWHM_ID * gaussian_fwhm_to_sigma\n",
    "        g_init = Gaussian1D(amplitude=A_init, mean=mean_init, stddev=stddev_init,\n",
    "                            bounds={'amplitude':(0, 2*np.max(cropped)) ,\n",
    "                                    'stddev':(0, TOL_REID)})\n",
    "        g_fit = fitter(g_init, x_cropped, cropped)    \n",
    "        fit_center = g_fit.mean.value\n",
    "        if abs(fit_center - peak_pix_init) > TOL_REID:\n",
    "            peak_gauss_REID.append(np.nan)\n",
    "            continue\n",
    "        peak_gauss_REID.append(fit_center)\n",
    "\n",
    "    peak_gauss_REID = np.array(peak_gauss_REID)\n",
    "    nonan_REID = np.isfinite(peak_gauss_REID)\n",
    "    line_REID[:, i] = peak_gauss_REID\n",
    "    peak_gauss_REID_nonan = peak_gauss_REID[nonan_REID]\n",
    "    n_tot = len(peak_gauss_REID)\n",
    "    n_found = np.count_nonzero(nonan_REID)\n",
    "    \n",
    "    if FITTING_MODEL_ID.lower() == 'chebyshev':\n",
    "        coeff_REID1D, fitfull = chebfit(peak_gauss_REID_nonan,\n",
    "                                        line_ID['wavelength'][nonan_REID], \n",
    "                                        deg=ORDER_WAVELEN_REID,\n",
    "                                        full=True)\n",
    "        fitRMS = np.sqrt(fitfull[0][0]/n_found)\n",
    "    \n",
    "    else:\n",
    "        raise ValueError('Function {:s} is not implemented.'.format(FITTING_MODEL_REID))\n",
    "\n",
    "    print('[{:04d}:{:04d}]\\t{:d}/{:d}\\t{:.3f}'.format(lower_cut, upper_cut,\n",
    "                                                      n_found, n_tot, fitRMS))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The printed output shows the sections ``[row_min:row_max]`` used for the ``i``-th reidentification, number of successfully found lines over all the lines, and the 4th order 1-D Chebyshev fitting RMS errors. The criterion for \"found\" lines is simple: ``abs(fit_center - peak_pix_init) > TOL_REID``, i.e., the fitted center from reidentify at given row should not be far from the result of the identify process (closer than ``TOL_REID``).\n",
    "\n",
    "FYI, IRAF results can be found after you do ``REIDENTIFY``:\n",
    "\n",
    "``` \n",
    "--> reidentify pcomp_medcomb pcomp_medcomb section=\"middle line\" override- trace- newaps- refit+ step=15 nsum=15 cradius=5 verbose+ \n",
    "```\n",
    "\n",
    "Now let me show you how to fit the 2-D Chebyshev surface to this reidentified data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Model is linear in parameters; consider using linear fitting methods. [astropy.modeling.fitting]\n"
     ]
    }
   ],
   "source": [
    "points = np.vstack((line_REID.flatten(),\n",
    "                    np.tile(spatialcoord, len(line_REID))))\n",
    "points = points.T # list of ()\n",
    "nanmask = ( np.isnan(points[:,0]) | np.isnan(points[:,1]) )\n",
    "points = points[~nanmask]\n",
    "values = np.repeat(line_ID['wavelength'], N_REID - 1)\n",
    "values = np.array(values.tolist())\n",
    "values = values[~nanmask]\n",
    "errors = np.ones_like(values)\n",
    "\n",
    "if FITTING_MODEL_REID.lower() == 'chebyshev':\n",
    "    coeff_init = Chebyshev2D(x_degree=ORDER_WAVELEN_REID, y_degree=ORDER_SPATIAL_REID)\n",
    "    fit2D_REID = fitter(coeff_init, points[:, 0], points[:, 1], values)\n",
    "    ww, ss = np.mgrid[:N_WAVELEN, :N_SPATIAL]\n",
    "else:\n",
    "    raise ValueError('Function {:s} is not implemented.'.format(FITTING_MODEL_REID))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You may ignore the possible ``WARNING``. What I do here is to flatten the ``x, y`` values, and fit it to [``Chebyshev2D`` of astropy](http://docs.astropy.org/en/stable/api/astropy.modeling.polynomial.Chebyshev2D.html#astropy.modeling.polynomial.Chebyshev2D). The plot is:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"1000\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(10, 6))\n",
    "gs = gridspec.GridSpec(3, 3)\n",
    "ax1 = plt.subplot(gs[:2, :2])\n",
    "ax2 = plt.subplot(gs[2, :2])\n",
    "ax3 = plt.subplot(gs[:2, 2])\n",
    "#plt.setp(ax2.get_xticklabels(), visible=False)\n",
    "#plt.setp(ax3.get_yticklabels(), visible=False)\n",
    "\n",
    "title_str = ('Reidentify and Wavelength Map\\n'\n",
    "             + 'func=Chebyshev, order (wavelength, dispersion) = ({:d}, {:d})')\n",
    "plt.suptitle(title_str.format(ORDER_WAVELEN_REID, ORDER_SPATIAL_REID))\n",
    "\n",
    "interp_min = line_REID[~np.isnan(line_REID)].min()\n",
    "interp_max = line_REID[~np.isnan(line_REID)].max()\n",
    "\n",
    "ax1.imshow(fit2D_REID(ww, ss).T, origin='lower')\n",
    "ax1.axvline(interp_max, color='r', lw=1)\n",
    "ax1.axvline(interp_min, color='r', lw=1)\n",
    "#ax1.text((interp_max+interp_min)/2, -10, \"Fitting Region\",\n",
    "#         horizontalalignment=\"center\",\n",
    "#         bbox={'facecolor':'red', 'alpha':0.8, 'pad':10})\n",
    "ax1.plot(points[:, 0], points[:, 1], ls='', marker='+', color='r',\n",
    "         alpha=0.8, ms=10)\n",
    "\n",
    "\n",
    "for i in (1, 2, 3):\n",
    "    vcut = N_WAVELEN * i/4\n",
    "    hcut = N_SPATIAL * i/4\n",
    "    vcutax  = np.arange(0, N_SPATIAL, STEP_REID) + STEP_REID/2\n",
    "    hcutax  = np.arange(0, N_WAVELEN, 1)\n",
    "    vcutrep = np.repeat(vcut, len(vcutax))\n",
    "    hcutrep = np.repeat(hcut, len(hcutax))\n",
    "    \n",
    "    ax1.axvline(x=vcut, ls=':', color='k')   \n",
    "    ax1.axhline(y=hcut, ls=':', color='k')\n",
    "    \n",
    "    ax2.plot(hcutax, fit2D_REID(hcutax, hcutrep), lw=1, \n",
    "             label=\"hcut {:d}\".format(int(hcut)))\n",
    "\n",
    "    vcut_profile = fit2D_REID(vcutrep, vcutax)\n",
    "    vcut_normalize = vcut_profile / np.median(vcut_profile)\n",
    "    ax3.plot(vcut_normalize, vcutax, lw=1,\n",
    "             label=\"vcut {:d}\".format(int(vcut)))\n",
    "\n",
    "ax1.set_ylabel('Spatial direction')\n",
    "ax2.grid(ls=':')\n",
    "ax2.legend()\n",
    "ax2.set_xlabel('Dispersion direction')\n",
    "ax2.set_ylabel('Wavelength change\\n(horizontal cut)')\n",
    "\n",
    "ax3.axvline(1, ls=':', color='k')\n",
    "ax3.grid(ls=':', which='both')\n",
    "ax3.set_xlabel('Fractional change (vertical cut)')\n",
    "ax3.legend()\n",
    "\n",
    "ax1.set_ylim(0, N_SPATIAL)\n",
    "ax1.set_xlim(0, N_WAVELEN)\n",
    "ax2.set_xlim(0, N_WAVELEN)\n",
    "ax3.set_ylim(0, N_SPATIAL)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the plot, I showed \n",
    "* The Chebyshev 2D fitting to the x, y coordinates. The color means fitted wavelengths (basically a 3-D plot).\n",
    "* The red ``+`` markers are the REID results. \n",
    "  * Although may not be clearly identifiable, I also showed two red vertical lines to show the min/max of interpolation regions. Outside this region, the fitting may not be very reliable.\n",
    "* The right panel shows the vertical cuts of 1/4, 2/4, and 3/4 of x-range. \n",
    "  * Along the y-axis, you can see how much the (fitted) wavelength changes as fraction. 0.001 means 6 Angstroms for the orange line (since wavelength ~ 6000).\n",
    "* The bottom pannel shows the wavelength with respect to the x-axis. \n",
    "  * It is seen that all the three horizontal cuts do not differ too much.\n",
    "\n",
    "From the right diagram, it is seen that the wavelength does not change much if the pixels changed ~ 100 pixel along the spatial direction (y-axis). \n",
    "\n",
    "In IRAF, you can use [``FITCOORDS``](http://stsdas.stsci.edu/cgi-bin/gethelp.cgi?fitcoords) to see similar, but different plots. If you run \n",
    "\n",
    "```\n",
    "--> fitcoords pcomp_medcomb interac+\n",
    "```\n",
    "\n",
    "You will see the residual with respect to x-axis. Please refer to the literature \"FOCAS cookbook\" section 5 to find information about ``FITCOORDS``. Basically I hit ``xxyyf`` to get the left panel of the figure below. This is similar to the ``+`` markers of the figure above. It's not identical, since this is the \"expected line peaks based on the 2D Chebyshev fit\". It is not showing you the re-identified positions but expected positions. \n",
    "\n",
    "Then I hit ``xxyrf`` and ``w``-``e``-``e`` as in ``IMPLOT`` or whatever plotting tools of IRAF to zoom in one of the vertical set of ``+`` markers (right panel). You can see that the re-identified line centers fluctuate in x-axis (~ 4 pixels). \n",
    "\n",
    "![](images/fitcoords_iraf.png)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "3. Aperture Extraction\n",
    "\n",
    "Since the position of the object along the dispersion axis is NOT necessarily on a single row, we have to find how it moves along the dispersion axis. This moving trace is called the aperture trace, and our goal is to find the position of the signal as a function of x-axis pixel.\n",
    "\n",
    "1. Crop the middle column and interactively set the initial peak location (rough guess) and sky estimation region.\n",
    "2. Check how the sky is fitted. Subtract sky and fit Gaussian to find the aperture center (center of the object).\n",
    "3. Iteratively find the aperture along the dispersion direction. \n",
    "   - From the Gaussian fits we found here, estimate the FWHM as the median of all the fitted Gaussians' FWHMs.\n",
    "4. Fit aperture trace (here I used Chebyshev order 3): y =y (x)\n",
    "5. Now, for all the columns, fit sky, subtract sky, and get the aperture sum. Use the fitted aperture trace as true center, sum the pixels around it (I used +- 2-sigma)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"640\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# apall (1): Plot a cut\n",
    "# =============================================================================\n",
    "lower_cut = N_WAVELEN//2 - NSUM_AP//2 \n",
    "upper_cut = N_WAVELEN//2 + NSUM_AP//2\n",
    "apall_1 = np.sum(objimage[:, lower_cut:upper_cut], axis=1)\n",
    "max_intens = np.max(apall_1)\n",
    "\n",
    "peak_pix = peak_local_max(apall_1, indices=True, num_peaks=10,\n",
    "                          min_distance=int(N_SPATIAL/100),\n",
    "                          threshold_abs=np.median(apall_1))\n",
    "\n",
    "x_apall = np.arange(0, len(apall_1))\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "title_str = r'Peak ($A \\geq {:.2f} A_\\mathrm{{max}}$, $\\Delta x \\geq {:.0f}$)'\n",
    "\n",
    "ax.plot(x_apall, apall_1, lw=1)\n",
    "\n",
    "for i in peak_pix:\n",
    "    ax.plot((i, i), \n",
    "            (apall_1[i]+0.01*max_intens, apall_1[i]+0.05*max_intens),\n",
    "            color='r', ls='-', lw=1)\n",
    "    ax.annotate(i[0], (i, 0.9),\n",
    "                xycoords = ('data', 'axes fraction'),\n",
    "                fontsize='xx-small', rotation=70)\n",
    "ax.grid(ls=':')\n",
    "ax.set_xlabel('Pixel number')\n",
    "ax.set_ylabel('Pixel value')\n",
    "ax.set_xlim(0, len(apall_1))\n",
    "ax.set_title(title_str.format(np.median(apall_1), int(N_SPATIAL/100)))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the plot, I can see that ``x=553`` can be a good guess of the object center. I will use some regions nearby it for sky estimation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAyAAAAH0CAYAAADFQEl4AAAgAElEQVR4nOydd3wcxfnGHyMXwLQAoZPzD0MoKYQ0WgimQwgkGEhCSWIgDZKQQEKObkPooYbegsGhmSbIGctFtmUsjOUGVrMtF9mWu2Vbsi1Zp/L8/phd6cqepDtLe7e7z/fz2Y9Oc7M77+y7szfPzjuzgBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIIIYQQQgghhBBCCNEFwwDQ+ttbnAFgGoBNVln5VvoU6/9cZQiMfSN6sYxBVhkje7GM7jIIuWNLLlJtbX7jNQBrAQzMtiG9wBTk9j0m1xgE790DPgKwGED/bBsihBBBJg/AbwEUAdgIoBnAOgDzALwE4MKE/MPQuwIkBKABwGoAj8J05n9hfTcFyZ2DIejdTv/3AbwMYAGALQCaACwD8C6An8GcP7dsAXLrB38QcseWXKQa2RUgewB4HMAnAFYB2A7TtksA/BWZCYjvAmgD8BcXyjoGwGjrONth2uBdAHbJ4FjdZQr8L0CqYerotK1J81iD4HwPGGmlD8rUyF7kWJhr+O/ZNkQIIYJKHoCxMD8UmwD8F8ADAF4AMAmmsz0tYZ9h6F0B8hvr+Jc7fPcVAEclpA1B73T6+wF41jp2C4BCGEF0P8wT4BXWd++6YEssg5A7nf5ByB1bcpFqZFeADIIR80UAXgRwH4CnAVTC+K0cRjikw3iYBxUDerms4wFsAxAF8AaABwHMtI41zaH8nsLpHuM3qgFshrlPJW7pdsr7wZyvAxPSRyJ3BQgAjIH5zfPjKJ4QQuQ8V8L8SHwOYE+H778E4MyEtGHoXQFyp3X8Id3MPwS90+l/0TruPABHOny/E4DLYJ7Q9rYtsQxC7nT6ByF3bMlFqpFdAZIH00F04nUY34XTON5XYZ4cP93LZeUBqLD2iR2B3QlG8BPAzd08lkimGr1/XY5EbguQn8HY95tsGyKEEEHEfsL/1zT2GQZnAfIlmPCLNgC3WGmfAWhF6h+hv1vH+hs6Ou9O2xAr/xTEh0eM7MY+mXCydYxaJD/ZSyT2SewQdAiQb8E8ZdsM82R4qnVcJ/oCuA7mfNVb+ecC+BNMpyuWQejo9B8FMz9mI8zT4mkAzk7I/wcr/50pyj4AJuyuNCZtTwDDYZ5abwGwFabD8g6A76SwZRCAtwBsgAmXmY3k8L1YLgMwGeYp5HaYJ+W3I/58Hgxz/czp5DgTLRu+3kmeWL4KYBRMmFDU+jvKSk9kBDqupV/BPIHfhvjOWx8YP5Vb9VgJ00HfE5139LpTfxvCXPsHAXgFJjyxFTv2EOAn1nFfSGOfB6x9ftjLZZ2Bjjoncpj13TKYc99dLoI532tgRnbXwLSXPybkmwLnEKwBMNfDEmv/pQDusdKdbB2BjmvnMpj20ABzvT2KDj+fBTNqtAXmWngNwN4O5Z8Gc/4qYO4RjTDXXCYhadXoOQEyCMkPIVLdkxPL3BtmRLkSpj51MCPNifcwIP5353yY+2k94n01BEAEQA1M214P02ZHOBxvAIw/pndePSGEEL3B3TA38GfS2GcYkgXIV2B+GKMwoyo2v7Ly3utwnD4AFsJ0vvaF+SEbgY4OwEh0hAUMsvaxv7P5KTpEyBTEhxIMislXjfSexr1q5b+/m/lthlj7RWB+3AoBPAwzStIKU9ejE/bpB6DA2q8SwHMwsfRfWGn/Tcg/yEovgumwfGLZORLmR7wVwM9j8u8G88O+HPHzVWxus473J+v/PjA/ygTwKUxn6SEAb8J0fP8Us69ty2SYOP3PADwGc/62w4jRMxzKfNnabznMPKNHABTHHCv2Sfo4K/0bDsc5xKrvLIfvnDgeptPSBuADmDChD6z/663vYxmBDn9uhxFgDwJ4PibPE1aeVQD+bdVlEUzHZxWcO3rp1B9WeilMp7sMwJMAngLwo27W24mXEO/37jAbRqym2+FNt6x70PkoxwLr+8HdPN61Vv7VMJ34+yybSmD8FMsUJAuQPjDXAGHuWY/A+LoG5vrpTIC8B3MveMPar9RKfwXmKXwTgPcB/Asd18BYhzoUwFxLb1h5n4J5SEGYznjfzk9BHNUw5+JKALfCzOc5Dc73h64YhGQBMgJmVJ0w97IR1hb7oCsEI+Lse9mjML5ZBdMef5dQzjAr7xiYNv8/mPvSO9b3P7L22wRz/7kP5l5aBLNgghPTrGPt1Z2KCiGE6DmOhRENbTAd3UsB/F8X+wxDvAA5FuZHow7J4VoDYJ5CrUbyD+Tp1nFeT0gfgdQjGFOQ2ST0aqQnQJZY+RPr0xW2LQTw64Tvfm+lP5uQPgIdP9SxHYA8dHRUfxqTPiimjH8lHOu7MB3ETYiPt3/Kyv/jhPw7wXQCtqEjBO+bVt58JLMTzEiXky3DE/KeA+fO1DAr/R0AOyd8NwLJI3KXW2kPO9hzs/Xdnx2+c7J9vpX/5wnfXWalL0D8iJNtzzYAxzkc8yTr+0WIf2q9MzpEXHXCPsOQXv2BjnP8GtLraNr0RUcn8N/o6BxOQPfnUgyE6ax94UJZ71j7XJzie1sMnNfN482B6ejv5/Ddvgn/T0HyPeaX6Ojox66ctBc6rqkpCfuMsNLrEP/QYQDMyEUrTDs9Nea7PugQ3N9KON5hcB7xuc/K/wuH71JRDecRiiUJ9nSHQUgWIEDXIVhTYH53Lk1I3wvmmmmEGZ21GWYdrw3AuQ7Hex/O5w1I9rHNY9Y+56f4XgghRC9yCYyAiP0h2gDz5M7pCeswdAiQM2GeHK+CESJOPGTlH5qQ/raVfkpC+gj0vAAZDBOulCpGPZEG65jpTka1bfnE4bt+MOIg9mn9TjDnehWcnz7uBfOD+05M2iCrjM0AdnfYZySSBdAxVtpHCXnPs9Jfjkn7hpX2hsOxE7FtWZrC/mUw9YtlLozodXrqmGflj30qvQtMJ261Qxn2qFuqDkYsP7BsTVxUwcZ++hwbXjQCHeLQCXue0FUO3w2BswBJt/6wjpOqA90ddkZyZ/NVmNGx7vJVa7/xLpQ1Hp0/ALDnlFzWzePNhhGRX+oqI5zvMXaYn1Po2RXoXIDc7bCPPc/tNYfv7FHjxAcYqdjHyv+fbuYHzMOC0wHsD2BXmPDF52DuNQ1IfS93YhDSFyDHWt+NdvgO6AjZiw2PG4bUD0YA83tFOIdSpiJs7fOHNPYRQgjRg/SDibu9G2Zo237/ht05jX3yNgwdYSlRmE7gVzo59mEwP2zjYtL2g+lQlTvkH4GeFyDpYgsQp8nnnWHb8liK72tg1p+3OQodYR0jUmzbED8/Y5C1z6QUZQyDc6d5CsxqXofEpOVbeb8fk5YH88TYFlI3wTzpd1oz37blgxS22CEONrvCXAvrkLq+K2Hi4WOxO/qxgvh7Vtr7KcpO5M9W/n+m+P4uJI8+jEDnHd3Z1veHOXyXByM4q2PSMq2/PTqzo/SBmVfzaxjROx/dHxU80bLjbRfK6kqAvIH4p/6D4HwubW608tvzL34K4Mspjj0FyfeYWpjr2Gn0KYTOBchPHPaxV/pLXMoYMHUmTGhkLANhwqVmwgjyNsSLvHHYcR5GcnsehM7P7SCkL0DseWnjUxz7aev7J2P2GWal3QJnhlrfb4IZZf4Z4u91TvzW2uf2LvIJIYRwiTyYG/hWJP+IDrPSatHx5K2ryaBjYX4w7fAu+8nT9Q55RyD7AmSxdUyn+Qud0ZUt1YjvkNqT3bvalsbsM8hKezNFGeda37+SkP5zK3249f9BMIJkrsMxvgQjouylhgnT6XkC8ctW2raMTGHLFMT762B0r76JPrZHL2I7v3ZYmVMHz4nbkfxUNRa7UzQ8Jm2ElXZain0WWd+nWspzDeL9nWn9CRPL3pMcj44HCd3hW1b+D10oK90QrCHo+hz+Ch2LYhDmflQI4NsJ+aY47NsCE0rqhD3iMyUhfQRS38eGWd8Nc/huCJLvIf0AzLDSS2HmStyLjg67U/mZcDg67u2J9qQ6t4OQvgCx5511tcXew4ZZaU6jjTbnwzyYicYcYyZS38evt/L8rZNjCiGEyAL/hLlB/zsmbZiV9nuYkB5bhCSu1hTLBVa++2DEyiKYUQankIgRyL4AsSeh35fmfl3ZUo34DunXkd5TfCDzEZB+MGFMK2AEph0G8vsuyjscwDUwE3YJc24SbRmZYt8piPfXbtb/na1qlYpFMBPB94KpywaYTmF3w+p2ZARkSIp90h0BybT+PdXBTGQTzEOG7nCQZUeqELaeLMuehJ7qaXe6k9Bj2QtmJO1FGDGyAfGjIVOQfI/ZiMxHQIY47DMM6QmQS5C6nR2YovxM2MM61vY09hmE9AXIn5D6AVQqhiH1OUtkIEyI2aMwc0mcFv8AOtr8L9OwQwghhAv8A+YG7TQUPgym82c/rXwdqVdR2QmmI7YG5se/s07rCKQnQE5B5x3LTLBHJjbAxEl3RqpleJ2oRnyHtC9Mx2wVut+RHmSVkc4cEBtbUP4EZgWmLSmO4cQuVv56B1tGpthnCpL9VQYTfue01GhnxAqmi6zPT6Sxvz2K4jQ/BzAjDITzHJAhKfbJZA5IJvXvDQGyO0ynenM38/eBCR1LNRLQk2XZC1Q4jfrYy/BWI71leJ2wF3mInZ82BcnXbCGSrw2bruaADHHYZxjSEyD2YgsXOOT/RYryM+Es61gVaewzCM73APvcOonE71rfvePwXSqGofsCJJY7rP1udPjOnofotMKeEEKIXuQymB8dp9GLAwBUwdygY1cqGYb4H4I8dEwKfQepO9K3WHnsCe8npMg3AukJEHuC9cgUxwPSn4QOmDAHwoQoHeHwfSYvIqxGcofUXgr5WTgvb3ogTB1tBqEjvCDVKlib4fzW6UNgwklsHzznkOf/AHzNIf0gmNCGdQ62jHTIDzj762orLR/OE7G/hOSwGMA8aW6DmSxuL33qtDJVKvqgY8WiSxK+s58wp1oFa0iKY9pCNZ1VsDKpf6YdzG+lKKM/Okb5Eleh6wz7JYCH91BZu8K0y8Q5ZJ29iNB+4JHOiwjPhfPoxf+sY8XOLZqC5Gv211Za4ipYe6LrVbCGOJQ7DOkJEFtkPJqQ9zCYhR7SuT6+Bmfxeyg6RpZu7eaxgNT3AHvxkSEp9psKI0qvTvH9NxC/6MIwdC5AzoDz/dMO1bzO4btlMIJ6R4WsEEKINHkc5ua8GuaH/SF0rK1uT8TOh/Mk9GExaTvBhGERJkbcacKyPfGc6HwpzxFIT4DkwUzu3g7zRPoOmHj/UEyeaqQOB0hFP3S8qLEFZiWcR2DCsl5Fx/yI2Kd4Q5C+AOkHc85o1eM1mPd6vIyOH+nYztYgdDwd3mTl6ew9IInYE89TdeB/an03yzrmfTDndZ2VHhsvbdsyMkVZU5DsL6BjkmktzITiB2AE3wSYa8RJGAHmHRmEEVnzUuTpDPs9IK0wYW/3waye04rO3wMypJNj/hsdwrq77wFJt/6ZCpDHYdrxxzDX8v0w1+5q65jz0fWLNmOxlyt2mkeTSVlDkLpux8MswBBFxzmaaeWfhu4v6QsYQb4GRkA9DOMjO6RwNuLvV1OQfM32gZnHRpgFIx6GGX2rQUd7SgyJHIGeEyAD0fEwaDzMu2jehBmRfAvpXR8jYO6VY2He//QgzD2s0TrOGDjfv1MxCM73AHsZ7ioY392O+PfAHAJzLgmz7O7zli2vo+NdKbEPqYahcwHyOYyf82GuxYfQMXK1DMmi60jru+chhBDCdQ6F6Ux8APP0qx7mB381TEfiSiSPjgyD8w9BH5gfNML8uCW+4wDoWKv92k5sGoH0BAhgVkQqRPzqMLH7VyN9AWJzPIwYWAgTw94EIz4+gJmoH3t+hiB9AQKYc/dLmDpshPHBSpiO1q0wfrIZhI4f/KNgxMsmmM5fMcwPf2dcaO0/I8X3h8B0zIvR8dboGhifJr53IdYWJ6bA2V+AeSdJBEbYRK2ySmDi/1MtfzwMHeIp04mjR8K8+Xw1jJBZDfMOHKcVz0agawFivwm9EuZcrUL33oSeTv0zFSAnw1y75TDXSAvM9TUNwN9hRiDSob9lp9O1k0lZQ9B53Y6B6RxvgDm3C5HZm7//ANNel8C0k40wI5v/QHII4hQ4X7M7w4xWLrVsqYaZCG4vLJC4EtwI9JwAAcw94HWY+4L9FvR/wIzspHN9nAojXubDdNibYUYBJsBM1E93NGAQUt8DbkRHu3AaDdwd5v42G+be2ghzfsfAvIgwdnGHYehcgPwMpl5V1rHqYcId74Xzimf2+1Oc3hsihBDCR9gvvdsK5/Ag4Q52yFeq0AchOsMOpUwn/M3P2PMm7s+2IaLbDIARchOzbYgQQojex45jfibbhgSYPWCeuG9A+k+RhQDMSMAymPkTQeIgh7R90LES2nfdNUfsAH+FCbtM54WLQgghPMbtMJMn62Bilg/tPLvoBS6ACV2xO0t/7Ty7EJ3yQ5j3paR6/4kfeQsmnOhlmDkN/0XHu5CezqJdIn2uRfqraQkhhPAYhIlxnwuzvKZwn5HoWGzgn+j8nS1CiGQuhQnZWQ0zp6EewKcw78gRQgghhBBCCCGEEEIIIYQQQgghhBBCCCGEEEIk0wdmzfU9tGnTpk2bNm3atOXsdjD0lnnhE+wXPmnTpk2bNm3atGnL7e1gCOED9gDAFStWsK6urte3TZs2sbS0lJs2bXKlPG25scnvwdv84vOioiICYFFRkafLcMOuTHyeq3XX1v3NL20917cVK1bYAmSPLPYZhegx9gDAuro6ukFzczOLi4vZ3NzsSnkiN5Dfg4dffL5w4UKeffbZXLhwoafLyIR07crE57lad9F9/NLWc526ujoJEOErXBUgQgghhBAiPSRAhN9wVYC0tLSwqqqKLS0trpQncgP5PXj4xectLS2sq6vr1Xq4UUYmpGtXJj7P1bqL7uOXtp7rSIAIv+F6CFZJSYmGagOG/B48/OLz2bNnEwBnz57t6TIyIV27MvF5rtZddB+/tPVcRwJE+A2FYAkhRAokQIJZdyFyDQkQ4TfSEiDRaJSNjY0Zb1u3bmV5eTm3bt26Q8fR5q1Nft/xzWvhDS0tLaysrPSc3YkEuROerl2Z+DxX6y66j1/aeq4jASL8RrcESF1dHRcvXsyKiood2srLy/n555+zvLx8h4+lzTub/L7jW2VlJVetWsW2tjaXfu52jJaWFs6ZM8fznRIJkPQESLo+z9W6i+7jl7ae60iACL/RpQCpq6tjRUUFly9fzrq6OjY0NGT9abA2bUHaGhoauH79elZUVHDTpk0u/uQJCZBg1l2IXEMCRPiNLgXI4sWLuXz58h558trW1saGhgbPPMUVPYP83jPU1NRwwYIFnjiPLS0tLC0t9fxT0Wg0ynXr1jEajXq6jExI165MfJ6rdRfdxy9tPdeRABF+o1MBEo1GWVFR0WOT1NURDSbye89QX1/PiooKT6w2o05J8JDPg4n87g4SIMJvdCpAGhsbWVFRwYaGBpebmhAikYaGBlZUVLCxsTHbpgSGRYsW8YILLuCiRYs8XUYmBLnuQuQaEiDCb3RLgPRUh6etrY3btm3Tk/CAIb/3DD3dHnsTv0xMDfI8CE1CF93BL20915EAEX7DdQGiUJzgIb/3DF4TIH5YmlMCRMvwis7xS1vPdSRAhN9wVYDkIp999hl/+tOf8tBDD2X//v2533778YQTTuCNN94Yly8UCvH888/vVVtCoZB9g0natmzZQpJ88cUXCYArVqxo32/UqFF84okn0iqrqamJhx9+OP/1r3/Fpc+aNYsXXnghDzjgAO6666486qij+M9//tMxDG/mzJk8/fTTOXDgQO61114cOnQoly5dmrLMCy64gEOHDm3/f+LEiTz99NO57777crfdduM3v/lNPvnkk44/ZOPGjePxxx/PnXfemfvuuy+vuuoqrl+/Pq065wJXXHEFBw8e3P7/hg0buPvuu/N///tfl/sGoT3mGhIgway7ELmGBIjwG66PgGzdujVnnoRHIhHutNNOPP300/nmm29yypQpfPPNN/m3v/2NBx98cFxetwTIySefzOnTpydtra2tJMl169Zx+vTpbGpqat/vnHPOievUdoeHH36YBx54YJywmDdvHgcMGMDjjjuOo0ePZmFhIe+44w7m5eXFCQeSLCsr48CBA3nqqafy448/5rvvvsujjz6ahxxyCDds2BCXt62tjatXr+aAAQP4+uuvkyQLCgrYp08fnnHGGfzwww85fvx4/vGPfySAJPE3ceLEdhsmTJjAUaNG8aCDDuKxxx4bdx68QKIAIcnbb7+dRx55ZJeTy70kQJqbm1lSUuKJCfOd4dVOeF1jlNubd+yJdLp2ZeJzCRDv45e2nutIgAi/4boAaWxszBkB8sMf/pCDBw92vHHaHX4btwRIJmWkK0Ci0SgPOOAA3n777XHp4XCYAFhdXR2XfvXVV7NPnz5x18lFF13E/fffv31khjRLNvft25e33npr3P5tbW0cOXIk+/fv336Mn//859x1112TRlZOP/107r333nFpxx13HL/5zW/GjYwUFRURAF944YVu1zsXcBIgK1euZF5eHt9+++1O9/WSAGlpaWFVVZXnwzLWrFnDRx55hGvWrPFMGdGWVobCEf75jTmu2pWJz904v6J38Utbz3UkQITfCHQI1te+9jUef/zx3crrJA6efvpp5uXl8c4772RbWxsPP/xwnn322Un7btmyhXvssQevu+66tMtIJDEE6+STT04K18rLy+v0GG+//TYBcMGCBXHpt99+OwFw48aNcek33ngj8/Ly2q+D7du3c8CAAfzjH/+YdOzTTz+dRx99dFL6xRdfHFe3K6+8knvttVeSGB06dCgPOuig9v+rq6sJIClUjCQHDx7M8847r9O6TpgwgQD4ySefxKVXVVURAEeNGhWXdumll/KAAw5g//79uf/++/OMM87gvHnz4vZ94403ePzxx3OXXXbhbrvtxnPOOYeff/55UtkvvfQSjzjiCPbv359HH300R40a5ShASPKss87iaaed1mld/N4eRc/wvy9WMhSO8CdPTcu2KUKIHkICRPiNHRMgTVtTb9HGpLxt27dwS+1atm3fkpC3IY3j9tySwL/5zW8IgH/+85/52WefdfoyrFhx0NbWxr/97W/s168fX3nllfY8TzzxBPv06cOFCxfG7fv0008TAMvLyzu1JxQK8Uc/+hGbm5vjttjRmEQBUl5ezhNOOIGHHHJIe7jWZ5991mk5v/rVr+I6+TaLFi3innvuyZ/97GdcsmQJ6+rq+OGHH3KPPfbgDTfc0J6vrKyMAPj8888nHeOvf/0r8/Ly4s7l1q1bueuuu/Lll19uT5s2bRr79+/P66+/nitXruSmTZv4yiuvsG/fvnz88cfb80UiEQLguHHjksqy5+50RjoC5PDDD+dXv/pVjho1ikVFRXzvvfd44403sqioqD3PXXfdxT59+vC3v/0tI5EI33vvPR5//PHcbbfdOH/+/PZ8tp8uuugiRiIRjho1ioMHD+ahhx7qKEDuvfde9u3bl/X19Snr4iUB0tzczOLiYs+HZWzcuJGjR49OEuW5XMbzRYsYCkd4w9tzXbUrE5+7cX5F7+KXtp7rSIAIv7FjAmT4Hqm3/14Sn/eeA1Ln/c+P4vM++H+p8z5/6g43ZJsNGzbwBz/4QfvIQb9+/XjSSSfx/vvvjwstIjsESENDAy+++GLuueeenDhxYlye+vp67r777vzLX/4Sl37MMcd0+XTbLgNInoB+2223tedxmoSebgjWEUccwR//+MeO35WXl/PII4+MKz9xToYd/vTOO+8k7X/33XcTANetW9ee9s4777Bv375Jk8anTp3KAw88sL2cvn378tFHH43L8+qrrxIAZ86cmVTW1VdfzV133bXTunZXgKxZs4YA+NRTT6U81tKlS5mXlxcnxkjj9/3224+XX345SROSsP/++/P73/9+3AiPHaLm5KuxY8cSACdMmJCyfC8JkNbWVlZXVyeFMnoNL84BebJwYY+EYKVrVyY+1xwQ7+OXtp7rSIAIvxFoAWIzc+ZMPvDAA7zkkku47777EgAHDRoU12EOhUI84YQTeMIJJ/DQQw9laWmp47Guv/567rnnnty6dStJsrCwkAD43nvvdWlHKBTiD37wA86cOTNuW7lyZXuenhAgAwcO5NVXX52UvnjxYh522GE85ZRT+P7773PKlCl84IEHuPvuu/N3v/tdez5bgLz77rtJx7AFSOy5u+yyy3jmmWfG5SspKeGXv/xl/uQnP2EkEuGkSZN46623sn///rz33nvb89kCZNasWUllXX311Rw4cGCnde2uAGltbeWgQYN4yCGH8LHHHuPcuXOTflCfffZZAuDcuXOTRqkuvvji9lGlefPmEUDcSI7NySef7OgruyM2cuTIlHXxkgDxC14UII+Mm89QOMJr/5vcZrJpV7bKEMIPSIAIv+F6CFZ97ZqcCcFyIhqN8oYbbiAA3nTTTe3poVCIe++9N/v06RPXGU+kqqqKO+20U3t40kUXXcRDDjmkWxP0MpkDQqYvQPLy8hzno1x88cU88MADuW3btrj0F154gQA4bZqJKU8nBGv79u3cfffd+dhjj8WNBnznO9/hcccdl9TJv/XWW8w+0W0AACAASURBVJmXl8dly5aRdDcEa+nSpbzqqqu43377EQD32Wcf/uUvf2kfDRsxYkTKZZIBsH///iTJyZMnEwDfeuutJHsuvvhiR1+Vl5cTAJ999tmUdfGSAGlubmZRUZHnwzK82Am/7+MKhsIR/ubV5FHD3rQrE59LgHgfv7T1XEcCRPgN11fBampqyplVsFKxefNmAoib4GyLg5EjRzIvL4+/+93vUg45n3/++fzmN7/J5cuXMy8vj/fcc0+3ynVLgOy33378xS9+kZQ+ePDgpJEKkpw7dy4B8LnnniNp3iEyYMAA/ulPf0rKe8YZZ8RNQv/oo4+40047cdmyZXF+79u3L3/zm98k7f/BBx8QAAsKCkiSy5YtIwA+/PDDjvZ2NQn9k08+IYCkcLnp06cnCZBY5s+fz7vuuos77bRT+2T7p556igCYn5+fNEo1c+bM9lGaTEZAbDudwtpsvCRAWltbWVNT4/mwDC8KkOEfljEUjvDX/5nhql2Z+FwCxPv4pa3nOhIgwm8EehWsVatWOabbndNrrrmmPS1WHLzzzjvs168fr7jiCseRjfHjxxMATzvtNPbv359r167tlj2ZCpALL7zQcVJ5Kn74wx/ye9/7XlL6KaecwgMOOKA9fMzmmWeeIQBGIpH2tKFDhyblXbp0Kfv27Rs3Z+XXv/41TznllKSyDj30UB577LFJP1r/+Mc/CIBlZWXtad/+9rd57LHHxp1ru8P+4osvdlrXFStWEEDS3JJbbrmlUwFi8/Wvf50nnngiSTNJPy8vj4888kin+7S0tHC//fbj97///bj0zuaA2KFmnS1U4Pf2mItUVFTwuOOOY0VFhWfKuOX9eQyFI7zixc4Xo3DbrmyVIYQfkAARfsP1EZC6urqcGQH5xje+wfPOO4/PPPMMJ02axIkTJ7a/oG+33XaLW341URyMGTOGu+yyC4cOHer4MrxjjjmGAHjllVd2255MBchtt93WHhI1Y8YMx/kSsdx5553s379/kl/fe+89AuBJJ53U/iLCe+65hwMHDuQ3vvGNuJWt7BcRDhkyhGPHjuV7773HY445Ju5FhNFolF/60pf42GOPJfn90UcfJQD++Mc/Zn5+PsePH8+bbrqJffv25TnnnBNnl/0iwosvvpgTJ07kf//7Xx588MHdfhHhkCFDuPfee/Oll15qL+eII46IEyCzZ8/mqaeeyieffJIFBQUsLCzkLbfcwj59+vDOO+9sP9bdd9/Nfv368dprr2V+fj6nTJnCt956izfeeCPvuuuu9nzPPfccAXDo0KEcM2ZM+ypYhxxyiKMAufbaa7n//vt3Wg8vCZDm5mYWFhYqLCML/G305wyFI/zZc5+6Wq58Hkzkd3eQABF+w3UBEo1Gc0aAvP3227z88st5xBFHcLfddmO/fv34la98hb/85S+Tnsg5iYPJkydzt91247nnnpv0Qj17vkBXS+J2VUYiTgKktraWQ4cO5V577dWt94AsWLCAffr04fvvv5/03cSJE3nmmWdy//335y677MIjjzySN910E2tra5PylpSU8LTTTuOuu+7KPfbYg0OHDuWSJUvavy8oKCAALlu2zNHvo0eP5sknn8x99tmHAwcO5Ne+9jXee++9SXNQ7GN9//vf584778x99tmHw4YNi1tpqzNWrlzJoUOHcu+99+Zee+3FX/3qV5wxY0acAFm9ejV//etf88gjj+TAgQO5++6789hjj+UTTzyRNErz/vvv89RTT+Uee+zBAQMGcNCgQbz00ks5adKkuHwvvPACDz/8cPbv359HHnkkX331Vcf3gLS2tvKQQw5JWl0rES8JkNbWVq5du1ZhGVngz2/MYSgc4UVPu/seEPk8mMjv7iABImyGIPVk1BMS8p4JYDqABgAbAIwEsJ/DMfsBGA6gGkATgPkA/pyi/MMAvA9gM4CtACYA+HYG9Qh0CFZv8p3vfIff/e53s21GSs4999yUS/H2FL/97W+TwpBEMgUFBczLy0t6f0wiao/uM2fOHPbv359z5uzYkrZulvH712YxFI7wgic/6Tqzi3Zlqwwh/IAEiLAZAnMh3AIjOGK33WLynQqgGUA+gLMAXAGgBkApgAEJx3wRwHYAN1nHvx9AG4BbE/J9GcBKAGUAhgL4EYBPANQDODLNergqQFpbW7l582bfPimpq6tjcXFx+/yCDz74INsmpeTzzz9n3759XZn86Xe/7yinnHIK//CHP3SZz0sCJBqNsqCgoNOXe3oBL05Cv+qVEobCEZ77+FRX7crE55qE7n380tZzHQkQYTME5kK4pIt8JQDKAfSNSTvJ2vfamLSvwYiNWxL2fwFm5GTvmLSHAEQBhGLS9gCwHsDb3bI+fj9XQ7Cam5tzJgSrp7GXX91nn304fPjwbJvTJa+++mrcxPLewu9+3xE2bNjA4cOHJ72k0QkvCZDW1lbW1tZ6XnR6UYBc8eJnDIUjPPORKa7alYnPJUC8j1/aeq4jASJshqBrAXKwledmh+8WABgf8/9tVt4DEvKdaKVfHpNWBaDA4ZjPw4iVvg7fpUIhWEJ4BLVH9/GiALn02U8ZCkd46kOTus7sol3ZKkMIPyABImyGwFwIawG0wIQ/jQPwg5g851h5fuSw/zsAVsX8/yaAdQ75BlrHuM/6fxeYkZKHHPL+0cr71W7WAchCCNamTZv0pCRgyO89g5cESDQaZSQS8XxYhhc74Rc+NY2hcIQn3V/oql2Z+FwCxPv4pa3nOhIgwuY4AI8D+CmAUwBcBaACRoycY+W5HM6T0gEzWtEU8/94mEnnTjRZ+QHgIKQeVbnM+u7ETuweAHPx2tvBALhx40aS5v0F9rsWWlpauHXrVlZUVLChoaE9fKatra3Lz62trSk/26E4drr92T5Gqs/plN8bnzurU2w9VCfnMm2/+6lObvupsbGR5eXl7Suu2SuLtbW1JX22j2F/bm1tjftsL5mZ6nNLS0vc59j7gtPn5ubmpM91dXWMRqPt57K5ubn9c2x67AppuVanLVu2sKysjPX19Y716Ik6bdu2jXPnzmVDQ0OP1Oncx6cyFI7w+/dM6NJPndWpoaGBc+fObX/XT1d1skNxEuvXWZ3q6+tZVlbGrVu39ti111t+ymZ7yuU6tbW1cdOmTV3Wz0t1ykU/bdiwQQJEpGQvACsAfGH9bwuQ4x3yPg8z4dxmPIDKFMdtAvCc9dkWIGGHfLYAcRI8NiPgsHLX1KlmsmJpaSlLS0tJmtVJysvLWVFRwdra2vanrlu2bOH27dtJkvX19e3vYbA7G6R5k7jdiDdt2tTeQO2n4LFPxO3PdsO2Pzc3N3Pz5s3tjdMepWlqamJ9fT1Jcvv27dyyZQtJ83TY/qFsaGhoX8q1oaGhvcO2bdu29s9bt25VnVQnT9WpoaGBs2fPbp8vkp+fz4aGBkajUebn5zMajbKhoYH5+fntx7Dn+NTW1ra/XX7t2rUsLDRPx2tqalhUVESSrK6uZnFxMUmyqqqKJSUlJMnKysr2VYoS7xGVlZUkzZLMVVVVJMni4mJWV1eTJIuKilhTU0OSLCwsbH8pZ0FBQfvSzpFIpP0cq047XqfTH57MUDjCb9z5sW/q5Ec/qU6qUzp1Gj16tASI6JRnYS6QXZCbIVhpjYDYIR+bN2/ukafQdifP7vx58Sl0qs9efrLe23WK7fT7pU7Z8FN9fT3Ly8vbhUkuPwncvn17+4+ql59uLl68mNdccw0XLVrUa083ly5dyquuuorV1dU9UqeTHyjk4beO4deHF3Tpp87qVF1dzauuuor2u326qlNTU1N756m7flq0aBGvueYaLlmyxBNPof34ZH1H62R3mu2HKn6oUy76SSMgoiueg7lAdkbHJHSn0Yr5iJ+EfiucJ6GfgORJ6AsBjE1Rdo9OQifJxYsXc/ny5e0NZUeI7fyJ4CC/9ww1NTVcsGCBJ85jW1tbXOimV/HiHJDv3TOBx941jkfdPtZVuzLxueaAeB+/tPVcR3NARGd8CeYdH3Nj0mbAvPMjLybNFhV/iEmzl+FNFCu2qIhdhvdBmLCsQ2PSdocZQXkrTZu7FCB1dXWsqKjg8uXLWVdXx4aGBjY2Nma02eEpO3IMbd7b5PcdP3/r169nRUVFe5hYrhP7JM/LeFGAHHvXOJ50vxkFcdOuTHwuAeJ9/NLWcx0JEGHzBoAHYJbhHQLgtzCjGs0wbz63GWKlvW+lXw5gOTp/EeHfYV5geC9Sv4hwFYB5MJPgzwNQBLMS11Fp1qNLAWJf+IsXL2ZFRcUObeXl5Zw1a1b73BJtwdjk9x3fKisruWrVKs/8yMfGMnsZLwqQo+8YyzMfmcJQOLJD10smq2Cl63MJEO/jl7ae60iACJubYUY6NsOsfLUORmR8zyHvWQCmA2gEUAvgVQD7OeTrBzNJfBnMCMcCAH9OUf5gAB8AqAOwDcBEAN/OoB7dEiA20Wh0h5/k1tfX60l4wDb5fcc3O87YK/jlqagXBcjgW8bwJ9ZSvNGWzJe+1giI6A5+aeu5jgSI8BtpCZAdRbGiwUR+Dx5+8XlNTQ1vvvnm9hVtcr2M1tY2hsIRXv7idIbCETY0ZS5c07UrE5+7cX5F7+KXtp7rSIAIv+GqANFQbTCR34OHfJ4dGqMtDIUj/M2rMxkKR7i5wb3zL58HE/ndHSRAhN9wVYAIIYSXqK+v5+TJk9vf/5LrZWxuiDIUjvD6N+cwFI5ww5btOWFXNssQwg9IgAi/4XoIVl1dnYZqA4b8Hjz84nOvzQHZsGU7Q+EIb35vHkPhCNfUNbpmVyY+1xwQ7+OXtp7rSIAIv+F6CFYkEtFQbcCQ34OHX3zuNQGypq6RoXCEd31UzlA4whUbt7lmVyY+lwDxPn5p67mOBIjwGwrBEkKIFHhNgCyv3cZQOMJ/FcxnKBzhkvVbc8KubJYhhB+QABF+w1UB0traytraWra2Zr40pPAe8nvw8IvPvdYJX7J+K0PhCJ+aVMVQOMKFazKfW5GuXZn4XALE+/ilrec6EiDCb7geglVQUKCh2oAhvwcPv/h83rx5PPjggzlv3jxPlLFwTT1D4Qhf/mQJQ+EIy1dmfm9P165MfO7G+RW9i1/aeq4jASL8hkKwhBDCJ5SvrGMoHOEbM5YxFI7w8+Wbsm2SEKIHkAARfsP1EKy1a9dqqDZgyO/BQz7PDl+s2MRQOMIP5tQwFI5wVvVG18qWz4OJ/O4OEiDCb7gqQJqbm1lYWMjm5mZXyhO5gfwePPzic6+FYM2qrmUoHOHY0lUMhSMsWVrrml2Z+FwhWN7HL20915EAEX5DIVhCCJECr01Cn754A0PhCCfNX8tQOMJPF23ICbuyWYYQfkACRPgN10OwampqNFQbMOT34OEXn3utE/7JwvXtwiMUjvCThetdsysTn0uAeB+/tPVcRwJE+A3XQ7CKioo0VBsw5Pfg4Refe02A2CMfc5ebuSCT5691za5MfC4B4n380tZzHQkQ4TcUgiWEECnwmgAZX76GoXCElavNalgTK9bkhF3ZLEMIPyABIvyG6yFY1dXVGqoNGPJ78PCLz+vr6zl58mTW12f+Qj83yxgzz0w+X2q9kLCgbLVrdmXiczfOr+hd/NLWcx0JEOE3XA/BKi4u1lBtwJDfg4d8nh3y55rld1dvbmQoHOGYeatcK1s+DybyuztIgAi/oRAsIYRIQU1NDW+++WbW1NR4oox3Z61gKBzh5m1RhsIRfvj5ypywK5tlCOEHJECE33BVgLS0tLCqqootLS2ulCdyA/k9ePjF516bB/FWiXkDemO0haFwhO/PWeGaXZn4XHNAvI9f2nquIwEi/IbrIVglJSUaqg0Y8nvw8IvPvSZARk2v5mG3jGFLaxtD4QhHz1zuml2Z+FwCxPv4pa3nOhIgwm8oBEsIIVLgNQHyn2lL+NXbPmZbmxEgb85YlhN2ZbMMIfyABIjwG66HYFVWVmqoNmDI78HDLz73Wif8haLF/PqdBSTJwbeM4ajp1a7ZlYnPJUC8j1/aeq4jASL8husCZM6cObpRBQz5PXj4xefV1dW85pprWF2deUfezTKenlzFb901jiT51ds+5sjipa7ZlYnP3Ti/onfxS1vPdSRAhN9QCJYQQviEJyYu5HfvmUCSPOaOsXzpkyVZtkgI0RNIgAi/4foISGlpqZ6UBAz5PXj4xecNDQ0sKytjQ0ODJ8p4ZNx8nnjfRJLkN4YX8PmiRa7ZlYnP3Ti/onfxS1vPdSRAhN+QABG9jvwePPzic6/NAbn/40qe8uAkkuRxd4/n05OrXLMrE59rDoj38Utbz3UkQITfUAiWEEKkwGsC5J//K+fpD08mSX73ngn898SFOWFXNssQwg9IgAi/oUnooteR34OHX3zutU748A/LeM5jRSTJE+6byEfHL3DNrkx8LgHiffzS1nMdCRDhN7QMr+h15Pfg4Refe02A3Pr+PJ7/76kkyZMfKOS/Cua7ZpeW4Q0mfmnruY4EiPAbCsESQogUzJkzh/379+ecOXM8UcY/3vmCP3lqGkny1Icm8f6PK3PCrmyWIYQfkAARfsNVAdLc3MySkhI2Nze7Up7IDeT34CGfZ4cb3prLS5/9lCR52sOTeU+k3LWy5fNgIr+7gwSI8Buuh2BVVVVpqDZgyO/BQz7PDn96Yw4ve2E6SfKsR6dwxEdlrpUtnwcT+d0dJECE31AIlhBCpKCiooLHHXccKyoqPFHGH0bN4i9fnkGSPPfxqbwjvzQn7MpmGUL4AQkQ4TdcD8EqLi7WUG3AkN+Dh1987rVJ6NeMnMmrXikhSf7435/wlvfnuWZXJj7XJHTv45e2nutIgAi/4aoAaW1tZXV1NVtbW10pT+QG8nvw8IvPvSZArnqlhL99dSZJ8idPTeM/3vnCNbsy8bkEiPfxS1vPdSRAhN9QCJYQQqTAawLkly/P4B9GzSJJXvxMMW98+/OcsCubZQjhByRAhN9wPQSrqKhIQ7UBQ34PHn7xudc64Ve8+Bn/+Lo5zs+e+5R/eTPz5W0zCcFK1+cSIN7HL20915EAEX7D9RCsmpoaDdUGDPk9ePjF5xs3buTo0aO5ceNGT5Tx8+c/5fWW6Lj8xentYsQNuzLxuRvnV/QufmnruY4EiPAbCsESQgifcOmzn/KGt+eSjA/HEkJ4GwkQ4TdcD8EqLCzUUG3AkN+Dh198vmbNGj7yyCNcs2aNJ8q46Olp/PtoM+8jdkK6G3Zl4nM3zq/oXfzS1nMdCRDhN1wPwVq7dq2GagOG/B48/OJzr80BufCpabz5PbPy1W9encmrrSV53bArE59rDoj38Utbz3UkQITfUAiWEEKkwGsC5Px/T+Wt1rs//jBqFn9lvZQw23Zlswwh/IAEiPAbrgqQaDTKgoICRqNRV8oTuYH8Hjz84nOvdcLPfXwq77Tefv7H12fzihc/c82uTHwuAeJ9/NLWcx0JEOE3XA/Bqq2t1VBtwJDfg4dffO41AXLWo1M44qMykuRf3pzDnz//qWt2ZeJzCRDv45e2nutIgIjO+A3MxbHV4btvA5hofbcZwPsADktxnD8DmA+gCcBSAMMB9HPItx+AkQA2AGgAMB3AGWnarBAsIYRIwaJFi3jBBRdw0aJFnijjtIcn855IOUnyxrc/5yXPFueEXdksQwg/IAEiUnEwjLBYiWQBchSAegBTAfwIwFAAZVbeLyfkvQ1AG4D7AAwBcBOMEHkhId8AAKUAVgC4AsBZAPIBNAM4NQ27XQ/BikQiGqoNGPJ78JDPs8OpD03i/R9XkiT/8c4X/OnT01wrWz4PJvK7O0iAiFT8D8BHMCMSiQJkNID1iL9oQgCiAB6MSdsHQCOA5xP2vxVGlBwTk3YdzIV4YkxaXwDlAGakYberAqStrY11dXVsa2tzpTyRG8jvwcMvPo9Go1y3bl2vdq56soyTHyjkQwVGgNzy/jxe8OQnrtmVic/dOL+id/FLW891JECEE1fCjHAcgmQB0hcmPOo5h/3GAVgY8/8VMBfXCQn5DrTSb41JmwATppXILVbeg7tpu0KwhBAiBV6bA3LifRP5yPgFJMk78kt53uNTc8KubJYhhB+QABGJ7AczB+M66/+RiBcgR8JcMNchmX/BjGzsbP1/v5V3oEPe9QDeiPl/NczISiLnW8c4u1vWZyEEKz8/X0+7Aob8Hjz84nOvdcK/d88EPj5hIUlyxEdlPPvRItfsysTnEiDexy9tPdeRABGJvAugGEAf6/+RiBcgJ8FcML9w2NcerTjQ+v8FANtTlLMAZsTEJgrnUZUTrWNeluI4A2AuXns7GAA3btxIkmxpaWFLS0vS5+bm5rjP9moXqT5Ho9G4z/bQbFNTE7dt28bW1tb29La2tvYbV+xnO4/92X7LaqrPLS0tcZ+d6tEbdUqsh+rkXI/6+nq2tbX5pk5+9FNP1qmlpYUNDQ1xtnuxTiUlJQTAGTNm9JqfZs2a1d4J39E6ffvu8XxqUhVbWlp49//KePrDkzO+9mxxMHPmzG7VqbW1tb2dd9dPM2bMaC9D7cmbdWpra+PWrVu7rJ+X6pSLftqwYYMEiGjnYpgJ4rFzM0bCWYD83GF/W4AcYP3/AswcECcWACiI+T8K4FmHfLYAcRI8ADDC+j5umzrVDNOXlpaytNSsIT9nzhxWVppY4pKSElZVVZEki4uLWV1dTZIsKipiTU0NSbKwsJBr164lSRYUFLC2tpYkGYlE2kdY8vPzWVdXx6ampvYnJg0NDczPzydpFH4kEiFJ1tbWsqCggCS5du1aFhYWkiRrampYVGSe6lVXV7O42KzyUlVVxZIS89bfyspKzpkzx7U62R0t1cm5TkuXLuW0adPY1tbmmzr50U89WaelS5cyGo16vk75+fkEwP/85z+95qeioqJ2AbKjdfraHWP4zORFLC0t5d9em8pTH5qU8bVnC5BJkyZ1q07btm1jfn5++5yA7vjpxRdfJABGIhG1J4/Wqa2tjWVlZb6qUy76afTo0RIgAgCwG4A1AB4GsFfM9gaMANkLJpQqkxCsXR3y9lQIVlZHQOwfqKampkA9uQh6nbZv395+U/VLnfzop56sk+3zhoYGT9fJayMgXx9ewBeKFrOlpYUPjq3gyQ8UujYCEvtgSSMgwblH2J3mxsZG39QpF/2kERBhMwgOIwkJWz46JqE7jVYUIH4S+uXWfscn5DsAyZPQxwOodDjmzVbeg7pZD9dXwYptdCIYyO/Bwy8+b2lpYV1dXXsnI9fLOPqOsXzpkyUkyUfHL+AJ9010za5MfO7G+RW9i1/aeq6jOSDCZmeY93QkbgUwYVRDAHzdyvs2gLUAdo/Z/ysw4VsPxKTtbe2bKFZuRvIyvNciWaz0hXm/yGdp1MN1AdLQ0KAbVcCQ34OHfJ4dvnrbxxxZvJQk+cTEhfzuPRNcK1s+DybyuztIgIiuGAnnFxFuAVAE4DwAF8G8RLCzFxHeC/NCwb/DTEx3ehFhGYDlMCMnZ8K8XT3nX0So1TKCh/wePPzi84ULF/Lss8/mwoULPVHG4FvG8LXpJg79qUlVPO7u8a7ZlYnP3Ti/onfxS1vPdSRARFeMRLIAAYDvAJgIYBuAOgAfABic4hjXw0w6bwKwDGbieD+HfPsDeBVALczIyXQYIZIOeg+IEEKkwGvL8A66OcI3ZiwjST43ZRG/MbwgJ+zKZhlC+AEJEOE39CZ00evI78HDLz73Uie8tbWNoXCEb5csJ0m+OHUxj7ljrGt2ZeJzCRDv45e2nutIgAi/4XoIViQS0VBtwJDfg4dffO4lAdLU3MpQOMJ3Z60gSb4ybQm/etvHrtmVic8lQLyPX9p6riMBIvyGQrCEECIFXhIgjdEWhsIR5s817x94bXo1B98yJut2ZbsMIfyABIjwG64KkNbWVtbW1ravgy2CgfwePPzi83Xr1vGpp57iunXrcr6MLdubGQpH+NHnK0mSb8xYxlA4knFoTLp2ZeJzN86v6F380tZzHQkQ4TdcD8EqKCjQUG3AkN+Dh3zuPpsbogyFI/x43iqS5NszlzMUjrCl1Z3YfPk8mMjv7iABIvyGQrCEECIFtbW1HDVqFGtra3O+jNqtTQyFIywoW02SfG/2CobCEW5vzuwlf16quxB+RwJE+A3XQ7DWrl2rodqAIb8HD7/43EvzINbVb2coHOHEijUkyfy5NQyFI9zW1OyKXZn4XHNAvI9f2nquIwEi/IarAqS5uZmFhYVsbs7sB1F4E/k9ePjF514SIKs3NzIUjnDS/LUkycgXqxgKR1jXmFloTLp2ZeJzCRDv45e2nutIgAi/oRAsIYRIgZcESM2mBobCEU5daCZ0jy1dzVA4wo1bm7JqV7bLEMIPSIAIv+F6CFZNTY2GagOG/B48/OJzL3XCl23YxlA4wuKq9STJCeVrGApHuK5+uyt2ZeJzCRDv45e2nutIgAi/4XoIVlFRkYZqA4b8Hjz84vP58+fzhBNO4Pz583O+jCXrtzIUjvCzxRtIkpPmr2UoHOHqzY2u2JWJz904v6J38Utbz3UkQITfUAiWEEL4gKq19QyFI5y51KwoNXXhOobCEa7YuC3LlgkhdhQJEOE3XA/Bqq6u1lBtwJDfg4d87j7zVxsBMmfZRpJk8aL1DIUjXLbBHQEinwcT+d0dJECE33A9BKu4uFhDtQFDfg8efvG5l+aAlK3czFA4wi9WbCJJzlhSy1A4wsXrtrhiVyY+1xwQ7+OXtp7rSIAIv6EQLCGESIGXBMgXKzYxFI6wbOVmkuSs6o0MhSNcuKY+q3Zluwwh/IAEiPAbrgqQlpYWVlVVsaUlszfzCm8ivwcPv/jcS53wOcuM4Ji/2giOz5cbQVKxKrP7e7p2ZeJzCRDv45e2nutIgAi/4XoIVklJiYZqA4b8Hjz84nMvCZCZS03IVdVaI0BKa0xIVmnNZlfsysTnEiDexy9tPdeRABF+QyFYQgiRbH10RQAAIABJREFUAi8JkM8Wb2AoHOGS9VtJkpWr6xgKRzh3+aas2pXtMoTwAxIgwm+4HoJVWVmpodqAIb8HD7/4vLGxkVVVVWxszOxdGm6WUVwVv+qVvSzvrOpaV+zKxOdunF/Ru/ilrec6EiDCb7guQObMmaMbVcCQ34OHfO4+9ns/ajY1kEx+MWFvI58HE/ndHSRAhN9QCJYQQqRgyZIlvOKKK7hkyZKcLyPxzefLa7cxFI6wuGp9Vu3KdhlC+AEJEOE3XB8BKS0t1ZOSgCG/Bw+/+NxL8yAmVqxhKBzhuvrtJMmaTQ0MhSMsWrDOFbsy8bnmgHgfv7T1XEcCRPgNCRDR68jvwcMvPveSACkoW81QOMLarU0kyTV1jQyFI5xUudYVuyRAgolf2nquIwEi/IZCsIQQIgVeEiAfz1vFUDjCzQ1RkuT6LdsZCkc4vnxNVu3KdhlC+AEJEOE3NAld9Drye/Dwi8+91An/6POVDIUj3LLdvI9h07YmhsIRji1d5YpdmfhcAsT7+KWt5zoSIMJvaBle0evI78HDLz5ftWoVhw8fzlWrMuvEu1lG/twahsIRNkbNOa9vjDIUjvB/X6x0xa5MfO7G+RW9i1/aeq4jASL8hkKwhBDCB7w7awVD4QibmltJkg1NLQyFI8yfW5Nly4QQO4oEiPAbrgqQ5uZmlpSUsLm52ZXyRG4gvwcPv/i8rq6OBQUFvXqP7Kky3i5ZzlA4wtbWNpJkU3MrQ+EI3521whW7MvG5G+dX9C5+aeu5jgSI8Buuh2BVVVVpqDZgyO/Bwy8+99IckDdmLOOgmyPt/7e2tjEUjvDtkuWu2JWJzzUHxPv4pa3nOhIgwm8oBEsIIVLgJQHy2vRqDr5lTFzaoJsjfP2zZVm1K9tlCOEHJECE33A9BKu4uFhDtQFDfg8efvG5lzrhI4uX8qu3fRyXdvitY/jap0tdsSsTn0uAeB+/tPVcRwJE+A1XBUhrayurq6vZ2trqSnkiN5Dfg4dffO4lAfLSJ0t49B1j49KOvP1j/mfaElfsysTnEiDexy9tPdeRABF+QyFYQgiRgrKyMg4ePJhlZWU5X8YLRYv59eEFcWlfu7OAL05dnFW7sl2GEH5AAkT4DddDsIqKijRUGzDk9+Ahn7vPM5MX8di7xsWlfXPEOD47ZZEr5cvnwUR+dwcJEOE3XA/Bqqmp0VBtwJDfg4d87j5PTarid/45Pi7t23eP51OTqlwpXz4PJvK7O0iACL+hECwhhEjBF198wX333ZdffPFFzpfx+ISF/N49E+LSvnfPBD4+YWFW7cp2GUL4AQkQ4TdcD8EqLCzUUG3AkN+Dh1987qVJ6I+MX8AT75sYl3bifRP5yLj5rtiVic81Cd37+KWt5zoSIMJvuB6CtXbtWg3VBgz5PXj4xedeEiAPFVTy5AcK49J+8GAhHxxb6YpdmfhcAsT7+KWt5zoSIMJvKARLCCFS4CUBcv/HlTz1oUlxaUP+NZn3janIql3ZLkMIPyABIvyGqwIkGo2yoKCA0WjUlfJEbiC/Bw+/+NxLnfB7IuU87eHJcWlnPDKFd/+v3BW7MvG5BIj38Utbz3UkQITfcD0Eq7a2VkO1AUN+Dx5+8fmWLVv46aefcsuWLTlfxoiPynjWo1Pi0s55rIjDP8zsHRvp2pWJz904v6J38Utbz3UkQITfUAiWEEL4gDvzS3nu41Pj0n70xFTe/kFpliwSQvQUEiDCb7geghWJRDRUGzDk9+DhF5+vWLGCN9xwA1esWJHzZdz6/jye/+94AXLhk5/w5vfmuWJXJj534/yK3sUvbT3XkQARNt8CMAbAcgCNADYCmA7gSoe83wYwEcBWAJsBvA/gsBTH/TOA+QCaACwFMBxAP4d8+wEYCWADgAar7DMyqIerAqStrY11dXVsa2tzpTyRG8jvwcMvPvfSHJCb3/uCFz41LS7toqen8aZ3PnfFrkx8rjkg3scvbT3XkQARNkMAPAcjOE4D8GMAb8JcHLfH5DsKQD2AqQB+BGAogDIAKwF8OeGYtwFoA3CfdfybYITICwn5BgAoBbACwBUAzgKQD6AZwKlp1kMhWEIIkQIvCZC/j/6cFz0dL0AuebaYN7w9N6t2ZbsMIfyABIjois9gRkVsRgNYj/gLJgQgCuDBmLR9YEZSnk843q0wouSYmLTrYC7CE2PS+gIoBzAjTXtdD8HKz8/XUG3AkN+Dh1987qVO+A1vz+Wlz34al/bz5z/l9W/OccWuTHwuAeJ9/NLWcx0JENEVEQBLrM99YcKjnnPINw7Awpj/r4C5sE5IyHeglX5rTNoEmDCtRG6x8h6chr2uh2A1NDRoqDZgyO/Bwy8+95IAuf7NOfz58/EC5IoXP+N1r2d23ExCsNL1uQSI9/FLW891JEBEIjvBCI0vw4xMNAP4vfXdkTAXy3UO+/0LZmRjZ+v/+628Ax3yrgfwRsz/q2FGVhI53zrG2WnY77oAiUajulEFDPk9ePjF58uWLeN1113HZcuW5XwZf3x9Nq948bO4tF+9PIO/f22WK3Zl4nM3zq/oXfzS1nMdCRCRyHMwFwRh5mtcG/PdSVb6Lxz2s0crDrT+fwHA9hRlLIAZMbGJwnlU5UTrmJd1Yu8AmIvX3g4GwI0bN5IkW1pa2NLSkvS5ubk57rO93neqz9FoNO6zfWPatm0b8/Pz2dTU1J5u37xIxn1ubW2N+9zc3Nzp55aWlrjPTvXojTol1kN1Sq7H9u3b24fo/VInP/qpJ+tk+7yhocE3dcp1P/3+tZm88qXP4upx9SslvOaVElfq1NTU1N7O5afg1MkOwWpsbPRNnXLRTxs2bJAAEXF8BcB3YSaYPwugFcDfre9sAfJzh/1sAXKA9f8LMHNAnFgAoCDm/6hVViK2AHESPDYj0CGY2repU83SjaWlpSwtNWvGz5kzh5WVlSTJkpISVlVVkSSLi4tZXV1NkiwqKmJNTQ1JsrCwkGvXriVJFhQUsLa2liQZiUTaR1jy8/NZV1cX90PV0NDA/Px8kkbhRyIRkmRtbS0LCgpIkmvXrmVhYSFJsqamhkVFRSTJ6upqFhcXkySrqqpYUlJCkqysrOScOXNcq1NDQ0NcHKzqFF+npUuXctq0aWxra/NNnfzop56s09KlSxmNRj1fp3nz5nH27NksLCzsNT+tXr2ajz/+OLdt27ZDdbrs6ckc+viEuDr99tWZHPr4hIyuvW3btvHJJ59sXyK3qzrZD5jsVZG646dx48Zx9uzZrKysVHvyaJ3a2tpYVlbmqzrlop9Gjx4tASI65VmYMKwvI7MQrF0d8vZkCFZWR0Campq4bdu29qcSQXlyEfQ6tbS0sL6+nm1tbb6pkx/91JN1amlpaf9B9XKdSkpKCIAzZszoNT/NmjWrfR7EjtTpqv/M4DWvlMTV49r/zuKVL07P6Nqz52fMnDmzW3VqbW1tb+fd9dOMGTPay1B78mad2trauHXr1i7r56U65aKfNAIiuuIqmAvkeHRMQncarShA/CT0y2P2i+UAJE9CHw+g0uGYN1t5D0rDXq2CJXod+T14+MXnXpqE/uv/JM/3+NMbc3jZC9NdsSsTn2sSuvfxS1vPdTQHRHTFazBhWPY7Pt4GsBbA7jF5vgIzX+SBmLS9YUKwEsXKzUhehvdaJIuVvjDvF/ksTXv1HhAhhEiBlwTIlS99xuv+G3+Mv741l5c+92mKPdyxK9tlCOEHJECEzQsAHgbwM5iX/10M4C2Yi+OhmHxHAdgCoAjAeQAugnmJYGcvIrzXOubfYSamO72IsAzmfSOXAzgT5u3qOf8iQjs22B52FMFAfg8efvG5lzrhv3h+Ov/8Rvw7P/42+nMOfabYFbsy8bkEiPfxS1vPdSRAcoMBMEvdvgnzTowjrPSfADjMJRuugnm7+XqYjv8mAFNg3oyeyHcATASwDUAdgA8ADE5x3OthJp03AVgGM2m8n0O+/QG8CqAWZuRkOowQSRfXQ7AikYiGagOG/B48/OLzuXPncvfdd+fcuZm9TdzNMi597lPe8Fb8McLvfsELn5qWYo+etSsTn7txfkXv4pe2nutIgGSffWFGENoArIIJd/q29d0rAJ7Jkl1eRSFYQgjhA4Y+U8y/jf48Lu3W9+fx/H9PzZJFQoieQgIk+/wHwAoY0dEXRojYAuRXAMqzZJdXcVWAtLa2sra2tn0VCBEM5PfgIZ+7z0+emsbwu1/Epd2ZX8pzHitypXz5PJjI7+4gAZJ91sGEPwFAHuIFyOkwIU6i+7geglVQUKCh2oAhvwcPv/i8vLycxxxzDMvLy3O+jB//+xPe8v68uLS7PirnmY9MccWuTHzuxvkVvYtf2nquIwGSfbbDCA0gWYCcAzPPQnQfhWAJIUQKvDQJ/bzHp/KO/NK4tHvHVPC0f03Oql3ZLkMIPyABkn0qYVaLApIFyB0A5mTDKA/jegjW2rVrNVQbMOT34OEXn3upE372o0Uc/mFZXNoDYyt5yoOTXLErE59LgHgfv7T1XEcCJPvcDqAeZsUrew7IcQC+B2ADgBuyZ5oncVWANDc3s7CwsP1NoyIYyO/Bwy8+95IAOeORKfzn/+JDmR4eN58n3V/oil2Z+FwCxPv4pa3nOhIg2acfgLEwwqPW+rsOZjWsMQB2yp5pnkQhWEIIkQIvCZAh/5rM+8ZUxKU9NmEBv3/vhKzale0yhPADEiC5QR8AvwAwCsB4AG/AvJBP4iN9XA/Bqqmp0VBtwJDfg4dffL5p0yZ++OGH3LRpU86XccqDk/jg2Mq4tCcLF/I7/xzvil2Z+NyN8yt6F7+09VxHAkT4DddDsIqKijRUGzDk9+Ahn7vPSfcX8uFx8+PSnpm8iMfeNc6V8uXzYCK/u4MEiPAbCsESQogUrF69mvfddx9Xr16d82Ucf+9EPjZhQVzaC0WL+fU7C7JqV7bLEMIPSIBkn6UAlnSyLc6eaZ7E9RCs6upqDdUGDPk9ePjF516aB/Gdf07gvycujEt7+ZMlPOr2sa7YlYnPNQfE+/ilrec6EiDZZySAVxK2/8FMSF9q/S+6j+shWMXFxRqqDRjye/Dwi8+9JEC+ddc4Pj25Ki7t1U+X8ohbP3bFrkx8LgHiffzS1nMdCZDcZR+Yd4Bckm1DPIZCsIQQIgVeEiDfGF7A56YsiksbNb2a/3dzJKt2ZbsMIfyABEhucxmA0mwb4TFcFSAtLS2sqqpiS0uLK+WJ3EB+Dx5+8bmXOuHH3DGWL32yJC7tzRnLGApH2Nra1ut2ZeJzCRDv45e2nutIgOQ2ZwHYlm0jPIbrIVglJSUaqg0Y8nvw8IvPFy9ezEsuuYSLFy/O+TKOuO1jjixeGpc2euZyhsIRRlvSj89P165MfO7G+RW9i1/aeq4jAZK79AMwGkBZtg3xGArBEkIIH/B/N0f438+q49Len7OCoXCEjVE9nRbCy0iAZJ9JDlsxgPUwb0O/MnumeRLXQ7AqKys1VBsw5Pfg4RefNzU1ccWKFWxqasrpMlpb2xgKR/h2yfK49A8/X8lQOMKt29N/Op2uXZn43I3zK3oXv7T1XEcCJPtMATA5YRsL4FkAJ2XPLM/iugCZM2eOblQBQ34PHn7xuVfmgDQ1tzIUjvDdWSvi0sfMW8VQOMLNDdFetysTn2sOiPfxS1vPdSRAhN9QCJYQQqTAKwJkW1MzQ+EI8+fWxKUXlK1mKBxh7db0Rxi8UnchgoAEiPAbro+AlJaW6klJwJDfg4dffO6VTnhdY5ShcIRj5q2KS59YsYahcIRr6xp73a5MfC4B4n380tZzHQmQ7PDDNDfRfSRARK8jvwcPv/jcKwKkdmsTQ+EIC8pWx6UXLVjHUDjCmk0NvW6XBEgw8Utbz3UkQLJDG8wE8642O5/oPgrBEkKIFHhFgKyta2QoHGFh5Zq49OJF6xkKR7h0/das2JULZQjhByRAssOpaW6i+2gSuuh15Pfg4Reft7a2cvv27WxtTf89Gm6WUbOpgaFwhEUL1sWllyytZSgcYdXa+l63KxOfu3F+Re/il7ae60iACL+hZXhFryO/Bw/53F2WbdjGUDjC4qr1celzl29iKBxh+crev8fL58FEfncHCRDx/+y9eXwb1bn///y+t7f3e+/t9u3eclvRUgqldIMulJatUFpKC90oZWkphS7Q0pZLqbKvJJCEEMhGNrLvC3GS8ZI4XhPvjh3vsWXHsi0vsi3vlqxtPr8/xhpJtuTIwhqNZp736zWvlzRzNOc5eubMnM+c55yjNTgEi2EYJgz19fW44447UF9fr+o8GruHYTAKKLpsC9pfZRmAwSigoq0/LnapIQ+G0QIsQNTBB4nor0S0gYi2T9jeiqNdiYiiAsTtdqO4uBhu9/QXxWISF/a7/tCKzxNlHMSlziEYjAIutPQF7a/vkvaXmvvC/HLm7IrG5zwGJPHRSl1XOyxA4s+nichKRMMkDTi3EpGbpAHoNiK6HD/TEhLFQ7BMJhN31eoM9rv+0IrPE0WAVLdLPR2VbQNB+5vGe0YKmnpjblc0PmcBkvhopa6rHRYg8Wc/EWUS0X+RJDpuIqJ3E9FfiKidiL4QP9MSEg7BYhiGCUOiCJCLYcZ6tNqksSHnGnrC/DK2dqkhD4bRAixA4k8zET1ERP+HJAFyc8CxJUQkxMOoBEbxEKy8vDzuqtUZ7Hf9oRWfJ0ojvNQszXbV0BU821XngDQ9b2adNeZ2ReNzFiCJj1bqutphARJ/7ET03fHPbiK6K+DY3UQ0oLhFiY2iAsTr9cJsNvOUizqD/a4/tOLznp4ebN26FT090+9BUDKPwqZeGIwCLk9Y76NneAwGo4DTExYojIVd0fhcif+XiS1aqetqhwVI/GkmogfGP5uI6MWAY08RUa/iFiU2HILFMAyT4Jw3SQsOttpGg/YP2F0wGAUIFR1xsoxhmJmABUj8OUBEC8c/ryBpMPo8IvoXEfUQ0eE42ZWoKB6ClZOTw121OoP9rj+04vNE6QHJumSFwSigY8AetH/U6YbBKCCp3BJzu6LxOfeAJD5aqetqhwVI/LmZiH4x/vm/iegkSbNhiUSUTUSfiI9ZCYviIVgWi4W7anUG+11/aMXniTIGJL2mCwajgO6hsaD9Lo8XBqOAI6VtMbcrGp/zGJDERyt1Xe2wAFEn7yOi98bbiASFQ7AYhmHCkCgCJLWqEwajgL4RZ9B+URRhMArYX9QSF7vUkAfDaAEWIPHnxngboDEUD8HKyMjgrlqdwX7XH1rxeaI0wk9VtMNgFDDkcE069rk5ydid3xxzu6LxOQuQxEcrdV3tsACJPyIRFRPRM0T0gTjbogUUD8GyWq3cVasz2O/6Qys+TxQBcrzMAoNRgMM1eTG46+elYtu5yzG3KxqfswBJfLRS19UOC5D48wwRFZIkROwkLUz4/bhalNhwCBbDMEwY6uvrcccdd6C+vl7VeRwpbYPBKMDlmdwIvHFhGjZlN8bFLjXkwTBagAWIevgCEa0kafVzkYhaSVqI8Jp4GpWAKCpAXC4X0tLS4HJNDhNgtAv7XX+wz5XlQFELDEYBoihOOnbTkjNYl9EQcxvY5/qE/a4MLEDUx/8hovtJmn53jIg88TUn4VA8BMtms3FXrc5gv+sPrfjc6/VibGwspuWYiTx2F5hxzezkkMe+uSwdr52Zfg/DdO2KxudK/L9MbNFKXVc7LEDUyXVE9AoRdZE0JS8TORyCxTAME4ZEGQOy4/xlXDcvJeSx77ySgZVpdXGxSw15MIwWYAGiHt5LRE8TUR5JosNORPuI6J54GpWAKB6CJQgCd9XqDPa7/tCKzxOlEb41twlfXJAW8tidq7KwLLk25nZF43MWIImPVuq62mEBEn/uIqLdRDRC0tiPIiL6MxG9P55GJTCKChBRFDE4OBgyTpnRLux3/aEVnyeKANmY1YivLD4d8tj3X8vGopPVMbcrGp+zAEl8tFLX1Q4LkPgjkhRq9SoR3RBnW7QAh2AxDMOEIVEEyLqMBty8ND3ksftez8Xc45VxsUsNeTCMFmABEn8eJKJ/i7cRGkLxEKykpCTuqtUZ7Hf9oRWfJ0oj/LUz9fjWsrMhjz2w7hyMRytiblc0PmcBkvhopa6rHRYgjI/vEdF2IrpERKMkTQd8gohuDpH2JiI6S1LY2AARvU1Enw1z3ufGz+kkomYiWkhE/x4i3UeJaCcR9ZI0/qWAiO6OohyKh2DZ7XbuqtUZ7Hf9oRWfO51OtLW1wel0qjqPlWl1+M4rGSGP/WJjHp4/VB5zu6LxuRL/LxNbtFLX1Q4LEMbHESLKJGlhxDuI6JckiQA3SeLEx/VENEREuUT0IyL6ORFVkyRYPjLhnHNJCjFbTkR3EtGLJAmRLRPS/QcRVRFRGxE9RtJCjEnjed8xzXIoLkBcLhffqHQG+11/sM+VZXlyLe5clRXy2MOb8/Hc/rKY28A+1yfsd2VgAcL4+GiIfe8haXzK2YB9h4moh4IvGAMRuYhoRcC+DxGRg4g2TzjnHJJESeB4l2dJugi/HbDvXURUQ9Kg/OnAIVhMzGG/6w+t+LypqQm//OUv0dTUpOo8lpyqwd2rs0Mee3xbIZ7ZWxpzu6LxuRL/LxNbtFLX1Q4LEOZKZBJR/fjnd5EUHrUpRLrTRNQQ8P0xki6sWyak+8T4/jkB+9JJCtOayOzxtFdNw17uAWFiDvtdf2jF54kyBmRBUhV+sCYn5LEndxTjqZ0lMbcrGp/zGJDERyt1Xe2wAGGm4v3kH+NBJC2QCJJ6LCayiqSejf87/v3l8bT/HSJtDxHtD/jeSVLPykTuHz/HvdOwmceAMDGH/a4/tOLzRBEgs9+uxI/Xngt57I+7S/DE9qKY2xWNz1mAJD5aqetqhwUIMxV7SRqH4RuIfitJF8uvQ6T19VZ8Yvz7FiIaC3PeepJ6THy4KHSvyrfHz/nIFDb+B0kXr2+7iojQ19cHAPB4PPB4PJM+u93uoM9er3fKzy6XK+iz78Y0OjqKpKQkOJ1Oeb/v7QmAoM9erzfos9vtnvKzx+MJ+hyqHLEo08RycJkml2NsbEzuotdKmbTop5ksk8/ndrs9octUXFwMIkJRUVHM/FRaWio3wqMt07+OVODB9edCluOZvaV4ZEtBSD9NVSafOCgpKYmoTE6nU67nkfqpqKhIzoPrU2KWyReC5XA4NFMmNfqpt7eXBUgc+PQ0t3iwlKQL468B+3wC5OEQ6X0C5OPj37eQNAYkFPVElBbw3UVEb4ZI5xMgoQSPj0XjaYK23NxcAEBVVRWqqqoAAGVlZairqwMAFBcXw2QyAQDy8vJgNpsBADk5ObBYLACAjIwMWK1WAEBaWhpsNhsAQBAEuYfF1yAJjBm12+1ISkoCICl8QRAAADabDWlp0sq+VqsVGRnSDC8WiwU5OVKogdlsRl5eHgDAZDKhuLgYAFBXV4eysjIuE5eJy8RlekdlSkpKAhFh+/btMStTTk6OLECiLdPzh8rxo9VnQ5bpNxsz8OPXzk7bTz4BkpmZGTM/bd26FUQEQRD42uMycZmmKNPhw4dZgMQBkYi809iUZiFNHqdBFF0I1n+FSDuTIVhx7QFxOp0YGBiQ30ro5c2F3svk8XjQ19cHURQ1UyYt+mkmy+TxeDA4OBhkeyKWyWKxYPny5Whra4uZnzo6OrB06VJ0dnZGXabn9pfh4c35IcvxwqFy/HT9uZB+mqpMnZ2dWLp0Kdrb2yMqk9frhc1mm1S+qcrU1taG5cuXo729netTgpZJFEX09/dfsXyJVCY1+ol7QOLD74joiWlsSuITHwtDHPMNQg/VW5FGwYPQHx0/z7cmpPs4TRY3Z4ioLsQ5Z42n/WQkho+j+CxYgiDIlYrRB+x3/cE+V5Zn917A49sKQx6bdSz8+JCZhH2uT9jvysBjQJhA5pN0MSydIs0hIrIS0XsD9n2apPU9XgnY90GSQrAmipVZNHka3mdoslh5F0nrixRGbj4RKSxAGIZhEon+/n6cOHEC/f39qs7jD7tK8LswA82nmiEr1napIQ+G0QIsQNTFf5IUQvSuOOT9AkkXQipJU+dO3HxcT0TDRJRDRPcR0c9IWkRwqoUIl5G0oOA/SRqYHmohwmoiaiWp5+QekmbeUv1ChL4uel8XJKMP2O/6Qys+T5RZsJ7cUYynd4WeanfJqRp879WsmNsVjc95FqzERyt1Xe2wAFEHd5G06rhnfLtpfP8GklYaV4JsCjGYO2AL5GaSFiccJaJBIjpORNeEOe/fSBp07iSiFpIGjf97iHQfI6JdRGQjqeekgCQhMl0UD8FKS0vjrlqdwX7XH1rxeaIIkN+8VRR2scHlKbW4fWVmzO2KxucsQBIfrdR1tcMCJP58j6Q3/eUkhTCJ5Bcgi4noZJzsSlQ4BIthGCYMiSJAHtlSgL/uLwt57NXTl/Dt5WfjYpca8mAYLcACJP6cJ6IT45/fRcEC5OckhSUxkaN4CJbVauWuWp3BftcfWvF5ojTCH3ozH88fLA957PX0Bnz9pfSY2xWNz1mAJD5aqetqhwVI/Bkloh+Nf/43ChYgt1H4xfyY0CgqQNxuNzIyMuRp7hh9wH7XH1rxeU1NDW644QbU1NSoOo+fbjiPF49cDHlsQ5YJX1l8OuZ2ReNzJf5fJrZopa6rHRYg8WeQiB4c/zxRgPyCpDUzmMjhECyGYZgE5yfrzmHWscqQx7bkNOGLC9IUtohhmJmEBUj8OUtER8c/TxQgB4jHgEwXxUOwLBYLd9XqDPa7/mCfK8sPX8/F/KSqkMd2nL+Mz89NibkN7HN9wn5XBhYg8eceklY7P07SlLZektbFWE/S4PTvxs+0hETxEKycnBzuqtUZ7Hf9oRWfl5eX473vfS/Ky0OPr1BLHveszsbik6HDmPYUmPGZWULM7YrG50r8v0xs0UpdVzssQNTB40TUS1Lvh2/rI6LH4mlUgsIhWAzDMGFIlEFYjLE4AAAgAElEQVTod63KwrLk2pDHDha3wGAU4PGKitulhjwYRguwAFEP/0lEd5O0EN+9RPTf8TUnYVE8BMtsNnNXrc5gv+sPrfg8URrh312RgRWpdSGPHbvQBoNRgMPliald0ficBUjio5W6rnZYgKifeKyKnsgoHoKVl5fHXbU6g/2uP7Ti80QRILcsP4vVZ+pDHjtxsR0Go4Dhsen5Yrp2ReNzFiCJj1bqutphARJ/dpHU+xGKq4moUDlTNAGHYDEMw4QhUQTI119Kx9qzDSGPpVZ1wGAU0DfiVNwuNeTBMFqABUj8GSGiaiK6YcL+n5I0DqRZcYsSG0UFiMfjgclkgsczvVAAJrFhv+sPrfh8dHQUFy5cwOjoqKrz+Ori09iQZQp5LL2mCwajAOuQI6Z2ReNzJf5fJrZopa6rHRYg8ecGIqohSYj8jqSQqzdIGoh+goj+X9wsS0wUD8EqLi7mrlqdwX7XH+xzZblxQRq25DSFPJZd3w2DUUB7vz2mNrDP9Qn7XRlYgKiD/ySinSRNwdtO0urn/xtPgxIYDsFiGIYJQ0tLC5599lm0tLSoOo/r56Vi+/nLIY/lmXpgMAow944obpca8mAYLcACRD08SER2kno+LhDRR+NrTsKieAhWXV0dd9XqDPa7/tCKzxNlHMTn5iRjd35zyGNFl20wGAWYrMMxtSsan/MYkMRHK3Vd7bAAiT//RkSvktT7cZKIfkFEXST1hNwZP7MSFsUFSFlZGd+odAb7XX9oxeeJIEBEUYTBKGB/UehehLKWPhiMAmo7pnefj0aATNfnLEASH63UdbXDAiT+5BGRi4heDNj3CSLKJmkl9PlxsCmR4RAshmGYMCSCAHF7vDAYBRwuaQ15vMoyAINRQGXbgKJ2qSUPhtECLEDiTysR3Rpi//8hopdIEiFM5CjeA1JVVcVvSnQG+11/aMXnidAId7g8MBgFvF3WFvL4pc4hGIwCSs19MbUrGp+zAEl8tFLX1Q4LkPjzoSscv0cRK7QDCxAm5rDf9YdWfN7W1obnn38ebW2hG/dqyGN4zA2DUcDJi+0hjzd2D8NgFFDY1BtTu6LxuRL/LxNbtFLX1Q4LEEZrcAgWwzBMAtM/6oTBKCC1qiPk8VbbKAxGAedNPQpbxjDMTMECJD7cTkTvCfh8pY2JHB6EzsQc9rv+0IrPh4eHkZ+fj+Hh6c0gpWQePcNjMBgFnKnpCnm8Y8AOg1FA5iVrTO2KxudK/L9MbNFKXVc7LEDig0hE3wz47A2z+Y4xkcPT8DIxh/2uP7Ti80QYA9I54JhSYHQPTS1QZsounoZXn2ilrqsdFiDx4Q7y94DcEcHGRA6HYDEMw4QhEQSIL8TqXEPoEKuBURcMRgHJlaFDtGJll1ryYBgtwAKE0RqKChC3243i4mK43W5F8mPUAftdf2jF54nQCG/uGYHBKKAgzCDzUac0SD2p3BJTu6LxOQuQxEcrdV3tsACJH/9JRI8Q0SwieoqIPhJfczSD4iFYJpOJu2p1Bvtdf2jF54kgQExWaZrdkmZbyONer7RQ4YEwCxXOlF3R+JwFSOKjlbqudliAxIdPElEj+cd5iETUT0S3xNMojcAhWAzDMGGoqKjAhz/8YVRUVKg2j9qOQRiMAspb+8OmuW5eCt46d1lRu9SSB8NoARYg8WE7ETmIaDER3UdEfyMiKxGVx9MojaB4CFZeXh531eoM9rv+YJ8rR2WbtNJ5lSX8SudfW3IG6zNNMbWDfa5P2O/KwAIkPrQR0fwJ++4nqUfkY8qboykUFSBerxdmsxler1eR/Bh1wH7XH+xz5ShptsFgFGCyhp/K9taXM7AyrS6mdrDP9Qn7XRlYgMQHN02e3eq/SQrF+rLy5mgKDsFiGIYJQ3V1Na655hpUV1erNo9zDT0wGAW02kbDprlndTYWnZze+ROh7AyjF1iAxIfAdUB8/Nv4/puUN0dTKB6ClZOTw121OoP9rj+04vNEGIR+trYLBqOA7qGxsGkeWHcOxqPTG2cRzSxY0/U5D0JPfLRS19UOC5D4IBLRb0gSG77tG+P7H52wnwXJ9FA8BMtisXBXrc5gv+sPrfg8EQSIUNEBg1HAoMMVNs3Dm/Px3P6ymNoVjc9ZgCQ+WqnraocFSHwIt/r5xP28Evr04RAshmGYMCSCADl2oQ0GowCnO3wD8MkdxXhqZ4midqklD4bRAixA4sMT09yYyFE8BCsjI4O7anUG+11/aMXnidAI31fYgqtnCRBFMWyaZ/dewKNbC2JqVzQ+ZwGS+GilrqsdFiCM1lA8BMtqtXJXrc5gv+sPrfh8cHAQaWlpMb1HvtM8tp+/jOvnpU6Z5oXDF/HTDedjalc0Plfi/2Vii1bqutphAcJoDQ7BYhiGSWA2ZjXiK4tPT5lmflIVfrAmRyGLGIaZaViAMFpDUQHicrmQlpYGlyv8YElGe7Df9YdWfN7R0YGFCxeio6NDtXmsSa/Ht5adnTLN8pRa3LYiM6Z2ReNzJf5fJrZopa6rHRYgjNZQPATLZrNxV63OYL/rD634PBHGgLycUofbV04tLl5Pb8DNS9Njalc0PucxIImPVuq62mEBwmgNDsFiGIYJQyIIkEUnq3Hva1OHV23JacIN86ceJzLTdqklD4bRAixAGK2heAiWIAjcVasz2O/6Qys+T4RG+KxjlfjJunNTptlTYL7iTFnv1K5ofM4CJPHRSl1XOyxAGK2hqAARRRGDg4PTeggyiQ/7XX9oxeeJIECeP1iOh97MnzKNb60Qh8sTM7ui8TkLkMRHK3Vd7bAAYbQGh2AxDMOE4fLly3jsscdw+fJl1ebx7N4LeHxb4ZRpUquk1dL7RpyK2aWWPBhGC7AAYbSG4iFYSUlJ3FWrM9jv+oN9rhy/j2CV8+z6bhiMAiz99pjZwT7XJ+x3ZWABwmgNxUOw7HY7d9XqDPa7/tCKzx0OB0wmExwOh2rzeGxrIZ7dN3UIU3GzDQajAJN1KGZ2ReNzJf5fJrZopa6rHRYgjNZQXIC4XC6+UekM9rv+0IrPE2EMyC825uH5Q+VTpqmyDMBgFFDR1h8zu6LxOY8BSXy0UtfVDgsQJpD3EtFKIjpDRD0kXRiLwqS9iYjOEtEIEQ0Q0dtE9NkwaZ8joktE5CSiZiJaSET/HiLdR4loJxH1EpGdiAqI6O5ploFDsJiYw37XH1rxeSIIkB+vPYfZb1dOmaapexgGo4CCpt6Y2RWNz1mAJD5aqetqhwUIE8jVJImJHCLaSuEFyPVENEREuUT0IyL6ORFVE1E7EX1kQtq5RCQS0XIiupOIXiRJiGyZkO4/iKiKiNqI6DEi+j4RJRGRm4jumEYZuAeEiTnsd/2hFZ8nggC5e3U2lpyqmTJN54ADBqOAzDprzOziHhB9opW6rnZYgDCB/H/jGxHRhym8ADlMUg9J4EVjICIXEa0I2PchInIQ0eYJv59Dkii5IWDfs+P5fTtg37uIqIaIiqZRBh4DwsQc9rv+0IrPE0GAfOOldLye3jBlmgG7CwajAKGiI2Z2ReNzFiCJj1bqutphAcKEI5wAeRdJ4VGbQvzmNBE1BHx/bPwct0xI94nx/XMC9qWTFKY1kdnjaa+K0G4OwWJiDvtdf2jF54kgQD4/NwXbz089ja3T7YXBKOBwSWvM7OIQLH2ilbqudliAMOEIJ0CuG9//bIjfrCKpZ+P/jn9/eTztf4dI20NE+wO+d5LUszKR+8fPcW+EdvM6IAzDMAmKw+WBwSjgaGnbFdNeMzsZu/KbY28UwzAzDgsQJhzhBMit4/t/HeI3vt6KT4x/30JEY2HOX09Sj4kPF4XuVfn2+DkfCXOe/yDp4vVtVxER+vr6AAAejwcej2fSZ7fbHfTZ6/VO+dnlcgV99nXNOp1ODAwMwOv1yvt98aMAgj770vg+u93uKT97PJ6gz6HKEYsyTSwHlyl0Ofr6+iCKombKpEU/zWSZPB4PBgcHg2xP9DKp0U9dA6MwGAWk13RdsUw3LkzDhsyGmJXJ6/XCZrNNKh/7SdtlEkUR/f39VyxfIpVJjX7q7e1lAcKE5EoC5OEQv/EJkI+Pf99C0hiQUNQTUVrAdxcRvRkinU+AhBI8NG4fJm65ubkAgKqqKlRVVQEAysrKUFdXBwAoLi6GyWQCAOTl5cFsNgMAcnJyYLFYAAAZGRmwWqUBjmlpabDZbAAAQRDkHpakpCScOnUKo6Ojcpet3W5HUlISAEnhC4IAALDZbEhLSwMAWK1WZGRkAAAsFgtycnIAAGazGXl5eQAAk8mE4uJiAEBdXR3KysoUK5Pdbg/qhuYyBZepqakJJ06cgMvl0kyZtOinmSxTU1MTBEFAdnZ2Qpfp9OnTuOWWW7B///6Y+am0tBTXXXcdLl26NO0yZZbUwGAUUHTZdsUyfXNZOl7YmRnxtXfp0iVcf/31KCwsjKhMg4ODQZ8j8dOePXtwyy23ICMjg+tTgpbJ5XLhxIkTKC0t1UyZ1Oinw4cPswBhQjKTIVj/FSLtTIVgxbUHRK9vLrhMXCYuU2KWqbi4GESEoqKimJWptLRUHgcx3TKVXO6FwSigrnPwimW6c1UWlpysjthPvvEZJSUlMfNTUVGRnAdfe1wmLhP3gDDT50qD0EP1VqRR8CD0R8fP8a0J6T5OkwehnyGiuhDnnDWe9pMR2q3oGBBfF72vAjL6gP2uP7Tic7UPQs+6ZIXBKKC9337FtPe9nos5V1gv5J3YFY3PeRB64qOVuq52eAwIE46ppuE9RERWkhYu9PFpktb3eCVg3wdJCsGaKFZm0eRpeJ+hyWLlXSStL1I4DbsVnwUrLS1NVvWMPmC/6w+t+FztAuTExXYYjAJGxtxXTPvLN/Pw/MGpV0x/J3ZF43MWIImPVuq62mEBwkzkPiL6JRE9SdKFcXj8+y/JH0p1PRENk7Rg4X1E9DOSFhGcaiHCZSQtKPhPkgamh1qIsJqIWknqObmHpNXVVb0QIcMwTCKhdgGyp8CMz85OlkM5puKJ7UX4w64SRexSUx4MowVYgDATMVOIQd3j29UB6W4morNENEpEg0R0nIiuCXPOv5E06NxJRC0k9ar8e4h0HyOiXURkI6nnpIAkITIdFA/Bslqt3FWrM9jv+kMrPrfZbNizZ488cFRteazPNOGri09HlPbZvRfw2NbCmNkVjc+V+H+Z2KKVuq52WIAwWkNRAeJ2u5GRkSEP8mL0Aftdf7DPlWF5Si1uX5kZUdoXj1zEg+vPx8wW9rk+Yb8rAwsQRmtwCBbDMEwYuru7sX79enR3d6syj1nHKvGTdeciSrvwRDW+/1q2InapKQ+G0QIsQBitoXgIlsVi4a5ancF+1x9a8bnax0E8uy/ysKqVaXW49eWMmNkVjc95DEjio5W6rnZYgDBaQ/EQrJycHO6q1Rnsd/2hFZ+rXYA8vq0Qz+wtjSjthiwTvhLheJFo7IrG5yxAEh+t1HW1wwKE0RocgsUwDBMGtQuQB9adg/FoRURpd+Y149o5KYrYpaY8GEYLsABhtIbiIVhms5m7anUG+11/aMXnam+E37kqC8uTayNKe6S0DQajAKc7Mp9EE4I1XZ+zAEl8tFLX1Q4LEEZrKB6ClZeXx121OoP9rj+04vOGhgbce++9aGhoUGUeNy05g/WZpojSplR2wGAU0D/qjIld0fhcif+XiS1aqetqhwUIozU4BIthGCYBEUUR18xOxu785ojS59R3w2AUYOm3x9YwhmFmHBYgjNZQVIB4PB6YTCZ4PB5F8mPUAftdf2jF5x6PB4ODgzEtR7R5jDrdMBgFJJVbIkpfarbBYBRQ3zUUE7ui8bkS/y8TW7RS19UOCxBGaygeglVcXMxdtTqD/a4/tOJzNY8B6Rp0wGAUkFHXFVH6us5BGIwCLrT0xcSuaHzOY0ASH63UdbXDAoTRGhyCxTAMEwY1C5DG7mEYjAIKm3ojSm/pt8NgFJBdH9mif2ouO8PoDRYgjNZQPASrrq6Ou2p1Bvtdf2jF52puhFe09cNgFFBlGYgo/ZDDBYNRwMmL7TGxKxqfswBJfLRS19UOCxBGayguQMrKyvhGpTPY7/pDKz5XswDJb+yFwSjgcs9IROlFUcRnZgnYW2iOiV3R+JwFSOKjlbqudliAMFqDQ7AYhmHCoGYBkl7TBYNRgHXIEfFvvrL4NDZkRTZtr5rLzjB6gwUIozUU7wGpqqriNyU6g/2uP7Tic5fLhe7ubrhcLtXlkVRugcEoYNQZ+eDf21Zk4uWUupjYFY3Plfh/mdiilbqudliAMFqDBQgTc9jv+oN9Hnv2FprxmVkCRFGM+Dc/XnsOs45VxsQe9rk+Yb8rAwsQRmtwCBbDMEwYGhsb8ZOf/ASNjY2qy2NzTiNuXJA2rd88urUAz+6LLNxJzWVnGL3BAoTRGjwInYk57Hf9oRWfq3kcxOoz9fjmsvRp/eaZvaV4fFthTOziQej6RCt1Xe2wAGG0Bk/Dy8Qc9rv+0IrP1SxAlp6qwV2vZk3rN8ajFXhg3bmY2MXT8OoTrdR1tcMChNEaHILFMAwTBjULkFnHKvCTCMWEj+Uptbj15YyY2jUdWIAwTGSwAGG0hqICxO12o7i4GG535LO2MIkP+11/aMXnam6E/3V/GX69uWBavzl2oQ0Go4AB+5VnnZquXdH4nAVI4qOVuq52WIAwWkPxECyTycRdtTqD/a4/tOLzrq4urF69Gl1dXarL47dvFeHpXSXT+k1D1xAMRgF5pp4Ztysanyvx/zKxRSt1Xe2wAGG0BodgMQzDJCB3rcrCklM10/qNxyvi+nmp2JzDs04xTCLBAoTRGoqHYOXl5XFXrc5gv+sPrfi8r68Phw8fRl9fn6rycHm8uGZ2MnYXmKed312vRiZcpmtXND5X4v9lYotW6rraYQHCaA1FBYjX64XZbIbX61UkP0YdsN/1h1Z8rtYxIJd7RmAwCjgfQSjVRB7alI+/Hyibcbui8TmPAUl8tFLX1Q4LEEZrcAgWwzBMGNQqQDLqumAwCrD026edX6Rrgai17AyjR1iAMFpD8RCsnJwc7qrVGex3/aEVn6u1Eb4puxE3zE+FKIrTzm/e8Sr88PXcGbcrGp+zAEl8tFLX1Q4LEEZrKB6CZbFYuKtWZ7Df9YdWfK5WAfL8oXI8uP58VPmtSa/HN1668grq0YRgTdfnLEASH63UdbXDAoTRGhyCxTAME4ba2lp87WtfQ21traryuH9tLv51pCKq/HYXmHHN7GR4vVP3nqi17AyjR1iAMFpD8RCsjIwM7qrVGex3/cE+jx0er4jr5qVga25TVL9PreqAwSigb8Q5o3axz/UJ+10ZWIAwWkPxECyr1cpdtTqD/a4/2Oexo3l8Bqzchu6ofl/XOQiDUUBhU++M2sU+1yfsd2VgAcJoDQ7BYhiGCUNZWRne/e53o6zsytPWKpVHWnUnDEYB1kFHVPm5Pd6IFiNUY9kZRq+wAGG0hqICxOVyIS0tDS6XS5H8GHXAftcfWvG5Ggehr8towJcXnY5qBiwfD72Zjz/tLp1Ru6LxOQ9CT3y0UtfVDgsQRmsoHoJls9m4q1ZnsN/1h1Z8rkYB8tf9ZXjozfx3lOeGLBM+PzcFA6PhG43RzII1XZ+zAEl8tFLX1Q4LEEZrcAgWwzBMGNQoQO59LQdzj1e+ozytQw58ZpaAg8UtM2ZXNLAAYZjIYAHCaA3FQ7AEQeCuWp3BftcfWvG52hrhLo8Xn5uTjF35ze843/vX5uL5Q+UzYhcQnc9ZgCQ+WqnraocFCKM1FBUgoihicHDwHcUuM4kH+11/aMXndrsd1dXVsNvtqshj0clqGIwC8hvf+QxWi05W47srMmbELiA6nyvx/zKxRSt1Xe2wAGG0BodgMQzDJABdg1LY1NO7SuB0v/N4+4y6LhiMAmo7+P7PMGqHBQijNRQPwUpKSuKuWp3BftcfWvG52WzGU089BbPZHPc83sxuxOfnpmDQMTP/qdPtxVcXn8bylNCrkE+37NH4XIn/l4ktWqnraocFCKM1FA/Bstvt3FWrM9jv+kMrPlfLGBBRFHH36mz87cDMrpcx73gVvrXsLLzeyX6abtmj8TmPAUl8tFLX1Q4LEEZrKC5AXC4X36h0Rjz9fq6hB6Vmm+L56h2t1HW1CJDy1v53tPp5OErNNhiMQsg6EsqujLouXO4ZmZS21TaKC+a+aft8uv+vw+WBJ4RYYuKHVuq62mEBwmiNuIZgNXYPY9axCvQMj8Hl8WLM7cGY24NFJ6vRMzwm/y4w3nnQ4ULnQPAKwCcutuPRrQX44+4SjLk9qG4fQMdA8KBGc+8IDEYBSeUW/GJjXtAgzt7hMaxIrcOdq7KQdckacXkq2vrx5z2l73hA6Jjbg9SqDlXdwK1DDrg8MzOvezy76A1GAQajgJbe0bBpWm2jWJlWp/j/39A1hKEZCqeZSaxDDrT3X3lQsNPthcPlkb87XB7szGuG2+PVTFjGTAiQxu5h1LSHv8eGyqO5ZwS7C8ywOz241DkEg1HAt5adnfHGt9vjxXXzUvBmdiPeONuAtWcbYLIOoa1vNMiuQyWtOD2+AvtXF5+edJ6H3syHwShg3V7J5zXtgzAYBTR0DU2Z/3T/X4NRwKxjFVGVNRSFTb1o7hnB3kJzSGE1XURRxKGS1qB6MZO8k3vUlX5bau4Luk7fONsw5csb24gTDpcn7nXdOuSAO8Jn1e4C84zMIBcPWIAwWiOuPSA/WXcOBqOAXfnN8gPsy4tOw2AUsPpMPQDgcEkrDEYB1kEHWm2juHnpGRiMAhwuD6yDDoiiiM/MEuSGpm97du8FiKKI2o5BPLvvAtak1wcdv25eCo5daAMAPLa1UN5/68sZEEURPeOiZOJgT69XxOnqTnxl8Wm8eOQiDEYBz+2fOiyiZ3hMLrPD5Zl0U9+V3zxpZptRpxuHS1onhUY43V4YjAIOl7TK+3wrI+c2dKOxexgHilpCPmw8XjGiwauiKMJgFPCvI/4HfattFJl1fnHWPTSGrEvWIPsWnqjGxqzGkOcL9Lvd6cGo0z0p3cHiFnx3RcYk2/cXtSC1qgOA1JgzWYMbNW6PN6hcXq/k957hMdmvJy+2T8qvf9SJhSeq8dTOkogaS12DDmw7dxljbg9Wn76EMzVdGHS4cKi49YoPd1EUYbIOy9+tQw5cNy8Fy5Ol+Pu9hWasSa+fJPp6h8dwsbV/ynNPLFNj9/AV0031wP75xjw8tOnKC93960gFHt9WKH/fW2iGwSggrbrzHb0VDXV9ON1e/HlPKS51Sj7qGnQgtapTPr8oijPaODdZh9Debw9qIA+PTb5mp0IURbm+GoyCvN824sQbZxtwvsGK0TGXnMcJQUBnfQmc5kK8sHI9fjt7GQ7u2oCkPWvxt9mzcKakxn/y9nKgcBOQvx44/waQuxrIXiltOauAvoAxFZ2VQN46KW3BRul3RVuA0p3AxQN4bPVR2cabjPvws1mv4VcLN2Ldpk0gIhw9dgxfNB7G1caTcrqJ//WtL2fAYBTw0qlqiKKIrblNMBgFHCgKv84I4BcghUUlSKnsgHVIernk9ngxMv5/e70i+kacGBh1TfovJxLY8BdFEU/tLMaZmq6QaTsHHEHPhK8tOTOlreWt/ThU3Bq0L6e+GyvT6uTv1e0DMBgFLDxRLe/75+GL8vMMkHqdfvNWEfpGnFPmN5EFSVX4+kvpcHm8+PuBMjyypQCvnakPGT4XiNPtxYtHLuJzc5Lx1rnLuNQ5hMq2Afl4W9+oLHINRgF/2XcBthGn/D3UYpW+Z8Tj2wrlul50uRezjlUG1clLnUNh7RNFESXNNnQPjeEfB8tR0Xbl+5zD5UGeqUf+PjzmhsEo4OUUvw9EUUSpuQ92Z7AI9HrFSdePr43QG/DCMxJ87Q8lYQHCaA3FBciD63Lx47W5sA46cM3sZBiMAr7xUvokAfHC4YsApMbQxGOB26HiVvnhF7h9aWEaUqs65O83zE/1i5N9F+Q8uwLs8G0LT1TjD7ukRmlJc7BYWHyyZlJed63KQkFTLwZG/Q2u4TE3vrggTX4QH7vQhuJmG767QrI1ubIDVZYB3L06G/88fDHo5u/2eLE8pRYGo4DHthbKvUH7i1pwaFyQGYwCzL0jGB5z477Xc+X/7K5VWdL/UtIK66ADewrMcHmkN9XP7ruAz89NgdvjhaXfjuNlFvkNvMcrYsmpGnQM2NHeb5dFmvFoBU5Xd+J324tgMAp4M7sR1iEHbll+FgajAKFCEga9AY39+gkN+T0FzfjGS2fw6JYCPL5NEnt/2FUCu9OD7PpuPLmjGA+Mi9HA3orKtgEcu9Am7y819wU9QFweL0zWIdy5Kgv3vZ6LkTE30mu6kFwp+f21M/VBPl2WXIshhwuttlH0Do9hVdolGIwCrp2bAoNRwDN7SzFgd+FwSSsOFbfiYHEL6ruGggTyxM33v1zqHMKWnCbcsvwsmrqHIYoi2vpG5Ya+T0jXdw0hv7EXD2+WzveDNTmYd7xKPt/eQjO2nbuMI6Vt6Bkew12vZuGmJWeQZ+pBUrkFgNSADaS2YxCz367Ed1dk4Onx6/Z324vG32D2oW/ECVEUsbvADKGiAxda+uT/s6ylD4D0QF2ZVodzDT0wGAV8YX4qmntG8Pi2wqDeozM1XThv6oHb48WXF53GtXNTZPHn+79fEmqQfcmKI0WX0dwzjPquIZyqaMeY298g2HbucpDAre0YhFDRgdSqDvx8Yx4+M0uQG/DfezULv3lL+p//efiiHI7ke2GxNbcJ33klA/e+lhMkeERRRHPPCP5xsDxIPPsoabZh9elL+Mu+CzheZpH3b8lpku8hb50uwfP/fBH/2pkJg1HA22VtgCgC9n6MdJqw/dBRVGYdwUDxAaB0B5C3Dh77APYWmvHjtefwwpwXkTXvNpTM/zrqF3wRbQs+i4EFH8fYgg8CC9+HJ1fuxqn8KprQ42cAACAASURBVLzvll9i0XMPAQvfF35rK/Ebf/6NqdM2ZfvTFm+dMu3O7RtgMAo4U9OFf8w2yvstz78Hs77zblief4+877nZs+X//sVVG1Gz7FYMbH0QJ+fdi31zf4rjK/8A77nXcXTrctw+axt+v6NYEgUeF3JrLbjUORT0ssBisWDWrFlYejBXfgHU2D0s3xO9XhFvZjdOqneiKOIloQYLT1RjZMyNg8UtWHSyGl9edBrDY24kV3bIPTYGo4DhMTeGHC75ftfYPRyyPu8tNOOOlZlyvS1s6sWvNuXj9zuK5TQPrj8viyPfPl+ZAp85KZUduHmp/9nm44fj9+vPz03BxdZ+1LT7p7DtH3XiUEkrfvtWEXqHxzBgdyG9pgsvBDwjdo+/sPrakjOTGt/nGnqwJr0eb5e1IbWqE8/sLQ377NyQZYInoFEeuJ2t7ZI/b8lpku9nPkxWv2BZllyLVan+5+Lp6k6UtfTh2X0XYDBKUQenKtrx9K4SrEmvx7qMBmzJacLVs/wv/Xy/LW62obCpN+heEcisYxWyfT/dcF7+3d2rs3HyYjuGx9zYmCVdL0tP1cDt8aJ/VLpf+u57vmfM07tKcLC4BQajgF9vLgiZXyDlrf3yS8LXztTj9pWZV/zNTMIChFET7yGi14mog4jGiOgiEf16mudQVIB09o9MutH97UAZvjA/FW+cbZAbZDctOYPr56Xisa2FuH5e6qTfBAqGp3eV4OrxFX3vWZ0dlO4Ha3JgMAqTzpHb0I2W3lH55hl47MYFaUHn353fDHPvCG5eegaPbCmYZMt3XgkWP7etyESrbVQuj29bnlKLn2/Mw9eWnMFNS87gxgVpk4STrxH+q035uHZOirx/ztuVcLg8k/L29Rb57PZ9vnZOCu59LQffezVL3ucTPoH5+LZd+c3yA+XxbYW4c1XWpLzCbY9vK4RtxIl/HamQH6rXzk3BbSsy8czeUrg8Xty1Smq8Pb2zOOhBHuijwO23bxXJDWHfA+rWlzPwl/EHmsEo9RZN/O1vxxupvu1zc5LxzWXp8gPfYBTw0KbQQuJK2/XzUkMK5XDb8uRafG5OMh7fVoghhwvLJ1xnE8t+w/zUSWL7oTfzJ/Xu/XG3JDCSKzvg8nhloTzV9tu3ioIaRoHXisEo4MkdxbJA8v1vBqO/nt0wPxXbz19GXmNPyPMvT6nF6tOXrmjH73cUo7p9ACmVflv2FJiDeiADt/WZJuwvagl5LNR14/PvqrRLaOkdneTro6VtEEURmZesQTYYjAJuNu5D9tlkoC4Zr7xkxNpFf8LmuY/g8NwfI3Ph9/Al40H5Pzn76m+mbND/ZNEO+bxvzP3tlGn/vHKbnHbTK/+L3oWfQtuCz6J/5Vcw8sa3UTL/6yiYfwt6N/4QsAbMVnUpFTj8O+DoU8DbfwKSngVOPDe+/RXo9r9xR2MmcPRp4MjvgSNPSr87+Diw72Fg90/haC6EZTzkru38fnS/dD06FlyN3gVXYWjBx+Be8AHZ3rrM/bK9f5s9a8qy/e/sf8FgFPD8oXK0FBwHFr4Pwws+itb516Bi0TdQ9PJ9eHvpw9j58rP44awNeGJ7kXTvn3sKnzWegMEohL0XHSn1v5jw9YpfqQ5cPy8V185JQVK5Bc/uvTApje+6921CRQfuez0XX1qYNimt72WA73t1+wAcLg9mHauEwSjg+69l44sT6tmY24NWm/TcWZfRIAsIg1F6CXeuoQd/3uMXDK+nN4Qtj6+n2PeC66FN+SGFmm+7fWUm0sYF2ZvZjbKdocoWuP3tQBluXpoui4ln9pZic04j7liZecX/3LctSKrCzwLEQqRbqbkPc49X4vuvZaO2YxBJ5Rb52LVzUsLeA3zb9fNScevLGbhxQRoutPThrlezZNET+Dw0GKXnVn3XEC629ocMPQ7sEQIgv0iLVahdKFiAMGriDBH1E9GfiOguItpK0sX56DTOoagAybxkhcEoYGtuE363vQiPbi2A0+0Pn/F11TZ2D8sLboXavCHe2lxokbpcfY0rX8Nt3vGqoDfhX1l8Gi6PF6Io4uvjDcovLzqNgqZeZNR1IWvcxj/tLsUP1uTgmb2l8ttq37lLmm04VCw12DbnTL7pL0uuxbfHewh8m++GfexCm/x26epZAv5xsBwGo/8NjC+sa/bblTD3juBfRypw3bwUHJjQEAtsMBqMQlDPiO/hfPvKTLx17nLYG/UXF6ThK4tP45rZyXg1TAPy9zuK5TJebO1HUrkFm3MakVlnxamK9qC0+wpbpJ6WEA/3wLE1eabghuysYxU4GtCgCPSfwSg1uvcUmCed86bxB/hNS87gie3B4sPX4/X8wXJZgPhCrUJtS07VTDreOeCQGze+cImJ/gzc7l+bG9TT5vuP71qVNamR9Nf9ZRgYdWF5Si02ZjWOhw1IA4K/utgvLMM1Qm5acgZPjou5u1dnB4kXX49Kek0XjEcrQv7++UPluGNlpiwyPhsgugOv6ad2Fk/6beBb3UBbDUYhqKdoTXo9bluRiSd3FOPkxfaQdvjyv/XlDOwrbJF71q6b5xfgd63Kwi825uHB9f5GjC/UZarty8aDeHnbAaQf3wHjnP/FqjlPoWLLH5A87x4Uzv+mLCpabaPIWv14yEb00Kz3IuuJ/8JjS7airKUPNy5Mw+tznwAWvg+jCz6C9gVXo3b+jSiY/y2kz78Tb8+7D/cv2YdjF9qwMasRw80lcJbuwR9nL0RG8iHAUoqxjlqI/a3ASC+q23rxqX8cxjW/W4n+gUG4Pd6gHq5W22hcxoaVt/Zj89lqfOyR5fjUPw7hhX0FwHA34LLjrvEXG9827sS2za9h05pF6Elfg55Ti7B97kM4Nu9HyJh3Oxas3SL74u9hxIrv//3nrOcx6HBhZMyNFevWwbvg/ehe8D8on/9VJM+7B1vm/hprlv4De3esx69WHZvk68e3FcqN8cDt6y+ly43FX2+e/AIp8DpeezZ0XVufaZLHxdhGnLh/ba58Twu833xrmXTtfmlhGira/L10Xxi/J/xxt9QD8NnZyRh0uCbViQfXn5fThtoeejMfZS19WJ9pku+noihO6p09VNwadD98ZEuBfE212vy9GBNfkoXa2vvtQXUxcPP1xPi2I6Vt8vPM9z/8/UAZ7ns9F5+fm4KtuU1wuDw4UtqGfx2pwOGSVqzPNMlln/gycOJ9IHD73JxkmKxDaOoextbcJty2IjMokmDitfC5Ocm4dm4KSs02zH67MujYRBF2/bxUrD59CcNjbrnHI1D8iKIoP3uaIgh3nSlYgDBq4UckXYiPTNh/hojaiejfIjyPogLk9fR6fGlhGrzeyAaMnTf14GBxC1p6R9HWNyrfAADAZB3Gzjz/DdDXZb70lNQV/OO15+SGStegAz9Yk4NXUuuCFt3yNdRmHauU9zndXqxIrUPP8Jjc3XvtnBRk13fD7fGiyiLFzoqiiIKmXrg9XnQPjcHSb4fD5cGcgJtbYIPJYJTeUjlcHjjdXsw9Xonq9gG5Mf6TdedkGwLj303W4FCBirZ+eUxA54ADNe2DcsjTvONVWJ9pgtPtxaGSVjlm3ffWzdeY++zsZFS3S+Vwebz40Ru5QXn4HlpPbC9C95AU/jUSJv797bI2+SHrE5Bjbg/Om3pwoaUPn5kl4N41ORgYGJAbUk63F4tOVuNX42+o12eaAEAOFbrUOYRHthQgpbIDv9iYJ9t6oKhFfnP1wPrzGHS48Nz+MmRdksTQ9fNSsfhkDVafqUdJsw1fmJ+Kkxfb5Z6rUacbhU29sPTbkXXJioy6LlS3D2DpqRqYe0fg9njROzwG24hTflD3jzqDYpMr2vpRZRmQwzuOl1lQ3zUkX1cXW/uDHnCBIu0fB8uRXd+Nvx0oC9uorGjrx8CoC2+cbcDylFqMjMc4f3XxadhGnPB4RaRWdeK2FZl4YP35oAGVrbZRDIy6MOb2IK+xR75ON2U34taXM7AgSbrefxUwvsPrFWUh8/DmfPmBWt81hOExNzxeEaNON6yDDuwuMOPFIxdhd0q9cd9/LRtjbg9y6rtRZRlAfdcQBh0udA064PF4MTg4CIfLDVEUIYoifrz2HH6wJge9w2NotY3KDbZt5y7L9vgaiaerO3Gqoh13rcqSx33Udw1hWXItPG4XMGBB88UsNGbvw6I5f8GmuY+ge/ujqGhsxe5xsbp//s+nfENfUHheHnTbceoltC/4DC7O/yrS590B59E/Y9PcR/C/T/8cRIS/v7pDvh4cI4NwjA6jf9QpC9Ga9kE4XB6cqmiXQz4CmUpEpGSdV+VUtOcKikBE+PgTr2NrbpO8v3/UKfdkBk4IIooi9p6XGrW+8LquQQceejMfn5t9Cn976yzc1gYcP3EMbecPYjh3I/75+/tBRChK3iefZ6xk15R+6zu/3S88Fm9A+rw74E42AkVbsH//DszdfgqHi5txsbUftR2DGHW6UdbSB7fHG3Qf9QkTg1F6geMLPU0qt+DugN70jLrgMSSB4vfqWVKP8k83nJeFgG8sya83F+CBdefg9nix+GSN3KsdOG6qpn0QfSNOuafv2rkp6B4akxv3vvDQwDCriWw7dzno3u0LXxJFEScutmPAHn5weKttVP4f7l+bG9TQ9tWNjVmNuOvVLGResuLz4/bUdUrHjpa24cH153HdvBQM2J1yw//GBWm42Nof9IwOteCl3enB9vOX0d5vl58lBqPUw/nQm/k419Aj/we+++3285expyD0ujEujxcplR3wekXcvjITd6/OxsIT1TAYBWw/L91nAsOFf7ExDymVUk/Xd17JwHlTj5ze9/z+4oI0WXQbjFIYre9zdv3Mzko3FSxAGLWwlYiGiehdE/Y/QtIFemuE51FUgJSbe7FwuxD1bBlJ5ZagwXNer4jr5qXg+UPl8j6PV8TAqAuXe0aQc4WbQ/fQGH7zVlHQW6FArEMOPLDunDwAOhIKm3rlm9OJ8TdcvkaKryEdyKhTamD+fGNeyPP5BvsZjJPHo0RK4DkmzlwEAJc6h3DtnBTcvjITxy60hRwgPhWjTnfIRhcADNhdGBixQxAm+903zuVEiAHiU3GhpS+iGboGHdKYnO6h6Q3kjgSvV0Ty+IMuFEnlFiRXStdNlWUAu/Kbw15nV+KOlZlYdLL6ygmvgE/MZE6Y6e1MTRduW5GJC+PjQSKhsKkXXVMMwnS5XJN87nR7gwa/P39Ielt6PmBQ6R92leDzxrdhqikHTGeBku2AK2BGrjPzgUUfCN9AtUoNtZMX29GStBhYdS2w6XZg70Oo3vg4sjf+Fc7zG4Cqo4A9+JpotY3isa2FcnjN5Z4RZJ0vBBGhoKg4ZDnNvSNBAioa1LoWhs+us7kFkwad+8ZbBeJyubDnmL+hGAkff+L1yWX3eqTelvZyoPYUUPAmkDYHOPQbYMtdQEuhHF7anfVm6OtgyYeBdV8HTOn+844NYevpUvz2rSIAwJDDhYut/UFjUnwvWjoG7Pj7gTIYjMKkGRUByC8ZAicDAaRxRb7JLHqHx4Jmk2u1jaKuczDkvcvrlWbP8g2wdnm8+OPuEpxr6MGxC21hXwD5fut7kbUsOfSikldib6FZFpOnqzsnTfThCXi5VNAUPOtjYF2/3DMCg1GQJ8LoGLAH9RxMhW/mtIkThuzOb8ZNS85gXUZDRJOo+LA7PbA7PegadODJHcXyyzQA+MXGPLwWMDFAIN1DfoEysRfFYBSCeo72Fiq3gCYLEEYtFBBRcYj9XyTpAv1jhOdRVIDEArXNC+8LD3t4cz5EUcSQw4XKtgG8JNSEvQGfqmgPGuA3kYut/UEzKEXDngLzpAfHRBvSqjvfUR7TxfemeqbXNtAa1kGHorHGMcftBLxeJJVbcM3sZAyVHZPGJ2y9B+4V14YVFQCAnJXSvsUfBF77IrDt+8DhJ4DU2dJMT0MBjeIZCF1Syzog8SBau2zjPZmRkJadF1UeJc02PL6tEN7uBmmQfdocaUzLum9I4sN37VzO9f+obI+0b9W1wM4fA8kvAiVvAeZ8wD5ZfIuiCHNv6Kl5u4fGIhZZSuGbFVJtGI9WYMf5yES6bwKPeCKKYtD4HN/m6539wZocfGXxaXznlYwpe6ZmGhYgjFpoIKK0EPs/QdIFOjvM7/6DpIvXt11FROjrk26+Ho8HHo9n0me32x302RdCFe6zy+UK+uy7oYyNjaG3txcej0fe75vCD0DQZ6/XG/TZ7XZP+dnj8QR9DlWOWJRpYjlEUUTngB39w3ZNlemd+MntdqO7uxterzeoTA6nC4dLWuD1iglXJi36aSbL5HYMo7+hEK6qE/CeXwuceh7enQ9AXPMlYNEH4OqohsfjQXPPCDxZKyaJDvGlT0Bc/y2I+x6C2FXjL9NIL1x9bYDXq0iZiouLpRChoqKY+am0tFRuhKvp2vMJkJKSkojK5PF4YLVa4fV6I772ioqK5DxmrExuF9DfAk99OrzjvVxutxverFemDO1yN5z1l8nWAtF2GWKA7XyPCF0mr9eLnp4e+bdaKJNvwpHHtxUiubIDZWYbeobs8qQzj2wpQM/wGBxjTsXK1NvbywKEUQUNRJQaYr9PgMwK87tF48eDttxc6S1RVVUVqqqqAABlZWWoq5PUfXFxMUwmKU4/Ly8PZrPU7ZiTkwOLRXoLlJGRAatVCu1IS0uDzSaFCwmCIPewJCUlITU1FaOjo/LCRXa7HUlJSQAkhS8I0hgPm82GtLQ0AIDVakVGRgYAadrGnJwcAIDZbEZenhS6ZDKZUFwshUnU1dWhrKxMsTLZ7fagxZi4TMFlampqwsmTJ+FyuTRTJi36abplKi/Ox+WCU0DtSbQeeAGXK/LlMtmOG6ds7F04sEwu07mDb8CRsRKoSULW/jWw2yxwOZ2q8JMgCLjqqquwe/fumPkpPz8fH/rQh1BZWamqa6+yshIf/vCHce7cuYjKNDg4GPQ5Ej9t374dV111FdLS0pSpT2NDKDiyHsPnNgFpc2B97TZ4X70eWPg+pB3eLpfp0kZpUgJx+afQvepbQOps2At2IPvIJsDrUZWf3sm1NxP1yeVy4dSpUygtLdVMmTLrrFidVDTJT77Zw9acKFK8TIcPH2YBwqiCaEOw4toDouY3tlwmLhOX6QplaiuBV3gB4s6fAKtvmCQqPHWpftsrDgHLroL45m0QDz0BnF0MT8lOeC+fA4Y64XI61VEmLfqJyzT9MjkG4XI65XJ4hH9CXPyh0AJ62SfhtTX7y+QaU2eZtOinOJfJNuLEgUKznEbJMnEPCKMWtlDoQei/JhUPQvd6vXIXPaMf2O8JgMsBdNUA1cel1bSPPg1svgNo9s/OhopDkxtjrxikgcFHnwJa/QO1vW4XrF1d7HMdobl67nYCHRXS2JHkfwLb7gVe+gSw/H+AwDIefQpYc6O0xkrBRmnBSPf0VjlPZDTnd5XCY0AYtXAfSRfiwxP2p5KKp+F1u93IyMiQ3x4w+oD9riIcg8BYwAw3jRnA618OP7NU4WZ/2t5GaRaqsj1AaxEwGn5WNq34vLKyEldddRUqKyuvnFjFeUTDdO2KxudqLXtYvB6gb8LMR298dXK9WfoxaaB79soZmQxBzWilrqsdFiCMmjhDRH1E9AeSFiLcQtLF+dg0zpHws2AxDBOCsWHAUiqJhbQ5wJ6f+8OmSt7yp2st9jealn8K2PI94O0/A7mrpSlQByOfglqL8CxY+iz7tHAMAk1Z0gxt+x4GXrnaX6c23hqctmS7NDXw2FCoMzFMWFiAMGriPUT0BhF1EpGTiCpICsGaDoqHYFksFu6q1Rns9xjisgMdFwFbwDSXrUVTDgBH+kJ/WueoFGY1bJ3RN7Va8bmeG+HTtSsan6u17O8IUZSmji7eBpT7F1iEywEs+YhUBxf9P2Db94HM5UBLAeCJbm0sNaCVuq52WIAwWkPxEKycnBzuqtUZ7PcZwOsBui8BlUeAs0uAA48Cb3zNHzp1eq4/7UiPtG/l5wLWO9guNXRCrHcQC7TicxYgkdsVjc/VWvaYMNwt9S6u+VKIge1XAdkr4m1hVGilrqsdFiCM1uAQLIZRG6M2oCk7aFA3BtrC92i8YggWIL5zMO8YFiD6LHvM6W8BSndKi2j6Qrby1vmPD7YDSX8Bak9yuBYDgAUIoz0UD8Eym83cVasz2O9h8HoBay1QcRg4s0Aap/HqdX5hceBRf1pRlMZwbL0bOPEcUPCmFHc+1KXKQa5a8fnQ0BCysrIwNBS7RqASeUTDdO2KxudqLbuieL1Aexkw1OnfV7bXfx9Y8mFg90+Bwk3BoZYqQSt1Xe2wAGG0huIhWHl5edxVqzPY75DCopqygPrT/n1ejzRbTqhejde/DAgvBJ9DhUIjHOxz/cE+n0E6KoAUo3QfmHhvWPcNoL083hbKsN+VgQUIozU4BIthZpruS9KaGafnAbt/Bqy61t942HBLcNqdP5YGo576hzRotaWQQy5UhMViwaxZs+RVjRM1j2jQc9lVgygC3fXA+TeAHfdLg9cXvh8Y6fWnacwA6tMA91j87GRiDgsQRmsoKkA8Hg9MJpO8ciijDzTrd5cDsFwA6oTg/eu+EbpX442vAod/l1A9GdGiFZ/reRzEdO2KxudqLbtqsfcDprPB+976wfg02v8jLYpYcwJwjihmklbqutphAcJoDcVDsIqLi7mrVmdowu+OQaD5vLTS8dt/AjZ8G1j8QenB/9InpHAqHyeeA7beA5x6Xlpzo7VYWpdDR2jC52ABMh27ovG5WsueMIgikPKv4LFjvoUQDz4GVL8dcxO0UtfVDgsQRmtwCBbDTGTYKoU1BPZU7PtVmBmorgZ2PcCzTmkUFiD6LHvC4fVK6/+kzQHW3Oi/P+37VXA6D4uERIUFCKM1FA/Bqqur465anaFav4uiNKtMTZK0tsbeXwKrPu9/eAeuAp65TJqFav+vpcXD6pKlqXF1EE4VDar1+TTRcyM8mhCs6fpcrWVPaERRGqR+dnFweGh/C/Dyp4ETfwUu5wT32r4DtFLX1Q4LEEZrKC5AysrK+EalM1Thd48b6KqRVg73cWZ+mLU13g+svTl4ppkZeljrBVX4fAYwm8146qmnYDabEzqPaJiuXdH4XK1l1yT5G4Lvc69eJ/WYtJe9oxcpWqnraocFCKM1OASL0R4uO9BWKo2/OPl3YPOdwNKPSg/d5vP+dBcPAIs/BGy6TVr0q2jL+CxU+hqvwTCMDvB6gMu50hi1lz8VLEbWf0t6QcOoFhYgjNZQvAekqqqK35TojJj63d4fPG1txaHxqSpD9Gws+yRQddSf1uUA3M6Zt4nRTF232+2orq6G3W6/cmIV5xEN07UrGp+rteyaxz0mhZEe/p30cualjwffR/uaAY8rolNppa6rHRYgjNZgAcLEnBnxuyhKYzLq04DsldIML2u+JAmLC7v96VqLpH0rPiutHnxmgSQ6ekzSQE1GEbRS13kMyPTGgEzX52otu66w90tjQnyIIrDxVmDlNVKIVlf1lD/XSl1XOyxAGK3BIViM+vB6g8dqtJcBKz8XZrzG+6QB5D7cTmCwnQeHMzMCCxB9ll3XDFsn32+33g2U7wu+LzOKwgKE0Ro8CJ2JOVP63eOW3rCV7wdSjMD2+6QFtdIX+dMMd0sPwUUfANZ/Ezj6NJC3VnprZ+9TriBMxGilruu5ER5ND8h0fa7Wsusejwu4lAIceNS/3tHC90mzaAX2OEM7dV3tsABhtAZPw8vEHNnvgQtVjdqkweFLPhK6V2PvL4NPYikFnKPKGs5EjVbqOgsQnoZX9wx1AbmvAq+Nry9Sn+Y/NjYEj9OuibqudliAMFqDQ7CY2OCLK85bBxz7gzTLypEn/cdFEVg+PhPLsquAt34o9YCU7wM6qyIeAMkwsaSsrAzvfve7UVZWltB5RIOey86EwOsBTOnBU5KfXSKFa2UsBYY642ebDmABwmgNRQWI2+1GcXEx3G5ejVWTiCJw5Pf+weETNvH1rwSnb8oGeht5cLgG4bquP9jnOkMUgTe/67+/L/6QFB5r4d6sWMAChNEaiodgmUwm7qpNVEQRsDUB1celVXb3/BzY/bPgNJtu84uO124EDjwKb+bLaM96C56+1vjYzSgO13X9wT7XIR4XPJVHYV9/W/ALp23fB2pPxds6TcEChNEaHILFXJncV6UQqeX/M7lnY/GHgtfSMJ2VejZGbfGzl2FmiNraWnzta19DbW1tQucRDXouOxMFlgtSuO3iD0nPhpN/j7dFmoIFCKM1FA/BysvL4y56NSGKUuxuQzqQu1oap7H5juCwqMNP+AXHkg8Dm24HTvxVWjm8tViayWoK2O/6Qys+50HokdsVjc/VWnYmcib5fagTyFwGdF/yJ+q4CKTOAgba4mOkBmABwmgNRQWI1+uF2WyGl2P+40/ZHmDXA9KCfaFmobI1+dM2ZQEXD0iDw6NYOZz9rj+04nMWIJHbFY3P1Vp2JnIi8vvh3433mH8QePtPQFeNcgZqBBYgjNbgECytMmqTZqHK3wAcf0YaLDjU5T9+drFfbCz6ALDuG1LvR+5qoOEM4ByJn+0MoxJYgOiz7MwMYzoL7Lh/wlTrDwHmPF40NkJYgDBaQ/EQrJycnIQPy1Atl1KlgeGrvxC6V8OU7k/bXg6U7pTW14jx6rbsd/2hFZ/ruREeTQjWdH2u1rIzkTMtv1tKgYOPAwvf738uHXw89kZqABYgjNZQPATLYrEkfFhGXPB6gB4TUJMEZL0s3bTX3gyY8/1pyvYEC441X5JWss1cLs1IEqeB4ex3/aEVn/f19eHw4cPo6+tL6DyiYbp2ReNztZadiZyo6nqPCTj5N2lM4fnXA0/G07KHgQUIozU4BEttiGLwoO6mLGlq26UfDd2rUbTFn7bPDBRvA1oKAMeA0pYzDMMwTOQMdQFjw/7v1ceBDd8Gqt9mITIBFiCM1lA8WJax5AAAGGtJREFUBCsjIyPhwzJmBFEE+lulsKj8DdLboLd+ALz8aeDCbn86c55fbCz9mDRD1fFnpBXGTWcTYrpb9rv+0IrPu7q6sHr1anR1dV05sYrziIbp2hWNz9VadiZyZrSuv/UD//Nu/TeByiPBK6/rGBYgjNZQPATLarUmfFjGtPC4pe7mAYt/X1spsOyToXs0Fr5Pmq7Qx9iwFD7V25iwN2Jd+l3naMXnPAZkerNgTdfnai07EzkzWtftfVLI8PJP+Z+H674OVBxK2OffTMEChNEaHII1U3hcQFc1UHVUuoEe+i2w4RYpxnXh+4D0Rf60Q53+KQnXfQM4+Bhwdol0k+2sBNxj8SsHwzAyLED0WXYmzjgGgOyVUkSAT4gceTLeVsUVFiCM1lBUgLhcLqSlpcHlcimS34zjcUs9EfWnpbCpSyn+Y/2t4Xs0ln4suFdDFKVFmjwJ+j9Mk4T3OzNttOJzPTfCp2tXND5Xa9mZyIlpXXcMAjmrJCFSf9q/3+vR3fS9LEAYraF4CJbNZkucsAznCJA2B9j3K2DtTVKPRaCwOPRbf1qvF3j1OmDrPUDSs0DeWumG2des+8F0Ced35h2jFZ+zAJleCNZ0fa7WsjORo0hddwwGC46cVdK6Iq1FsctTZbAAYbSGPkOwXHapB6I+DSjcBKTOBvY/Is2+kfQXfzqvxx9CJfdmfFRKd/Bx/P/t3XuUHGWZx/Fv1kSRS0BAJRtXZtEDLO66IHqMNxhWLorginIUCXvOuCsqrOBxBYV4YVblJh6XeAMSTeKJcgkLjGGUMSaBCcaYMQZ0YJNMckJuGCbJBAbCTOb67B9PVaa7pmamuzPdU131+5xTJ9PVb1e/b56u7nqq3vct+8NdE9cGESm7TZs22YUXXmibNm2q6vcoRZbbLgnW12P2nTcP/Sb/4hNmO1snulZlpwRE0qbiXbAaGxvL3y1jYMCs869+j4wn7/Gp/XKf++ZrR+4udecZ+dta8V2zlrk+He4L2zN/NaMUFYu7JIZinj2KeTZNSNyf3+Y9DeqPCn67jzR78HP5k72kjBIQSZuKJiCDg4PW2dlpgwfbd7NnnycYuRZ/wWzBBWazTx2eYMw5K7/s7NPMbpxudsd7/EZ9TbP8fhptv/UxHjKuxi3uUjXSEvPe3l7btWtXWQ+uKvEepSi2XqXEPKltl8JN6L6+u827Quf2UPjL/ZWvRwUoAZG0SXYXrNYHzB7/H7NfXWN29yWeMNxyfHxScfs/5ycd9a/xO4EvuMATjFz7X8zcADYRKZ7GgGSz7VJltq8x++kH/Hd/d9tE16YslIBI2lS8C9bjC2+xvj/93Ozx75n9+it+9uKn55nd/lb/N9fsU0fuKjX7tPyyf77P7Ml7zZ75nd8RPCMzTFWD3t5ea2ho0FnODElLzLN8EF7KLFjFxjypbZfCJWZfHxwcPhbksVt9tsoUnHBUAiJpU/EuWAO3j5JU3HZi/gseud7sgcv9HhotP/FZpZ572ucIl6oxODhoXV1dVd8dRwqXlpgrASm8XqXEPKltl8Ildl/f2epjQ26Y6jNm7fzLRNfooCgBkbSpfALywGdscMGFZg98xmzJ181W/djsqQd9wHjH5orUQyprcHDQent7k/cDJWWTlpgrASkuASk25kltuxQusft69wtmS74xNCa0/iizh79o9nLHRNesJEpAJG0q3gUrEZdqpaIU9+xJS8yVgKgLlowu8fv681vNFtUN9bS4+Y0+6cxA/0TXrChKQCRtKn4FJJFnSqSsFPfsSUvM+/v7rbOz0/r7y3ewUon3KEWx9Sol5kltuxSuavb1Zx73e3jdMNUns6myKyFKQCRtKp6AJLKvqJSV4p49inn2KObZVFVx7+/zqx9P/GJo3eCg2UvtE1enAikBkbRRFywpO8U9e9IS87a2Njv33HOtra18U3tW4j1KUWy9Sol5Utsuhav6ff3/Hjb79nF+0+G+nomuzYiUgEjaJPs+ICIiE0hjQLLZdsmQBy4fGh/yg3f4VP4JpAREQkcA3wGWALvxD0X9KOXfBiwF9gEvAA8CJ4xQ9ipgPdADPAPcAEyJKfc6YAGwB+gCVgHvL6oV1XondKkqinv2pCXmWT4IL2UWrGJjntS2S+Gqfl8fHDR78h6zW08YSkQeutJs356JrlkeJSASqsETiWZgLqMnICcDLwIrgPOBjwJPAc8Cr42U/SowCNwE1ALX4onInEi5VwGtwHZgJnAO0AD0AWcW0Y6Kd8FqbGys3ku1UhLFPXvSEnMlIMXNglVszJPadilcWvZ1e7nDbPHVQ0nILcebPfXQRNfqACUgEpoULADHMnoCsgi/SpL7oTke6AVuzVl3DNAN3BV5/Sw8KTklZ92VwXu+K2fdZOBpYHWBbQB1wRIRGZESkGy2XTJs6x+GZstq++1E1+YAJSASZ7QEZDLePerOmOd+A7TlPJ4ZbGdGpNy0YP2snHW/xbtpRV0flJ1eQL2hwgnIwMCAdXR02MDAQEXeT5JBcc+etMR8165d9sMf/tB27dpV1e9RimLrVUrMk9p2KVxa9vU8/b1m6xrz121fY9bbNTH1MSUgEm+0BOSk4LkrY567Db+ycUjw+Oag7GExZXcDd+c83olfWYn6ULCNcwuoN0xAF6ympqbqv1QrRVHcs0cxzx7FPJsyEffOZ81ueoPZ7NMmbJC6EhCJM1oC8u7guUtinguvVkwLHs8B9o/wHhvwKyahXuKvqrwr2OYnR9jOq/APb7hMB2zv3r1m5jeFCm8Ilft3X19f3t/hmY6R/u7t7c37OxycFv6de+Oi8G8zy/t7YGAg7+++vr5R/+7v78/7O64dapPapDapTcW0adeuXbZw4UJrb28vW5v27Nlj8+fPP3AWOSlx6ujosPnz59vu3bvLFqf29nZbuHCh7d69W589tSm5bXpmpQ3eduKB8SEDi79g1t1Z0Tbt2bNHCUgK1eJBLWQ5Neb1hSQgn4h5LkxAjgsez8HHgMTZADTlPO4F7ogpFyYgcQkPQR2HtWvFihVmZtba2mqtra1mZrZ27Vpbt26dmZm1tLTYxo0bzcxs5cqVtmXLFjMza25uth07dpiZ2bJly6y93W/m09TUZB0dfpfRxsbGA1dYGhoabNu2bbZ///4D84Z3dXVZQ0ODmXmG39jolz07OjqsqanJzMza29tt2bJlZma2Y8cOa25uNjOzLVu22MqVK83MbOPGjdbS0mJmZuvWrbO1a9dWrE1dXV15c6GrTflt2rx5sz366KM2MDCQmjalMU7j2abNmzdbe3t71bepoaHBAJs3b17Z4tTc3HxgHESSPnvh+Izly5cX1KZ9+/ZZQ0ODDQwMFBynuXPnGmCNjY3an6q0TQMDA7ZmzZpUtSkuTk//6fdmv7xqaJD6d0+2Py9/sGJtWrRokRKQFJoGfLrA5eiY1493F6xDY8qOVxesCb0C0tXVZUuXLrXe3l6djclQm3p6emzp0qUH6peGNqUxTuPZpp6eHlu2bJl1d3dXdZtaWloMsNWrV5ctTmvWrDmQgCTpsxcmIH/84x8LalNvb++B/bzQOK1evfrAe2h/qs429fX12dKlS23//v2padOocdr0qA3OPtXszvdZX0+3roDIhCpkEHrc1Yom8gehXxps552RcscxfBD6EmBdzDavC8r+bQH1Bs2CJSIyoizPBJXltouMqrfL7PmtFX1LjQGROGNNw3sf0I7fvDD0Rvz+HrfkrDsa74IVTVauY/g0vFcwPFmZjN9f5A9F1L3is2Dt2LHjwBkAyQbFPXvSEvMsH4QXW69SYp7Utkvh0rKvJ50SEMn1QeBi4FP4h2JR8Phi8rtRnQy8hN+08IPARfhNBEe7EeGN+A0Fr8EHpsfdiPApYBt+5eRs/O7qib4RYV9fnzU3Nx+4xCnZoLhnT1pivn79epsxY4atX7++qt+jFMXWq5SYJ7XtUri07OtJpwREcm1h5MHqNZGypwNLgZeBTuAh4E0jbPdqfNB5D7AVv7IyJabc64GfAR34lZNVeCJSDHXBEhEREUkwJSCSNhXvgrVlyxZdqs0YxT17FPPsUcyzSXGvDCUgkjYV74K1cuVKXarNGMU9e9ISc40BKbxepcQ8qW2XwqVlX086JSCSNuqCJSIyAiUg2Wy7SNIoAZG0qWgC0t/fbxs3bjwwb7Zkg+KePWmJeZYPwoutVykxT2rbpXBp2deTTgmIpE3Fu2C1tLToUm3GKO7Zk5aYKwEprgtWsTFPatulcGnZ15NOCYikjbpgiYiMQAlINtsukjRKQCRtKt4Fa926dbpUmzGKe/akJebd3d22ceNG6+7urur3KEWx9Sol5kltuxQuLft60ikBkbSpeAKydu1afVFljOKePYp59ijm2aS4V4YSEEkbdcESERnB5s2bbebMmbZ58+aqfo9SZLntIkmjBETSpuJXQFpbW3WmJGMU9+xJS8yzPA6ilFmwio15UtsuhUvLvp50SkAkbZSASNkp7tmTlpgrAVECIqNLy76edEpAJG3UBUtEZARKQLLZdpGkUQIiaTMVsO3bt1tnZ2fZl71799qKFSts7969FXk/LclYFPfsLWmJeXNzswHW3Nxc1e9RiXqVEvOktl1L4Uta9vWkL9u3b1cCIqkyHf9Aa9GiRYsWLVq0aEn2Mh2RFJiEf5inVmgJE55KvqeWiV8U9+wtinn2FsU8m4viXtn/60mISNGm4l9UUye6IlJRinv2KObZo5hnk+IuIomnL6psUtyzRzHPHsU8mxR3EUk8fVFlk+KePYp59ijm2aS4i0jivQqoD/6V7FDcs0cxzx7FPJsUdxERERERERERERERERERERERERERERERERERKdnhwO3AX4H9wJPAJRNaIynFvwDzgPXAy8CzwC+B02PKvg1YCuwDXgAeBE4YYbtXBdvsAZ4BbgCmjGfFZVx9Gp/1Zl/Mc4p7urwX+DXwPNANbAS+HilzNrAK6AL2AAuA18Vsawoe4y14zNfjnwFJjtOABvy3uguP0TeAQyPltJ+LSFVYgv+AfRY4C5iLH8BcOpGVkqLdDywHrgDOBC7GDzz68OQkdDLwIrACOB/4KPAUnrC8NrLNrwKDwE1ALXAt/kM1p0xtkIMzHT/geJbhCYjini6XAgPAPcCF+Hf3p/ED0tCZ+P7fAJwDzAR2AK0MnxVpLn4C6lo85jfjn4FZ5WqAFOUUPMl8Evg4/p1eD/TjJ5pC2s9FpCqcjycbn4ysX4J/Yb2i4jWSUsWd1TwceA4/GxZaBOwmf17444Fe4NacdcfgP3h3RbY5C//ROuUg6yvj72FgMX6WO5qAKO7pMR2P74/HKNcCPA1Mzln3bvw7/4qcdW/BY3t95PVz8DPtRx9MZWVcfBuP25si6+8K1r8meKz9XESqwlzgJfJ/oMATEsN/rKS6LQc2BH9Pxg8o7owp9xugLefxTPwzMCNSblqwXmdGk+Uy/MznGxiegCju6XIDHovjRykzPShzXcxzG/CTTKGvBmWPi5R7F7oanhRhzI+NrL8VvxJ2GNrPRaSKrMLPkkW9Bf8S+kxlqyPj7EiG+gADnITH9cqYsrfhZ74OCR7fHJQ9LKbsbuDuca2pHIzX4f37w7guID8BUdzTZRnQAZyHd8npB3bhB57hme/z8DieH/P6+/FxBKF7gtdHHRZs46ZxqbUcjBq8q/T9+HiOI4AL8O/37wdltJ+LSNVoA5pi1odnQaKX5KW6/BzvAx4ORA+7X8RNMnB98Ny04PEcvE94nA34GTVJhv8FVgKTgscLyE9AFPd0WY93n3kRj18t3n+/C/gd/jm4lPgz3ODdbnpyHi8Jthmnh+HddGRinAysw+MaLrMZ2u+1n4tI1WgDHolZHyYgcZfvpTp8C4/h53PWhT9Qn4gpH/5Ahd0w5uAHOXE2EJ+4SuV9DD9IzO2zvYD4BERxT4c24r+fvxCsP5uhBOSdMa+/i/yDzyX4gW2cHuK79Ehl1eCznP0O3+fPwJPOTuCnQRnt5yJSNdQFK53C/sLRfrylXKKPTvEIukSfFOEkA98FjspZ7sYTkKPwLhaKe7qswmN0WmT9icH6L6MuWGlzL9DO8C5Tn8JjdCbaz0WkiswhfhD6JWgQerUKk48bYp4LByneEfNcE/mDFEc6g3ocGqSYFDXkd8eIWxpQ3NMmnPkomoCEB6DXMDQI/Ssxr19P/iD0WcQPQp+BBqEnxXrg0Zj1/4jH6D/Rfi4iVeSDxF+yfQRNw1uNvo7H81ujlLkPP5N2RM66N+JdLW7JWXc0fok++mN2HZqmMSkOwfv/R5cmPHa1+AEKKO5pci7xB4lfDNa/N3i8Gr/nR+73eJhUfC5nXTgNbzRZuRNNw5sUy/GrVIdH1l+Ox/Nfg8faz0WkaiwB9uJfZGfhV0UMn6ZPqseX8Lg9gh9kRJfQyfhVr2Y8Ab0IP0gZ7UZVN+KX+K/B+47rRlXJtoD4GxEq7umxGI/J1/AxH9fhB5QP55SpxSeheJChcSHbGP1GhNfgMb8R3YgwST6Mx2MVQzcinIXv008DrwzKaT8XkapxOD6Txk78LMmfiZ9FQ5LtMUbvipPrdPzmhC/jgxgfYvgNrkJX44MSe4Ct+N13p4xrzWW8LWB4AgKKe5q8Gj+jvQ1PMrbiYzWiicU5+EFrNz5178+Iv2npFDzGW/GYbwCuKkO9pXRn4bNT7cSvTG3Ax38dEymn/VxEREREREREREREREREREREREREREREREREREREREREREREREREREREREREREREREREREREZALVkX8zyX5gBzAfmJ5TrjZ4vrYCdakp43uUSy1e94snuB4iIiIiIolWhx841wEz8Dsh3wDsBzYDhwXlpgbPT61AXWrK+B7lUosSEBERERGRMdXhB85vj6z/ZrB+5gTUpaaC7zleaqlMAvJqYFKZ30NEREREpGzqiE9Azg/Wzwoe15LfBetYYDvwe2BKzutOAV4GFka2dzawDHgR6AJWAu8foS41Y9S5Pij3FuAeoBNoB+YBR+aUq2Ho6k6UBduJbvOtwP3BNvcC3wMmAycBTcBLwBbgy5Ht1Qavvyx4zXNAN9AMnBbz/m8HFgfvsR94Avh4pExdsM1zg7btDh4fErM9EREREZGqUEd8AnJ1sP7y4HEtw8eAvAfoww+4AQ4FngbWMdR1C/ygfBB4CLgIuAB4GB9vkpuEhHWpGaPO9UG59cB/48nNF/ED+Xk55WooPgFZD3wt2OatwbofBG26Klg/L1j/0ZzX1wbrtgENQRtnAhvxZOaEnLJnAT3ACjzpOA8fcxOta12wbgdwF/AB4GPAK2LaIyIiIiJSFerwg9x34mf6Dwc+BOzCr1a8PihXS/wg9C8H6z8CLMCvbvxTzvOHAh342f5cfwM8CayOqUvNGHWuD8pdG1n/I/yqQ9hFqYbiE5D/ipR7Ilh/Uc66yfj/zwM562qDcn8iv4vU8UAvMDdn3TpgbbCdXA8Df8X/b2Do/+NnMfUXEREREalKdeTPghUuf8GvcIRqiU9AJgGN+IG/Af8Ref7sYP3H8APu3OUW/MpIeLUkrEvNGHWuD8qdFFn/2WB9mDTVUHwCcmKk3N1BHaPdnn4PrMl5XBu8/ksx7/UYsCn4+8055aL/H1cEz/1DULYuePzhmG2KiIiIiFSlOvwg99/wblinAtNiytUy8jS8FwfP7QReGXluJvEJTu7yd5G61IxR5/qg3LEjtCV8fQ3FJyDRbS4A9sW8/jHgqZzHtYw8aP9e4Png7/cw9v/H+yLteUfMNkVEREREqlId8WNAomqJT0Cm4QPAn8C7Gn0/8vx5wes+H7xH3BImLWFdasaoSz2FJSDHBY8/Gyl3DOVLQMa6AnJSUO4mRv7/OCLSnrFiIyIiIiJSNeooPQF5BX5w/Rx+sP8lhg/OPhw/+//jIupSM0a5egpLQCbhXcN+FCn375QvAVlD/BiQn+SsawN+FbPNqDqUgIiIiIhIytRRegLybWCA/JmsFuMJx9/nrLssKHcv3l3rDHxMyDeBO2LqUjNGXeopLAEBH/zdjQ8ufz9wPdBK+RKQcBasDwGX4rNgvQi8KafsWfiMXb8BPon/f3wkqNv9Me1RAiIiIiIiqVFHaQnIOXhSUR8pdzSwFWghfzzIGfhg9Q78isCO4HHujfvCutSMUZd6Ck9ApuJJyHN4IrEYvypRrgTkMmA2PkvWfnyq3dNjXv9W4D68+1ovPn5mGfndxcL2KAERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERERkeT4f8jWjEHvtp06AAAAAElFTkSuQmCC\" width=\"800\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# apall(2): manually select sky, see how the fit works\n",
    "# =============================================================================\n",
    "ap_init = 553\n",
    "\n",
    "ap_sky = np.array([480, 520, 580, 620])\n",
    "# Regions to use as sky background. xl1 - 1, xu1, xl2 - 1, xu2. (0-indexing)\n",
    "#   Sky region should also move with aperture center!\n",
    "#   from ``ap_center - 50`` to ``ap_center - 40``, for example, should be used.\n",
    "# TODO: accept any even-number-sized ap_sky.\n",
    "\n",
    "# Interactive check\n",
    "x_sky = np.hstack( (np.arange(ap_sky[0], ap_sky[1]), \n",
    "                    np.arange(ap_sky[2], ap_sky[3])))\n",
    "sky_val = np.hstack( (apall_1[ap_sky[0]:ap_sky[1]], \n",
    "                      apall_1[ap_sky[2]:ap_sky[3]]))\n",
    "\n",
    "fig = plt.figure(figsize=(8, 5))\n",
    "ax = fig.add_subplot(111)\n",
    "title_str = r'Skyfit: {:s} order {:d} ({:.1f}-sigma {:d}-iters)'\n",
    "ax.plot(x_apall, apall_1, lw=1)\n",
    "\n",
    "if FITTING_MODEL_APSKY.lower() == 'chebyshev':\n",
    "    # TODO: maybe the clip is \"3-sigma clip to residual and re-fit many times\"?\n",
    "    clip_mask = sigma_clip(sky_val, sigma=SIGMA_APSKY, iters=ITERS_APSKY).mask\n",
    "    coeff_apsky, fitfull = chebfit(x_sky[~clip_mask], \n",
    "                                   sky_val[~clip_mask],\n",
    "                                   deg=ORDER_APSKY,\n",
    "                                   full=True)\n",
    "    fitRMS = np.sqrt(fitfull[0][0]/n_found)\n",
    "    n_sky = len(x_sky)\n",
    "    n_rej = np.count_nonzero(clip_mask)\n",
    "    sky_fit = chebval(x_apall, coeff_apsky) \n",
    "    ax.plot(x_apall, sky_fit, ls='--',\n",
    "            label='Sky Fit ({:d}/{:d} used)'.format(n_sky - n_rej, n_sky))\n",
    "    ax.plot(x_sky[clip_mask], sky_val[clip_mask], marker='x', ls='', ms=10)\n",
    "    [ax.axvline(i, lw=1, color='k', ls='--') for i in ap_sky]\n",
    "    ax.legend()\n",
    "    ax.set_title(title_str.format(FITTING_MODEL_APSKY, ORDER_APSKY,\n",
    "                                  SIGMA_APSKY, ITERS_APSKY))\n",
    "    ax.set_xlabel('Pixel number')\n",
    "    ax.set_ylabel('Pixel value sum')\n",
    "    \n",
    "else:\n",
    "    raise ValueError('Function {:s} is not implemented.'.format(FITTING_MODEL_REID))\n",
    "ax.grid(ls=':')\n",
    "ax.set_xlabel('Pixel number')\n",
    "ax.set_ylabel('Pixel value')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although the skyfit does not seem to work at far from the object (since it's extrapolation), it should work for the object region since it's interpolation. \n",
    "\n",
    "Now I do the aperture trace fit:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/aaa/anaconda3/lib/python3.6/site-packages/astropy/modeling/functional_models.py:172: RuntimeWarning: divide by zero encountered in true_divide\n",
      "  return amplitude * np.exp(- 0.5 * (x - mean) ** 2 / stddev ** 2)\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# apall (3): aperture trace\n",
    "# =============================================================================\n",
    "# within +- 100 pixels around the aperture, the wavelength does not change much\n",
    "# as can be seen from reidentify figure \n",
    "# (in NHAO case, ~ 0.01% ~ 0.1 Angstrom order).\n",
    "# So it's safe to assume the wavelength is constant over around such region,\n",
    "# (spatial direction) and thus I will do sky fitting from this column,\n",
    "# without considering the wavelength change along a column.\n",
    "# Then aperture extraction will map the pixel to wavelength using aperture\n",
    "# trace solution.\n",
    "\n",
    "aptrace = []\n",
    "aptrace_fwhm = []\n",
    "#coeff_apsky = []\n",
    "#aptrace_apsum = []\n",
    "#aptrace_wavelen = []\n",
    "\n",
    "# TODO: This is quite slow as for loop used: improvement needed.\n",
    "# I guess the problem is sigma-clipping rather than fitting process..\n",
    "for i in range(N_AP - 1):\n",
    "    lower_cut, upper_cut = i*STEP_AP, (i+1)*STEP_AP\n",
    "    \n",
    "    apall_i = np.sum(objimage[:, lower_cut:upper_cut], axis=1)\n",
    "    sky_val = np.hstack( (apall_i[ap_sky[0]:ap_sky[1]], \n",
    "                          apall_i[ap_sky[2]:ap_sky[3]]))\n",
    "    \n",
    "    # Subtract fitted sky\n",
    "    if FITTING_MODEL_APSKY.lower() == 'chebyshev':\n",
    "        # TODO: maybe we can put smoothing function as IRAF APALL's b_naverage \n",
    "        clip_mask = sigma_clip(sky_val, sigma=SIGMA_APSKY, iters=ITERS_APSKY).mask\n",
    "        coeff, fitfull = chebfit(x_sky[~clip_mask], \n",
    "                                 sky_val[~clip_mask],\n",
    "                                 deg=ORDER_APSKY,\n",
    "                                 full=True)\n",
    "        apall_i -= chebval(x_apall, coeff)\n",
    "#        fitRMS = np.sqrt(fitfull[0][0]/n_found)\n",
    "#        n_sky = len(x_sky)\n",
    "#        n_rej = np.count_nonzero(clip_mask)\n",
    "    \n",
    "    else:\n",
    "        raise ValueError('Function {:s} is not implemented.'.format(FITTING_MODEL_APSKY))\n",
    "\n",
    "    #TODO: put something like \"lost_factor\" to multiply to FWHM_ID in the bounds.\n",
    "    search_min = int(np.around(ap_init - 3*FWHM_AP))\n",
    "    search_max = int(np.around(ap_init + 3*FWHM_AP))\n",
    "    cropped = apall_i[search_min:search_max]\n",
    "    x_cropped = np.arange(len(cropped))\n",
    "    peak_pix = peak_local_max(cropped, \n",
    "                              min_distance=FWHM_AP,\n",
    "                              indices=True,\n",
    "                              num_peaks=1)\n",
    "    if len(peak_pix) == 0:\n",
    "        aptrace.append(np.nan)\n",
    "        continue\n",
    "    peak_pix = peak_pix[0][0]\n",
    "    \n",
    "    #TODO: put something like \"lost_factor\" to multiply to FWHM_ID in the bounds.\n",
    "    g_init = Gaussian1D(amplitude=cropped[peak_pix], \n",
    "                       mean=peak_pix, \n",
    "                       stddev=FWHM_AP * gaussian_fwhm_to_sigma,\n",
    "                       bounds={'amplitude':(0, 2*cropped[peak_pix]) ,\n",
    "                               'mean':(peak_pix-3*FWHM_AP, peak_pix+3*FWHM_AP),\n",
    "                               'stddev':(0, FWHM_AP)})\n",
    "    fitted = fitter(g_init, x_cropped, cropped)\n",
    "    center_pix = fitted.mean.value + search_min\n",
    "    std_pix = fitted.stddev.value\n",
    "    aptrace_fwhm.append(fitted.fwhm)\n",
    "    aptrace.append(center_pix)\n",
    "#    coeff_apsky.append(coeff)\n",
    "#    aptrace_apsum.append(apsum)\n",
    "#    apsum_lower = int(np.around(center_pix - apsum_sigma_lower * std_pix))\n",
    "#    apsum_upper = int(np.around(center_pix + apsum_sigma_upper * std_pix))\n",
    "#    apsum = np.sum(apall_i[apsum_lower:apsum_upper])\n",
    "\n",
    "aptrace = np.array(aptrace)\n",
    "aptrace_fwhm = np.array(aptrace_fwhm)\n",
    "#coeff_apsky = np.array(coeff_apsky)\n",
    "#aptrace_apsum = np.array(aptrace_apsum)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You may ignore ``RuntimeWarning``.\n",
    "\n",
    "Here I left some part of the codes for those who want to check/test the code.\n",
    "\n",
    "Now, for every ``STEP_AP`` columns (in our tutorial 10), find the object center and plot it:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"1000\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# apall(4): aperture trace fit\n",
    "# =============================================================================\n",
    "x_aptrace = np.arange(N_AP-1) * STEP_AP\n",
    "\n",
    "coeff_aptrace = chebfit(x_aptrace, aptrace, deg=ORDER_APTRACE)\n",
    "resid_mask = sigma_clip(aptrace - chebval(x_aptrace, coeff_aptrace), \n",
    "                        sigma=SIGMA_APTRACE, iters=ITERS_APTRACE).mask\n",
    "\n",
    "x_aptrace_fin = x_aptrace[~resid_mask]\n",
    "aptrace_fin = aptrace[~resid_mask]\n",
    "coeff_aptrace_fin = chebfit(x_aptrace_fin, aptrace_fin, deg=ORDER_APTRACE)\n",
    "fit_aptrace_fin   = chebval(x_aptrace_fin, coeff_aptrace_fin)\n",
    "resid_aptrace_fin = aptrace_fin - fit_aptrace_fin\n",
    "del_aptrace = ~np.in1d(x_aptrace, x_aptrace_fin) # deleted points\n",
    "\n",
    "fig = plt.figure(figsize=(10,8))\n",
    "gs = gridspec.GridSpec(3, 1)\n",
    "ax1 = plt.subplot(gs[0:2], sharex=ax2)\n",
    "ax2 = plt.subplot(gs[2])\n",
    "\n",
    "title_str = ('Aperture Trace Fit ({:s} order {:d})\\n'\n",
    "            + 'Residuials {:.1f}-sigma, {:d}-iters clipped')\n",
    "plt.suptitle(title_str.format(FITTING_MODEL_APTRACE, ORDER_APTRACE,\n",
    "                              SIGMA_APTRACE, ITERS_APTRACE))\n",
    "ax1.plot(x_aptrace, aptrace, ls='', marker='+', ms=10)\n",
    "ax1.plot(x_aptrace_fin, fit_aptrace_fin, ls='--',\n",
    "         label=\"Aperture Trace ({:d}/{:d} used)\".format(len(aptrace_fin), N_AP-1))\n",
    "ax1.plot(x_aptrace[del_aptrace], aptrace[del_aptrace], ls='', marker='x', ms=10)\n",
    "ax1.legend()\n",
    "ax2.plot(x_aptrace_fin, resid_aptrace_fin, ls='', marker='+')\n",
    "#ax2.plot(x_aptrace, aptrace - chebval(x_aptrace, coeff_aptrace_fin), \n",
    "#         ls='', marker='+')\n",
    "ax2.axhline(+np.std(resid_aptrace_fin, ddof=1), ls=':', color='k')\n",
    "ax2.axhline(-np.std(resid_aptrace_fin, ddof=1), ls=':', color='k', \n",
    "            label='residual std')\n",
    "\n",
    "ax1.set_ylabel('Found object position')\n",
    "ax2.set_ylabel('Residual (pixel)')\n",
    "ax2.set_xlabel('Dispersion axis (pixel)')\n",
    "ax1.grid(ls=':')\n",
    "ax2.grid(ls=':')\n",
    "ax2.set_ylim(-.5, .5)\n",
    "ax2.legend()\n",
    "plt.show()\n",
    "#plt.savefig('aptrace.png', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see, the 3-sigma 5-iters clipping for the residual worked and some traces are removed by this process. Althougn there still seem to be a small amount of non-linear residual so order 4 or 5 will work better, let me stick to order 3 for this tutorial.\n",
    "\n",
    "For IRAF, the following code with [this](https://drive.google.com/open?id=0B-MLFRYnMxUvWkxDa244X2E5MVU) aperture extract file will give the following image (put it in ``database/`` as IRAF default)\n",
    "\n",
    "```\n",
    "--> apall pobj_mls170213_0025 find- resize- nsum=10 b_func=cheb b_order=3 b_nav=1 b_nit=5 t_nsum=10 t_step=10 t_func=cheb t_ord=3 t_nav=1 t_nit=5 background=fit skybox=1 pfit=fit1d\n",
    "```\n",
    "\n",
    "![](images/apall_iraf.png)\n",
    "\n",
    "I tried to match the sky regions and aperture size as my python code, but few pixels differ. As you can see from the aperture trace and residual plot, the results are different, but the order of residuals are similar (much less than 1 pixel).\n",
    "\n",
    "Now I want to use the fitting function as the true center of the object, and sum the aperture values after sky subtraction:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# apall(5): aperture sum\n",
    "# =============================================================================\n",
    "apsum_sigma_lower = 2.104 # See below\n",
    "apsum_sigma_upper = 2.130 \n",
    "# lower and upper limits of aperture to set from the center in gauss-sigma unit.\n",
    "ap_fwhm = np.median(aptrace_fwhm[~resid_mask])\n",
    "ap_sigma = ap_fwhm * gaussian_fwhm_to_sigma\n",
    "\n",
    "x_ap = np.arange(N_WAVELEN)\n",
    "y_ap = chebval(x_ap, coeff_aptrace_fin)\n",
    "ap_wavelen = fit2D_REID(x_ap, y_ap)\n",
    "ap_summed  = []\n",
    "ap_sky_offset = ap_sky - ap_init\n",
    "\n",
    "for i in range(N_WAVELEN):\n",
    "    cut_i = objimage[:, i]\n",
    "    ap_sky_i = int(y_ap[i]) + ap_sky_offset\n",
    "   \n",
    "    lower = y_ap[i] - apsum_sigma_lower * ap_sigma\n",
    "    upper = y_ap[i] + apsum_sigma_upper * ap_sigma + 1\n",
    "    # + 1 required since python regards, e.g., list[1:3] = list[1], list[2] (NOT list[3] included).\n",
    "    \n",
    "    x_obj_lower = int(np.around(y_ap[i] - apsum_sigma_lower * ap_sigma))\n",
    "    x_obj_upper = int(np.around(y_ap[i] + apsum_sigma_upper * ap_sigma)) \n",
    "    x_obj = np.arange(x_obj_lower, x_obj_upper)\n",
    "    obj_i = cut_i[x_obj_lower:x_obj_upper]\n",
    "   \n",
    "    l_pix_w = x_obj_lower - lower - 0.5\n",
    "    u_pix_w = upper - x_obj_upper - 0.5\n",
    "   \n",
    "    x_sky = np.hstack( (np.arange(ap_sky_i[0], ap_sky_i[1]),\n",
    "                        np.arange(ap_sky_i[2], ap_sky_i[3])))\n",
    "    sky_val = np.hstack( (cut_i[ap_sky_i[0]:ap_sky_i[1]],\n",
    "                          cut_i[ap_sky_i[2]:ap_sky_i[3]]))\n",
    "    clip_mask = sigma_clip(sky_val, sigma=SIGMA_APSKY, iters=ITERS_APSKY).mask\n",
    "   \n",
    "    coeff = chebfit(x_sky[~clip_mask],\n",
    "                    sky_val[~clip_mask],\n",
    "                    deg=ORDER_APSKY)\n",
    "\n",
    "    obj_i -= chebval(x_obj, coeff)\n",
    "\n",
    "    l_pix_val = obj_i[0] * l_pix_w   # negative value (hopefully)\n",
    "    u_pix_val = obj_i[-1] * u_pix_w  # negative value (hopefully)\n",
    "\n",
    "    ap_summed.append(np.sum(obj_i)+l_pix_val+u_pix_val)\n",
    "#    print(lower, upper, ap_summed[i])\n",
    "   \n",
    "ap_summed = np.array(ap_summed) / EXPTIME"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I then compare it with the IRAF result. I saved it as ``~~~.ms.fits`` file, so"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "/* Put everything inside the global mpl namespace */\n",
       "window.mpl = {};\n",
       "\n",
       "\n",
       "mpl.get_websocket_type = function() {\n",
       "    if (typeof(WebSocket) !== 'undefined') {\n",
       "        return WebSocket;\n",
       "    } else if (typeof(MozWebSocket) !== 'undefined') {\n",
       "        return MozWebSocket;\n",
       "    } else {\n",
       "        alert('Your browser does not have WebSocket support.' +\n",
       "              'Please try Chrome, Safari or Firefox ≥ 6. ' +\n",
       "              'Firefox 4 and 5 are also supported but you ' +\n",
       "              'have to enable WebSockets in about:config.');\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure = function(figure_id, websocket, ondownload, parent_element) {\n",
       "    this.id = figure_id;\n",
       "\n",
       "    this.ws = websocket;\n",
       "\n",
       "    this.supports_binary = (this.ws.binaryType != undefined);\n",
       "\n",
       "    if (!this.supports_binary) {\n",
       "        var warnings = document.getElementById(\"mpl-warnings\");\n",
       "        if (warnings) {\n",
       "            warnings.style.display = 'block';\n",
       "            warnings.textContent = (\n",
       "                \"This browser does not support binary websocket messages. \" +\n",
       "                    \"Performance may be slow.\");\n",
       "        }\n",
       "    }\n",
       "\n",
       "    this.imageObj = new Image();\n",
       "\n",
       "    this.context = undefined;\n",
       "    this.message = undefined;\n",
       "    this.canvas = undefined;\n",
       "    this.rubberband_canvas = undefined;\n",
       "    this.rubberband_context = undefined;\n",
       "    this.format_dropdown = undefined;\n",
       "\n",
       "    this.image_mode = 'full';\n",
       "\n",
       "    this.root = $('<div/>');\n",
       "    this._root_extra_style(this.root)\n",
       "    this.root.attr('style', 'display: inline-block');\n",
       "\n",
       "    $(parent_element).append(this.root);\n",
       "\n",
       "    this._init_header(this);\n",
       "    this._init_canvas(this);\n",
       "    this._init_toolbar(this);\n",
       "\n",
       "    var fig = this;\n",
       "\n",
       "    this.waiting = false;\n",
       "\n",
       "    this.ws.onopen =  function () {\n",
       "            fig.send_message(\"supports_binary\", {value: fig.supports_binary});\n",
       "            fig.send_message(\"send_image_mode\", {});\n",
       "            if (mpl.ratio != 1) {\n",
       "                fig.send_message(\"set_dpi_ratio\", {'dpi_ratio': mpl.ratio});\n",
       "            }\n",
       "            fig.send_message(\"refresh\", {});\n",
       "        }\n",
       "\n",
       "    this.imageObj.onload = function() {\n",
       "            if (fig.image_mode == 'full') {\n",
       "                // Full images could contain transparency (where diff images\n",
       "                // almost always do), so we need to clear the canvas so that\n",
       "                // there is no ghosting.\n",
       "                fig.context.clearRect(0, 0, fig.canvas.width, fig.canvas.height);\n",
       "            }\n",
       "            fig.context.drawImage(fig.imageObj, 0, 0);\n",
       "        };\n",
       "\n",
       "    this.imageObj.onunload = function() {\n",
       "        fig.ws.close();\n",
       "    }\n",
       "\n",
       "    this.ws.onmessage = this._make_on_message_function(this);\n",
       "\n",
       "    this.ondownload = ondownload;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_header = function() {\n",
       "    var titlebar = $(\n",
       "        '<div class=\"ui-dialog-titlebar ui-widget-header ui-corner-all ' +\n",
       "        'ui-helper-clearfix\"/>');\n",
       "    var titletext = $(\n",
       "        '<div class=\"ui-dialog-title\" style=\"width: 100%; ' +\n",
       "        'text-align: center; padding: 3px;\"/>');\n",
       "    titlebar.append(titletext)\n",
       "    this.root.append(titlebar);\n",
       "    this.header = titletext[0];\n",
       "}\n",
       "\n",
       "\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(canvas_div) {\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_canvas = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var canvas_div = $('<div/>');\n",
       "\n",
       "    canvas_div.attr('style', 'position: relative; clear: both; outline: 0');\n",
       "\n",
       "    function canvas_keyboard_event(event) {\n",
       "        return fig.key_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    canvas_div.keydown('key_press', canvas_keyboard_event);\n",
       "    canvas_div.keyup('key_release', canvas_keyboard_event);\n",
       "    this.canvas_div = canvas_div\n",
       "    this._canvas_extra_style(canvas_div)\n",
       "    this.root.append(canvas_div);\n",
       "\n",
       "    var canvas = $('<canvas/>');\n",
       "    canvas.addClass('mpl-canvas');\n",
       "    canvas.attr('style', \"left: 0; top: 0; z-index: 0; outline: 0\")\n",
       "\n",
       "    this.canvas = canvas[0];\n",
       "    this.context = canvas[0].getContext(\"2d\");\n",
       "\n",
       "    var backingStore = this.context.backingStorePixelRatio ||\n",
       "\tthis.context.webkitBackingStorePixelRatio ||\n",
       "\tthis.context.mozBackingStorePixelRatio ||\n",
       "\tthis.context.msBackingStorePixelRatio ||\n",
       "\tthis.context.oBackingStorePixelRatio ||\n",
       "\tthis.context.backingStorePixelRatio || 1;\n",
       "\n",
       "    mpl.ratio = (window.devicePixelRatio || 1) / backingStore;\n",
       "\n",
       "    var rubberband = $('<canvas/>');\n",
       "    rubberband.attr('style', \"position: absolute; left: 0; top: 0; z-index: 1;\")\n",
       "\n",
       "    var pass_mouse_events = true;\n",
       "\n",
       "    canvas_div.resizable({\n",
       "        start: function(event, ui) {\n",
       "            pass_mouse_events = false;\n",
       "        },\n",
       "        resize: function(event, ui) {\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "        stop: function(event, ui) {\n",
       "            pass_mouse_events = true;\n",
       "            fig.request_resize(ui.size.width, ui.size.height);\n",
       "        },\n",
       "    });\n",
       "\n",
       "    function mouse_event_fn(event) {\n",
       "        if (pass_mouse_events)\n",
       "            return fig.mouse_event(event, event['data']);\n",
       "    }\n",
       "\n",
       "    rubberband.mousedown('button_press', mouse_event_fn);\n",
       "    rubberband.mouseup('button_release', mouse_event_fn);\n",
       "    // Throttle sequential mouse events to 1 every 20ms.\n",
       "    rubberband.mousemove('motion_notify', mouse_event_fn);\n",
       "\n",
       "    rubberband.mouseenter('figure_enter', mouse_event_fn);\n",
       "    rubberband.mouseleave('figure_leave', mouse_event_fn);\n",
       "\n",
       "    canvas_div.on(\"wheel\", function (event) {\n",
       "        event = event.originalEvent;\n",
       "        event['data'] = 'scroll'\n",
       "        if (event.deltaY < 0) {\n",
       "            event.step = 1;\n",
       "        } else {\n",
       "            event.step = -1;\n",
       "        }\n",
       "        mouse_event_fn(event);\n",
       "    });\n",
       "\n",
       "    canvas_div.append(canvas);\n",
       "    canvas_div.append(rubberband);\n",
       "\n",
       "    this.rubberband = rubberband;\n",
       "    this.rubberband_canvas = rubberband[0];\n",
       "    this.rubberband_context = rubberband[0].getContext(\"2d\");\n",
       "    this.rubberband_context.strokeStyle = \"#000000\";\n",
       "\n",
       "    this._resize_canvas = function(width, height) {\n",
       "        // Keep the size of the canvas, canvas container, and rubber band\n",
       "        // canvas in synch.\n",
       "        canvas_div.css('width', width)\n",
       "        canvas_div.css('height', height)\n",
       "\n",
       "        canvas.attr('width', width * mpl.ratio);\n",
       "        canvas.attr('height', height * mpl.ratio);\n",
       "        canvas.attr('style', 'width: ' + width + 'px; height: ' + height + 'px;');\n",
       "\n",
       "        rubberband.attr('width', width);\n",
       "        rubberband.attr('height', height);\n",
       "    }\n",
       "\n",
       "    // Set the figure to an initial 600x600px, this will subsequently be updated\n",
       "    // upon first draw.\n",
       "    this._resize_canvas(600, 600);\n",
       "\n",
       "    // Disable right mouse context menu.\n",
       "    $(this.rubberband_canvas).bind(\"contextmenu\",function(e){\n",
       "        return false;\n",
       "    });\n",
       "\n",
       "    function set_focus () {\n",
       "        canvas.focus();\n",
       "        canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    window.setTimeout(set_focus, 100);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items) {\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) {\n",
       "            // put a spacer in here.\n",
       "            continue;\n",
       "        }\n",
       "        var button = $('<button/>');\n",
       "        button.addClass('ui-button ui-widget ui-state-default ui-corner-all ' +\n",
       "                        'ui-button-icon-only');\n",
       "        button.attr('role', 'button');\n",
       "        button.attr('aria-disabled', 'false');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "\n",
       "        var icon_img = $('<span/>');\n",
       "        icon_img.addClass('ui-button-icon-primary ui-icon');\n",
       "        icon_img.addClass(image);\n",
       "        icon_img.addClass('ui-corner-all');\n",
       "\n",
       "        var tooltip_span = $('<span/>');\n",
       "        tooltip_span.addClass('ui-button-text');\n",
       "        tooltip_span.html(tooltip);\n",
       "\n",
       "        button.append(icon_img);\n",
       "        button.append(tooltip_span);\n",
       "\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    var fmt_picker_span = $('<span/>');\n",
       "\n",
       "    var fmt_picker = $('<select/>');\n",
       "    fmt_picker.addClass('mpl-toolbar-option ui-widget ui-widget-content');\n",
       "    fmt_picker_span.append(fmt_picker);\n",
       "    nav_element.append(fmt_picker_span);\n",
       "    this.format_dropdown = fmt_picker[0];\n",
       "\n",
       "    for (var ind in mpl.extensions) {\n",
       "        var fmt = mpl.extensions[ind];\n",
       "        var option = $(\n",
       "            '<option/>', {selected: fmt === mpl.default_extension}).html(fmt);\n",
       "        fmt_picker.append(option)\n",
       "    }\n",
       "\n",
       "    // Add hover states to the ui-buttons\n",
       "    $( \".ui-button\" ).hover(\n",
       "        function() { $(this).addClass(\"ui-state-hover\");},\n",
       "        function() { $(this).removeClass(\"ui-state-hover\");}\n",
       "    );\n",
       "\n",
       "    var status_bar = $('<span class=\"mpl-message\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.request_resize = function(x_pixels, y_pixels) {\n",
       "    // Request matplotlib to resize the figure. Matplotlib will then trigger a resize in the client,\n",
       "    // which will in turn request a refresh of the image.\n",
       "    this.send_message('resize', {'width': x_pixels, 'height': y_pixels});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_message = function(type, properties) {\n",
       "    properties['type'] = type;\n",
       "    properties['figure_id'] = this.id;\n",
       "    this.ws.send(JSON.stringify(properties));\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.send_draw_message = function() {\n",
       "    if (!this.waiting) {\n",
       "        this.waiting = true;\n",
       "        this.ws.send(JSON.stringify({type: \"draw\", figure_id: this.id}));\n",
       "    }\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    var format_dropdown = fig.format_dropdown;\n",
       "    var format = format_dropdown.options[format_dropdown.selectedIndex].value;\n",
       "    fig.ondownload(fig, format);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.figure.prototype.handle_resize = function(fig, msg) {\n",
       "    var size = msg['size'];\n",
       "    if (size[0] != fig.canvas.width || size[1] != fig.canvas.height) {\n",
       "        fig._resize_canvas(size[0], size[1]);\n",
       "        fig.send_message(\"refresh\", {});\n",
       "    };\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_rubberband = function(fig, msg) {\n",
       "    var x0 = msg['x0'] / mpl.ratio;\n",
       "    var y0 = (fig.canvas.height - msg['y0']) / mpl.ratio;\n",
       "    var x1 = msg['x1'] / mpl.ratio;\n",
       "    var y1 = (fig.canvas.height - msg['y1']) / mpl.ratio;\n",
       "    x0 = Math.floor(x0) + 0.5;\n",
       "    y0 = Math.floor(y0) + 0.5;\n",
       "    x1 = Math.floor(x1) + 0.5;\n",
       "    y1 = Math.floor(y1) + 0.5;\n",
       "    var min_x = Math.min(x0, x1);\n",
       "    var min_y = Math.min(y0, y1);\n",
       "    var width = Math.abs(x1 - x0);\n",
       "    var height = Math.abs(y1 - y0);\n",
       "\n",
       "    fig.rubberband_context.clearRect(\n",
       "        0, 0, fig.canvas.width, fig.canvas.height);\n",
       "\n",
       "    fig.rubberband_context.strokeRect(min_x, min_y, width, height);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_figure_label = function(fig, msg) {\n",
       "    // Updates the figure title.\n",
       "    fig.header.textContent = msg['label'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_cursor = function(fig, msg) {\n",
       "    var cursor = msg['cursor'];\n",
       "    switch(cursor)\n",
       "    {\n",
       "    case 0:\n",
       "        cursor = 'pointer';\n",
       "        break;\n",
       "    case 1:\n",
       "        cursor = 'default';\n",
       "        break;\n",
       "    case 2:\n",
       "        cursor = 'crosshair';\n",
       "        break;\n",
       "    case 3:\n",
       "        cursor = 'move';\n",
       "        break;\n",
       "    }\n",
       "    fig.rubberband_canvas.style.cursor = cursor;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_message = function(fig, msg) {\n",
       "    fig.message.textContent = msg['message'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_draw = function(fig, msg) {\n",
       "    // Request the server to send over a new figure.\n",
       "    fig.send_draw_message();\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_image_mode = function(fig, msg) {\n",
       "    fig.image_mode = msg['mode'];\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Called whenever the canvas gets updated.\n",
       "    this.send_message(\"ack\", {});\n",
       "}\n",
       "\n",
       "// A function to construct a web socket function for onmessage handling.\n",
       "// Called in the figure constructor.\n",
       "mpl.figure.prototype._make_on_message_function = function(fig) {\n",
       "    return function socket_on_message(evt) {\n",
       "        if (evt.data instanceof Blob) {\n",
       "            /* FIXME: We get \"Resource interpreted as Image but\n",
       "             * transferred with MIME type text/plain:\" errors on\n",
       "             * Chrome.  But how to set the MIME type?  It doesn't seem\n",
       "             * to be part of the websocket stream */\n",
       "            evt.data.type = \"image/png\";\n",
       "\n",
       "            /* Free the memory for the previous frames */\n",
       "            if (fig.imageObj.src) {\n",
       "                (window.URL || window.webkitURL).revokeObjectURL(\n",
       "                    fig.imageObj.src);\n",
       "            }\n",
       "\n",
       "            fig.imageObj.src = (window.URL || window.webkitURL).createObjectURL(\n",
       "                evt.data);\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "        else if (typeof evt.data === 'string' && evt.data.slice(0, 21) == \"data:image/png;base64\") {\n",
       "            fig.imageObj.src = evt.data;\n",
       "            fig.updated_canvas_event();\n",
       "            fig.waiting = false;\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        var msg = JSON.parse(evt.data);\n",
       "        var msg_type = msg['type'];\n",
       "\n",
       "        // Call the  \"handle_{type}\" callback, which takes\n",
       "        // the figure and JSON message as its only arguments.\n",
       "        try {\n",
       "            var callback = fig[\"handle_\" + msg_type];\n",
       "        } catch (e) {\n",
       "            console.log(\"No handler for the '\" + msg_type + \"' message type: \", msg);\n",
       "            return;\n",
       "        }\n",
       "\n",
       "        if (callback) {\n",
       "            try {\n",
       "                // console.log(\"Handling '\" + msg_type + \"' message: \", msg);\n",
       "                callback(fig, msg);\n",
       "            } catch (e) {\n",
       "                console.log(\"Exception inside the 'handler_\" + msg_type + \"' callback:\", e, e.stack, msg);\n",
       "            }\n",
       "        }\n",
       "    };\n",
       "}\n",
       "\n",
       "// from http://stackoverflow.com/questions/1114465/getting-mouse-location-in-canvas\n",
       "mpl.findpos = function(e) {\n",
       "    //this section is from http://www.quirksmode.org/js/events_properties.html\n",
       "    var targ;\n",
       "    if (!e)\n",
       "        e = window.event;\n",
       "    if (e.target)\n",
       "        targ = e.target;\n",
       "    else if (e.srcElement)\n",
       "        targ = e.srcElement;\n",
       "    if (targ.nodeType == 3) // defeat Safari bug\n",
       "        targ = targ.parentNode;\n",
       "\n",
       "    // jQuery normalizes the pageX and pageY\n",
       "    // pageX,Y are the mouse positions relative to the document\n",
       "    // offset() returns the position of the element relative to the document\n",
       "    var x = e.pageX - $(targ).offset().left;\n",
       "    var y = e.pageY - $(targ).offset().top;\n",
       "\n",
       "    return {\"x\": x, \"y\": y};\n",
       "};\n",
       "\n",
       "/*\n",
       " * return a copy of an object with only non-object keys\n",
       " * we need this to avoid circular references\n",
       " * http://stackoverflow.com/a/24161582/3208463\n",
       " */\n",
       "function simpleKeys (original) {\n",
       "  return Object.keys(original).reduce(function (obj, key) {\n",
       "    if (typeof original[key] !== 'object')\n",
       "        obj[key] = original[key]\n",
       "    return obj;\n",
       "  }, {});\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.mouse_event = function(event, name) {\n",
       "    var canvas_pos = mpl.findpos(event)\n",
       "\n",
       "    if (name === 'button_press')\n",
       "    {\n",
       "        this.canvas.focus();\n",
       "        this.canvas_div.focus();\n",
       "    }\n",
       "\n",
       "    var x = canvas_pos.x * mpl.ratio;\n",
       "    var y = canvas_pos.y * mpl.ratio;\n",
       "\n",
       "    this.send_message(name, {x: x, y: y, button: event.button,\n",
       "                             step: event.step,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "\n",
       "    /* This prevents the web browser from automatically changing to\n",
       "     * the text insertion cursor when the button is pressed.  We want\n",
       "     * to control all of the cursor setting manually through the\n",
       "     * 'cursor' event from matplotlib */\n",
       "    event.preventDefault();\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    // Handle any extra behaviour associated with a key event\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.key_event = function(event, name) {\n",
       "\n",
       "    // Prevent repeat events\n",
       "    if (name == 'key_press')\n",
       "    {\n",
       "        if (event.which === this._key)\n",
       "            return;\n",
       "        else\n",
       "            this._key = event.which;\n",
       "    }\n",
       "    if (name == 'key_release')\n",
       "        this._key = null;\n",
       "\n",
       "    var value = '';\n",
       "    if (event.ctrlKey && event.which != 17)\n",
       "        value += \"ctrl+\";\n",
       "    if (event.altKey && event.which != 18)\n",
       "        value += \"alt+\";\n",
       "    if (event.shiftKey && event.which != 16)\n",
       "        value += \"shift+\";\n",
       "\n",
       "    value += 'k';\n",
       "    value += event.which.toString();\n",
       "\n",
       "    this._key_event_extra(event, name);\n",
       "\n",
       "    this.send_message(name, {key: value,\n",
       "                             guiEvent: simpleKeys(event)});\n",
       "    return false;\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onclick = function(name) {\n",
       "    if (name == 'download') {\n",
       "        this.handle_save(this, null);\n",
       "    } else {\n",
       "        this.send_message(\"toolbar_button\", {name: name});\n",
       "    }\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.toolbar_button_onmouseover = function(tooltip) {\n",
       "    this.message.textContent = tooltip;\n",
       "};\n",
       "mpl.toolbar_items = [[\"Home\", \"Reset original view\", \"fa fa-home icon-home\", \"home\"], [\"Back\", \"Back to  previous view\", \"fa fa-arrow-left icon-arrow-left\", \"back\"], [\"Forward\", \"Forward to next view\", \"fa fa-arrow-right icon-arrow-right\", \"forward\"], [\"\", \"\", \"\", \"\"], [\"Pan\", \"Pan axes with left mouse, zoom with right\", \"fa fa-arrows icon-move\", \"pan\"], [\"Zoom\", \"Zoom to rectangle\", \"fa fa-square-o icon-check-empty\", \"zoom\"], [\"\", \"\", \"\", \"\"], [\"Download\", \"Download plot\", \"fa fa-floppy-o icon-save\", \"download\"]];\n",
       "\n",
       "mpl.extensions = [\"eps\", \"jpeg\", \"pdf\", \"png\", \"ps\", \"raw\", \"svg\", \"tif\"];\n",
       "\n",
       "mpl.default_extension = \"png\";var comm_websocket_adapter = function(comm) {\n",
       "    // Create a \"websocket\"-like object which calls the given IPython comm\n",
       "    // object with the appropriate methods. Currently this is a non binary\n",
       "    // socket, so there is still some room for performance tuning.\n",
       "    var ws = {};\n",
       "\n",
       "    ws.close = function() {\n",
       "        comm.close()\n",
       "    };\n",
       "    ws.send = function(m) {\n",
       "        //console.log('sending', m);\n",
       "        comm.send(m);\n",
       "    };\n",
       "    // Register the callback with on_msg.\n",
       "    comm.on_msg(function(msg) {\n",
       "        //console.log('receiving', msg['content']['data'], msg);\n",
       "        // Pass the mpl event to the overriden (by mpl) onmessage function.\n",
       "        ws.onmessage(msg['content']['data'])\n",
       "    });\n",
       "    return ws;\n",
       "}\n",
       "\n",
       "mpl.mpl_figure_comm = function(comm, msg) {\n",
       "    // This is the function which gets called when the mpl process\n",
       "    // starts-up an IPython Comm through the \"matplotlib\" channel.\n",
       "\n",
       "    var id = msg.content.data.id;\n",
       "    // Get hold of the div created by the display call when the Comm\n",
       "    // socket was opened in Python.\n",
       "    var element = $(\"#\" + id);\n",
       "    var ws_proxy = comm_websocket_adapter(comm)\n",
       "\n",
       "    function ondownload(figure, format) {\n",
       "        window.open(figure.imageObj.src);\n",
       "    }\n",
       "\n",
       "    var fig = new mpl.figure(id, ws_proxy,\n",
       "                           ondownload,\n",
       "                           element.get(0));\n",
       "\n",
       "    // Call onopen now - mpl needs it, as it is assuming we've passed it a real\n",
       "    // web socket which is closed, not our websocket->open comm proxy.\n",
       "    ws_proxy.onopen();\n",
       "\n",
       "    fig.parent_element = element.get(0);\n",
       "    fig.cell_info = mpl.find_output_cell(\"<div id='\" + id + \"'></div>\");\n",
       "    if (!fig.cell_info) {\n",
       "        console.error(\"Failed to find cell for figure\", id, fig);\n",
       "        return;\n",
       "    }\n",
       "\n",
       "    var output_index = fig.cell_info[2]\n",
       "    var cell = fig.cell_info[0];\n",
       "\n",
       "};\n",
       "\n",
       "mpl.figure.prototype.handle_close = function(fig, msg) {\n",
       "    var width = fig.canvas.width/mpl.ratio\n",
       "    fig.root.unbind('remove')\n",
       "\n",
       "    // Update the output cell to use the data from the current canvas.\n",
       "    fig.push_to_output();\n",
       "    var dataURL = fig.canvas.toDataURL();\n",
       "    // Re-enable the keyboard manager in IPython - without this line, in FF,\n",
       "    // the notebook keyboard shortcuts fail.\n",
       "    IPython.keyboard_manager.enable()\n",
       "    $(fig.parent_element).html('<img src=\"' + dataURL + '\" width=\"' + width + '\">');\n",
       "    fig.close_ws(fig, msg);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.close_ws = function(fig, msg){\n",
       "    fig.send_message('closing', msg);\n",
       "    // fig.ws.close()\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.push_to_output = function(remove_interactive) {\n",
       "    // Turn the data on the canvas into data in the output cell.\n",
       "    var width = this.canvas.width/mpl.ratio\n",
       "    var dataURL = this.canvas.toDataURL();\n",
       "    this.cell_info[1]['text/html'] = '<img src=\"' + dataURL + '\" width=\"' + width + '\">';\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.updated_canvas_event = function() {\n",
       "    // Tell IPython that the notebook contents must change.\n",
       "    IPython.notebook.set_dirty(true);\n",
       "    this.send_message(\"ack\", {});\n",
       "    var fig = this;\n",
       "    // Wait a second, then push the new image to the DOM so\n",
       "    // that it is saved nicely (might be nice to debounce this).\n",
       "    setTimeout(function () { fig.push_to_output() }, 1000);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._init_toolbar = function() {\n",
       "    var fig = this;\n",
       "\n",
       "    var nav_element = $('<div/>')\n",
       "    nav_element.attr('style', 'width: 100%');\n",
       "    this.root.append(nav_element);\n",
       "\n",
       "    // Define a callback function for later on.\n",
       "    function toolbar_event(event) {\n",
       "        return fig.toolbar_button_onclick(event['data']);\n",
       "    }\n",
       "    function toolbar_mouse_event(event) {\n",
       "        return fig.toolbar_button_onmouseover(event['data']);\n",
       "    }\n",
       "\n",
       "    for(var toolbar_ind in mpl.toolbar_items){\n",
       "        var name = mpl.toolbar_items[toolbar_ind][0];\n",
       "        var tooltip = mpl.toolbar_items[toolbar_ind][1];\n",
       "        var image = mpl.toolbar_items[toolbar_ind][2];\n",
       "        var method_name = mpl.toolbar_items[toolbar_ind][3];\n",
       "\n",
       "        if (!name) { continue; };\n",
       "\n",
       "        var button = $('<button class=\"btn btn-default\" href=\"#\" title=\"' + name + '\"><i class=\"fa ' + image + ' fa-lg\"></i></button>');\n",
       "        button.click(method_name, toolbar_event);\n",
       "        button.mouseover(tooltip, toolbar_mouse_event);\n",
       "        nav_element.append(button);\n",
       "    }\n",
       "\n",
       "    // Add the status bar.\n",
       "    var status_bar = $('<span class=\"mpl-message\" style=\"text-align:right; float: right;\"/>');\n",
       "    nav_element.append(status_bar);\n",
       "    this.message = status_bar[0];\n",
       "\n",
       "    // Add the close button to the window.\n",
       "    var buttongrp = $('<div class=\"btn-group inline pull-right\"></div>');\n",
       "    var button = $('<button class=\"btn btn-mini btn-primary\" href=\"#\" title=\"Stop Interaction\"><i class=\"fa fa-power-off icon-remove icon-large\"></i></button>');\n",
       "    button.click(function (evt) { fig.handle_close(fig, {}); } );\n",
       "    button.mouseover('Stop Interaction', toolbar_mouse_event);\n",
       "    buttongrp.append(button);\n",
       "    var titlebar = this.root.find($('.ui-dialog-titlebar'));\n",
       "    titlebar.prepend(buttongrp);\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._root_extra_style = function(el){\n",
       "    var fig = this\n",
       "    el.on(\"remove\", function(){\n",
       "\tfig.close_ws(fig, {});\n",
       "    });\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._canvas_extra_style = function(el){\n",
       "    // this is important to make the div 'focusable\n",
       "    el.attr('tabindex', 0)\n",
       "    // reach out to IPython and tell the keyboard manager to turn it's self\n",
       "    // off when our div gets focus\n",
       "\n",
       "    // location in version 3\n",
       "    if (IPython.notebook.keyboard_manager) {\n",
       "        IPython.notebook.keyboard_manager.register_events(el);\n",
       "    }\n",
       "    else {\n",
       "        // location in version 2\n",
       "        IPython.keyboard_manager.register_events(el);\n",
       "    }\n",
       "\n",
       "}\n",
       "\n",
       "mpl.figure.prototype._key_event_extra = function(event, name) {\n",
       "    var manager = IPython.notebook.keyboard_manager;\n",
       "    if (!manager)\n",
       "        manager = IPython.keyboard_manager;\n",
       "\n",
       "    // Check for shift+enter\n",
       "    if (event.shiftKey && event.which == 13) {\n",
       "        this.canvas_div.blur();\n",
       "        event.shiftKey = false;\n",
       "        // Send a \"J\" for go to next cell\n",
       "        event.which = 74;\n",
       "        event.keyCode = 74;\n",
       "        manager.command_mode();\n",
       "        manager.handle_keydown(event);\n",
       "    }\n",
       "}\n",
       "\n",
       "mpl.figure.prototype.handle_save = function(fig, msg) {\n",
       "    fig.ondownload(fig, null);\n",
       "}\n",
       "\n",
       "\n",
       "mpl.find_output_cell = function(html_output) {\n",
       "    // Return the cell and output element which can be found *uniquely* in the notebook.\n",
       "    // Note - this is a bit hacky, but it is done because the \"notebook_saving.Notebook\"\n",
       "    // IPython event is triggered only after the cells have been serialised, which for\n",
       "    // our purposes (turning an active figure into a static one), is too late.\n",
       "    var cells = IPython.notebook.get_cells();\n",
       "    var ncells = cells.length;\n",
       "    for (var i=0; i<ncells; i++) {\n",
       "        var cell = cells[i];\n",
       "        if (cell.cell_type === 'code'){\n",
       "            for (var j=0; j<cell.output_area.outputs.length; j++) {\n",
       "                var data = cell.output_area.outputs[j];\n",
       "                if (data.data) {\n",
       "                    // IPython >= 3 moved mimebundle to data attribute of output\n",
       "                    data = data.data;\n",
       "                }\n",
       "                if (data['text/html'] == html_output) {\n",
       "                    return [cell, data, j];\n",
       "                }\n",
       "            }\n",
       "        }\n",
       "    }\n",
       "}\n",
       "\n",
       "// Register the function which deals with the matplotlib target/channel.\n",
       "// The kernel may be null if the page has been refreshed.\n",
       "if (IPython.notebook.kernel != null) {\n",
       "    IPython.notebook.kernel.comm_manager.register_target('matplotlib', mpl.mpl_figure_comm);\n",
       "}\n"
      ],
      "text/plain": [
       "<IPython.core.display.Javascript object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<img src=\"data:image/png;base64,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\" width=\"1000\">"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "TESTIRAF = fits.open(os.path.join('45P', 'pobj_mls170213_0025.ms.fits'))\n",
    "TESTIRAF = TESTIRAF[0].data[0,0,:]\n",
    "\n",
    "fig = plt.figure(figsize=(10,5))\n",
    "gs = gridspec.GridSpec(3, 1)\n",
    "ax1 = plt.subplot(gs[0:2])\n",
    "ax2 = plt.subplot(gs[2])\n",
    "#plt.setp(ax1.get_xticklabels(), visible=False)\n",
    "\n",
    "title_str = '{:s}\\nobject = {:s}, EXPTIME = {:.1f} s'\n",
    "label_str = r'aperture width $\\approx$ {:.1f} pix'\n",
    "plt.suptitle(title_str.format(os.path.split(OBJIMAGE)[-1],\n",
    "                              OBJNAME, EXPTIME))\n",
    "ax1.plot(ap_wavelen, ap_summed, lw = 1,\n",
    "         alpha=0.5,\n",
    "         label=label_str.format(apsum_sigma_lower * ap_sigma\n",
    "                                + apsum_sigma_upper * ap_sigma))\n",
    "ax1.plot(ap_wavelen, TESTIRAF/EXPTIME, lw = 1,\n",
    "        alpha=0.5, label='IRAF')\n",
    "ax2.plot(ap_wavelen, (ap_summed - TESTIRAF/EXPTIME)/(TESTIRAF/EXPTIME), \n",
    "         ls='', alpha=0.5, marker='+')\n",
    "ax1.set_ylabel('Instrument Intensity\\n(apsum/EXPTIME)')\n",
    "ax2.set_ylabel(r'$\\frac{\\mathrm{python} - \\mathrm{IRAF}}{\\mathrm{IRAF}}$')\n",
    "ax2.set_xlabel(r'$\\lambda$ ($\\AA$)')\n",
    "ax2.set_ylim(-0.05, +0.05)\n",
    "ax1.grid(ls=':')\n",
    "ax2.grid(ls=':')\n",
    "\n",
    "ax1.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the code, ``apsum_sigma_lower`` and ``upper`` are set as such strange floating number to match the IRAF manual reduction. When I did the IRAF manual spectrum extraction, the generated ``apall`` file (stored in ``database/apBLAHBLAH``) says\n",
    "\n",
    "    ...\n",
    "    center\t1024. 553.6911\n",
    "\tlow\t-1023. -8.617371\n",
    "\thigh\t1024. 8.722351\n",
    "    ...\n",
    "\n",
    "i.e., the sigmas should be selected such that ``apsum_sigma_lower * ap_sigma = -8.617371`` and ``apsum_sigma_upper * ap_sigma = 8.722351``. So I set ``apsum_sigma_lower = 2.104`` and ``apsum_sigma_upper = 2.130``. \n",
    "\n",
    "\n",
    "Some large residuals at short and long wavelength limits are most likely due to low S/N ratio, which results in non-identical aperture trace and thus aperture sum. At these wavelength limits, aperture sums are dominated by noises, so it's not unrealistic to have this kind of errors.\n",
    "\n",
    "A slight deviation is acceptable since my python code may not EXACTLY resemble the IRAF code. \n",
    "\n",
    "* **NOTE**: I am not sure why yet but if you run the ``apall(5): aperture sum`` again and plot it, the resulting plot is not as desired. You have to load the image again and do ``apall(5): aperture sum`` again. Maybe I made a mistake..??\n",
    "\n",
    "* **NOTE**: I am still not sure about the \"Jumps\" in the residual plot. At wavelength of ~ 5900 and 7100, for example, there are small yet visible jumps. Maybe IRAF is doing some round off approximations?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}