PostageEyeballer.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "import pickle\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib as mpl\n",
    "try:\n",
    "    import matplotlib.pyplot as plt\n",
    "except:\n",
    "    mpl.use(\"Agg\")\n",
    "    import matplotlib.pyplot as plt\n",
    "    \n",
    "import fitsio as fio    \n",
    "import ngmix.medsreaders\n",
    "import ngmix\n",
    "import galsim\n",
    "import galsim.des\n",
    "import images"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prepare DATA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MEDS file paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "meds_g_path = \"/e/eser2/vargatn/DES/Y3_DATA/DEEP3/SN-C3_C01_r3688p01/SN-C3_C01_r3688p01_g_meds-Y3A2_DEEP.fits.fz\"\n",
    "meds_r_path = \"/e/eser2/vargatn/DES/Y3_DATA/DEEP3/SN-C3_C01_r3688p01/SN-C3_C01_r3688p01_r_meds-Y3A2_DEEP.fits.fz\"\n",
    "meds_i_path = \"/e/eser2/vargatn/DES/Y3_DATA/DEEP3/SN-C3_C01_r3688p01/SN-C3_C01_r3688p01_i_meds-Y3A2_DEEP.fits.fz\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "meds_paths = [\n",
    "    meds_g_path,\n",
    "    meds_r_path,\n",
    "    meds_i_path,\n",
    "]\n",
    "mms = [ngmix.medsreaders.NGMixMEDS(path) for path in meds_paths]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PSFex files path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "psfex_g_path = \"/e/eser2/vargatn/DES/Y3_DATA/DEEP3/SN-C3_C01_r3688p01/SN-C3_C01_r3688p01_g_psfcat.psf\"\n",
    "psfex_r_path = \"/e/eser2/vargatn/DES/Y3_DATA/DEEP3/SN-C3_C01_r3688p01/SN-C3_C01_r3688p01_r_psfcat.psf\"\n",
    "psfex_i_path = \"/e/eser2/vargatn/DES/Y3_DATA/DEEP3/SN-C3_C01_r3688p01/SN-C3_C01_r3688p01_i_psfcat.psf\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "psfex_paths = [\n",
    "    psfex_g_path,\n",
    "    psfex_r_path,\n",
    "    psfex_i_path,\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MOF path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "fname = \"/e/eser2/vargatn/DES/Y3_DATA/DEEP3/run-d02-SN-C3_C01_r3688p01.fits\"\n",
    "allmofs = fio.read(fname)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Renderer class"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "nbands = allmofs[\"bdf_flux\"].shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "pars = allmofs[\"bdf_pars\"][0, :-nbands]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'<'"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mms[1][\"id\"].byteswap().newbyteorder().dtype.byteorder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'>'"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "allmofs[\"id\"].dtype.byteorder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {},
   "outputs": [],
   "source": [
    "class ComparePostage():\n",
    "    _icutout = 0\n",
    "    _bands = (\"g\", \"r\", \"i\", \"z\")\n",
    "    def __init__(self, moftab, mms, psfex_files=None, mode=\"bdf\"):\n",
    "        \"\"\"\n",
    "        Currently this works in both bdf and cm\n",
    "        \n",
    "        It is assumed that the psf is not in the MEDS file, but is in an external psfex file\n",
    "        \n",
    "        Parameters:\n",
    "        -----------\n",
    "        moftab: record_array\n",
    "            MOF rows to render\n",
    "        mms: list\n",
    "            list of ngmix.medsreaders.NGMixMEDS objects, one for each band\n",
    "        psfex_files: list\n",
    "            list of psfex file paths corresponding to each MEDS files\n",
    "            \n",
    "        \"\"\"\n",
    "        \n",
    "        self.mof = moftab\n",
    "        self.mms = mms\n",
    "        self.psfex_files = psfex_files\n",
    "        \n",
    "        self.medsrow = self._get_indexarray()\n",
    "        \n",
    "        self.mode = mode\n",
    "        if self.mode == \"bdf\":\n",
    "            self.fluxcol = \"bdf_flux\"\n",
    "        elif self.mode == \"cm\":\n",
    "            self.fluxcol = \"cm_flux\"\n",
    "        else: \n",
    "            raise KeyError(\"only bdf or cm mode is supported in the moment\")\n",
    "            \n",
    "        self.nbands = self.mof[self.fluxcol].shape[1]\n",
    "        \n",
    "    def _get_indexarray(self):\n",
    "        col1 = mof[\"id\"]\n",
    "        col2 = self.mms[0][\"id\"]\n",
    "        \n",
    "#         if col\n",
    "        \n",
    "        imof = pd.DataFrame()\n",
    "        imof[\"mofid\"] = col1.byteswap().newbyteorder()\n",
    "        \n",
    "        imeds = pd.DataFrame()\n",
    "        imeds[\"medsid\"] = col2.byteswap().newbyteorder()\n",
    "        imeds[\"medsrow\"] = np.arange(len(col2)) \n",
    "        \n",
    "        tmp = pd.merge(imof, imeds, how=\"inner\", left_on=\"mofid\", right_on=\"medsid\")\n",
    "        return tmp[\"medsrow\"].values       \n",
    "        \n",
    "    def get_cutouts(self, iobj, iband, seg=False):\n",
    "        \"\"\"\n",
    "        iobj should refer to the id in the mof file,\n",
    "        and we should look up here which meds line it corresponds to. this is stored in self.medsrow\n",
    "        \"\"\"\n",
    "        m = self.mms[iband]\n",
    " \n",
    "        imeds = self.medsrow[iobj]\n",
    "#         print(imeds)\n",
    "        obs_im = m.get_cutout(imeds, self._icutout)\n",
    "\n",
    "        jac = m.get_jacobian(imeds, self._icutout)\n",
    "        wcs = galsim.JacobianWCS(jac[\"dudcol\"], jac[\"dudrow\"], jac[\"dvdcol\"], jac[\"dvdrow\"])\n",
    "\n",
    "        obs_ii = galsim.Image(obs_im, wcs=wcs)  \n",
    "        \n",
    "        # re color segmentation maps to catch same objects in different bands\n",
    "        seg_ii = None\n",
    "        if seg:\n",
    "            seg_ii = m.get_cutout(imeds, self._icutout, \"seg\") \n",
    "\n",
    "        orig_x = m[\"orig_col\"][imeds, 0]\n",
    "        orig_y = m[\"orig_row\"][imeds, 0]    \n",
    "        orig_pos = galsim.PositionD(orig_x, orig_y)\n",
    "        des_psf = galsim.des.DES_PSFEx(self.psfex_files[iband], wcs=wcs)\n",
    "        psf = des_psf.getPSF(orig_pos)    \n",
    "\n",
    "        flux = self.mof[self.fluxcol][iobj, iband]\n",
    "        band_pars = list(self.mof[self.mode + \"_pars\"][iobj, :-nbands])\n",
    "        band_pars = np.array(band_pars + [flux,])\n",
    "        \n",
    "        if self.mode == \"bdf\":\n",
    "            galmaker = ngmix.gmix.GMixBDF(band_pars)\n",
    "        elif self.mode == \"cm\":\n",
    "            galmaker = ngmix.gmix.GMixModel(band_pars, mode=\"cm\")\n",
    "            \n",
    "        gal = galmaker.make_galsim_object()\n",
    "\n",
    "        gal = galsim.Convolve(gal, psf)\n",
    "\n",
    "        gal_ii = gal.drawImage(nx=obs_ii.array.shape[0], ny=obs_ii.array.shape[1],\n",
    "                               wcs=wcs)   \n",
    "        return obs_ii, gal_ii, seg_ii        \n",
    "        \n",
    "    def draw_postage(self, fname_root=None, seg=False):\n",
    "\n",
    "        for iobj in np.arange(len(self.mof)):\n",
    "            g_obs, g_gal, seg_g_ii = self.get_cutouts(iobj, 0, seg=seg)\n",
    "            r_obs, r_gal, seg_r_ii = self.get_cutouts(iobj, 1, seg=seg)\n",
    "            i_obs, i_gal, seg_i_ii = self.get_cutouts(iobj, 2, seg=seg)\n",
    "\n",
    "            mv = np.min((g_obs.array, r_obs.array, i_obs.array))\n",
    "            mos_g = np.hstack([g_obs.array - mv, g_gal.array- mv, g_obs.array - g_gal.array - mv])\n",
    "            mos_r = np.hstack([r_obs.array - mv, r_gal.array- mv, r_obs.array - r_gal.array - mv])\n",
    "            mos_i = np.hstack([i_obs.array - mv, i_gal.array- mv, i_obs.array - i_gal.array - mv])\n",
    "\n",
    "            factor = 0.02\n",
    "            scales = np.array([1., 1., 1.3]) * factor\n",
    "            nonlinear = 0.5\n",
    "            clip = None        \n",
    "\n",
    "            obs_im = images.get_color_image(mos_i, mos_r, mos_g,\n",
    "                                            nonlinear=nonlinear, clip=clip, scales=scales)    \n",
    "\n",
    "            fig, axarr = plt.subplots(ncols=3, nrows=3, figsize=(9, 9))\n",
    "            fig.subplots_adjust(hspace=0.04, wspace=-0.01)\n",
    "            faxarr = axarr.flatten()\n",
    "\n",
    "            for ax in faxarr:\n",
    "                \n",
    "                ax.set_xticks([])\n",
    "                ax.set_yticks([])\n",
    "            \n",
    "            if seg is not None:\n",
    "                            \n",
    "                objids = list(set(list(np.unique(seg_g_ii)) + list(np.unique(seg_r_ii)) + list(np.unique(seg_i_ii))))\n",
    "                \n",
    "                for i, oseg in enumerate(objids):    \n",
    "                    seg_g_ii[seg_g_ii == oseg] = i\n",
    "                    seg_r_ii[seg_r_ii == oseg] = i\n",
    "                    seg_i_ii[seg_i_ii == oseg] = i\n",
    "                \n",
    "                cmap = plt.cm.inferno\n",
    "                axarr[0, 0].imshow(seg_g_ii, cmap=cmap)\n",
    "                axarr[0, 0].text(0.05, 0.05, \"SEGMAP\", color=\"white\", transform=axarr[0,0].transAxes)\n",
    "                axarr[0, 1].axis(\"off\")\n",
    "                axarr[0, 2].axis(\"off\")\n",
    "#                 axarr[0, 1].imshow(seg_r_ii, cmap=cmap)\n",
    "#                 axarr[0, 2].imshow(seg_i_ii, cmap=cmap)\n",
    "\n",
    "            ysize = int(obs_im.shape[0])    \n",
    "            xsize = int(obs_im.shape[1])\n",
    "            axarr[1, 0].imshow(obs_im[:, :ysize] / obs_im.max())\n",
    "            axarr[1, 1].imshow(obs_im[:, ysize:2*ysize] / obs_im.max())\n",
    "            axarr[1, 2].imshow(obs_im[:, 2*ysize:] / obs_im.max())\n",
    "\n",
    "            mos = mos_g + mos_r + mos_i\n",
    "\n",
    "            vmin = mos.min()\n",
    "            vmax = mos.max()\n",
    "        \n",
    "            cim = axarr[2, 0].imshow(mos[:, :ysize], vmin=vmin, vmax=vmax)\n",
    "            cim = axarr[2, 1].imshow(mos[:, ysize:2*ysize], vmin=vmin, vmax=vmax)\n",
    "            cim = axarr[2, 2].imshow(mos[:, 2*ysize:], vmin=vmin, vmax=vmax)\n",
    "            \n",
    "            position=fig.add_axes([0.9,0.125,0.01,0.24])\n",
    " \n",
    "            cb = fig.colorbar(cim, position)\n",
    "            cb.set_label(\"g + r + i flux\")\n",
    "\n",
    "            fig.text(0.4, 0.65, \"ID: {:d}\".format(mof[\"id\"][iobj]),\n",
    "                     fontsize=16)\n",
    "            \n",
    "            if fname_root is not None:\n",
    "                fig.savefig(fname_root + \"_\" + str(iobj) + \".pdf\", dpi=300, bbox_inches=\"tight\")    \n",
    "                fig.savefig(fname_root + \"_\" + str(iobj) + \".png\", dpi=300, bbox_inches=\"tight\")            \n",
    "#             return fig, axarr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Postage rendered"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {},
   "outputs": [],
   "source": [
    "mof = allmofs[[10353, 7424, 9237]]\n",
    "# mof = allmofs[[10353,]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n",
      "Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers).\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAisAAAH+CAYAAABdtU4pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XecnWd55//vPb33kTTqVrUtuUiuYBsDAYyBQAgBbEogyYYku+wGlmxCgABL8vtRNmVJsmQXQs2PGAzYhICN8boX3OQm2ZJlWV0aSVM00jRNO/fvjzM2c+b7SFM8kp7R+bxfr3lJc839lPPM8a3Lz3Od6w4xRgEAAKRVwek+AQAAgBMhWQEAAKlGsgIAAFKNZAUAAKQayQoAAEg1khUAAJBqJCsAACDVSFYAAECqkawAAIBUK5rK4BAC7W5xurTHGJtP90kAAE69KSUrWYUzfxbAhEZ2ne4zAACcHjwGAgAAqUayAgAAUo1kBQAApBrJCgAASDWSFQAAkGokKwAAINVIVgAAQKqRrAAAgFQjWQEAAKlGsgIAAFKNZAUAAKQayQoAAEg1khUAAJBqJCsAACDVSFYAAECqkawAAIBUI1kBAACpRrICAABSjWQFAACkGskKAABINZIVAACQaiQrAAAg1UhWAABAqpGsAACAVCNZAQAAqUayAgAAUo1kBQAApBrJCgAASDWSFQAAkGokKwAAINVIVgAAQKqRrAAAgFQjWQEAAKlGsgIAAFKNZAUAAKQayQoAAEg1khUAAJBqJCsAACDVSFYAAECqkawAAIBUI1kBAACpRrICAABSjWQFAACkGskKAABINZIVAACQaiQrAAAg1UhWAABAqpGsAACAVCNZAQAAqUayAgAAUo1kBQAApBrJCgAASDWSFQAAkGokKwAAINVIVgAAQKqRrAAAgFQjWQEAAKlGsgIAAFKNZAUAAKQayQoAAEg1khUAAJBqJCsAACDVSFYAAECqkawAAIBUI1kBAACpRrICAABSjWQFAACkGskKAABItaLTfQKYXYYz3572tkUFH5jBMwEA5AvurAAAgFQjWQEAAKlGsgIAAFKNZAUAAKQaBbYnSccfLpvWdo3/e/sMn8n0vZxi2snuj6JbAMBEuLMCAABSjWQFAACkGskKAABINZIVAACQaiQrAAAg1UhWAABAqpGsAACAVCNZAQAAqUZTuNMoTQ3gkiQ1bJvpRnEAAEyEOysAACDVSFYAAECqkawAAIBUI1kBAACpRoHtSZL24tnTgRWWAQDTwZ0VAACQaiQrAAAg1UhWAABAqpGsAACAVCNZAQAAqUayAgAAUo1kBQAApBrJCgAASDWSFQAAkGp0sMWUJHWhHc58e1LjAACYDu6sAACAVCNZAQAAqUayAgAAUo1kBQAApNopK7D9+ML/OKlxX9j7lZN8JphpFNMCAE4m7qwAAIBUI1kBAACpRrICAABSjWQFAACkGskKAABINZIVAACQaiQrAAAg1UhWAABAqpGsAACAVCNZAQAAqUayAgAAUo1kBQAApBrJCgAASDWSFQAAkGohxjj5wSFEqfAkng5wPCMbYowXn+6zAACcetxZAQAAqUayAgAAUo1kBQAApBrJCgAASLWi030CaXb7ZW+f9ravf/jmGTwTAADyF3dWAABAqpGsAACAVCNZAQAAqUayAgAAUo0OtqNeTjHtZFBwO3lDP5trseI376eDLQDkKe6sAACAVCNZAQAAqUayAgAAUo1kBQAApBodbE+RpAJeim6Ti2kBABiLOysAACDVSFYAAECqkawAAIBUI1nBGS+E8MEQQgwhrBgT+9ZoLIYQMiGEIyGEZ0MIXw8hvGIGjlkeQvhsCOH5EMJACOFgCOGnIYSSceP+KISwZXTM7hDCX4YQiseNWRtC+D8hhA0hhMFsc8YTHvvyEMLPQwhdIYTeEMLGEMJ1Y37+2TGvffzXsRPs9/rRMXsTfnb3cfb3kXHjCkMIHw0hbBo9t9YQws0hhPPHjfvgcfb35IleO4AzEwW2yGdtkt46+vdKSaslvU/SgyGEL8QY/3w6Ox1NNm6VdJakz0t6VlKzpNdrTAvoEMKfS/p/JP2dpJ9LulDSf5fUIuk/jNnlRZLeJOkxSQOSjptMhRDeLOlmSf8q6T2SBiWdK6lszLB/Hj3eWJWjsZ8cZ791o+d54HjHlvS0pD8YF9s57vu/lPRnyl6XOyU1SfqUpLtCCBfEGMcnQu+UNDbWe4LjAzhDkawgnw3GGB8a8/0dIYR/UvYf5Y+HEB6LMf5oGvv9mKT1ktbEGPeMib+0rxBCmaRPSPpOjPFjo+HbR++afCmE8HcxxmdG4/8SY/z26HZ/peMkKyGEaknflPSVGOPYOxr/d+y40YRg77ht36/sfPDt47ymL0l6SlKrpNcdZ0z3uOuZ5IOSvh9j/NSYYz8tabOkN0v6P+PGPxlj3DbBPgGc4XgMBIwRs4tl/amkg5I+MsHw4/mPkn4wLlEZb62kKmXvwIz1c0lB0m+MOafMJI/7TmXv4PzN5E/1JR9Q9jXfNv4HIYQrlL3j9J+msd/xSiQdHRfrGv2T+QhAIiYHYJwY46CkOyRdGkJ46e7jaM3Et060bQhhsaRFkraHEL4WQjgaQjgWQrgjhHDhmKEjo38OjtvFwOifa6dx6ldK6pR03midynAIYU8I4TMhhOOuQBpCWCjpNZK+G2McHvezYklflfQ/JnGHY91o7c9QCOHpEMLvJYz5iqT3hRDeFkKoCSEsG43tlfT9hPH3hxBGRmtb/ncIoWGCcwBwBuIxEJBst7J3ARqVveMgZROMkeNukTV/9M8/k/SopOsklSpbi3J3COH8GONuSc9Lyki6XNkakxe9+IhnOv8oz5dUoWy9yl9K2qDsI5u/kFQn6aPH2e79yv6PS9IjoD8bPf/PT3DseyV9V9LW0WP9tqR/DiG0xBj/6sVBMcZPhxAGJN2kX/3P0lZJr44xdo7ZX6ukz0l6WFK/pCtGz+WKEMIlMcbjFgIDOPPM+J2VT3ziz7Vp01N66qnH9cQTj+nSSy/VXXfdoS1bntETTzymJ554TD/4wa/+B+q9732PnnrqcW3a9JSefHKDvva1/6Pa2lpJ0l133aFdu7bn7P/mm3+k7u6unNhHPvLH6u/vUU1NzUuxq6++Wl1dHXr88Uf17LMb9elP/8VMv1Sc2cLony998ibGWBRjTLpbMNaL/031Sfr1GOMtMcabla3HKNfoo5QYY4+kb0j6cAjhuhBCXQjhNcomBSPKJjJTVaBsIe3nYox/E2O8e7Q25GuS/lMIofY42/22pCdijE+PDY5+euqTkj48UXIQY/x0jPFrMcZ7Yoz/FmN8h6QfS/pkCKFqzD7/SNmC2r9S9m7OOyV1S/pFCGH+mP3dFmP8zOj1u2s04Xmfsnec3jeFawLgDDCjycrll1+ut7zlzVq//hJdcMF6ve5112jPnuxj+/e+97e1bt3FWrfuYr3zne+WJF1zzTX66Ef/WNde+xatXXuB1q+/RA8++EvNnfurFuxdXV264oorJEm1tbVqaWmx415//XV69NHH9Pa3/0ZO/L777tf69Zfo4osv0/ve9x6tX79+Jl8uzmyLlH1E0znRwHE6Rv98IMbY92JwtH5li6R1Y8Z+TNkakX+VdFjSLZL+5+jfW6dxzi8e+/Zx8V9IKpa0ZvwGIYRLJZ2t5Lsqf6/sJ3YeGk2m6pS92xRGvy+f4HxuUDZ5Om/0WA3KFi//9WgicneM8YeS3qBsrc1/m2B/P1H200CXTDAOwBlmRpOVlpZ5am9v1+Bg9jF8R0eHWluPP+d+8pN/rj/5kz/T/v37JUmZTEbf/Oa3tHXr1pfGfO97N+q6694lSfrN33y7bropdz2dZcuWqaqqUp/61Kd1/fXXKUlfX582bHhcy5cve1mvD/lhtBfK6yQ9NL6GYxK2K/vYIqkXStCYOyYxxqMxxt+UNFfS+ZLmSPqOsh/nvX8ap/7ip4fGH/vFu0RJd2s+IGlY2YRpvHOV/cj04TFf1yv7uOmwJn40NP7u1CplHyk9OnbQ6OOfFySdM8H+XtpkkuMAnCFmNFn5xS9u16JFC/Xcc8/qf/2vf9CrXvWql3723e9+56XHQF/60hclSWvWnKvHH3/8hPu844479apXXaWCggJdd9279f3v35jz8+uvv0433PB93XfffVq9epWam5ttHw0NDbr88sv0zDPPzsCrxJkshBCU/ZjuHGXvAkxJjHFI0s8kXRVCqByz38XK9nF5NGGbthjjxhhjt7J1Je2SfjCN0//x6J9vHBe/RtIxSZvGBkeTsusk3RJjbEvY33XKPqoZ+3Xb6Pm9RtI/TnA+71E2cds4+v2LPVouHXceDZJWSNo3wf5+Q9l+MA9PMA7AGWZGC2x7e3t10UWX6qqrrtJrXvNqff/7/6qPf/wTkrKPgTZs2HDcbdeuXat/+Zdvqbq6Wp/4xKd0443ZuXpkZET33/+A3v3ud6m8vFy7du3K2e66696lt7/9txRj1E03/VjvfOdv6Stf+SdJ0lVXXanHH39UmUxGX/jCl/TssyQryFESQrh89O8V+lVTuFdI+qsY44/HDg4hDEv69iTqVj4j6RFJPwsh/I2yj0I+o+xHdF/6Bz6E8G5lC2mfk1Qv6e2S3i3pHaOJy4vjKpS9wyFlH9kohPBbo9/vjDE+Jkkxxk2jn1b6XAihQNLjyt4h+g+S/nK0Tmast4weP7G3SlLPlBDCByUNxBjvHhO7StLHlS2a3SmpVtk7Nm+V9PEYY+/o/naGEH4q6b+FEDKS7lG2gPlPlb3j8k9j9nm7pLuUTbBeLLD9E2V7vSTdBQJwBpvxTwNlMhndc889uueee7Rx40Z94AO/fdyxzzzzrNavX6+7775bmzZt0rp1F+sf/uHLKi/PfRT+ve/dqJtv/qE++9nP5cTPO+88rVy5Urffnm3GWVJSou3bd7yUrNx33/369V9/26TO+/UP32yx2y97+6S2xeQM/WzuxINOrWZJv1T2sUKvsh+ffVDSx47T3KxQYzrQHk+M8dkQwmslfVHZj+MOKfsP72/EGA+OHapsT5blyj6KeUjZT8U8MG6Xc+R3Wl78/tvKNlp70R8oe4fiPyv7eGmnpP8aY/xywql+QNmanJ9O9Jom0KrsXdrPKfsIa0jZbrbviTHeMG7su5Wt1bl+9M+jyiZVV76YdI16RtlPKS1UNpHZI+l/K5t0DQhAXpnRZGXVqlXKZDLati3bjuHCCy/Url27tHZtcsuIz3/+i/rrv/6i3va239S+fdk7wOMTFUm677779PnPf1E33PC9nPj111+nz372c/rCF774Umz79ue1ePHimXpJOAPEGL8l6VvjYh9U7j/yk9lPmHjUS2MfUfZRyYnG3CjpxhONGR23U7+q/5ho7KCyn7b51CTGTi6Tz93mgwmxbZKuneT2fcp+rPovJxg33YZ8AM5AM5qsVFVV6R/+4cuqq6vV8PCwtm17QR/60B/qhz+8Ud/97nfU398vSWpv79DrX3+Nbr31VjU3N+nWW3+qwsJCdXUd0aZNm3Tbbb+wff/N3/ytxa677l269tq35MRuvvnfdN1179bDDz8yky8NAACcJiHbXXySg0OIk7gLfsY42Y+Bkh49nclezmOg4jfv3xBjvHgGTwcAMEvQbh8AAKQa7fZPkjPhrsn68vdMatzDP7zjJJ8JACCfcWcFAACkGskKAABINZIVAACQaiQrAAAg1ab60eU2SbsmHAjMvCUxRl/4CQBwxptSsgIAAHCq8RgIAACkGskKAABINZIVAACQaiQrAAAg1abUbr+soipW1TXmxMKwr1xfEXy3PZlhi5UXl1lspOCYxeLQoJ/MiB+jrNiLhY8VeCyMlPq5JNQZ7/FhWjjSa7Gh6K9juGAg5/vi7hIbk9GQxQariy02oBGLFQz4669KeK39hRmLVSYtRjnieWtf9N/ZsJ+elHB+NdHPb8gvgTKH/f0zVOTnd7hze/uJPg1UUVEda2sbj/dj4KQ4cqRDfX3d/iYexZzJnJnWOXO2mVKyUlXXqDf//idzYiUdfuHWFfo/Gvf3d1ns/PkrLHak8jmLDbbusVjBkSaLnTPX38hbyvzNU3T0LIudlzCBfHSFx/77kYcttn/oXIt1lL2Q8/2C+xbZmKPxgO/rqnkW2y6/dhW7/BpfXuIT1MY6j70y1lgsc8QnjydH2i12aEHCDJXxyeh1Q35+hxb6JNBzs78FD9bVWuz7N7z7hB+Zr61t1O/8zqdPNASYcd/85udO+HPmTObMtM6Zsw2PgQAAQKqRrAAAgFQjWQEAAKk2tZqVWKgrR6pyYk9mvLirt7zNYk1zqyxW3r/bYjt3dvu2w/48bneZPy9cXtVgsU0JFV/zKub4MY76c9XfOuDjnt9+ucX2Vu+z2EULc8/v0Cp/DvpCf4XF1vb4619UtdRiXx88aLG31HiR1TndLRarCYcs9kD0Z9f3jngR2NmbveKrpckryHrmeB5c3Orjii8asFj/Qn+eqxs8BKQdc+bJnzPXMGfmxZzJnRUAAJBqJCsAACDVSFYAAECqkawAAIBUm1KB7fBQUEdrbrHQihpveNPVU2mx55q8qc5QQnOkXy7yoqpV1d6k6PJnvMhoa7E3TDpa8oLF3lDiBW5Fy+ssVt7nBVm7qjzWfKU3Wyr8vwtzvu9b6uc7V76v5/q9GGvxoDdCWnzM88zH1WexJdU+bkvjAouVPnTYYisu8Ov0dvl16hucb7HV/b7tnV3+Hmir8qZHn93ojZV+YhEg/ZgzT/6cuTVhzlzEnHnGzZncWQEAAKlGsgIAAFKNZAUAAKQayQoAAEi1KRXYDhZG7avOLea6dEm5jdtf4sVTryj3VTDrirzbX8lhHzeYUED1XJHHtj25yWJnVXiHwo29my22osJXFb162Iug7ur3oqr2Tf0Wu39B7jW4rNOvU2OBv4Y5wa/djvCExd6y3jtU1j7sq5S2LvfXX9Llq7SWlnhBXv9BP+cD53oR2LH25y3WUL7aYs+WHrHYC11HLfZIix8XmI2YM5kzmTNnBndWAABAqpGsAACAVCNZAQAAqUayAgAAUm1KBbaVBVGXVuYWi9212TvnzW/w5bRbM97Fr7Fpv8VWHqu32CJ5YdRTc7dbbEmnhbSp1rdd1H+Bxdq8Rk3/ctCLwHY0e4Hb+3cusdhljbnX4O7g1+l1e30J+P2L/TrtPZhwvg2+LPqK4h6LFR7wpcN7Gr0grbPai7ZeccSXJ3/mF75E+6HLvIDswLZWi62u9tfbHH3c4i4v0gNmI+bMUzBnLkqYM9uZM8803FkBAACpRrICAABSjWQFAACkGskKAABItSkV2A6EoG0luQVE9Su8sKfxkT0W6z7Xl9jelbnIYiHu83Ejvoz53IN+jM1D3o2wosqLkUrnWEh7+8os1lm1zWINRxst1jvfL+PZVblFVXuLSm3MHWXe2bEjYXn2WO6FXAv6/bqXrvTcM2z25ePPWlFtsd7nfKn0FaHZYqtXegHZ18u8IK2t1ZcxL23zQsCDV3hs8zEvrANmI+bM6c+Ze17OnFnBnHmm4c4KAABINZIVAACQaiQrAAAg1UhWAABAqk2pwDaUBpUuyy2qevjpDhvXssY7Cg4Pe/fA54e9WGrFfo/tyfzIYo0t3gGxf4G3Y5zb60tnxw4vlmo4tMNij8e1Flu8xCvNyl/w5b4/1ZDb8fCC87yLYUujF15dmfHrdGubL4u+p8i3XfbCLou1LvKCvJFCL8hTzVwLbYneebK+2wvX5m/34q51q7zjY2n7Rt/fAf9dFBX7uQCz0emaM3czZ1qMOXN2484KAABINZIVAACQaiQrAAAg1UhWAABAqk2tg+3wiF44mLsueFNbtHHlld6J79oiL6i6c593aFwxz7sitgx7B8QtdV5UdsGCcyw2+OB9Fts7Eiy2aMF6i11d6h0K24d8KfebaudZbG5FbifHVSVtfh5HrrFYb6EX383pL7TYvnrvRlk44oVsG7sSujtmPEcdqPZYR78Xcq1s2mSx18m7NvY957+zTIm/3c4b8IK0f+pbaDFgNmLOlNoS5swfneQ5s5k584zDnRUAAJBqJCsAACDVSFYAAECqkawAAIBUm1KBbU9Bie6ryF1S/EM13olv24MDFtt4/mqLDdQndG2s90Kusmcf8XMZOWaxPQNekHak3wvX+td48dW2+iGL1WvYYq/e7QVku+u9Q+FQe24RVF2nv9bifu9O2J7xAq1Dc/x8a3Z6QdW2Yi+ga6rwArLKat9fc6W/hpFB/922bltssZ8ss5DeONeXWb+z2gsLDyYUrq3r226xu/wQQOoxZ0pXJ8yZe+oS5syO3OMmzpl9zJn5OmdyZwUAAKQayQoAAEg1khUAAJBqJCsAACDVplRgWzvUrze15Xbje2HAC4A2rvGCr8a2HovNbfBtS6q2WuxAkXf7qxqstdjWRbsttr3Ml85+x4AvFd66d7/FGoa90Ky3MaFbYnuLxUoW53aV3LnXz3d4uMti9fIiuJ5yP9+zohdjdbX4dV894gVkw0mFdiP+Vrhs0Avj5kRfPv5wqce2lPu1qxr0IsLlhZ4v9xZ5kRowG+XbnFmfMGf2NSTMmR1Jc2Zpzvc7EubMkcnOmRXMmWca7qwAAIBUI1kBAACpRrICAABSjWQFAACk2pQKbEuHS7W8fXnuDoIXMsX53mWwpMWXtV52V6+PK1tisa1LvKhsSbcXd/U+3mSxN8kLmaqrd1js4Fle3PTUC75kd32XF4YVDfpy58+tyD1u40Ff7nxrwrLwc7uqLFZ7sNRiXQ39FlOnd3J8em6Fxa4t88K99iGPPdVfY7GCCn8dVfsrLRYXenfLLQcaLNZU7ce9quw8i33FIkD6pX3O7JnhOfPphDmzIWHOLJzmnLkxYc6cx5yZF3Mmd1YAAECqkawAAIBUI1kBAACpRrICAABSbUoFtgXFhSpryS1maklYYnzTsRKLvX2Px3651A+/Y493Cix7pRc8bTriy3OXL/Tis/OHzrZY94gvAf6mJ7z46pZCL+bqLFxksdqMF8IN7Ms9v33zvGjt97XSYg83tFqsu9qvSdEBL1BbXe3XZGH/Jott2ufb9i7w30X7/E6LLWnwa1c27MunP7nTOyr+xVx/vf9fb0JBWsKy6Jge7885OfwGZsZMz5kPnoI587yEObNneJJzZsHk5syaER/Xv7c75/u9c33O/FBYZbGH6n3O7Knxa1LYypw5m3FnBQAApBrJCgAASDWSFQAAkGokKwAAINWmVGDbNTKgn3bmdjJcO+zFTVUHvIvhUwnLhDfUe5FVQa2fUuEOLwJbMt8Lj2476AVKzyUssT3fG0jq9grvgni418sTC6IvUf50aZnFVu7Zl/P9pQ11NmbffN+uYJ8X65437NvecbTeYpcf8m37L/VYZY8XvFX90i9K+Vpfen6k3K/J0S7vFlnb7MVx+3q9k+MF+7xr4x0tfl0wsaRi2hC88G78uEmX5kU/wplf1vfyTHrObE2YMxt9zmxMmDMLZ3jO3Do4aLGkOfMX5T5ndvYlvAszPmc+Uerzw8rd4+bMxlobsydxzhy22NqRyc2Zr2jz+fEYc2YqcWcFAACkGskKAABINZIVAACQaiQrAAAg1aZUYFtUMKSGygM5se42z3d6q/ssdqTMlyJvOuqFRw/M8UKmH4x4Qdr/6vJlspv2eLlfrPJYx5AXKJ1X5t0i/235PostLptrsXdv846CD+3PLeZ6OByxMc1NXqB1Vp13VLzzef81nVXxlO/vCi+qu23LuT5uoV+7gWKPnVPuBWm7R/z81tZ7cVfYlbAEfKW/3l9bc9BiPdUHLHaHRfLbZItpCxMGFhbkjosJVbIjCcW0I0nltAnj8CuTnjNr0jRnehFv0px5fsKc+eNlPmcuTJgz37nV58wH9jbkfH+/fM5safBjLq/rtthdSXNmJXPmbMadFQAAkGokKwAAINVIVgAAQKqRrAAAgFSbUoHtsRi0eSC38941zd7F8IVe78S3rt+X7G4v9qKy9qOHLfbtUi8y+vfeZotduOQJi80v9KXSu/q8qOzpNl+y++3b1lhs0yrviHuo0rsWzlmXe13mrfaCsvIuL4rq27nUYv3F8y3WcPEhiz29faPFjpV698h5T3gny561XsxWL++qubDOtx0e8IKvpWv8dzuU8aXntw77MvM72umLOh0FCbWuxQV+LcuLJi6w7fcmnoojfgAvxaSr7Vhnypx5eJJz5m8kzJlPr/SOuHvLn7dY5XnVOd+3rPYOtvXdCcWlzy+xWF+Jz5mNF/mc+dQOnzP7mTNTiTsrAAAg1UhWAABAqpGsAACAVCNZAQAAqTalAtvakSK9uTu3y2DTiBcZvXKud/Fb2+WdDb816MVD72j3ZdGfy3hB0X9t8GXB+0ZaLPaLVi/ual7qRbKFfV6Qta1st8VqM170tbt/scWqL8o9xvBBL4B6crlXMZ6337sxXtzm1+Qf7/Nxy5t8WfRzmv21Hpnry8Ifkr/+OV3eBfPZBV4c+JaihRZbnfECulsyP7HYgfuXW+zK+gUWu8Ei+c7fT0m9ZEsKfVx10cTFeEmFs8OZhALbM7+u72WZnXOmd6ttWurFvgW9k5sza0Z8Xnqh14tiR87LnatL9/qba+s5Pme+4oB3kr24o9Ni/3h/wpzZOMNz5mHmzJOFOysAACDVSFYAAECqkawAAIBUm1LNynBhVFtd7jPDgmf9edyOwjm+cbE/Q12Y0NxGS7050hu3+TPUb3R4k7UV5f7s9hx/NKqWCl/x8paMPwstqPRt5/sinSpc5atq/s6G3Dzw4Bp/5rlsk9e63DHir6Fqvjedu7y42mKHO/w56FXBn91+Yb+F9LYl/jx3zVJvUnRsvtcArd7mtQx7oz8L3lJ2lcXeWukns7thh58gxkmqUJlcHYuPSdrOtwwJO0ta6TmyEvNLTteceU3CnPnNSc+Z/jtNmjN/ljBnhoQ5c07CnFmQMGf+9obcJnMH1vqcqY2+wvTDIz6uNnHO9Ove2elz5pUJc+YXJztnnsWcebJwZwUAAKQayQoAAEg1khUAAJBqJCsAACDVplRgWzA4pOo7eWJoAAAgAElEQVRduY2K2pd5wdOB/l0W2922xWLvbPLiplsPzbPYQ4VetbWizFcsPlTnRWAF23xF0sOVTRYbOstX2nx1vxc8lZX6yqWdG3wFzfvenFtA1TLgK5Qe2uoNhN50wAtn7x30821Z84jFzm/2Yqwle73Z1O8ffsZiD9dfYLGbO9ZbrLnHC7luWOarme7bvdRi9Ye9cO2x/b46bNsWb16FiSX1ZxtIaOTWbZfXxwxOsgEcxbQnNhvnzJAwZ3ZWJMyZy3zOvLrf55uSEp8zW5/wOePJN+Z+3zLg/y/dtcuLS9/c5oWzDwz76sfzEufMV1lsKXNmKnFnBQAApBrJCgAASDWSFQAAkGokKwAAINWmVGA7Ulqko8tzC5eWZ7x4quWAd/HrrvfVQvct9Fwp80yDxc5Z+ILFauu8LeJgxotYFwx5d8fOWl9T9vxWP5ejvV7s2ldxv8VWtngh7t0v5BZpHSzwbrWdBV48VXWBVzEu2u8rPT/23GUW++Vi7265scI7WQ4X+jU5XOgFWqV7vSBtzYi/Zc7f5yu3HlnhBYN3Fe212KH5Xmw4sPvMLxY7GZIKYAdHvAB2OGlJZdsXKyzPhNk4Zy6c5Jx5QcKc2d3rxa6F5fdZbO1CL2K9Y1durK/Qi4m7in3ObLrQz23xJOfMhxb7dd9Y4XMcc+bpx50VAACQaiQrAAAg1UhWAABAqpGsAACAVJtSge1AQUYvlOUWcbYOe3Hpa+q8aKm3xDsqbnhswGKXnNNmsS3dvhT59zJeoLS6w4+7qNW7yy6u82KpW5u9mO3cfb58+vpjXtz0dJMXqe3anFugOK/ECxZ3HfVitCM9z1usboV3Y6wtWGCxt57l3R07n/duh8MrKyx22W4/7tFGL9h9YcVSi7W0+XLsm3q8M+ZIlRe9Ldvnv9unVpVaDLkSa10TimJHEgYmNKf1XSUewDek5vbE0jRn3vBy5szaSc6Ze33OvKjf58ynmvyfnn1bc99fi0psiPYe9WvS0ze5ObMuoWD3rUv3WYw5M524swIAAFKNZAUAAKQayQoAAEg1khUAAJBqUyqwLR0u0NLO3KLQ53p9+esnVvm29UXHLFZVttRPaJUvT77+oSW+v93ejfHfwhyLvXGJFzz1bPNipIo5XtzUvuZRi/1LwrjlrV5A9o7FuedS2O4dBjcUenfZ8oXnWezn+49abGnwTo4tXYMW21Pv1/2C7Z6jdnZ7h93Y4MVddQf9XJ7u8NcW670QrrbfO02Wd/u5lBf4dcHEJlvsGhMKcXFynDFz5guTmzPb1k5/znzn4rk53yfNmU9Mcs68LXHO9OLfliMJc2aDFzEzZ55+3FkBAACpRrICAABSjWQFAACkGskKAABItSkV2BYVlam5eXVObPCYL2tdU+2FR3P6vbjpw2Xeie/zbV6MteDoEYvtG/T9XVLky47fUuKdB5c2e6HZG/q6LPZ47xsstmT3sxbbVO3nrJqdOd+eNeJ5YVNjtcXObttpsfLF/rr6u+dZbGd9k8Va2r1Ib8+Qn+/aRu9kubj8gMUOdXvHx+aFXsy3c8SL9O7r9HOZs8yLymoO+3UBZqM0zZl7E+bMixPmzFsT5swlTT5nvn6Sc+bShDnzmWrviDt+zlyW8TmzscnnhnMS5swK5swzDndWAABAqpGsAACAVCNZAQAAqUayAgAAUm1KBbZDg4Nq3bMnJ3bWnmU2LjZ44dXaQi8o2rfAi8VK5IVM91+9wWLL7vbz21XpXV3rB7yoqrTAl+d+ZNCP29Dkr2NOOMtiT3R6YVTBuP19e7DBxqzc7kVmey/2/LGl2DsqPtbphVzLMz5uRYEvd154vq+9vm33dov1NXiRWji20mJlXd5Bc/8hf21zypot1tLQbbGKPu/4CMxGZ8qcWZYwZz6aMGfWv4w5s3C6c+ZFzJn5MGdyZwUAAKQayQoAAEg1khUAAJBqJCsAACDVplRgqwJJpblLWz/V6AVaWwq9CKxv3n6L1Rd6IdNlz3vx0PpaXz/9SLV3gWwa8m1Hoh/j8WpfFv3NNcFidxz0AqpXF3h+98VVvtz30xW5XSs7d3jx1OE13lGyrcH3nznkxVirxy07L0kFc73g7bng2x4+5NfklduXW+zAMS9Im1PrhVyPlfhr65nrnSHnzyu0WMfWhCK6pqctBsxKzJlqPtlzZuP058zAnDlrcGcFAACkGskKAABINZIVAACQaiQrAAAg1aZUYFs1HHXl4UxO7HsJBaGvLmizWEmvFyjdWn+pxX59ZIfFNkXvnthd5oVRnS8kxFb6+b3Wa7vUcY4XPJ3T9Dofd7efy43LvDiutDLmfD/4a3famMw9XqDV4Suxq3FXtNjSdb4k+LOx1mKdz3ux2K7oseU1Cy325JB30Oxrfd5i66r9bVTa7J0sd3V718qSZl8CvdXrAIFZiTmTOZM5c2ZwZwUAAKQayQoAAEg1khUAAJBqJCsAACDVplRgmymU+mpyC6P2BS+yWlTiRVtlbd7t8KJe77rXHX1/c6t9mexNpd4Fckmt517n6JDFFhd4B8AXNlVa7A19PRY7eq1v29rbarFnHs7tZNhxyTk2pnmx72tdQufFoaOXWGyg/fsW27fyAos9cOFci53b4QVaA93eZbG2zrsnLt7vv8f2Vr92a2r8d9F7wPfX0uBLtD+w2peoB2Yj5sx0z5l7E+bMB5kzU4k7KwAAINVIVgAAQKqRrAAAgFQjWQEAAKk2pQLbkYGgI9tyi7nqanfauLaEDo3FjV6MtHeHd+y7otSXNi/b7kVGS0u8yKh2jhdGNfb4S3w0tFts2QV+zhtrvMDrlfu9VeCDlfMttvKq3CK11XtLbExbhRfV9bV4J8uFe7ZYbNdmv07XXLzNYn/0Y++ouLnBu0d2V/r1nLvUi7YWBb/GtUu9wO2WR5darGV+n8We3O0dKYuryKFxZmDOTPec+caEOfM/Mmem0pn/CgEAwKxGsgIAAFKNZAUAAKQayQoAAEi1KRXYFpYE1S3N3eR3vf5Hdz7tOdCyuQssNqeqymLP9HkRVPu6zRY7u907AA5vqbDY0Uy3xc6a6+udrxzwLoNl83157s4GL25aVurjlnXX53x/sMvHPHv0gMWaNpRa7Bfy7omXrJ1jsb4n/Nyeas5YbMcC31/BRu+CufKod5n84UIv+nvjU36M2hVefNaaUJD2tvP9ui8N+yx2o0WA9GPOPM6cWZIwZ/YwZ47HnPkr3FkBAACpRrICAABSjWQFAACkGskKAABItal1sC2K6qrPLTR6JmF58rPXdlhsw2O+tHfNMj98X793Hmy82wvDzp7rnRwPl//SYnU96yz2i0XeGbGjvNFir+8dtNjKhiaLhaMHLVbQMZTzfWzy5dkH93sh29HqFRarWeIdII+EnRbramix2N5Sfw2Ln00oqvMmizpa5NfpzZ1e4Pa1wnMt9qZWz4MvO8eP2zpSa7FFRV58BsxGMz1nVp+CObM2Yc68nTnTj8uceUpxZwUAAKQayQoAAEg1khUAAJBqJCsAACDVplRg219Qoqcrl+bEBvcesnE7no4WW3e2L7H9bGmdxZoPeaFZbbEXKJUXn22x52sspPuLei1Wfdi7EVb2e1HV7XXe3XDVRu9G+LYWf71fqX0q5/udO4ZszEhZvcXuWLffYu/d7q/huTXejbF7x1MWay7xa7z7wjUWKxl6xGJbj/lS6XVVZ1nsXYu9CKz5mLfp3NLg12lO3RGLLWsrthgwGzFnJs+Zb02YM/+JOdNizJm/wp0VAACQaiQrAAAg1UhWAABAqpGsAACAVJtSgW3BcL8qD23MiZUXeze9knav2iqM2yzW94wXS+1oXGSxKm/4qAd3b7LYe5r9XO4t8sKwZa1eaFVU5R0V19d4YdTBhb5s+UCnL9nd35H7OjbstCFqrPNCu8vlS7E/uNav59aNFtLF3d5RsmZxQj7a6wV5W5Z78d1rvQmknm73jo81Jf67LVvmBV9znvSumh21fg3ipd5BE5iNmDOZM5kzZwZ3VgAAQKqRrAAAgFQjWQEAAKlGsgIAAFJtagW2RUFVTYU5satKvcvibQVePLWz63yLtSxqt9hFPd618FZv9qeBBl8WfE5xlcXqdnr3xP/Z4kuv/5cFvnz67qeetNjhLn9tbT1eBDX3WG4HxbcNenfCTKcXtw3M8/zxxwN+bu+t866ND8314rtX9fjvZ6jHl16/YJcXxh0Y9LfHtUf89R85xwvIBna1Wey5wlaLra/y4sDQ+rTFgNmIOZM5kzlzZnBnBQAApBrJCgAASDWSFQAAkGokKwAAINVCjL4E9XEHh9AmKaFHH3DSLYkxNh/vh7w3cZrwvkRanfC9OdtMKVkBAAA41XgMBAAAUo1kBQAApBrJCgAASDWSFQAAkGpTardfXFoZSytz2y6HjBfoDlV4e+HinoRxNT4uDPtxi45Nrgh4uNz3V+CdqBUTUrRMicdKujIWG6pK2NgPq4LBcUMSXkLBoO9/pMz3Xzjg45Swv+HyhG2HJvf7CQmHKOpLOEjCa016bUOVPjAmbFtyNOHACeO6e/a3n6iyvai8MpZUe0tw4GQa7O7UcH9vwjs2izmTOTOtc+ZsM6VkpbSyQee94SO5O+j3C9d2vu+25ZcDFtvzBn+3lx/0q16/1f/riQlraXSs8eNWtCa88RLeFL0LfdzSn/raFK1X+JoWSW+Uyv251yVp8qjYd8xih8/2RT3qtvu4MOTXvfNc37bygK9z0XahX6dC//Wo+QkPZor9P+6iY36M1svLLJY04S28q9fHFfrAO+/95Ak//llS3aDV7/joiYYAM+65H/3dCX/OnMmcmdY5c7bhMRAAAEg1khUAAJBqU3oMBAAAZqdrXlMZOzr9EZQkbXh64LYY4xtP8SlN2pSSleFyqf383IeNcx73h48l3b5t5zmlfvAeH9ew2Z+1Dlf4DaDiHr/gg7X+jHPeQ/4MsWuln0vjJt/28OqE55mt/tyzat+gxQ6ty33+WNbh+69M6B6cVLS29+pyiy35fx+x2LFXXWqx4Qr/FRckFOTVb/Vg6YObLRaWLrRY/+JaH5dUfOaPkROfNydtq3sTYkDKMWcyZ6ZpzmzvHNaDP1+QMFgqm7+jKfEHKcGdFQAA8kCUlEn6WNQsQLICAEAeiIoaismPgdKOZAUAgDzBnRUAAJBaUdKQkgpc0m9KyUrBkFSxP7c4rGtZoY0rOeqZW3FCZ7/inoSujUe88KpvjhcU9TckdB48lrC/w/2+bZM33+md79vWbfVfas0Ob3q0+w1VftxxhXBD1UlNlbwIrHqvF20VDiQ0JGr2WqiGLb5te0LTp1hsIRUOJDSCuuxsi+16oxfaJXZjbPCiv7qn/MBJhWFtlyfcpvymh4C0Y85kzkzTnBklDSUUKU8khLBI0nckzZOUkfTVGOOXR3/2nyV9WNKwpJ/FGP90ygeYBO6sAACQB6KiRqb3GGhY0sdijI+HEKolbQgh3C5prqS3STo/xjgQQpgzg6ebg2QFAIA8EKOUsPTRJLaLrZJaR//eHULYLGmBpN+X9IUY48Dozw7N3NnmooMtAAB5ICpoKCZ/SWoKITw25utDSfsIISyVtE7Sw5JWSboqhPBwCOGeEMIlJ+vcubMCAECeGElaRTKrPcZ48Ym2DSFUSfqRpI/EGI+GEIok1Uu6XNIlkm4MISyLcRqFMROYUrISMlJJd+45dF3gBUpFXb7bTEKBUtJS6UUDXkBVMOzjyg57ldHcn++z2OBZ/ghtyU86LLbvDY0WO+TNDVXW6e0S5zyecA36cgueuhf7dgPVfmOre5EX3zU/4W0Md79vmcWSlpRPWhm09JCPS1p6veSAt9UsHPBisYpWP8bhBo9VtPkxah8/aLGuVfN8Y2AWYs5kzkzTnJktsJ3eA5UQQrGyicp3Y4w3jYb3SrppNDl5JISQkdQkqW1aBzkBHgMBAJAHMgoaVGHi14mEEIKkr0vaHGP82zE/+rGk146OWSWpRFL7yTh3HgMBAJAnMvG4j4FO5ApJ75e0MYTw5GjsE5K+IekbIYRNkgYlfeBkPAKSSFYAAMgLUUGD8cR3URK3i/F+6bjFLu97WSc1SSQrAADkgWwH26knK2kwpWRlpEzqGtegb9n3vQDoyHLftnDAk7LuRQmxxb5t74KEIqOtCV0GC3zp64KED5UfWeaVTKWdPq7l/l6LFXZ4AdX+a+dbLFOUe34tDxy1MQdeUWOxYw0JBXQ93qGy7gWvvisc8Os0XOZlSZliv+4Fg94BsWd1vcWW/NTXqB+s9wKymp1+jJ6FHhtc5MeIpNA4QzBnMmemac6MMWhkmgW2pxv/LAAAkAfy5s4KAACYrbizAgAAUizbZ4U7KwAAIKWy7fZn5z/7U+tgOyyVHM4t+Dl0sRcK1W73wqPuBZ7NJS27XXrQi6UyhQlLmycsz13U78VSx+r9JfbP9aKlyn2+bUGfF2llqr1b5FC1hdTXkru/lgd8TN12XxK8ZrefWyzwWH+9X5OhGr/GNbv8d9GxxvdXtcuvZ9cyv3atl/vS7gvu9W6UlZt9PatM6VyLDdb4MSr2T6sPAJA6zJnMmWmaM6OkkVnaC3Z2plgAAGBKsndWeAwEAABSKkaSFQAAkGJRmrWfBpqdZw0AAKbkxcdASV8nEkJYFEK4K4SwOYTwTAjhj8f9/E9CCDGE0HSyzn1Kd1YyJVLv4tzio+U3+nra+6+qsFh5mxcj1SYUMpV0eeHR3J/6Mub73+EtHzOFXmQ0UuqxIm8oqJqdvqS49vg63j2/do7Fard7odmS7+3P+b7tau8U2bPYz22g3vc1/z6/nqVHE5Yn7/H9HV3sb8ICr1FTf0uZxZJ+Z8XeoFKlnQnvgWv99SYtUT9UkfA781MBZqV8mzPj7v0W63nduRarfcFfx5Lv5c63zJkzP2dmP7o8rQcqw5I+FmN8PIRQLWlDCOH2GOOzIYRFkl4vafd0djxZ3FkBACAvBI0c5+tEYoytMcbHR//eLWmzpBezq7+T9KfK5kInDTUrAADkgQmawjWFEB4b8/1XY4xfHT8ohLBU0jpJD4cQ3ippX4zxqRBObssJkhUAAPJAjEFDmeP+s98eY7z4RNuHEKok/UjSR5R9NPRJSW+Y0ZM8Dh4DAQCQB6KkjELi10RCCMXKJirfjTHeJGm5pLMkPRVC2ClpoaTHQwjzTsa5T+nOSmG/1PB0bn4TEp5SJdXvdC/1WOlRz5UOrfdKodKzllmsar8XaO1+s+9vyU+9+KzhKa8W2/8aX3a7dMUai5V3+HGrdvgS6LG0JOf7Ya/3UoE3e1RZm7+G3rl+kRu2eIFW1wrvjDn/zk6LjVT5uFic1PHSf5FDA34Lcc/rvENjWbufc1JhWO8Cj81/wF/bZosA6Zdvc2bZ8rUe60yYM7czZ453KubMqKChzNT7rITsM56vS9ocY/xbSYoxbpQ0Z8yYnZIujjG2T/kAk8CdFQAA8sB0P7os6QpJ75f02hDCk6Nfbzr5Z/wr1KwAAJAnMtO4RxFjvF868bOiGOPSaZ7SpJCsAACQB2KUhjKz84EKyQoAAHkgKigzS9vtTylZKRiOqmjPLZY6eIlXQS28s89iR5b5MuEDtX5XqWeJd+yr8maMqn7usMVaihssVrH5gMUGF3tH4IqDftyREj+/rpV+yQZraixWtSe34CmpK2RplxdUFSY00g1+ahop82eMVfu8MK57Za3F+pv8zTpQ5+dXfsjPr26bn2Bxt1+Ttgt9LfumTQmFe5u9MKx3ob9XgNmIOfN4c6bPS8yZJ3/OzPZZyYNkBQAAzFZBw6y6DAAA0ipGaSSe3E6zJwvJCgAAeSAqaHgafVbSgGQFAIA8ECUN50PNSiwMGqjOzcoaN3mxT1Gnr4nd2NVvsa7zvANi5T6/kLHAq6X6lngRVHGPd0pUge+va6V3fKxs9UKmkXLfdqQsodCqxm+r9V2UW/BUvXdyS7sP1PuvJKnQrKfFxxUNeHHXsXo/34JhH1e532ODCcV8R5b7teuf4+MqDk5uAc7DZ3uxYeVBvy7AbMScKY2Uemyw2ueMQ+vHzZl7/NxKE+bMYw3MmVOZM/Pi00AAAGB2ijHkx50VAAAwO0VJw7O0KdzsPGsAADBlmRgSv04khLAohHBXCGFzCOGZEMIfj8b/RwhhSwjh6RDCzSGEupN13iQrAADkgajsY6CkrwkMS/pYjPEcSZdL+k8hhHMl3S5pbYzxfElbJf35yTr3KT0GyhRLPQtzM7Duxb50ds2uRt+2yDO3kh4vAktaFjxp+aRjjf7xq5rtXpB2ZP08ixUdSyiMqvH91W7uslh/gxe4Zcr8BEfG1VRlChO6Mbb7+YbonQh753lnw4KEuriQECs94te4/UIfV/+Mn1/tjiE/7oDvL4YSi41//ZJUvteLCPuavfvv7msS/sP5qYeQaybbJ4TJ1fphAqmaMxsS5swdp2DOrPc5c2QSc2ZMmDNLOvx8JebMyc6ZMU7vMVCMsVVS6+jfu0MImyUtiDH+YsywhyT91lT2G0IoiTEOTmYsd1YAAMgTJ3gM1BRCeGzM14eStg8hLJW0TtLD4370u5JuPd5xQwh3j2774veXSnp0sudNgS0AAHkgKmjk+HdW2mOMF59o+xBClaQfSfpIjPHomPgnlX1U9N0TbP55ST8PIfy9pAWSrpX0O5M9d5IVAADyQIzTbwoXQihWNlH5bozxpjHxD0h6i6RfizEe9wFyjPG2EMIfKlvn0i5pXYzRV808DpIVAADyRJxGcVsIIUj6uqTNMca/HRN/o6Q/k3R1jNGXDs/dx19IepekV0k6X9LdIYSPxRh/NplzmFqBbUlU36LcTnkFAwlZ2i4PHVnhsepdvu28h7xA6fBKL5ZqudeXOx+q90KrtnV+jNJO/2UlLSleuX9yy24XDHkyWbMzN5aUzG75gyqLNTzpRWtJ762k8+1Z4Aep3uMDF9ydVKTn2xYf9a6I7RckLFuf8GG1JT/ptNiWj1ZabN4vEjpD7p2da1ecSid7LbLJ7p9C3BM7JXPmL70+8fAqL+A8JXPmvoQq0QTTnjM/lDBnPjXJ+SLhvZp/c+YJHwOdyBWS3i9pYwjhydHYJyT9vaRSSbdn8xk9FGP8w+Pso0nSpTHGfkm/DCH8XNI/S5r5ZAUAAMxOUZqwp0ridjHer8TPmOmWKezjj8d9v0vS6ye7PckKAAD5IEojJ/u27HGEEO5Swv2tGONrJ7M9yQoAAHlggk8DnWx/MubvZZLeoewniCaFZAUAgDxx/M/rnOzjxg3jQg+EEO6Z7PZTSlZKDkvLfpTb8m+wxm8pFfd6W8AYvEi2d35CF8NSH1fW6cVNu97qXRGX3tRhsZoXvFvkgG+qum2e4BX2eeFaRbufX2m7jwsjue+I/a/yQqnChNrp4h5/J3Wt8ky46Wm/xkPVXmQ1UJOwPHtVwvLkbX6NB+v8tRYmdLIsa7eQepb7cvQVzycVgflx5/3Su1Q+m7AlJiHhjm8ME98GDkkzWlIoqfibotuXnJo504tpyztO05zZ7x+QqGj38ytrG7BYwVDuOe+7utr3nzBnlnT7G+7waubMpDkzRilzmu6shBAaxnxbIOkiSd4u+Ti4swIAQJ4YyZyemhVJG5T9X56g7OOfHZJ+b7Ibk6wAAJAnptNnZWaOG896OduTrAAAkAeiwrQ+uvxyhBB+80Q/H9sN90RIVgAAyAdRiqf+MdCvn+BnUdJJSFYKpJGy3OKckoSOfYX9HiupTCgU2u+FPiW9Xjw09B+8CKz0ljm+vwEvdC3u9+Km0qN+jCNn+aUoO+Tn19Pir6N9ja/RvvSmtpzv599nQ3TwEi+67Z3v45b+5IjF+hb6tj3LvIBsyb/7a03qCJzULbL9PL8m5Yf8eh5Z7bGiYwldMA/7uANX+/nVbE7oHHyvh/LFpP9HKKmYttCDmXFv4cSC2JGkjqWTK7rFGHk2Z5Yf9PPrneuvo/Nsn7+WfG9PzvcL7vBjHnylF6H2tlhIS3+cMGcuTpgzz0qYM396Zs+Zp+Ex0JMxxi+HEK4cbS43LaftA9cAAODUiZIymZD4dRK9uLLy37+cnfAYCACAfDDNx0AhhEWSvqPsR40zkr46erekQdL3JS2VtFPSu2KM4xeh2hxC2CmpOYTw9NjdSooxxvMncw4kKwAA5IvpPbodlvSxGOPjIYRqSRtCCLdL+qCkO2KMXwghfFzSx5VdhflXh4vx+hDCPEm3SXrrdE+bZAUAgLwQpnVnJcbYKql19O/dIYTNkhZIepukV48O+7akuzUuWRnd5oCkC6Z1yqOmlKwUDIyoYlvuHZ7W13nRVsNm33ao0i9QeacXlVXsPGqx3ff6MRY81mOx/uWNFute6GU55e2eWjY854VmB16Z0EExoRth9R6PtV3elPN919k2RMX+UlXqdWEaaPYC3v1XetFa42O+bVGvv66iY/5r72v2/R2b44VcBZf4SWfavXCt8zzf30id/76L2rxwrWq/F71hYrEgoZjWL6+Gy3K/TyqwLTzmsYLBhKLbhE63dLX9lTNmzmyb5Jz5Cp8zixIKdmu3+3/j3RfmVsp2nOvzVKEfUkV9vv+BOV5wmjhnjm8ArzN8zjzxY6CmEMLYf0W+GmP86vhBIYSlktZJeljS3NFERjHG1hBCQhX3zODOCgAAeeO4yUp7jPHiE24ZQpWkH0n6SIzxaJjE0h0zhU8DAQCQLzLH+ZpACKFY2UTlu2MauR0MIbSM/rxF0qGTcMaSuLMCAEB+iJpC46ZfCdlbKF+XtDnG+LdjfvQTSR+Q9IXRP/8tYdv/GWP8SAjh35VQ3htjnFTRLckKAAB5Ik7iLkqCKyS9X9LGEMKTo7FPKJuk3BhC+D1JuyW9M2Hbfxn986+ndeRRU0pWhqqKdOjK5pxYUgHf/lf5EuNJSp5M6BR4vq9F3vKAL3+9+5oqixUkFOjd5MUAABOxSURBVF8tut2Lm46u8G0H6vxSVO/1oqWkDrbFfT6uc3XuuGKvbVOB104lXs+e+X5uy//bLy1WuHqFxY5c0GSxgVrPrIsSll6v2JfQwbfCC8PedcmjFtvXX2exR+88x2J1W/24By9PyPxv9FBeS/qfo4RYYoFtxcTVriGhCC/p/Zr4Mcg8LaZNwpwp9c7xeaQsoVC4e1FJzvcVB/2NNFJiIcUif6/2zPeLzJyZlfTf9kRGO88eb8Nfm2DbDaN/3jPlA4/BnRUAAPJB1Kz9nwmSFQAA8kKQTv1ChjOCZAUAgHwxvZqV025KyUqIUuFQbqx8V8IzyoTVI4crPJsbrPJxfXMTVu2c7w1+Ft+W0OBoXpnFDl3iTYr6ElbpXHjngMWSVqwdLvNYX7Ofc/+icc9kC/zeW+j3+pcFd/u4TMJvafCNl1ispMtfQ9uFfr4r/3m/xXb/li/33LvGO4PNbfbn2WeVtllsV1+DxWLCIrLtFyasPto3OzP/NJpU4X/Sas2TjPGbOrF8mzOT3hAjxV5o0jfXa0p6FuZunDRfxEKfL+Y9NGSx4XK/JsyZmvangWZSCOHPY4yfn+p29FkBACBPhEzy1ymU9ImhCfEYCACAPDFbl74gWQEAIF+chsdAIYQdyj6ECpJaQgjbR/8eY4zLJrMPkhUAAPJB1GkpsI0xnvXi30MIT8QY1011H1NbdXkwqmpvbheh3hYvlGrc7EVGxc/sttie311tsaanvViq+Ih3LjrW5IVhSRo3+blUtfo5D1X5pehY47GBRr+H1rxhEvfVEpLZ2ue9ZKhrpY+b85gXgRUM+Tuu9Qpv3DRc6eN6zvWFMav2+riRYr/Gh/d486ov7n2TxUra/NpV7veLUNrl165uqxcCbrMITNLqyQlNv9Q78f9ZJW13ip9rnxHO5DlzuNKrPzvW+LiBen9jtjzgRcZDVRMX2NYnrE7debYfc84T/hrCsJ9HPs6Zs/W/YwpsAQDIF/E4XxMIIXwjhHAohLBpTOzCEMJDIYQnQwiPhRAuncQZPDCd0yZZAQAgD4SYbbef9DUJ35L0xnGxL0n67zHGCyV9evT7E4oxfniq5y2RrAAAkD+meWclxnivpM6EvdWM/r1WkjekmSEU2AIAkCdmuGblI5JuCyH8tbI3P145o3sfY0rJSqY4qHfcapZdqxK6LNZ4kVE4e5XFhhPqvUZKE1bQPLvCYkPVCatg9np62DPPq7SSOtOWHvHfYHmb769hS8IKy2f7ZZzzQO4xehb5MSsO+TEPXeTj2tZ5gdb8e7wr4sJbvMjq2KJai5Ue8HGF87zQrOJAQkHaK/xcCsp9BdWRpR4r3prwC0+6+xhnaSOAUynhEoWMBwsG/QKHpNWTbV+T2/9sXRTtVEn7nFncM7NzZlnSnPmsz5mHV/uc2fxE7v66Fyes1tzl+zqyyscdWucXav49Ryy28KfdFju2xFc/PmPmzHjCZKUphPDYmO+/GmP86gR7/CNJH40x/iiE8C5JX5f0uvGDQghB0sIY457JnajjzgoAAPni+HlNe4zx4inu7QOS/nj07z+Q9M+Jh4wxhhB+LOmiKe7/JdSsAACQJ0JM/pqm/ZKuHv37ayU9f4KxD4UQfIGmSeLOCgAA+eDEj4FOKIRwg6RXK/u4aK+kz0j6fUlfDiEUSTom6UMn2MVrJP1BCGGXpF79qoPt+ZM5PskKAAD5Ypp3UWKM1x/nR5N9tHPt9I6cNbUOtkNRlftzuyUOVfjy3/3NCcWv/b6/Jbd4kejua2os1rTJC4+aHvVtM1V+LiPl/hIPXuRFS+Xtfn5lh/24Hed6t8SyTv/tNz50MOf7WDjXxtQ82+Wxrf5kru0iL5LtXOvLuPfO92tXs9PT6KOL6y1WdMxfQ1IBXYHXt0ntXkBWeihhiXY/vcTP9/curvSBjyYcN08k3aJNXN5jkkW30/40wCQnudm6UNrJkHdzZufk5szSwwldWDfkzpkjpfNsTPVzPmdWvZAwZ17sRbKda/069c73616z68ydM4NOXwfbGOOul7M9d1YAAMgHL+Mx0OlGsgIAQL6YpXc+SVYAAMgTs/XOCh9dBgAgH0RJmeN8nSIhhB9NZ7sp3VmJRUHHGnM3aX7cOwAeusSLP4crvCiofZ2P61/kBVpthX6ag1VeQNXwyCE/l9d7kVZSZ9qYkLYVDCV1sPXz653rHR+3/JfmnO+L/DKpYZO/rraL/JrMu8Wb/g0vaLDYYI13VCzr9OquTML1TCoW617gryskFIs1bPTfbb+vqJ58jROWbS/qm6Wpfxqd5Fu+FNOe2Jk8ZyZ1Ui0Y9P92E+fMhC65W/5L7qRRnDBn1j/rxbpt6xPmzFsT5swWL5IdrPVtz/Q5MwX/zS6bzkY8BgIAIE+cjsdAIYTFL/5VUnEIYdHo3xVj3D2ZfZCsAACQD158DHTqfXv06EHSktHvw2jstZPZAckKAAB5ICh5HcSTLcb4mpfOIYQnYoyTSlDGIlkBACBPzNZPA00pWRmukNrW5Vb8FA75UuTdS5O29qqeZW/dbrH+ny+3WFmHb1u/yZf77l3dZLG5D/u4gaZyix1a750chyq8mKtqvxeL1T9/zGIjJbnHGE5oMNizxINJxVj737bYxyUUWQ37y9KRs/xXXJbQPbJrhReGDdT5uJrtHutYl7BU/CHfX8svvSVn52rvjNnf6NsiVwqK5DAJ+TZnDlZ6jDkzZXPmNOeOEMI3JL1F0qEY49ox8f8s6cOShiX9LMb4pxPs6svTOT53VgAAyAcvr4PttyT9o6TvvBgIIbxG0tsknR9jHAghJHymadwpxPit6RycPisAAOSJkEn+mkiM8V5JnePCfyTpCzHGgdEx/ln4GUKyAgBAvojH+ZKaQgiPjfn60CT2tkrSVSGEh0MI94QQLjlZp81jIAAA8sGJHwO1xxgvnuIeiyTVS7pc0iWSbgwhLIsxznhV3ZSSlaJeae6jua+0r8kLe0qO+IejRi70doRFCWtnH5vjV7LigO/vWIt3ay1M6J54bJ4XsyUVWtU/7+eS1I2w/QIvul387770enln7hLgdfcctjGZct/XYLV3VOzzxouaf9+AxY4s82XHexb6tTu83gve5t3lx8gU+o23/mYfV97q16lhs1/P/uaEIuYqP78ir70DZqVZOWfOTZgzRyY3Z/bMZ84cL01zZtCMfxpor6SbRpOTR0IIGUlNktrs2CH814Ttj0jaEGN8cqID8RgIAIB8cfzHQNPxY402dQshrJJUIqn9OGMvlvSHkhaMfn1I0qslfS2EMNEniHgMBABAXohSyEwvMwkh3KBsctEUQtgr6TOSviHpGyGETZIGJX3gBI+AGiWtjzH2jO7vM5J+KOlVkjZI+tKJjk+yAgBAnpjuY6AY4/XH+dH7JrmLxcomNC8akrQkxtgfQvBndOOQrAAAkCdOY0PJf5X0UAjh30a//3VJN4QQKiU9O9HGU0tWghTH1QXV7B6yYfXPeaHQ7kov7trQ4StFFw0mLIt+pR/j2GYvPCrr9N9CJuEV1uzy/XWs9YKns37Q4edySYPFNn+41mLji3hHin27yoN+HknPDuuf81R4qCphCfhqv3bVuxM6Ku7wbfsbfdvK/b5t/5zJjRus9nKoIysspKqE8+tecjpWrwBOgnybM2/0coX2SxstljhnDs3gnLnFrydzpl5uU7iXJcb4lyGEWyRdqWyt7x/GGB8b/fF7J9qeOysAAOSBk/BpoCmJMW5Qtj5lykhWAADIFzPfAuWUIFkBACAfnMbHQC8XyQoAAHkiaZXq2WBKycpIidS9KLeoqn6rp2lHF/kS1kU9XgBUMJBw+IQ6ofp7vWthabdf8ZonD1psaJ4XcnWe6x0aB+v8dWz9XW+D2HK/j6t+3l/HQMO4YjFvlKiChO6R1XsGE8b5a93xNl/bvCyhFU/dI30WK271zpCt1y6wWEWbH7f5Id+2/TJfZj6pcK1xo+8voSGnivsosMWZIe/mzN/zoljmzHTNmafx00AvC3dWAADIBy+jKdzpRrICAEC+mJ25CskKAAD5IMQ4a++ssJAhAAB5ImSSvybcLoRvhBAOja4DNP5nfxJCiCEEL8aZIVO+sxLH1fEUH/GOgnGh77bQa6BU5HVMynhdWGIXv/n3e5XR4Pw6i3Uv8cK13vkJXQtf8NiRc3xZ8IPv8iUMwjYvPlt8e+6a3d2LvFqs/Xw/t6RbdEV9Hmx4xscNl/m4kPCZ+v5Vc3xcQtFWf8JS9gpedFy1398Dra/0bpnVe5OO4flySffszPyBJMyZzJnjJc6Zrzg1c+bLKLD9lqR/lPSdnP2FsEjS66X/v717iY2rvuI4fk7sjD3xK048dggkzqOJK0CwIEJqpUKVSiVVo64qVSwKVbvphm0RYtEFsKi66N4SiA1CrCLxkKB0hUhAgoYSqJS4pTFNXEgwGT8S/J7TBQFi//7xPDJO7uh+P9Is5vjOvf+5vvrr6N4z52//bXjPNeDOCgAAeRBmthLpV7WPRrxlZpcSf/qLmf3BNrgahpoVAAByYp2alQF3f/+a96MRMbruvtx/YWYTEfGh+8a2nCBZAQAgJ9Z5DDQZEYdq3o/7FjN70sx+2oRhVcVjIAAAcsCv9llJvRqw38z2mtmH7j5uZneY2Ul339HEIX+rrjsrbQtmfeOrq4oK576U7YpDOyW2aVHzotKJLyQ28TMtZOoo6+2laNPY1EHtULjUrdsN/V2LmyYe0FOxZeiKjmWzFpB9FVosdvbo6kKwwkxtt8g6tNmhVbTuyvrGtPpu/KhW2s3t0GXmC4ljtC3qxVq+U7ebXtICskq7xna/oUV1M3v0i/SPzUtssSdRMQi0IObMWzRnJqaQvnHmTDMzr6E+pRYR8ZGZfXvxXU1YDkVEoi/wjePOCgAAeRDrvKpw9xfN7B0zG3H38+7+u40bqKJmBQCAXGi8KVxEPFzl73sa2nGNSFYAAMiDaN5joJuNZAUAgLxINL1rBXUlKyudZuUDqwuDpvbtku12vDcnsem9Wsj1+eGSxDZfTnQUTLQCLo9oQVFhSj9bnNQPTw/rZ5cHtPjq53vOSKy7TYug3usallipeHnV++P/PKDbvK2nf+tYYnnyC9MSG/9VoiBvIXERJs5dJbH0emei1U/xcy1wK/1Dv/9cSc/n9D4tDGuf1/H9+1EtNGvrTLTufEVDQNYxZzJnZm3ObNW1gbizAgBAHnzTwbYFkawAAJADbpFc+6gVkKwAAJAXlRqWWM4gkhUAAPIgLFmT0wrqS1bCbNOaZoT9Y9qd8LMfaGHYbce1gGx2WKuWes5pMdJyUYd5cUgLlIpLentrZlj73hUndbv2Sd3fpUXtsvin3e9I7CmJmJ0srymia9crZMuknruFAT0n84PaobJ7Qr/DlYoWd/V+qsetJP7ryx2JjpeJ7WZ3aRHYwOufSKx8eJ/ELvwwcftxXv8/Pdu1QyPQkpgzmTMzNmc6d1YAAEBmRfAYCAAAZFxr5iqsDQQAQF54pZJ8Vf2c+3PuftHdP74m9md3P+3up9z9mLtv3ahxk6wAAJAHEWYrlfSruufN7Mia2JtmdndE3GNmY2b2RHMH/J26HgN5mG1a0yivMKNLh992XGP/+5EWkBWm9BgdiaWuU78Lny/pyS1e1IKnwQ90LPP9iQ6AiaKlE2e14Gm091OJPT34kcT2fnD/qve9p7QIbLFbv0Pfae28OHugV2Kpgq/dL+vy8XPDmuiWD+o57jm3IrG2xBL189v1HH/y2P6axldILFu/cLv+f/Zv0xXGT+nugMxjzmzunLnUpd+h9wxzZl1zZoN9ViLiLXffsyb212vevmtmv2xo5zWgZgUAgDwIW+8uyoC7v3/N+9GIGK1j7781s5caHVo1JCsAAORCmFX0jtBVkxFxqJG9uvuTZrZsZi80OrJqSFYAAMiDMLMmL2To7o+a2VEz+0nExvXyJ1kBACAvmthnxd2PmNnjZvZgROjy101UV7LSthDWd3Z1B8GlLt1F+5zeZuooa8I1dEKrxZb6tais0q5FRiPPaAfA87/5vh73lHb2m729W2LFCxKy9vv03M+udErs6Uk9rl9efV76z2hR1GKPFq1N3dmn40gsE146dlpiVtqmx+jVY2xaTHRyHNLCsLlBPe9diS6QZrrdzF36fQv/0SK1rSe1u+MjD5yQ2LHEUYGsY85kzszUnBlhtnLdx0DrcvcXzezH9nVty3kz+6N9/eufDjN7093NzN6NiN83dIAquLMCAEBeNP5roIcT4WdvbDC1I1kBACAXotaeKplDsgIAQB6EWQTJCgAAyLI83FlZ2ex2Zcfq4qPpA7rdzre1eKgtsYL1pXu1U+D8Nv3s9o91CfTK3p0Sm92vy4f7G1q0NPiqFpqd+/X39LOvbZfY88OHJVaY0jF3rqnRKh/Uoq1lrYuzxX59nrjrb/q9ph4akVjXeT3JqSIwT1yr0yOJAr9LiWI2rYuzlW4dX8dnemn1/0uPMXFYv+/oxIN6EHrYogUxZzJnZmrOvIEC21uNOysAAORENPGnyzcTyQoAAHkQFNgCAIAMCzMLHgMBAIDMijBr0V8DeT2t/N39CzPT9b6BjTccEaXr/ZFrE7cI1yWySq5Nd3/dzAaus/1kRBzZ+GE1pq5kBQAA4GbT32gBAABkCMkKAADINJIVAACQaSQrAAAg00hWAABAppGsAACATCNZAQAAmUayAgAAMo1kBQAAZNr/AaKzGrqJIWl8AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 648x648 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x648 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cpost = ComparePostage(mof, mms, psfex_paths)\n",
    "cpost.draw_postage(seg=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {},
   "outputs": [],
   "source": [
    "tmp = fio.read(\"/e/eser2/vargatn/DES/Y3_DATA/COSMOS_C01_r3766p01_mof.fits\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 757.,  228.,  312.,  683., 8311., 2987.,  154.,   79.,   68.,\n",
       "         194.]),\n",
       " array([0. , 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1. ]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "inds = allmofs[\"flags\"] == 0\n",
    "plt.hist(allmofs[\"bdf_pars\"][inds, 5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1, 4)"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mof[\"bdf_flux\"].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10,)"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mof[\"bdf_pars\"][-1].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9,)"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tmp[\"cm_pars\"][-1].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-9999.        , -9999.        , -9999.        , -9999.        ],\n",
       "       [-9999.        , -9999.        , -9999.        , -9999.        ],\n",
       "       [-9999.        , -9999.        , -9999.        , -9999.        ],\n",
       "       ...,\n",
       "       [  101.66138442,   181.18842708,   208.17364311,   192.30484706],\n",
       "       [ 4127.13066122, 14833.16786719, 23692.08332877, 31529.87687425],\n",
       "       [   50.49053744,    86.41434677,   142.5037418 ,   214.84054715]])"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tmp[\"cm_flux\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype([('id', '>i8'), ('number', '>i4'), ('ra', '>f8'), ('dec', '>f8'), ('nimage_tot', '>i4', (4,)), ('fofid', '>i8'), ('flags', '>i4'), ('time_last_fit', '>f8'), ('box_size', '>i2'), ('obj_flags', '>i4'), ('psf_flags', '>i4', (4,)), ('psf_flux', '>f8', (4,)), ('psf_flux_err', '>f8', (4,)), ('psf_mag', '>f8', (4,)), ('psf_flux_s2n', '>f8', (4,)), ('nimage_use', '>i4', (4,)), ('mask_frac', '>f8'), ('psfrec_T', '>f8'), ('psfrec_g', '>f8', (2,)), ('cm_flags', '>i4'), ('cm_pars', '>f8', (9,)), ('cm_pars_cov', '>f8', (9, 9)), ('cm_T', '>f8'), ('cm_T_err', '>f8'), ('cm_T_s2n', '>f8'), ('cm_flux_cov', '>f8', (4, 4)), ('cm_flux', '>f8', (4,)), ('cm_flux_s2n', '>f8', (4,)), ('cm_mag', '>f8', (4,)), ('cm_logsb', '>f8', (4,)), ('cm_g', '>f8', (2,)), ('cm_g_cov', '>f8', (2, 2)), ('cm_weight', '>f8'), ('cm_max_flags', '>i4'), ('cm_max_pars', '>f8', (9,)), ('cm_max_pars_cov', '>f8', (9, 9)), ('cm_max_T', '>f8'), ('cm_max_T_err', '>f8'), ('cm_max_T_s2n', '>f8'), ('cm_max_flux_cov', '>f8', (4, 4)), ('cm_max_flux', '>f8', (4,)), ('cm_max_flux_s2n', '>f8', (4,)), ('cm_max_mag', '>f8', (4,)), ('cm_max_logsb', '>f8', (4,)), ('cm_s2n_w', '>f8'), ('cm_chi2per', '>f8'), ('cm_dof', '>f8'), ('cm_flags_r', '>i4'), ('cm_s2n_r', '>f8'), ('cm_T_r', '>f8'), ('cm_psf_T_r', '>f8'), ('cm_fracdev', '>f4'), ('cm_fracdev_noclip', '>f4'), ('cm_fracdev_err', '>f4'), ('cm_TdByTe', '>f4'), ('cm_TdByTe_noclip', '>f4'), ('cm_mof_flags', '>i4'), ('cm_mof_num_itr', '>i4'), ('cm_mof_abs_diff', '>f8', (9,)), ('cm_mof_frac_diff', '>f8', (9,)), ('cm_mof_err_diff', '>f8', (9,)), ('fofind', '>i8')])"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tmp.dtype"
   ]
  },
  {
   "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": "3.6.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "widgets": {
   "state": {},
   "version": "1.0.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}