{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Data Inspection- Preprocessing - Unsupervised ML"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This tutorial demonstrates the most important basic steps involved in the analysis of scanning electron diffraction data."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The data was acquired from a GaAs nanowire adopting the zinc blende structure and exhibiting type I twinning (i.e. on {111}) along its length. The region of interest contains a single nanowire comprising multiple crystals each in one of the two twinned orientations and near to a <1-10> zone axis."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This functionaility has been checked to run in pyxem version 0.14.1 (April 2022). However, bugs are always possible, do not trust the code blindly, and if you experience any issues please report them here: https://github.com/pyxem/pyxem-demos/issues"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Contents"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"1. Loading & Inspection\n",
"2. Alignment & Calibration\n",
"3. Virtual Diffraction Imaging\n",
"4. Machine Learning SPED Data\n",
"5. Peak Finding"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Import pyxem and other required libraries"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Changing the matplotlib background will give you interactive \n",
"#%matplotlib qt5"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING:silx.opencl.common:Unable to import pyOpenCl. Please install it from: https://pypi.org/project/pyopencl\n"
]
}
],
"source": [
"%matplotlib inline\n",
"import hyperspy.api as hs\n",
"import pyxem as pxm\n",
"import numpy as np"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1. Loading and Inspection"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Load the SPED data acquired from the nanowire"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/carterfrancis/mambaforge/envs/pyxem-demos/lib/python3.11/site-packages/hyperspy/misc/utils.py:471: VisibleDeprecationWarning: Use of the `binned` attribute in metadata is going to be deprecated in v2.0. Set the `axis.is_binned` attribute instead. \n",
" warnings.warn(\n",
"/Users/carterfrancis/mambaforge/envs/pyxem-demos/lib/python3.11/site-packages/hyperspy/io.py:572: VisibleDeprecationWarning: Loading old file version. The binned attribute has been moved from metadata.Signal to axis.is_binned. Setting this attribute for all signal axes instead.\n",
" warnings.warn('Loading old file version. The binned attribute '\n"
]
}
],
"source": [
"dp = hs.load('./data/01/twinned_nanowire.hdf5', reader=\"hspy\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Inspect the dp object"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dp"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Specify that the data is electron diffraction data"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"dp.set_signal_type('electron_diffraction')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Inspect the signal type"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dp"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Inspect the data type of the object"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"dtype('uint8')"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dp.data.dtype"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Inspect the metadata associated with the object 'dp'"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" - \n",
" Acquisition_instrument\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" TEM\n",
"
\n",
" \n",
" \n",
" - beam_energy = 300.0
\n",
" \n",
" \n",
" - camera_length = 0.21000000000000002
\n",
" \n",
" \n",
" - scan_rotation = 277.0
\n",
"
\n",
" \n",
" \n",
" - \n",
" General\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" FileIO\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" 0\n",
"
\n",
" \n",
" \n",
" - hyperspy_version = 1.7.5
\n",
" \n",
" \n",
" - io_plugin = hyperspy.io_plugins.hspy
\n",
" \n",
" \n",
" \n",
" \n",
" - timestamp = 2023-10-26T10:27:31.436805-05:00
\n",
"
\n",
" \n",
" - original_filename = nanowire_precession.blo
\n",
" \n",
" \n",
" - time = (2014, 12, 8)
\n",
" \n",
" \n",
"
\n",
" \n",
" \n",
" - \n",
" Signal\n",
"
\n",
" \n",
" \n",
" \n",
" \n",
" - signal_type = electron_diffraction
\n",
"
"
],
"text/plain": [
"├── Acquisition_instrument\n",
"│ └── TEM\n",
"│ ├── beam_energy = 300.0\n",
"│ ├── camera_length = 0.21000000000000002\n",
"│ └── scan_rotation = 277.0\n",
"├── General\n",
"│ ├── FileIO\n",
"│ │ └── 0\n",
"│ │ ├── hyperspy_version = 1.7.5\n",
"│ │ ├── io_plugin = hyperspy.io_plugins.hspy\n",
"│ │ ├── operation = load\n",
"│ │ └── timestamp = 2023-10-26T10:27:31.436805-05:00\n",
"│ ├── original_filename = nanowire_precession.blo\n",
"│ ├── time = (2014, 12, 8)\n",
"│ └── title = \n",
"└── Signal\n",
" ├── signal_origin = \n",
" └── signal_type = electron_diffraction"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dp.metadata"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Set important experimental parameters using the built in function"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"dp.set_experimental_parameters(beam_energy=300.0,\n",
" camera_length=21.0,\n",
" scan_rotation=277.0,\n",
" convergence_angle=0.7,\n",
" exposure_time=10.0)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"See how this changed the metadata"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" - \n",
" Acquisition_instrument\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" TEM\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" Detector\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" Diffraction\n",
"
\n",
" \n",
" \n",
" - camera_length = 21.0
\n",
" \n",
" \n",
" - exposure_time = 10.0
\n",
"
\n",
" \n",
" - beam_energy = 300.0
\n",
" \n",
" \n",
" - camera_length = 0.21000000000000002
\n",
" \n",
" \n",
" - convergence_angle = 0.7
\n",
" \n",
" \n",
" - scan_rotation = 277.0
\n",
"
\n",
" \n",
" \n",
" - \n",
" General\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" FileIO\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" 0\n",
"
\n",
" \n",
" \n",
" - hyperspy_version = 1.7.5
\n",
" \n",
" \n",
" - io_plugin = hyperspy.io_plugins.hspy
\n",
" \n",
" \n",
" \n",
" \n",
" - timestamp = 2023-10-26T10:27:31.436805-05:00
\n",
"
\n",
" \n",
" - original_filename = nanowire_precession.blo
\n",
" \n",
" \n",
" - time = (2014, 12, 8)
\n",
" \n",
" \n",
"
\n",
" \n",
" \n",
" - \n",
" Signal\n",
"
\n",
" \n",
" \n",
" \n",
" \n",
" - signal_type = electron_diffraction
\n",
"
"
],
"text/plain": [
"├── Acquisition_instrument\n",
"│ └── TEM\n",
"│ ├── Detector\n",
"│ │ └── Diffraction\n",
"│ │ ├── camera_length = 21.0\n",
"│ │ └── exposure_time = 10.0\n",
"│ ├── beam_energy = 300.0\n",
"│ ├── camera_length = 0.21000000000000002\n",
"│ ├── convergence_angle = 0.7\n",
"│ └── scan_rotation = 277.0\n",
"├── General\n",
"│ ├── FileIO\n",
"│ │ └── 0\n",
"│ │ ├── hyperspy_version = 1.7.5\n",
"│ │ ├── io_plugin = hyperspy.io_plugins.hspy\n",
"│ │ ├── operation = load\n",
"│ │ └── timestamp = 2023-10-26T10:27:31.436805-05:00\n",
"│ ├── original_filename = nanowire_precession.blo\n",
"│ ├── time = (2014, 12, 8)\n",
"│ └── title = \n",
"└── Signal\n",
" ├── signal_origin = \n",
" └── signal_type = electron_diffraction"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dp.metadata"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Set another metadata item and check it"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" - \n",
" Acquisition_instrument\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" TEM\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" Detector\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" Diffraction\n",
"
\n",
" \n",
" \n",
" - camera_length = 21.0
\n",
" \n",
" \n",
" - exposure_time = 10.0
\n",
"
\n",
" \n",
" - beam_energy = 300.0
\n",
" \n",
" \n",
" - camera_length = 0.21000000000000002
\n",
" \n",
" \n",
" - convergence_angle = 0.7
\n",
" \n",
" \n",
" - scan_rotation = 277.0
\n",
"
\n",
" \n",
" \n",
" - \n",
" General\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" FileIO\n",
"
\n",
" \n",
" \n",
" \n",
" - \n",
" 0\n",
"
\n",
" \n",
" \n",
" - hyperspy_version = 1.7.5
\n",
" \n",
" \n",
" - io_plugin = hyperspy.io_plugins.hspy
\n",
" \n",
" \n",
" \n",
" \n",
" - timestamp = 2023-10-26T10:27:31.436805-05:00
\n",
"
\n",
" \n",
" - original_filename = nanowire_precession.blo
\n",
" \n",
" \n",
" - time = (2014, 12, 8)
\n",
" \n",
" \n",
" - title = GaAs Nanowire
\n",
"
\n",
" \n",
" \n",
" - \n",
" Signal\n",
"
\n",
" \n",
" \n",
" \n",
" \n",
" - signal_type = electron_diffraction
\n",
"
"
],
"text/plain": [
"├── Acquisition_instrument\n",
"│ └── TEM\n",
"│ ├── Detector\n",
"│ │ └── Diffraction\n",
"│ │ ├── camera_length = 21.0\n",
"│ │ └── exposure_time = 10.0\n",
"│ ├── beam_energy = 300.0\n",
"│ ├── camera_length = 0.21000000000000002\n",
"│ ├── convergence_angle = 0.7\n",
"│ └── scan_rotation = 277.0\n",
"├── General\n",
"│ ├── FileIO\n",
"│ │ └── 0\n",
"│ │ ├── hyperspy_version = 1.7.5\n",
"│ │ ├── io_plugin = hyperspy.io_plugins.hspy\n",
"│ │ ├── operation = load\n",
"│ │ └── timestamp = 2023-10-26T10:27:31.436805-05:00\n",
"│ ├── original_filename = nanowire_precession.blo\n",
"│ ├── time = (2014, 12, 8)\n",
"│ └── title = GaAs Nanowire\n",
"└── Signal\n",
" ├── signal_origin = \n",
" └── signal_type = electron_diffraction"
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dp.metadata.set_item(\"General.title\", 'GaAs Nanowire')\n",
"dp.metadata"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": null,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dp"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Plot the data to inspect it"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAOoAAADECAYAAACRMr0IAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABVZElEQVR4nO2dd3xUVfr/P1Pu9J7JZNILQQQCCChlVWDpKIIFC+xXAVlXxYbYUFYR1wXERVkRXVcQVEAEBFRQpKMICCKINGkhddIm03s5vz/yO2czmQSCBDLofb9e89K598y9Z0KePOc+53mej4AQQsDDw5PQCFt7Ajw8POeHN1QenisA3lB5eK4AeEPl4bkC4A2Vh+cKgDdUHp4rAN5QeXiuAHhD5eG5AuANlYfnCuA3GeqhQ4cwYcIEtGnTBnK5HHK5HG3btsWDDz6IH3/88aIn9cUXX0AgECApKQmBQOCir1efnJwcCAQCPPTQQ3Hntm/fDoFAgFWrVrXoPS83OTk5GDdu3CW/z9mzZyEQCCAQCLB8+fK48y+//DIEAgFqamou2RzoPS4lXq8XL7/8MrZv335J73MuLthQ33vvPXTv3h0//PADnnjiCaxbtw7r16/HpEmTcOTIEVx33XU4ffr0RU1q4cKFAIDa2lqsXbv2oq51rnv8+uuvl+Tarc2aNWvw4osvXtZ7Tp06FaFQ6LLeEwD++te/Yvfu3Zf0Hl6vF9OnT29VQwW5AHbu3EmEQiG55ZZbSCAQaHTMihUrSFlZ2YVcNgaLxULEYjHp378/kclkZNCgQb/5Wo2RnZ1NevfuTbRaLbn99ttjzm3bto0AICtXrmzReyYiwWCQhEKhi7pGYWEhAUCGDRtGAJC33nor5vy0adMIAFJdXX1R92ltqqurCQAybdq0Fr2u1+sl0Wi0WWMvyFBvuukmwnEcKS8vb/Zn9u3bR+6++26SnZ1NZDIZyc7OJvfccw85e/Zso+NnzZpFAJAvv/ySjBkzhgiFwkbHrlixgvTo0YNoNBoil8tJbm4uGT9+/Hnnk52dTW6++WYyc+ZMAoDs3r2bnWvMUE+ePEnGjRtH8vPziVwuJ2lpaWT48OHk0KFDMdeln122bBl54YUXSGpqKlGr1WTAgAHk+PHjcfNYuHAh6dy5M5FKpUSv15Nbb72VHD16lJ1ft24dAUD27t3Ljq1atYoAIDfddFPMtTp16hTzRyc7O5uMHTs2bm4fffQRmTx5MklLSyMCgYAcO3aMEELIpk2bSP/+/YlarSZyuZz86U9/Ips3bz7vz5Ia6uuvv06GDBlCkpOTidPpZOcbM9SNGzeSESNGkPT0dCKVSkmbNm3I3/72t5gxa9asIQAancM777xDAJCff/455h718fv9ZPLkySQlJYXI5XJy4403kh9//DHu51JVVUUefvhh0r59e6JUKklycjL585//TL799tu479jwVf863333Henfvz9RqVRELpeT3r17k3Xr1sXMadGiRQQA+eabb8j48eOJ0WgkAIjP5zvvz5kQQpq99I1EIti2bRuuvfZapKamNttjnz17Fu3atcPcuXPxzTff4LXXXoPFYsF1113X6LPLBx98gNTUVAwbNgz3338/otEoFi9eHDNm9+7duPvuu5GXl4fly5dj/fr1eOmllxAOh5s9ryeeeALp6el49tlnzzmuvLwcSUlJmDVrFjZs2ID58+dDLBajZ8+ejS6dX3jhBRQVFWHBggX473//i5MnT+KWW25BJBJhY2bOnIkJEyagY8eOWL16Nf7973/j0KFD6N27N06ePAkA6Nu3LziOw+bNm9nnNm/eDLlcjh07drBlZlVVFQ4fPoyBAwee9zs///zzKC4uxn/+8x98+eWXMJlMWLJkCQYPHgyNRoMPP/wQK1asgMFgwJAhQ7Bly5Zm/SwB4LXXXkNNTQ1ef/31c447ffo0evfujXfffRcbN27ESy+9hB9++AE33HAD+07Dhw+HyWTCokWL4j6/ePFidOvWDZ07d27yHuPHj8fcuXMxfvx4fP7557jjjjtw2223wW63x4yrra0FAEybNg3r16/HokWLkJeXh379+rFlbmpqKjZs2AAAmDBhAnbv3o3du3ezR4sdO3agf//+cDgcWLhwIT755BOo1Wrccsst+PTTT+Pmdv/994PjOHz88cdYtWoVOI4758+L0SxzJoRUVFQQAOSee+6JOxcOh0koFGKvc7nzcDhM3G43USqV5N///nfMuW+//ZYAIFOmTCGEEBKNRklubi7Jzs6Ouea//vUvAoDY7fbmTp9BPSohhLz//vvMexPSvKVvOBwmwWCQtG3bljz55JPsOP1sQ2+3YsWKGM9ts9mIXC6PG1dcXEykUikZM2YMO3bDDTeQ/v37s/f5+fnkmWeeIUKhkOzYsYMQQsjSpUsJAHLixImY79iYR+3Tp0/MPT0eDzEYDOSWW26JOR6JREiXLl1Ijx49mvw5EBLrUQkh5C9/+QtRKpXEYrEQQs6/9I1GoyQUCpGioiICgHz++efs3OTJk4lcLo/5Nz569CgBQObNm8eONfSoR44cIQDIc889F3OvTz75JM4TNoT+Hg8YMIDcdttt7Pi5lr69evUiJpOJuFyumOsUFBSQjIwM9ntLPep9993X5P3PRYtsz3Tv3h0cx7HXnDlz2Dm3243nnnsO+fn5EIvFEIvFUKlU8Hg8OHbsWMx1aBDp/vvvBwAIBAKMGzcORUVFMX/dr7vuOgDAXXfdhRUrVqCsrOw3zXv8+PHo0KEDpkyZgmg02uiYcDiMGTNmoEOHDpBIJBCLxZBIJDh58mTc/AFgxIgRMe/pX/6ioiIAdasBn88XF5XNzMxE//79Y77ngAED8P3338Pn86GoqAinTp3CPffcg2uuuQabNm0CUOdls7Ky0LZt2/N+3zvuuCPm/a5du1BbW4uxY8ciHA6zVzQaxdChQ7Fv3z54PJ7zXpfy6quvIhQKYfr06U2OqaqqwkMPPYTMzEyIxWJwHIfs7GwAiPl53n///fD5fDFeadGiRZBKpRgzZkyT19+xYweAut+N+owaNQpisThu/H/+8x9069YNMpmMzWfLli2N/ts2xOPx4IcffsCoUaOgUqnYcZFIhHvvvRelpaVxq66G/wbNpdmGajQaIZfL2S9cfZYtW4Z9+/bhiy++iDs3ZswYvP322/jrX/+Kb775Bnv37sW+ffuQnJwMn8/HxrlcLqxcuRI9evRAcnIy7HY77HY7brvtNggEAmbEANCnTx+sXbsW4XAY9913HzIyMlBQUIBPPvnkgr68SCTCjBkzcOTIEXz44YeNjpk8eTJefPFF3Hrrrfjyyy/xww8/YN++fejSpUvM/ClJSUkx76VSKQCwsVarFQAafXxIS0tj5wFg4MCBCAQC2LlzJzZt2gSj0YiuXbti4MCBbEm8ZcuWZi17G7tnZWUlgLpf4vp/aDmOw2uvvQZCCFseNoecnBxMnDgRCxYsYEv4+kSjUQwePBirV6/Gs88+iy1btmDv3r3Ys2cPAMT8PDt27IjrrruOLX8jkQiWLFmCkSNHwmAwNDkH+vNLSUmJOS4Wi+P+bd544w08/PDD6NmzJz777DPs2bMH+/btw9ChQxv9t22IzWYDIaTJf8v686FcyGNjzPybO1AkEqF///7YuHEjLBZLzA07dOgAoO55tD4OhwPr1q3DtGnTMGXKFHY8EAjE/QJ88skn8Hq92Lt3L/R6fdz916xZA5vNxs6NHDkSI0eORCAQwJ49ezBz5kyMGTMGOTk56N27d3O/FkaOHInrr78e06ZNw3//+9+480uWLMF9992HGTNmxByvqamBTqdr9n0o9JfFYrHEnSsvL4fRaGTve/bsCZVKhc2bN+Ps2bMYMGAABAIBBgwYgDlz5mDfvn0oLi5utqE23G+k95o3bx569erV6Gca/sKfj7///e/44IMP8MILL6Bjx44x5w4fPoyff/4ZixcvxtixY9nxU6dONXqt8ePHY+LEiTh27BjOnDkDi8WC8ePHn/P+9OdbWVmJ9PR0djwcDscZzZIlS9CvXz+8++67McddLtf5vygAvV4PoVDY5L8lgJh/TyD+36C5XNDS9/nnn0ckEsFDDz3UrD0zgUAAQgjzKpQFCxbEBFeAumWvWq3Gli1bsG3btpjX66+/jkAggKVLl8bdQyqVom/fvnjttdcAAAcOHLiQrwSgLhBSUlKCt956q9Hv0HD+69ev/83L7d69e0Mul2PJkiUxx0tLS7F161YMGDCAHeM4Dn369MGmTZuwdetWDBo0CABw4403QiwW4+9//zsz3N/C9ddfD51Oh6NHj+Laa69t9CWRSC7omklJSXjuueewatUq7N27N+Yc/SVt+PN87733Gr3W6NGjIZPJsHjxYixevBjp6ekYPHjwOe/fp08fAIgL5KxatSou2NjYv+2hQ4fi9mUbroooSqUSPXv2xOrVq2PORaNRLFmyBBkZGbjqqqvOOd/m0myPCtT9w86fPx+PPfYYunXrhr/97W/o2LEj+6vy2WefAQA0Gg37b58+ffD666/DaDQiJycHO3bswMKFC2O80eHDh7F37148/PDD6N+/f6P3nTNnDhYuXIhHH30UL730EkpLSzFgwABkZGTAbrfj3//+NziOQ9++fS/4h3D99ddj5MiR+Pzzz+PODR8+HIsXL8bVV1+Nzp07Y//+/Xj99deRkZFxwfcBAJ1OhxdffBEvvPAC7rvvPowePRpWqxXTp0+HTCbDtGnTYsYPGDAATz31FAAwzymXy/GnP/0JGzduROfOnWEymX7TXFQqFebNm4exY8eitrYWo0aNgslkQnV1NX7++WdUV1fHeZvmMGnSJMyfPx9ff/11zPGrr74abdq0wZQpU0AIgcFgwJdffsmetxui0+lw2223YfHixbDb7Xj66achFJ7bt3Ts2BGjR4/GnDlz2CrwyJEjmDNnDrRabcznhw8fjn/84x+YNm0a+vbti19//RWvvPIKcnNzY4xarVYjOzsbn3/+OQYMGACDwcB+n2fOnIlBgwbhz3/+M55++mlIJBK88847OHz4MD755JOWy5r6LRGogwcPkvHjx5Pc3FwilUqJTCYj+fn55L777iNbtmyJGVtaWkruuOMOotfriVqtJkOHDiWHDx+OiUxOmjSJACAHDx5s8p5TpkwhAMj+/fvJunXryLBhw0h6ejqRSCTEZDKRm266iXz33XfnnXv9qG99jh49SkQiUVzU12azkQkTJhCTyUQUCgW54YYbyHfffUf69u1L+vbty8Y1FTGmkdFFixbFHF+wYAHp3LkzkUgkRKvVkpEjR5IjR47Ezevnn38mAEjbtm1jjv/zn/8kAMjkyZMb/Y6NRX2bimbv2LGD3HzzzcRgMBCO40h6ejq5+eabz5v40TDqW5///ve/bM+xftT36NGjZNCgQUStVhO9Xk/uvPNOUlxc3GRUdePGjew69SPblHPto5pMJiKTyUivXr3I7t27iVarjYnUBwIB8vTTT5P09HQik8lIt27dyNq1a8nYsWNJdnZ2zDU3b95MunbtSqRSaZP7qEqlksjlctKrVy+2k0ChUd99+/ad60faJAJC+C6EPE3j9/sRDAbjjkskEshkslaY0W9j165duP7667F06dJzRo0TFd5QeZrE7/cjNzcXFRUVcefMZjMKCwsT0lg3bdqE3bt3o3v37pDL5fj5558xa9YsaLVaHDp0KCHnfD4u6BmV549FMBhERUUFzp49y+IOAOB0OpGTk4NgMJiQv/QajQYbN27E3Llz4XK5YDQaMWzYMMycOTMh59sceI/K0yROpxNarRY1NTVxhmo0GuFwOGKO81w6+MLxFuadd95Bbm4uZDIZunfvju+++661p3TRRKPRuBfP5YU31Bbk008/xaRJkzB16lQcOHAAN954I4YNG4bi4uLWntpFwRtq68MvfVuQnj17olu3bjF7j+3bt8ett96KmTNntuLMfht06VteXh639E1LS+OXvpcRPpjUQgSDQezfvz8mVRIABg8ejF27djXrGtFoFOXl5VCr1Ze8vci5IITA5XKxRPOGXpT3qJcf3lBbiJqaGkQikbjc2JSUlEa3N4C6nOf6PaHKysrQ/v/nTbu0Wnz46qsA6jKR8vPzUVFRge7du+P48eMA6vKvDQYDBAIBKisrEY1GWcpfSUkJ8vLy4HQ6EQ6HEYlEUFZWhqSkJHg8HpSXl8Pr9cLpdOLkyZOoqKiAUqmEyWTC8ePHUVJSgqNHjwLgDTUR4J9RW5iGnpAQ0qR3nDlzJrRaLXvR4gZ6HZqmlp2dDZfLBalUCovFAqlUCqFQCJfLBY/HA7FYDK1WC0IIOI6DSqWCQqFAaWkphEIhBAIBHA4HJBIJPB4PqqqqIJPJUFpaCpvNBpFIxM55vV5kZWUBQJxH5Z9RWw/eUFsIo9EIkUgU5z2rqqqarEB5/vnn4XA42KukpAQ1UimcGg18/9/wgLoSL5fLhVAoBJ/Ph1AohNraWgiFQni9XhQWFuL06dMIBAIQi8UghECn0yEpKQkKhQIGgwHp6emsY6REImFjotEoqxGWSqUghLDnTvoHhhASY6R8WOPywy99WwiJRILu3btj06ZNuO2229jxTZs2YeTIkY1+RiqVxlVv3Dd0KLp27QqTyQQd6jrgCQQC6HQ6hMNhqFQqeL1eVuQslUrBcRwIIfD7/fD5fPD5fGwJTA1RJpOxgv3MzEzYbDbU1NSwKiihUIhwOIxgMAiRSBQzJ37p2/rwhtqCTJ48Gffeey+uvfZa9O7dG//9739RXFzcaA/hpohEIuyZMhwOM4OkXiwQCCAajUKj0bCOBSKRCCqVCqFQCGKxmJUQSqVShMNh+P1+AP8zRo7j4Pf7WSUJrRQRiUTw+/1x9ZW8obY+vKG2IHfffTesViteeeUVWCwWFBQU4KuvvmKtRpqDQCBAOByGSCSCz+eDXC6HWq2G3W6HVCpFMBhEIBCASqUCIQQ+nw/hcBgCgQAKhQJSqRRyuRwikQhisRjRaBSBQAChUAihUIg9r/r9fuY5afsVsVgMoVAY1wSMN9TWh39GbWEmTpyIs2fPIhAIYP/+/ayQublQzxiJRNizayQSgVQqhdFohFarRSgUYkYajUYRDofhcDjAcRx8Ph/rbGe1WiESiZhnFQqFkMlkIITAbDZDJBIhEokwgw0EAtDpdEhOTo6ZEx9Man14Q00wwuEwJBIJbDYbhEIh29OUyWQwGo04e/YsRCIRlEolotEohEIh9Ho9y711uVwQCoWsmL+6uhpA3R8AgUAAjuPg9Xrh9XphsVig0WigVqshkUjAcRzOnj0b1yaHN9TWhzfUBMPlcuHEiRPgOA4pKSkwm83QaDSIRCI4duwYgsEgQqEQJBIJ7HY76xIoEAgQiUSQm5sLv9+P6upqdOvWDVlZWSxIpNVqUVVVBb1ej7S0NNTW1kIkErFlsU6ng0ajaXGPOmTIECiVSggEAgiFQqSmpsZ1f6ioqEDnzp0hEolYi5TRo0fHjHE6nejcuTPbcjKbzdi3b1/MmMLCQuTl5TFNnLy8vLiGfLt370ZKSgqbT5cuXeB2u2PGfPbZZ9DpdBAIBKxTRGv+geINNcGg3fKoN3W73bDZbGxbpGvXrtDr9Th8+DBUKhXkcjkCgQAUCgXy8/MhkUjgcDig1WrhcrngdrvBcRxkMhm8Xi9ycnJQXl6OQ4cOwWw2IxgMwmq1wuPxgOM4CIVC5oUpF2uo+/fvx+jRo7F27VosX74c0WgUt9xyC6qqqtiYfv364dixY5gzZw6+++473HnnnVi+fDleeOEFNqZ37944cuQIZs2ahWXLliEQCKBPnz4xhe09evRAeXk53nvvPbz33nsoLy9Hjx492PlgMIg///nPCAaDWLZsGWbNmoXDhw/jT3/6ExtTWlqKu+66C1qtFqtXr8ZTTz2Fbdu2NRm9vxzwwaQEg+M4iEQiBINB2O12FvXlOA46nQ4nT56E1+tl2y3BYBBCoRAikQgnT56ERqNBXl4eqqqqIBAIoNVqIRAI4Pf7EYlE4PV64Xa7UVFRAYPBAL/fj/T0dNjtdlRWVkIsFse147zYYFJDRYROnTqhQ4cO+PTTT/HYY48BqOtg2adPH0yaNAkAcMMNN2D16tWs+qi4uBhHjx7F448/ztQN2rVrh+7du+P111/H1KlTsW7dOtTU1GDBggWYMGECgLpI9wMPPICvv/4aw4YNw6xZsxAIBLBr1y5069YNQJ1hzps3D6WlpcjIyMAzzzyDaDSKX375BRqNBrfddhsOHTqEr776ij1uXG54j5pgVFVVQSKRwOv1wuFwMEMMBoPweDyQyWTQarUwGo1ISkqCSqViASXaHtNut8cEpJxOJ4LBIMRiMUQiEWQyGRQKBdLS0qDRaEAIQSQSgUwmg8FgiGuX2VTCQ1lZGUpLS9nL6XQ26zvSVpr1G8Tl5eVhz549+PHHHxGNRvHGG2/A5/Ox5e/KlSsBgBkyANY4e+PGjQDAmutRIwXq1N7qf37z5s2QyWTMSAHgySefBACsWLECALB37172yEEZO3YsotEodu7c2azv2NLwHjXBEIlEEAqFLLhDn9kEAgFUKhXcbjd0Oh3bB6UdC4RCIUKhEDQaDaLRKHw+HztPEyE4jkM0GoXf70daWhrsdjvcbjcbR/doG+qhNOVR66c8AnV6OeeTJoxGo/jLX/7CPBVl79696NKlC1NBAIAHH3wQEydOBACcOXMGAJCbmxtzPYVCwZbQZWVljWq5cBzH2rtWV1dDoVDEnKfXLCwsBFD3h65hZ8d27doBAI4fP37BkfyWgDfUBMPtdsPn80EkEkEkEkGhUEAoFLLsovpbMzTCC9T1N/J6vdDpdJBIJJBKpYhGo5BKpSxgRJMoCCEQiURwu90IBALw+XzsvlSyoz5NGerRo0ehVqvZ8eaUvHXp0gU1NTVxvXPvuecelJSU4O9//zu6dOmCTz/9FO+99x7y8vLOK+TVnEqj+mOaGn+u69Dv3BrLXoBf+iYcNTU18Hg8zNvR5AdqbBqNBjabDQKBAIFAAMFgED6fDw6HA4QQhMNhuN1uSKVSeL1eRCIR9owrEokQCASgVqvh8XjYLx19bg0EAvD7/XHPoE0Fk9LT05GRkcFe5zPUzp0749ixY9i6dWuM56ytrcX69evxwgsv4B//+AdGjRqFlStX4qqrrsLs2bMB1C2Ngf95PYrX62VR6vT09EYbw4dCISYxkZycHKenQ6+Zk5MDoK6fcMOkDyrR0VINtS8U3lATDIPBAK1Wi5SUFFYFo1QqEYlE4Pf7IRaLYbfboVKpoFQqY9IIlUolVCoVioqKEI1G4fF44Pf74XQ6YbfbEYlE4PF4oNFooNFomJHr9Xrk5ubCaDTCarXGiSldbNQ3Go2ic+fOOHLkCL7++uu4paPX6wUQ761o5BsA7rzzTgCIUTM4ePAg/H4/655PBZjqyzVSzSL6+YEDB8Lv9+PgwYNszNy5cwH8T1iqR48eqKysjNmy+eijjyAUCnHDDTdc0HdvKfilb4JBl6RKpZJVwBiNRvj9fsjlcjidTlZ/GgqFkJycDK1Wy5L7PR4PpFIpIpEIkpKSWJmdQqGA2+1GUlISjh8/zmpPbTYbysvLEQgE2JK4ofTDxUZ9qZHOnTsXKSkpOHToEIC6YJLBYEBGRga0Wi1mzJgBjUaDbt264aOPPsLx48dxzz33AACysrLQoUMHzJs3D2lpacjIyMAjjzwCmUyGZ555BkBd53uj0YiJEyeyfOfHHnsMJpMJw4YNAwBMmTIFM2bMQP/+/TF//nyUlpZi/vz56NSpEwtuvfbaa1ixYgUKCgrw5ptvYs+ePdiwYQOGDx/eaktfvhVLAuF0OnHfffchNzcXKpUKarWaRXdpwrzZbEZaWhqqq6tRXV0NsVgMuVzOAlDBYBDJyckoLi5GMBhkZW3RaBTBYBCRSAQHDhxgOcXl5eUoLy+H3+9n5W+RSARff/01SkpKkJmZiW+//TZGVtDtdqNPnz7NbsXS1LPfhAkTsGDBAgB1mi933HEHzpw5wwrghw4dijVr1jDjsNvt6NOnDw4fPgxCCEwmE7744gv07NmTXfP06dMYOHAgEyzLycnB9u3bY/Ktd+/ejVtvvZVtYRUUFGDnzp0x3+Wzzz7D/fffD6fTCaFQiD59+mDLli2tZqi8R00w9Ho9OI5DamoqSktLEQwGoVAoIJFIYDQaIRQKceDAAahUKnAcB4FAAKfTCZ/PB6PRiEAgwLZkgDphI4FAgFAoxOpVgbrKmpqaGshkMshkMrjdbpb8QJ8HKRfrUZvjCzp37tyoVGN9dDod88ZN0aZNm7jn2Ib07t2bSU42xR133PGbtUwvBfwzaoLh9/uhVCrh9Xqh1+uh0Wjgdrvh9XphNBpRXl6O7Oxsti0TDAYhEAig1+uZtKBOp4PL5UI0GoVarUZaWhqys7NZATsthdNqtSyxX6PRwGQy4eqrr252MInn8pFwHtXhcGDNmjX47rvvcPbsWRbV69q1K4YMGRKT6vV7xGQysQoalUoFnU7HytbsdjsIITh9+jT0ej0zMqVSCblcDpfLBbFYjNraWiQnJ0MikUAoFMLhcLBoqFgsRnZ2NqxWK2pqakAIYfejyRUNDZEmPNR/z3N5SRiParFY8MADDyA1NRWvvPIKPB4PrrnmGiatuG3bNgwaNIilnv1ekUgk8Pl8bJlKW6vQ7CODwQCHwwG5XA6VSoWkpCTo9XpIJBLWO0kul8NgMMRcC6jzjDSSTEWekpOTodPpIJPJIBKJEI1GIZfLY+bEe9TWJ2E8apcuXXDfffdh7969KCgoaHSMz+fD2rVr8cYbb6CkpARPP/30ZZ7lpcfv9zMdT/pcKRAIIBaLEQwGYTAYkJKSAkIICCEsWES9KV0Si8ViBAIBRCIRKBQKqFQq+Hy+mG0eo9GIaDTKtm3qJ1TUhy8cb30SxlCPHDkSV17VELlcjtGjR2P06NFxFR6/F/x+P+vrG4lEEAgEWG5vZWUl9Ho98vLycPr0aZZGKJVK4Xa7mVf1eDwssV8ikUChUEAul4PjOFaIHo1GodPpWNCJphHKZDJYrdaYOfGG2vokzNL3fEZ6seOvFAQCASoqKiCXy5kqeygUQkpKCqtBpR0IDQYDSxGUy+UIh8OQSqVQKpUA6vZUw+EwbDYbioqK4PV6kZmZCYFAALfbDZFIxAxVo9EwQ6UJCBR+6dv6JIxHbUhZWRm+//57VFVVxf1iPP744600q0tPamoqvF4vXC4XK1FzuVw4efIk3G430tLSUFZWhkgkwsrR6BaMQqFAZWUlnE4n8vLyYDAYEA6HIZfLodFoIJVKWe1pcnIyS9Cv71WLiopi8ncB3qMmAglpqIsWLcJDDz0EiUSCpKSkuITq37OhnjlzBh06dEAoFEJRUREyMzPRpk0b+P1+qFQq1kWQEILa2loEg0GWDwyA7Zna7XYkJSWxnsC0cTdQlxMbCARgs9lYTStNnPD5fNBqtTFz4g219UlIQ33ppZfw0ksv4fnnn2+1TJDWQiqVwm63QyAQwOv1suJur9cLs9mMkpIShMNhlu7GcRwikQjrrZScnAxCCGvWTfsfSaVS+Hw+pKSksEqZzMxMttfq9XpRW1sLvV7fZK5v/fc8l5eENFSv14t77rnnD2ekQN0+atu2bSEUCllSAm1KZjAYoFarYTAYUF5ezp4naR6wTCaDQCDA1VdfDalUiqKiIsjlcrbP6na7Wb1pJBJBUVERLBYLotEo6x/kdrvj0gJ5Q219EtISJkyYwCry/2iIRCJmSCqViklNaLVapjNDG3PTjKX6vXuBuk6GlZWVUKlUSE9Pj0mgyMzMZHmvkUgERqMRHMfB6XTC4XDA4/Hg1KlTMXPiJS1an4T0qDNnzsTw4cOxYcMGdOrUKa5q/4033milmV163G43ampqoFQqWQ0qIQRKpZJJIdIsIrrHKpPJmMcE6qK9gUCAdcQnhEAikbCke5vNBqVSySpmaCI+IQRqtRo+ny9mTrxHbX0S0lBnzJiBb775hrW/aE51/u+FQCCAiooKZGRksA74tMueSqVi2Uoul4sVhCsUCuh0OkilUthsNmbYoVAINpsNarUaJpMJgUCAVcvQKDAtLqdtMSUSCSuypvCG2vokpKG+8cYb+OCDDzBu3LjWnsplR6lUsi4NdC+VtlWhbUTp0lipVDI9Gao7QyUZqaekQlJCoRA+nw92u531VKof6RUKhSxolZqaGjMn3lBbn4R8RpVKpbj++utbexqtAq0hzcnJYR4vHA5Do9GgqqqK1Z5eddVVyMjIYHWiDocDP/30E9RqNVwuF2vTQnv/0sbdSUlJEIvFUCgU8Pl8IIQgOTkZaWlp4DgOgUAgrgshn/DQ+iSkoT7xxBOYN29ea0+jVdBoNDAYDMzQgLrgUFVVFRQKBcrLy/Hrr7+ioqKC1aFSKYv8/Hz2PB8IBFhbl7KyMhw6dAgqlQpmsxlqtRpqtTqm35LX64VCoYBer+cNNQFJyKXv3r17sXXrVqxbtw4dO3aMCyatXr26lWZ26aEtO6VSKStPUygUrMsDrX4pLi6GzWbD1VdfjfT0dJw8eRKBQABnzpxB+/btIRAIUFpaCoVCgTZt2rAtGOplI5EIMjMzcfLkSdTU1DDv2vD5FOCXvolAQhqqTqfD7bff3trTaBVoAXhRURFkMhlro6LVamG1Wlm+rslkgkQiYcfos6tEImE5vrQfMN1Soe1cAMBmswGo62lrNptRVVWFs2fP4vTp07yQcQKSkIZav4vcHw2n08mKvmUyGQsClZSUIC0tjfXmDYVCqKiogFKpZDm/VMXtxIkT7NmVGibHcVCr1fjll19w9dVXM6OlSQ52ux3hcBgmkymu4z1vqK1PQj6j/pEJhUJMhNhutyMQCLAytmAwCI7jUF1dzTKUdDodFAoF659E9101Gg2MRiMEAgErHqcNwWhbl+rqakSjUfYZvV7P9lLrwz+jtj4JaaiVlZW49957kZaWxvRS6r9+79CoLSEEfr+f1ajSpATa7Z6WpoVCIXi9XjidTlZATvdE6+87C4VCZGRksNRMWhpHE/3rb9XUh89Man0Scuk7btw4FBcX48UXX0RqaurvPsmhPoFAAF6vF1KpFDqdDoFAAB6PB1lZWUyXRqlUwm63s5YptMM97WVLlcdpZQ1dRlNvLRQK4XQ6YTAYUFtbyzKXqD4Nbd1C4Ze+rU9CetSdO3di6dKlePjhh3Hrrbdi5MiRMa+W5uWXX2aBF/oym83sPCEEL7/8MtLS0iCXy9GvXz8cOXIk5hqBQACPPfYYjEYjlEolRowYgdLS0gueS01NDQsUicViJCcnw2AwoKSkBF6vl0Vs6ZLW5XLB4/Gw+lS5XM76Jvn9fgSDQSbR6PP5UFFRAYvFAqBOPDgQCLDuD6FQCFlZWU0aKr/0bT0S0lAzMzMv+/KqY8eOsFgs7PXLL7+wc7Nnz8Ybb7yBt99+G/v27YPZbMagQYNi9hsnTZqENWvWYPny5di5cyfcbjeGDx/OvFxzyc3NRXFxMVwuF1wuF9tOSU9PZxHd2tpaaLVaqFQqZGRkoE2bNtDpdKynks1mA8dxSEpKQlJSEgAw6YpIJAKdTodotE7nU6/Xw+/3w2KxsOQK2neJcrGG2hzFcQBYt24dzGZzjHpdfTEpXnE8wZg7dy6mTJnCup1fDsRiMcxmM3vRVi+EEMydOxdTp07F7bffjoKCAnz44Yfwer1YtmwZgLqsoIULF2LOnDkYOHAgunbtiiVLluCXX37B5s2bL2ge9BmQijuJRCJ4vV7YbDYEg0G4XC4IBAJUV1dDKBQiEAiwtiqEEJSWlqJdu3YsQESNPC0tDYQQZGRkwGazobS0FHq9HkqlEikpKcjMzIREIsHp06fjVgIXa6jNURzfunUrRowYgaysLCxbtgzbtm3D008/HVNyxyuOJxh33303vF4v2rRpwyKa9aHF0C3JyZMnkZaWBqlUip49e2LGjBnIy8tDYWEhKioqmBARUJfi2LdvX+zatQsPPvgg9u/fj1AoFDMmLS0NBQUF2LVrF4YMGdLsebRt2xaFhYVQqVSs+yDtwiASiZCVlcWCR1RJXCQSwWAwwO12Q6/X48yZM1AoFBCLxfD5fGy7hVbcpKWloVevXkwWo6amhinFlZWVtXhSfnMUx8eNG4ecnBzs3buXjevXrx/7/z+64nhCGipV17pc9OzZEx999BGuuuoqVFZW4tVXX8Wf/vQnHDlyBBUVFQDAusxTUlJS2JKqoqICEokEer0+bgz9fGMEAoGY50Gn04nq6mpkZGTA5XJBJpOxViopKSmIRCKorq5mBkwDS8FgENXV1ZDJZHA4HFCr1VCpVLDZbGzvtby8HF6vF6FQCGq1GqdOnWL1pzU1NfB6vZDL5TCZTHG/iJFIJGYJT/+/rKwsZs+VqsSdj4aK4+FwGCUlJejfvz+MRiNqa2uhUCjw2GOPYebMmQDOrzg+derUJhXHH3jgAaxcuRLDhg1rUnF83rx5WLFiBSZPntyk4vg333yDnTt38kLGlLFjx17W+1GlL6Dur33v3r3Rpk0bfPjhh+jVqxeA+PI6qpJ2Ls43ZubMmZg+fXrMsWnTpiEajTJtVKPRyLKS6B+LcDiMzMxMVixOm3NrtVrmadVqNcRiMaxWK3tPn9uo5gzdkjEajWxrx+PxxP3BacqjtpTi+NGjRwHULX9HjhyJ0aNHY9GiRZg1axbMZjOeeOKJP7zieMI8ozYUl23p8ReCUqlEp06dcPLkSRb9begZq6qqmOGYzWYEg0GWltfYmMZ4/vnn4XA42KukpAROpxMejwcqlYqVndEECOB/3o0Wd4dCIbbtEgwGWUkbNeBAIMA6QlANVbVajdTUVEgkEoTDYUQiEQSDQdjtdmaw9WlqH/Xo0aMoKSlhry+++OK8P1uqOF7/2Z3KPGZnZ2Pt2rW4++67sWHDBqSkpDSrSQCvOH4Zyc/Px4wZM9iyqDEIIdi0aROGDRsWI2jb0gQCARw7dgypqaksF3bTpk3sfDAYxI4dO1gAonv37uA4LmaMxWKJC1I0RCqVsuUifbndbiZZQXN9ae9emplEu+B7vV74fD6WAFFfOhGoM2Kackh7/kokEuh0OkgkEgB1RkKFjkOhECQSSbO3Z1pKcZyqeOfn58eMz87OZn/8eMXxBGH79u04cOAAcnNz0bNnTzzyyCP45z//iTlz5uDvf/87br/9dqSlpWHChAkYMWIECyi0BE8//TR27NiBwsJC/PDDDxg1ahScTifGjh0LgUCASZMmYcaMGVizZg0OHz6McePGQaFQYMyYMQAArVaLCRMm4KmnnsKWLVtw4MAB/N///R86deqEgQMHXtBcaK/dQCDA0gSpnAUhhLVdoVlEfr8f4XCYnacRUOpBqbcEwKQagbqAXCQSYUn/tHN+/f3j+nO6mKhvNHpuxXHaG4oubyklJSUsj5lXHE8Q2rVrh5UrV6K0tBQrV67Et99+i127djHdz65du+L999/HTTfd1OLLj9LSUowePRo1NTVITk5Gr169sGfPHtYE7Nlnn4XP58PEiRNhs9nQs2dPbNy4MSYn9s0334RYLMZdd90Fn8+HAQMGYPHixRec8piUlAS1Wo1wOAylUgm/388ak1GBKJVKBbfbzfYTaaUMTYDIzc1FaWkpE5KiIsfUq4bDYSgUCjgcDnAcB61Wy4JTpaWlcdqhFxv1PZ/iOABMnDgRs2fPxn333YexY8fiP//5DywWC+bPnw+AVxznFccTCKfTiddff50tbWmRN8dxLJ0wPz8fZrMZbrebLZPpXmhZWRlLVjAYDKiqqoJSqWSVNLQZmsvlgtfrZTo1J06cQE1NDaRSKSorK2EymfDMM88wxfHp06fHJEH4/X5MmzatRRXHAWD8+PFYunQpQqEQZDIZnnzyScyYMYOd5xXHeRIGrVbLeiQFAgGmNJ6cnAy5XI6ysjKIRCIEAgEolUqYzWbWfNvv90Ov17O2K4QQOBwOlsPLcRyi0ShKSkqgUCjg8Xig1Wqh0WhYwXo4HI7bp75Yj9pcX7Bo0aJzljj+kRXHeUNNMHQ6HctEkkqlCAQCsNvtUKvVcLvdyM7Ohk6ng9vthslkgtvtRmVlJYLBIJKTk/Hrr7/CbDazRAiRSMTah9KglFarhdlsRlFREdxuN4xGI3Q6HSwWC4xGI184noDwhppglJeXo23btiypQSaTsedKu92Odu3asS0bq9XKtl7oc2hubi6LWspkMgSDQSgUCggEAjgcDpjNZhw7dgyVlZVQKBSw2WxMwtLv96OysjIuKsobauuTMFFfnjoqKipQU1PDpCeUSiUrHJfJZFAqlfB6vczTUg1U2gmCBrjC4TCTT6TLaNpalFbUKBQKmM1mJpwskUjQrl27JiUt+OqZ1oP3qAmGWq1m0hVUXFgulyMQCEAmk7EACFUNp8tZKoVBjdjv90OhULCEiGg0yoJUBoMBKpUK1dXVbKvHZrOx6hmqy0qhCQ/13/NcXhLSo27YsAE7d+5k7+fPn49rrrkGY8aMicv++b1Bt0qo4BNNZhCLxSCEoLKykpVeqVQqVreqUqmYt3U6nWzpTJXdaPYP3fynASsaeAqHw/B4PKioqIiLbPIetfVJSEN95plnWLL3L7/8gqeeego33XQTzpw5g8mTJ7fy7C4tdImq0WiYzqnNZoPNZmPGF41GoVQqQQhhLVuoN5VKpVCr1WzflHYfpP2UBAIB5HI5QqEQ2+ahS2HaCaKhx+QNtfVJyKVvYWEhS/j+7LPPMHz4cMyYMQM//fQTbrrpplae3aVFq9Xi1KlTKC4uRkpKCnQ6HSKRCPOgQqEQWq0W5eXl8Pl8LHOJBo8UCgUCgQCsVis4joPb7YZCoYBCoYDdbofZbEZ1dTXTqKGCVCqVClarFcnJyTH1nQAfTEoEEtKjSiQSFgjZvHkzSxEzGAxxrSx/b5SUlMBqtUKhUMDpdCIQCECj0bBu+bSlikQiQUFBAbKzsyGVSllSRFVVFaxWKyQSCatfzczMhFqtZh0f6JJZIpGw8jkarLLb7SwPmMJ71NYnIT3qDTfcgMmTJ+P666/H3r178emnnwIATpw4wdK8fq8YDAbm8bRaLRwOB/x+P5RKJZxOJyoqKhCNRmEymVBWVgapVAoALEspKSkJTqcTGRkZKC8vZ5U5tLWJUChEZWUlXC4X7HY7kpKSkJycDJVKhaqqKpw5cyau0Jv3qK1PQnrUt99+G2KxGKtWrcK7776L9PR0AMDXX3+NoUOHtvLsLi0WiwUCgQCEELjdblitVng8HvZMWllZCZvNxhL1abdBoC5VjzbSLisrg91uh1gshk6ng0qlgsfjwfHjx5mmaqdOnaBUKpmRi0QitG/fPq6mk/eorU9CetSsrCysW7cu7vibb77ZCrO5vESjURw/fhwpKSms2NvpdCIYDCIlJQVpaWmQSCRwuVwxbUFoBJcuaemeKxVGlkgkUCqVqK2thcFggMfjYYXitFdwdXV1TElY/TnxHrV1SRhDdTqdbKP9fM+hzUkEv1IRCoUssuvxeFgj7fp7pUKhkC1bOY5j1TPhcBjJyckoLCyE2Wxm0hcCgYC1A6XPp1QBTiqVQiwWs89XV1fzmUkJSMIYql6vh8VigclkYm0aG0Jbm1xoC84rCWo8VKSYfmeas2u1WlnjM9rLVywWsyJyrVaLQCCAmpoapkxODZ22Hq2urobX62XF5jRbiaq9NUzK5xMeWp+EMdStW7ey2sStW7f+obrj18ftdsPn8yEUCrFO+DTPly6BzWYz5HI5KisrWcE3TSmsqqqCWCxm9apUdyYYDLJsprKyMmg0GkilUlgsFtYlwu12M83V+vAetfVJGEPt27cv+//6bSL/aGi1Whw7dgxerxcymQx6vR4Gg4F5UtoN32AwsH5Hfr8fKpWK9Ueqra1FcnIySktLodPpIJfLWUPvnJwcRCIRtuQF6pbbIpGI9UPiq2cSj4SM+r744ouNLm8dDgdGjx7dCjO6fJSWliI5OZml9fl8PrjdbiiVSgB1UWGVSsXOBYNBOJ1OlJeXQyQS4ezZs7Db7aisrIRIJGKtQ5VKJXQ6HfR6PfR6PRQKBcvrpa/09HRkZmbCaDTGzImP+rY+CWmoH330Ea6//nqcPn2aHdu+fTs6dep0WbvntwbUsxmNRubpZDIZyyQyGAys4bdAIIDH42FVNBzHISMjA1qtFllZWUhLS2MNuGlTtG+//Zbtq9KgnM1mg8ViQXl5eaM/X95QW5+ENNRDhw4hJycH11xzDd5//30888wzGDx4MMaNGxeTrP97RCKRsP1Tp9PJmmQbDAbk5+dDJBLBbDYjEolAr9cjPz8fqampTLWturoaJpMJHo+HGTqN/spkMqapGo1GsWfPHqZjk5KSwrxtUymEvKG2HgnzjFofrVaL5cuXY+rUqXjwwQchFovx9ddfY8CAAa09tUsOx3FM97SmpobVkNbU1ECv14PjODidThiNRoTDYZSXl8NqtbLPms1mlJaWok2bNnA6nawhN71uNBrFqVOnWBO0SCSCmpoaRCIRqNVqqNXquFaZ/DNq65OQHhUA5s2bhzfffBOjR49GXl4eHn/8cfz888+tPa1LTjQaZYXiYrGYae+EQiFUV1fHBICkUilUKhUyMzPRrl07+P1+cByHlJQUJrmoUqmg1WrBcRzb6mnTpg1rxE2j6+FwGFarFXa7vclgEu9RW4+ENNRhw4Zh+vTp+Oijj7B06VIcOHAAffr0Qa9evTB79uzWnt4lhWYV0a0VugylzbRra2uhUqlQWVnJOuTJ5XKIRCL4/X7W+xeoa/BN90BpwoRQKIxpMxoKhVh2E40a84aaeCSkoYbDYRw6dAijRo0CUCdh/+6772LVqlW/+zRCoVAIj8fD8ng9Hg/zlEqlkmUR+f1+2O12NpY2M6PnxWIxq0mNRqOsgz4dS6Ux5HI5hEIh8+RisTgu4t6UpAXP5SMhn1HrS0PU5+abb44RGP49Eg6Hmc6NUChEJBJBIBBAcnIyZDIZzGYzXC4X0tPTWeE4NUqFQsEamkWjUVYUTiPJXq8XXq+XJTsAYPKMtPhcKpW2eLtQnosnIT3quWi4x/d7gzYZo6mUYrGYpfjV1taitrYWhBCkpKQgPT0dSUlJkEgkzAuGQiH4/X74/X4UFhaiqqoK1dXVKC0tRW1tLTO6pKQk+P1+2Gw2SCQSqNVqhEIhcBwX51EvdunbXMVxSvv27SEQCJjaG4VXHE8wIpEI/vWvf6FHjx4wm80wGAwxr98zGRkZEAqFCAaDcToyCoUCkUgEhBCUlJTA6/UyZXKdTgdCCCoqKlgjtNzcXOTm5iI1NZWpiyuVSlRUVOD48eNo164dDAYDgsEgIpEIax/aUH/mYg21OYrjlOeffx5nz55ttCM9rzieYEyfPh0LFizA5MmT8eKLL2Lq1Kk4e/Ys1q5di5deeqm1p3dJCQaDyM7ORmlpKeudRCUpMjMzWfCHVtHQ51KFQgGlUgm5XI7q6mpoNBoQQphIFNWy8Xg8SEpKQigUYk2+8/PzYbPZcOrUKdZUrT4Xu/RtjuI4APz444+YPXs2Vq9eHdel/o+uOJ6QHnXp0qV4//338fTTT0MsFmP06NFYsGABXnrpJezZs6e1p3dJ8Xg8OHnyJMrLy9k+qNfrhcPhwE8//YSzZ8+yGlLasMxut6O8vByhUAhWqxVGoxGnT5+GQCBgejPUgF0uF4qKiuDxeHD27FkmCkVL6qxWKxP9pTTlUcvKylBaWspezW2T01BxHKh7Nh80aBBGjhzZqOc6n+I4gCYVx+t/vinFcQBYsWIFADSpOB6NRlst4SYhDbWiogKdOnUCUCfJ53A4ANSpda1fv741p3bJoV4tMzMTJpMJ0WiUqa7V1NQgPz8ffr8fZWVlOHPmDEKhEIxGI9RqNU6cOAGr1YrKykpwHMeeTy0WC6xWKwwGA7RaLfLy8pjuDJXHqK2thUKhYP2V6tOUoXbo0AGZmZnsNWLEiPN+v2g0XnEcqAsUCoVCrFq1qtHPXW7F8Ya9jesrjrcGCbn0zcjIgMViQVZWFvLz87Fx40Z069YN+/btY9HK3ytqtZr1M4pEIhAIBNBqtRCJRGxJnJqaimAwyJLr61e96PV6JgqlVCrZ/ivdqgkGg2ypHIlE4HQ6EQ6HwXEc/H4/PB5PnGdsaul79OjRGKNuTkE/VRzfvXs3O7ZkyRJs3rwZ+/bt+03Lyj+C4nhCGuptt92GLVu2oGfPnnjiiScwevRoLFy4EMXFxWyZ8nuFihhT5TWFQsH0UGUyGfMYcrkcfr8/JlopEAig0+lY3i9tvwLUJVIEg0HWvYF2kaBRzVAoxLKhmlvmlp6efkHdNppSHP/ss88QjUbRvXv3mPFr165lUe/6iuP1vepvURxvWHjAK47/RmbNmoUXXngBADBq1Cjs3LkTDz/8MFauXIlZs2a18uwuLTKZDNFolKlwazQaSCQS2O125mVdLhfr3kBbitIEBo7jmKEJBALWKYLKNIpEIhBCmNwFNfxAIACgrtNGc5e+zSUaPbfiOA0g1X8JhUL07NkTX375JQBecTwhPWpDevbsGSNW+3umoKAA5eXlzPPRyK1QKIRarYbNZoNMJoNOp2Nd7amB0pxgGkSiHSIcDgeLJtN2K1qtluUOUy1VmpnUsK/vxbZiOZ/ieNu2bdG2bduYzwgEAqSmpjKl8D+64vgVYah/JPx+P7RaLUpLSxGN1klXaDQaKJVK9lzq9/tZs2ya10sIQXp6OqxWK1Mqp8s1msxQXFwMpVIJvV4Pp9OJaDTK9k+FQiEsFgtTkqvPxW7PHDlyBEBsxBaIVxw/H99//z369OmD5557jimOf/311zF/WPbs2YOBAwfigQceAPA/xXGKRCLBtm3bcOutt2LMmDExiuOUrKwsrFixAvfffz9uv/12CIVC9OvXD59//vkFfe+WhDfUBIN2Y8zOzmb9i6xWK7Kzs2EymeB0OqHT6RAMBnH11VdDo9EwL0ifQU+dOoXk5GSo1WpEo1GmoQrU7Wm63W6IRCJkZGSgqqoKLpcLQqEQOp0OPp8PJSUlMXO6WEP9LbnBVNSqPn9kxfGEfEZtSb799lvccsstSEtLg0AgwNq1a2POE0Lw8ssvIy0tDXK5HP369WMegBIIBPDYY4/BaDRCqVRixIgRKC0tjRljs9lw7733QqvVQqvV4t57740LSDQHnU6H8vJyVFdXw2q1wuVyIRKJIBQK4fjx42ypGggEUFxcjMLCQpSWlqKiogJerxdt2rRBTk4OkpKSYLPZEAwGmZyFTqdDXl4etFotS84PhULw+XwsPVEqlSIrKytmTnz1TOvzuzdUj8eDLl264O233270/OzZs/HGG2/g7bffxr59+2A2mzFo0CC4XC42ZtKkSVizZg2WL1+OnTt3wu12Y/jw4TE5sWPGjMHBgwexYcMGbNiwAQcPHsS99957wfM9deoU84I6nQ7Z2dlITU1lKYS07afL5WJtWuhLKBTixIkTCAaDsNvtkEqlTMLi9OnTsFqtKC4uRjAYBCGELYVVKhULUjWWuMAbauuTkEvfcePG4f7774+LDv4Whg0bxgIJDSGEYO7cuZg6dSpuv/12AMCHH36IlJQULFu2DA8++CAcDgcWLlyIjz/+GAMHDgRQt++XmZmJzZs3Y8iQITh27Bg2bNiAPXv2sKDX+++/j969e+PXX39lm+XNQSKRMGW1+rWjNEPJaDQiGo2ybg0ikQhyuRzRaBQ+n4+lEno8npjn2XA4DIfDwfRlk5KS2P3EYjFLVTxXUn799zyXl4T0qC6XC4MHD0bbtm0xY8aMuJS2lqKwsBAVFRUsvA/UFVv37dsXu3btAgDWSKz+mLS0NBQUFLAxu3fvhlarjYlM9+rVC1qtlo1pjEAgAKfTGfMKhUKwWCysyLu+vgyN/tIMI7lcDo/Hg0AgwAw4HA6jsrKSRXIrKipQXl7O1PEUCgU8Hg+rLvF6vaw7fyAQYF3068N71NYnIQ31s88+Q1lZGR599FGsXLkSOTk5GDZsGFatWtXohvZvhdZ9pqSkxBxPSUlh5yoqKiCRSKDX6885xmQyxV3fZDKxMY0xc+ZM9kyr1WpZ0j3dKgHqNuvplotYLIbH42H9fCUSCaLRKNtPdTgc8Hq98Pv9zAjdbjcrLheLxUzqwufzsa0G6rVpHSsvEpV4JKShAnVLsyeeeAIHDhzA3r17kZ+fj3vvvRdpaWl48skn2dZDS9AwdYwmvJ+LhmPOJcHRFM8//zyTRXQ4HCgpKQHHcTAajZDL5SCEsBRA2qOXPn/a7XbWdoUQwhqcRaN1nfNpyiBt2i2VShEKhaBQKCCVShEMBtm2D22gRgNMWq02Zp68obY+CWuoFIvFgo0bN2Ljxo0QiUS46aabcOTIEXTo0OGi27LQusuGXq+qqop5WbPZjGAwyJ7tmhrTWLi/uro6zlvXRyqVQqPRxLyokdEsokgkwrYqqJYpVRKnz59GoxFpaWlo06YN5HI5tFota7+SkpICs9kMqVQKh8MBu90Oo9EIqVQKg8HAvDJtURoOh+O2RvhWLK1PQhpqKBTCZ599huHDhyM7OxsrV67Ek08+CYvFgg8//BAbN27Exx9/jFdeeeWi7pObmwuz2RzT+iUYDGLHjh2skLh79+7gOC5mjMViiSk27t27NxwOB/bu3cvG/PDDD3A4HDEFyRcCbZBtMpmg1Wrhcrlw+PBhCIVCpiFDpRNpk+2qqir4/X5UVVWxAnSbzQar1QqBQIDs7GwmtRgOh1FYWAin0wm/349oNAq1Wo20tLS4gm7eo7Y+CRn1TU1NRTQaxejRo7F3715cc801cWOGDBkSV4rUGG63G6dOnWLvCwsLcfDgQRgMBmRlZWHSpEmYMWMGS2ObMWMGFAoFxowZA6Cux/CECRPw1FNPISkpCQaDAU8//TQ6derEosDt27fH0KFD8cADD+C9994DAPztb3/D8OHDLyjiC9R1yi8tLYVKpYJYLGYR2/Lyclx77bWw2+0oKChAdXU15HI5KioqEA6H0bFjR3i9XsjlcggEApSUlECn00Gr1bLnWVp0npGRgePHj7OWozQ9USwWMy9dH5q8X/89z+UlIQ31zTffxJ133hn3C1MfvV5/3gwUoK5rwJ///Gf2fvLkyQDqCoEXL16MZ599Fj6fDxMnToTNZkPPnj2xcePGmMT0N998E2KxGHfddRd8Ph8GDBiAxYsXx1SZLF26FI8//jiLDo8YMaLJvdtzYbFYWC6v3W5HbW0tJBIJDAYDysrKkJSUhIMHDyIzMxOBQIAl7dfU1CA1NRUKhQIVFRWQy+VMihGoW6V4vV6mOp6RkYFTp06xjCXa1VAmk8VVxDTcrvk9y14mKgLCP3AkDE6nE8888wwOHjyI2tpaZGdnIxqNwmKxMBHirKwsJCUlwWQysR5SdGuFRnUrKiqQnJwMQggIIawJN22y7XA4oNVqIZPJWGYSbU1aUVEBj8eD119/HSUlJcjMzET79u1j/ihFIhEcO3YMDofjdy0qnUgkpEf9I0O7CAKA1WqFRCKBTqeDRqNBJBJhNaNyuRwmk4mpvul0OuTk5MBut0OlUiE5OZllG4XDYfj9ftbhUK1Ws2JxGlV2uVyoqKiAzWaLqxDhl76tD2+oCUZ5eTmToqB7qHK5HG63G9FoFHK5HMnJyVCpVKzvL+16TwWJOY5DMBhkRk+Lr2lBOT1OdVKpEhwtRucNNfFIyKjvHxnaMRAAa69Cl7ahUAgajYbl+NIuhLQw3GazsVxgt9sNl8vFEiVoVhOtXQ2FQggGg5BKpZBKpew+tH1Lffiob+vDG2qCYTKZmFQiXapKpVLI5XKoVCooFArY7Xb2XEnHUUOjS2TqQeVyOau4of8ViUSsaTf1ntTLchwX1ymf30dtffilb4JhMBig1+shk8ngdrvh8/ng9/uhUCiYkebk5LASOK1Wi/T0dKSlpUGn00EqlcJkMsFut0MoFMJut0MsFrOtF1o0ThXePB4PvF4vxGIxkpKSEIlEGt2eOdd7nksP71ETDLFYjJqaGpw5cwYWiwUCgQAGgwHhcBgCgQDV1dU4ffo0Uwygeb1yuRy1tbU4ffo0TCYTqqqqYLVaWUmbXC5HRkYG69/rcDig0+ngcDgQiUTYNo9IJIrLpuKXvq0P71ETDK/Xy1Tb6J6mVCpFNBply9e8vDzI5XJwHAeFQgGBQACbzQaTyQSxWIzKykpcddVV8Hq9zKholQ39XDAYRElJCZKSkuDxeFjXh+rq6karZ871nufSw3vUBIN2syeEsHTA6upqRKNROJ1O5ObmorS0FFKpFOFwmKUQhsNhBAIBFiBSqVQIBAKsmoYQwrKxaPqhVCpFRUUFS/AXiUTQ6XRxDap5j9r68IaaYKSnpzNxYlodYzQawXEcq5rR6/U4c+YMCCGsSiYcDsNoNLLWLYFAAJ06dWK9d7VaLXw+H06cOAGdToc2bdpAIBAgMzMT6enpMYn+jW3P8IbauvCGmmBUVFRAp9PB6XSy3rJUB1WlUqG4uJhFeWnlCxUkJoSwjoMOh4O1XVEqlay+Va1WswCVUChkCQ9isRhSqZQ15K4Pb6itD/+MmmDQzCMqpUjL08LhMPR6PZNdpMtX+jxbU1MDqVTKosV06SsQCMBxHDNqn8/HtmmUSiUIIcwjBwIB1sy7PvwzauvDe9QEg+bkymQypKSkQKlUMg9I0/9cLhcCgQB8Ph/kcjkUCgVLuKfF3zTZnnYwDAQCcLlcqKmpYWpwfr+fpRfSrvk0EaI+F+tRzydk7PV60bNnTyb5KBKJ0KZNG/z0008x1+GFjHkShrS0NJZpRAhhv5TRaBSBQABWqxXV1dWsHE0mk0GpVMJgMLDkBo7j4PP5WFlbJBKBzWZDVVUVE4YC6nRBa2pq4HA4IBAIkJSUBLFYHJfQcLEJD+cTMq6pqcGpU6fw8MMP46uvvsJ7772HyspK3HjjjTHX4YWMeRIGmr5Hm5nRv/ocx8Hr9cJutyMzMxNarRZKpZIpsCmVShw/fhz5+fnQarUsi8lms8HtdoPjOLRt25ZFglNSUlBTU4NwOAyDwYBQKISysjKIRKI4VfeGub4XaqjnEzLOyspirWQoHMdh3Lhx2L17N3r37s0LGV/2O/I0CSEELpcLSqUS0WiUdXDwer1s68RgMLAIrUAggFQqhUQiYdU0Op0OSqUSVqsVR48eRWlpKUvAt9vtcLlcKC4uRlFREUwmE5ODoEn5VJiKzgdo+WBSY0LGDaGtbTIzMwHwQsa8R00gXC4XlixZct5xDZ/dLhX0ua0pj1pWVhbTrJv2fToX0WjjQsb1sdvtmD59OnJycpgxX24h44ZdJesLGbdEv+kLhTfUBCItLQ1Hjx5Fhw4dUFJSclmKsp1OJzIzM2PuRz27SqUC0LShdujQIeZaffv2jRFkaozGhIzr4/V60b59exBC8N133zXrO/BCxjyXFaFQiPT0dADN804tScP7abVa5i2bMtQLVRxvSsiY4vV6kZ+fD7vdjkOHDsUsjf/oQsYgPAmFw+EgAIjD4Wj1+9FzHMcRiUTCXhzHXdAcI5EI6dSpExEKhWTjxo2NjvF4PCQ1NZVIpVJy9OjRuPNFRUUEAJk0aRI7duDAAQKAvPrqq4QQQr788ksCgHzwwQdszIIFCwgA8tVXXxFCCJk+fToBQA4cOMDGPP744wQAKSkpIYQQcs899xCBQEBcLhcbM3ToUCIUCkkkEmnWd25peENNMBLRUEUiERGLxewlEokuaI4dO3YkAMjcuXPJzz//zF5Wq5UQQojP5yNms5mIRCLy6aefxoypbywdOnQgIpGIzJ49myxbtozo9Xoik8lIIBBgY4xGI5HJZOT9998n77//PpHJZMRkMrHzgUCASKVSotfrybJly8js2bOJSCQinTp1YmOKioqIUCgk2dnZZPXq1eTZZ58lAMjw4cMv+OfbUvCGmmD4/X4ybdo04vf7W/1+1FAFAgERCoXsJRAILshQATT6mjBhAiGEkO+++67JMW+++Sa7js1mI506dWL3N5lMZM+ePTH3OnXqFMnJyWGfz8nJIWfPno0Zs2vXLmIymdh369SpU9x3WbVqFdFoNAQAEQqFpF+/fq3mTQkhhO9CyNMktMC8KfguhJcPfh+Vp0kkEgmT/WiI2Wxme7A8lx7eo/KcE7/fH5f7C9QZ8bkapPO0LLyh8vBcAfBLXx6eKwDeUBOId955B7m5uZDJZOjevXuzM3MaMnPmTFx33XVQq9UwmUy49dZb8euvv8aMGTduHCsFo69evXrFjAkEAnjsscdgNBqhVCoxYsQIlJaW/ubvx/Pb4Q01Qfj0008xadIkTJ06FQcOHMCNN96IYcOGobi4+IKvtWPHDjzyyCPYs2cPNm3ahHA4jMGDB8Pj8cSMGzp0KCwWC3t99dVXMecnTZqENWvWYPny5di5cyfcbjeGDx/Oi0S1Bq22McQTQ48ePchDDz0Uc+zqq68mU6ZMuehrV1VVEQBkx44d7NjYsWPJyJEjm/yM3W4nHMeR5cuXs2NlZWVEKBSSDRs2XPSceC4M3qMmAMFgEPv372eSjZTBgwdj165dF3192v6zYZ3p9u3bYTKZcNVVV+GBBx6IETDev38/QqFQzJzS0tJQUFDQInPiuTB4Q00AampqEIlE4hpfp6SkoKKi4qKuTQjB5MmTccMNN6CgoIAdHzZsGJYuXYqtW7dizpw52LdvH/r3749AIACgrsmaRCKBXq9v8TnxXDh89UwC0bDMihDSrBKuc/Hoo4/i0KFDcQXPd999N/v/goICXHvttcjOzsb69etx++23N3m9lpgTz4XDe9QEwGg0QiQSxXmqqqqqOC97ITz22GP44osvsG3btnN2UwCA1NRUZGdns3Ius9mMYDAIm83WonPi+W3whpoASCQSdO/eHZs2bYo5vmnTppimW82FEIJHH30Uq1evxtatW+O6IjSG1WpFSUkJUlNTAQDdu3cHx3Exc7JYLHGNwHguE60by+KhLF++nHAcRxYuXEiOHj1KJk2aRJRKZVzlR3N4+OGHiVarJdu3bycWi4W9vF4vIYQQl8tFnnrqKbJr1y5SWFhItm3bRnr37k3S09OJ0+lk13nooYdIRkYG2bx5M/npp59I//79SZcuXUg4HG6x783TPHhDTSDmz59PsrOziUQiId26dYvZTrkQ0ETJ2KJFiwghhHi9XjJ48GCSnJxMOI4jWVlZZOzYsaS4uDjmOj6fjzz66KPEYDAQuVxOhg8fHjeG5/LA5/ry8FwB8M+oPDxXALyh8vBcAfCGysNzBcAbKg/PFQBvqDw8VwC8ofLwXAHwhsrDcwXAGyoPzxUAb6hXGNu3b4dAIIjTRrkQfv31V5jNZrhcrpabWCNUVVUhOTmZKanx/Hb4zKQrjGAwiNraWiZt/1sYNWoUunTpghdffLGFZxfP5MmT4XQ6sWDBgkt+r98zvKH+wSgtLUVeXh7OnDlz3tK3luCXX35Bjx49UF5eHleEztN8+KXvb6S6uhpmsxkzZsxgx3744QdIJBKmgN0Yzz33HK666iooFArk5eXhxRdfZFKBhBAMHDgQQ4cOZdKGdrsdWVlZmDp1KoD4pW9RURFuueUW6PV6KJVKdOzYMa5JWX1WrFiBLl26xBjp4sWLodPp8M0336B9+/ZQqVSs8Rll3LhxuPXWWzFjxgykpKRAp9Nh+vTpCIfDeOaZZ2AwGJCRkYEPPvgg5n6dOnWC2WzGmjVrmvmT5WmU1qsHuPJZv3494TiO7Nu3j7hcLpKfn0+eeOKJc37mH//4B/n+++9JYWEh+eKLL0hKSgp57bXX2PnS0lKi1+vJ3LlzCSGE3H333eTaa68lwWCQEELItm3bCABis9kIIYTcfPPNZNCgQeTQoUPk9OnT5Msvvzxn1c3IkSPjmqgtWrSIcBxHBg4cSPbt20f2799P2rdvT8aMGcPGjB07lqjVavLII4+Q48ePk4ULFxIAZMiQIeSf//wnOXHiBPnHP/5BOI6Lq7C56667yLhx48778+RpGt5QL5KJEyeSq666ivzlL38hBQUFxOfzXdDnZ8+eTbp37x5zbMWKFUQqlZLnn3+eKBQK8uuvv7JzDQ21U6dO5OWXX272/bp06UJeeeWVmGOLFi0iAMipU6fYsfnz55OUlBT2fuzYsSQ7OztG0axdu3bkxhtvZO/D4TBRKpXkk08+ibn+k08+Sfr169fsOfLEw/dMukj+9a9/oaCgACtWrMCPP/54Xj2WVatWYe7cuTh16hTcbjfC4XCcItqdd96JNWvWYObMmXj33XfPqXL9+OOP4+GHH8bGjRsxcOBA3HHHHejcuXOT430+X6NzVCgUaNOmDXufmpoa05UQADp27Aih8H9PSykpKTEN00QiEZKSkuI+J5fL4fV6m5wTz/nhn1EvkjNnzqC8vBzRaBRFRUXnHLtnzx7cc889GDZsGNatW4cDBw5g6tSpcSJMXq8X+/fvh0gkYj2MmuKvf/0rzpw5g3vvvRe//PILrr32WsybN6/J8UajMa4PEgBwHBfzXiAQsOfkc41p7Fg0Go05Vltbi+Tk5HN+D55zwxvqRRAMBvGXv/wFd999N1599VVMmDABlZWVTY7//vvvkZ2djalTp+Laa69F27ZtGzXup556CkKhEF9//TXeeustbN269ZzzyMzMxEMPPYTVq1fjqaeewvvvv9/k2K5du+Lo0aPN/5ItwOHDh9G1a9fLes/fG7yhXgRTp06Fw+HAW2+9hWeffRbt27fHhAkTmhyfn5+P4uJiLF++HKdPn8Zbb70VFw1dv349PvjgAyxduhSDBg3ClClTMHbs2Ea9IFAnO/HNN9+gsLAQP/30E7Zu3Yr27ds3OYchQ4Zg9+7dl02Wgq4OGjYX57kweEP9jWzfvh1z587Fxx9/DI1GA6FQiI8//hg7d+7Eu+++2+hnRo4ciSeffBKPPvoorrnmGuzatSsm6aC6uhoTJkzAyy+/jG7dugEApk2bhrS0NDz00EONXjMSieCRRx5B+/btMXToULRr1w7vvPNOk/O+6aabwHEcNm/efBHfvvl8/vnnyMrKwo033nhZ7vd7hU94+APyzjvv4PPPP8c333xzye/Vo0cPTJo0CWPGjLnk9/o9w0d9/4D87W9/g81mg8vlglqtvmT3qaqqwqhRozB69OhLdo8/CrxH5eG5AuCfUXl4rgB4Q+XhuQLgDZWH5wqAN1QenisA3lB5eK4AeEPl4bkC4A2Vh+cKgDdUHp4rAN5QeXiuAP4fQm3hFwPIWesAAAAASUVORK5CYII=",
"text/plain": [
"