{ "cells": [ { "cell_type": "markdown", "id": "27d7d8f2-a943-481c-be10-365391d39458", "metadata": {}, "source": [ "# W to Graph Migration Guide\n", "\n", "Author: [Serge Rey](http://github.com/sjsrey)" ] }, { "cell_type": "markdown", "id": "57537223-f1c3-4771-b8f0-f7b32962e995", "metadata": {}, "source": [ "## Introduction\n", "\n", "Beginning in the fall of 2023, the PySAL project released a new `graph` module that offers a modern implementation of spatial weights. This module's [Graph](../../generated/libpysal.graph.Graph.html) class is set to eventually replace the [W](../../generated/libpysal.weights.W.html) class, which has been the cornerstone for spatial weights in PySAL for the past 15 years. The `W` class has significantly contributed to the library's success, but as the scientific landscape evolves, new opportunities necessitate updated interfaces and designs for spatial weights.\n", "\n", "While the application programming interfaces (API) of the `W` and `Graph` classes are similar, there are important [differences](../../migration.rst)\n", "to consider when transitioning from weights-based resources to graph-based implementations.\n", "\n", "This guide is designed to provide users with an overview of migrating from the `W` class to the `Graph` class.\n", "\n", "Beyond the specifics that we outline below, it is important to note two utility methods are available to convert between the two classes:\n", "\n", "- `Graph.to_W()` will generate a `W` instance from a `Graph` object\n", "- `Graph.from_W()`will generate a `Graph` instance from a `W` object\n", "\n" ] }, { "cell_type": "markdown", "id": "2d4e2096-0b7c-4acc-9875-32321256693b", "metadata": {}, "source": [ "## Imports\n", "To access the `W` and `Graph` class, use the following imports:" ] }, { "cell_type": "code", "execution_count": 1, "id": "f46293f2-1362-4e44-840f-eef77a67285d", "metadata": {}, "outputs": [], "source": [ "from libpysal import graph, weights" ] }, { "cell_type": "markdown", "id": "c5d73325-f533-466d-8108-ef034705fa58", "metadata": {}, "source": [ "## Example Data Set\n", "\n", "To illustrate the migration from `W` to `Graph` we will utilize a built-in data set from `libpysal`. In addition to the relevant `libpysal` modules we will also import the other packages needed:" ] }, { "cell_type": "code", "execution_count": 2, "id": "a2254e95-de33-4856-bc71-2817e52af346", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Last updated: 2024-07-18\n", "\n", "Python implementation: CPython\n", "Python version : 3.12.2\n", "IPython version : 8.21.0\n", "\n", "libpysal: 4.2.3.dev1352+gcfa4e0ce\n", "\n" ] } ], "source": [ "%matplotlib inline\n", "\n", "import geopandas as gpd\n", "import pandas as pd\n", "import seaborn as sns\n", "\n", "from libpysal import examples\n", "\n", "%load_ext watermark\n", "%watermark -v -d -u -p libpysal" ] }, { "cell_type": "code", "execution_count": 3, "id": "8c3bbda5-833a-4e73-82cc-eb6cfc906856", "metadata": {}, "outputs": [], "source": [ "dbs = examples.available()" ] }, { "cell_type": "code", "execution_count": 4, "id": "acd84693-21e0-49fc-b878-840ba66840e9", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "sids2\n", "=====\n", "\n", "North Carolina county SIDS death counts and rates\n", "-------------------------------------------------\n", "\n", "* sids2.dbf: attribute data. (k=18)\n", "* sids2.html: metadata.\n", "* sids2.shp: Polygon shapefile. (n=100)\n", "* sids2.shx: spatial index.\n", "* sids2.gal: spatial weights in GAL format.\n", "\n", "Source: Cressie, Noel (1993). Statistics for Spatial Data. New York, Wiley, pp. 386-389. Rates computed.\n", "Updated URL: https://geodacenter.github.io/data-and-lab/sids2/\n", "\n" ] } ], "source": [ "examples.explain(\"sids2\")" ] }, { "cell_type": "code", "execution_count": 5, "id": "8422e23d-a7f4-48b9-a31c-725a06dde2f6", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 100 entries, 0 to 99\n", "Data columns (total 19 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 AREA 100 non-null float64 \n", " 1 PERIMETER 100 non-null float64 \n", " 2 CNTY_ 100 non-null int64 \n", " 3 CNTY_ID 100 non-null int64 \n", " 4 NAME 100 non-null object \n", " 5 FIPS 100 non-null object \n", " 6 FIPSNO 100 non-null int64 \n", " 7 CRESS_ID 100 non-null int64 \n", " 8 BIR74 100 non-null float64 \n", " 9 SID74 100 non-null float64 \n", " 10 NWBIR74 100 non-null float64 \n", " 11 BIR79 100 non-null float64 \n", " 12 SID79 100 non-null float64 \n", " 13 NWBIR79 100 non-null float64 \n", " 14 SIDR74 100 non-null float64 \n", " 15 SIDR79 100 non-null float64 \n", " 16 NWR74 100 non-null float64 \n", " 17 NWR79 100 non-null float64 \n", " 18 geometry 100 non-null geometry\n", "dtypes: float64(12), geometry(1), int64(4), object(2)\n", "memory usage: 15.0+ KB\n" ] } ], "source": [ "# Read the file in\n", "gdf = gpd.read_file(examples.get_path(\"sids2.shp\"))\n", "\n", "gdf.info()" ] }, { "cell_type": "code", "execution_count": 6, "id": "c116377a-e1c6-4a0c-90ed-928db6f08220", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0 POLYGON ((-81.47276 36.23436, -81.54084 36.272...\n", "1 POLYGON ((-81.23989 36.36536, -81.24069 36.379...\n", "2 POLYGON ((-80.45634 36.24256, -80.47639 36.254...\n", "3 MULTIPOLYGON (((-76.00897 36.31960, -76.01735 ...\n", "4 POLYGON ((-77.21767 36.24098, -77.23461 36.214...\n", " ... \n", "95 POLYGON ((-78.26150 34.39479, -78.32898 34.364...\n", "96 POLYGON ((-78.02592 34.32877, -78.13024 34.364...\n", "97 POLYGON ((-78.65572 33.94867, -79.07450 34.304...\n", "98 POLYGON ((-77.96073 34.18924, -77.96587 34.242...\n", "99 POLYGON ((-78.65572 33.94867, -78.63472 33.977...\n", "Name: geometry, Length: 100, dtype: geometry" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "gdf.geometry" ] }, { "cell_type": "code", "execution_count": 7, "id": "248f154c-d874-496b-88ad-b8ad2f367bbc", "metadata": {}, "outputs": [], "source": [ "gdf = gdf.set_crs(\"epsg:4326\")" ] }, { "cell_type": "code", "execution_count": 8, "id": "592bad5b-95fc-4f62-9dfe-8139188c6eef", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
Make this Notebook Trusted to load map: File -> Trust Notebook
" ], "text/plain": [ "" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "gdf.explore()" ] }, { "cell_type": "markdown", "id": "ed1f23f4-968f-4ab8-a876-2d1b07fdd794", "metadata": {}, "source": [ "## Building Spatial Weights\n", "With a GeoDataFrame in hand, we can build spatial weights using the `W` class as:" ] }, { "cell_type": "code", "execution_count": 9, "id": "4e4ea913-d462-4943-ae66-41a3878c60b0", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/tmp/ipykernel_616347/4235704840.py:1: FutureWarning: `use_index` defaults to False but will default to True in future. Set True/False directly to control this behavior and silence this warning\n", " w_queen = weights.Queen.from_dataframe(gdf)\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen = weights.Queen.from_dataframe(gdf)\n", "w_queen" ] }, { "cell_type": "markdown", "id": "d6f48b63-6a66-4bc5-97b6-00a276dcdb1f", "metadata": {}, "source": [ "For the `Graph`, weights are constructed from the dataframe as:" ] }, { "cell_type": "code", "execution_count": 10, "id": "3e77b0cb-313c-4960-9cc6-f5deb602748a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen = graph.Graph.build_contiguity(gdf, rook=False)\n", "g_queen" ] }, { "cell_type": "markdown", "id": "c601e671-09b3-41cf-860d-92db22326e1b", "metadata": {}, "source": [ "Two things to be aware of here are:\n", "\n", "- the methods have different names for the two classes\n", "- the `W` relies on different methods to generate `Rook` or `Queen` contiguity weights, while the `Graph` relies on the `rook` keyword argument to do so." ] }, { "cell_type": "markdown", "id": "4192da75-3d27-4a61-99aa-8b9f11e11bd1", "metadata": {}, "source": [ "## Neighbors\n", "From the output in the previous cells, we see different information reported in the two cases.\n", "\n", "The neighbors of a spatial unit are those units that satisfy the specific contiguity relationship specified by the user. In our case of `Queen` contiguity, and pair of polygons that share at least one vertex are considered neighbors.\n", "\n", "This information is encoded differently in the two classes. For the `W` class, information on the neighbors is stored in the `neighbors` attribute which is a `dict` using the unit's id as the key, while the `list` of neighbor ids is the value:" ] }, { "cell_type": "code", "execution_count": 11, "id": "302e07a0-6bad-4ba2-9866-1194caa08c8f", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "dict" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "type(w_queen.neighbors)" ] }, { "cell_type": "code", "execution_count": 12, "id": "4e035e8a-6ac7-46aa-96d8-866cd3f01b4e", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[96, 97, 98]" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen.neighbors[99]" ] }, { "cell_type": "markdown", "id": "5f75edf4-9cab-4e29-9246-fa71365c057a", "metadata": {}, "source": [ "For the `Graph` class the neighbor information is stored in the `adjacency` attribute:" ] }, { "cell_type": "code", "execution_count": 13, "id": "f4e3cae4-854f-4a56-8afc-f0fb2524b67a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "focal neighbor\n", "0 1 1\n", " 17 1\n", " 18 1\n", "1 0 1\n", " 2 1\n", " ..\n", "98 96 1\n", " 99 1\n", "99 96 1\n", " 97 1\n", " 98 1\n", "Name: weight, Length: 490, dtype: int64" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.adjacency" ] }, { "cell_type": "code", "execution_count": 14, "id": "3be036b5-97ef-4eb2-8429-93d867d017da", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "pandas.core.series.Series" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "type(g_queen.adjacency)" ] }, { "cell_type": "markdown", "id": "b686532a-fb2e-4b45-a362-c7f7bdb13609", "metadata": {}, "source": [ "This is encoded as a pandas Series, with a multi-index. The first index is for the focal unit, and the second is for the neighboring unit.\n", "So we see here that the observation with the identifier of 99 has three neighbors: 96, 97, 98.\n", "This agrees with what we had for `W` so the question is why the need for the change?\n", "\n", "Part of the answer is in facilitating easier access to this information:" ] }, { "cell_type": "code", "execution_count": 15, "id": "075208ba-584e-403e-bc01-87834241fcc6", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "neighbor\n", "96 1\n", "97 1\n", "98 1\n", "Name: weight, dtype: int64" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen[99]" ] }, { "cell_type": "markdown", "id": "23bfd089-9b67-493f-81f0-5900ed4368f6", "metadata": {}, "source": [ "here we can query the graph with an id to get the neighbor information, along with the weights attached to each neighbor in the form of a pandas series:" ] }, { "cell_type": "code", "execution_count": 16, "id": "589290ec-aec6-4f93-b70e-1f3a80d63643", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "pandas.core.series.Series" ] }, "execution_count": 16, "metadata": {}, "output_type": "execute_result" } ], "source": [ "type(g_queen[99])" ] }, { "cell_type": "markdown", "id": "81380f43-5245-4502-a2bf-567d47ec7a6d", "metadata": {}, "source": [ "While we could also query the `W` object with an id, we get back a `dict`." ] }, { "cell_type": "code", "execution_count": 17, "id": "5e1d8999-5d5d-4136-904a-05df8889919c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{96: 1.0, 97: 1.0, 98: 1.0}" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen[99]" ] }, { "cell_type": "markdown", "id": "ec3c3a30-982a-4568-b942-e3f5204d29b1", "metadata": {}, "source": [ "As we will see below, the pandas series will offer substantial gains in efficiency and scope over encoding the weights as `dicts`." ] }, { "cell_type": "markdown", "id": "3591c67e-d29b-4630-a401-0b2892e6ab6d", "metadata": {}, "source": [ "At the same time, if the neighbors are needed in the form of a `dict`, the graph has such an attribute:" ] }, { "cell_type": "code", "execution_count": 18, "id": "afc97b96-a76a-4001-a59e-a1d027c7c653", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{0: (1, 17, 18),\n", " 1: (0, 2, 17),\n", " 2: (1, 9, 17, 22, 24),\n", " 3: (6, 55),\n", " 4: (5, 8, 15, 27),\n", " 5: (4, 7, 27),\n", " 6: (3, 7, 16),\n", " 7: (5, 6, 16, 19, 20),\n", " 8: (4, 14, 15, 23, 30),\n", " 9: (2, 11, 24, 25),\n", " 10: (11, 13, 26, 28),\n", " 11: (9, 10, 24, 25, 26),\n", " 12: (13, 14, 23, 29, 36),\n", " 13: (10, 12, 28, 29),\n", " 14: (8, 12, 23),\n", " 15: (4, 8, 23, 27, 30, 32, 35),\n", " 16: (6, 7, 19),\n", " 17: (0, 1, 2, 18, 22, 33, 38, 40),\n", " 18: (0, 17, 21, 33),\n", " 19: (7, 16, 20),\n", " 20: (7, 19),\n", " 21: (18, 31, 33, 42, 45),\n", " 22: (2, 17, 24, 38, 39),\n", " 23: (8, 12, 14, 15, 30, 36, 53),\n", " 24: (2, 9, 11, 22, 25, 39, 41),\n", " 25: (9, 11, 24, 26, 41, 46),\n", " 26: (10, 11, 25, 28, 46, 47),\n", " 27: (4, 5, 15, 35, 43),\n", " 28: (10, 13, 26, 29, 47),\n", " 29: (12, 13, 28, 36, 47),\n", " 30: (8, 15, 23, 32, 36, 48, 53),\n", " 31: (21, 34, 45),\n", " 32: (15, 30, 35, 48, 50),\n", " 33: (17, 18, 21, 40, 42, 51),\n", " 34: (31, 37, 45, 52),\n", " 35: (15, 27, 32, 43, 50, 56),\n", " 36: (12, 23, 29, 30, 47, 53, 62),\n", " 37: (34, 52, 54),\n", " 38: (17, 22, 39, 40, 49, 51, 64, 67, 68),\n", " 39: (22, 24, 38, 41, 49),\n", " 40: (17, 33, 38, 51),\n", " 41: (24, 25, 39, 46, 49, 69, 70),\n", " 42: (21, 33, 45, 51, 60, 63, 64),\n", " 43: (27, 35, 44, 56, 86),\n", " 44: (43, 86),\n", " 45: (21, 31, 34, 42, 52, 60),\n", " 46: (25, 26, 41, 47, 66, 69),\n", " 47: (26, 28, 29, 36, 46, 59, 62, 66),\n", " 48: (30, 32, 50, 53, 58, 61),\n", " 49: (38, 39, 41, 68, 69, 70),\n", " 50: (32, 35, 48, 56, 58, 73, 90),\n", " 51: (33, 38, 40, 42, 63, 64),\n", " 52: (34, 37, 45, 54, 60, 71, 74),\n", " 53: (23, 30, 36, 48, 61, 62, 78),\n", " 54: (37, 52, 57, 65, 71, 74),\n", " 55: (3, 86),\n", " 56: (35, 43, 50, 79, 86, 90),\n", " 57: (54, 65, 72, 77),\n", " 58: (48, 50, 61, 73),\n", " 59: (47, 62, 66),\n", " 60: (42, 45, 52, 63, 71, 76),\n", " 61: (48, 53, 58, 73, 78, 87),\n", " 62: (36, 47, 53, 59, 66, 78, 81),\n", " 63: (42, 51, 60, 64, 75),\n", " 64: (38, 42, 51, 63, 67, 75),\n", " 65: (54, 57, 74, 77),\n", " 66: (46, 47, 59, 62, 69, 81, 85, 88, 91),\n", " 67: (38, 64, 68, 75, 83),\n", " 68: (38, 49, 67, 70, 83),\n", " 69: (41, 46, 49, 66, 70, 84, 88),\n", " 70: (41, 49, 68, 69, 83, 84),\n", " 71: (52, 54, 60, 74, 76),\n", " 72: (57, 77, 80),\n", " 73: (50, 58, 61, 82, 87, 90),\n", " 74: (52, 54, 65, 71),\n", " 75: (63, 64, 67),\n", " 76: (60, 71),\n", " 77: (57, 65, 72, 80, 89),\n", " 78: (53, 61, 62, 81, 87, 95, 96),\n", " 79: (56, 90),\n", " 80: (72, 77, 89),\n", " 81: (62, 66, 78, 85, 93, 95),\n", " 82: (73, 87, 90, 92, 94),\n", " 83: (67, 68, 70, 84),\n", " 84: (69, 70, 83, 88),\n", " 85: (66, 81, 88, 91, 93),\n", " 86: (43, 44, 55, 56),\n", " 87: (61, 73, 78, 82, 92, 96),\n", " 88: (66, 69, 84, 85, 91),\n", " 89: (77, 80),\n", " 90: (50, 56, 73, 79, 82, 94),\n", " 91: (66, 85, 88, 93),\n", " 92: (82, 87, 94, 96),\n", " 93: (81, 85, 91, 95, 97),\n", " 94: (82, 90, 92),\n", " 95: (78, 81, 93, 96, 97),\n", " 96: (78, 87, 92, 95, 97, 98, 99),\n", " 97: (93, 95, 96, 99),\n", " 98: (96, 99),\n", " 99: (96, 97, 98)}" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.neighbors" ] }, { "cell_type": "markdown", "id": "5d9c48c6-9b9e-45da-bcab-ebf1302ff4ce", "metadata": {}, "source": [ "## Weights\n", "\n", "The value of a weight specifies the \"strength\" of the neighbor relationship between to geographical units.\n", "For our contiguity weights, these will be binary valued.\n", "\n", "In the `W` case, these values are stored in the `weights` attribute:" ] }, { "cell_type": "code", "execution_count": 19, "id": "81111346-a12e-4ada-a9b4-de7318450d8b", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[1.0, 1.0, 1.0]" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen.weights[99]" ] }, { "cell_type": "markdown", "id": "e0bfecb4-2b20-47ff-87d2-c92226132977", "metadata": {}, "source": [ "For the `Graph`, the values of the weights are stored in the `adjacency` attribute:" ] }, { "cell_type": "code", "execution_count": 20, "id": "75126118-bd99-4467-967a-6e91a745acdc", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "neighbor\n", "96 1\n", "97 1\n", "98 1\n", "Name: weight, dtype: int64" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen[99]" ] }, { "cell_type": "markdown", "id": "ebee5f1c-88b3-4d23-a9cd-aee2eb42ce1b", "metadata": {}, "source": [ "Again, the underlying types of these attributes need to be kept in mind. `weights` is a `dict` for `W`, and as part of the `ajacency` attribute of the `Graph`, which is of type:\n" ] }, { "cell_type": "code", "execution_count": 21, "id": "1814d24f-3283-45a0-8ff5-f14a16ee22b5", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "pandas.core.series.Series" ] }, "execution_count": 21, "metadata": {}, "output_type": "execute_result" } ], "source": [ "type(g_queen[0])" ] }, { "cell_type": "markdown", "id": "d2a6f2fb-7dce-4668-9529-b080207ad373", "metadata": {}, "source": [ "And, the helper `weights` attribute on the `Graph` mimics that on the `W`." ] }, { "cell_type": "code", "execution_count": 22, "id": "ce459560-5b7b-433a-bede-5bfaac273bb5", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(1, 1, 1)" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.weights[99]" ] }, { "cell_type": "markdown", "id": "ef3bccc1-a36c-4b47-afd3-c02cd1015b4f", "metadata": {}, "source": [ "Individual weight values will be identical between the two implementations:" ] }, { "cell_type": "code", "execution_count": 23, "id": "b6f6d449-1d55-4d22-b889-76088cb106e6", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "True" ] }, "execution_count": 23, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen[99][97] == w_queen[99][97]" ] }, { "cell_type": "markdown", "id": "30c912f4-f367-405d-bd9e-f9d3589e4cc0", "metadata": {}, "source": [ "As well as the neighbor sets for a given unit:" ] }, { "cell_type": "code", "execution_count": 24, "id": "6fa4e245-68e2-4dc2-bbdf-b05ed7527d76", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "neighbor\n", "96 True\n", "97 True\n", "98 True\n", "Name: weight, dtype: bool" ] }, "execution_count": 24, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen[99] == w_queen.weights[99]" ] }, { "cell_type": "code", "execution_count": 25, "id": "b3240892-ab52-4f99-a106-c9ebee1b3131", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[1.0, 1.0, 1.0]" ] }, "execution_count": 25, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen.weights[99]" ] }, { "cell_type": "markdown", "id": "bb9e8401-d6ab-47c2-92cf-5617dbee8b69", "metadata": {}, "source": [ "We are implicitly assuming that the order of the neighbor ids in the `W` matches that of the `Graph`.\n", "Here we see one advantage of the `Graph` in that the information about the neighbor ids comes along for the ride in the adjacency attribute, or any pandas like queries on that attribute.\n", "\n", "To be safe, we would have to double check the ordering of the weights in `W`:" ] }, { "cell_type": "code", "execution_count": 26, "id": "57ef256a-8130-4124-9916-5fbe288a76db", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{96: 1.0, 97: 1.0, 98: 1.0}" ] }, "execution_count": 26, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen[99]" ] }, { "cell_type": "markdown", "id": "ea43bd1a-9525-4b5a-8d24-22fc37db423d", "metadata": {}, "source": [ "So in this case our equality check above happened to be comparing the values for the same $i,j$ observations, but this is not guaranteed to always be the case. Handling the proper alignment of the ids and the weights was a key motivation for developing the `Graph`." ] }, { "cell_type": "markdown", "id": "a11ed77e-1f95-406e-8ad1-72ffaf7f37a3", "metadata": {}, "source": [ "The key take-away here is that the `Graph` combines the information about who are the neighbors *and* the values of the associated weights in the *same* data structure, the `adjacency` attribute, while in `W` there are *two different* `dicts` that handle the neighbor information and the weights information (`neighbor` and `weights`, respectively)." ] }, { "cell_type": "markdown", "id": "b60d44ce-d9a8-4a3c-a4a2-6d2f9b1336c5", "metadata": {}, "source": [ "## Cardinalities\n", "The `cardinalities` attribute contains information on the number of neighbors for each unit." ] }, { "cell_type": "code", "execution_count": 27, "id": "f420b789-0ede-4ade-8b17-ff7f88e3c9e0", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{0: 3,\n", " 1: 3,\n", " 2: 5,\n", " 3: 2,\n", " 4: 4,\n", " 5: 3,\n", " 6: 3,\n", " 7: 5,\n", " 8: 5,\n", " 9: 4,\n", " 10: 4,\n", " 11: 5,\n", " 12: 5,\n", " 13: 4,\n", " 14: 3,\n", " 15: 7,\n", " 16: 3,\n", " 17: 8,\n", " 18: 4,\n", " 19: 3,\n", " 20: 2,\n", " 21: 5,\n", " 22: 5,\n", " 23: 7,\n", " 24: 7,\n", " 25: 6,\n", " 26: 6,\n", " 27: 5,\n", " 28: 5,\n", " 29: 5,\n", " 30: 7,\n", " 31: 3,\n", " 32: 5,\n", " 33: 6,\n", " 34: 4,\n", " 35: 6,\n", " 36: 7,\n", " 37: 3,\n", " 38: 9,\n", " 39: 5,\n", " 40: 4,\n", " 41: 7,\n", " 42: 7,\n", " 43: 5,\n", " 44: 2,\n", " 45: 6,\n", " 46: 6,\n", " 47: 8,\n", " 48: 6,\n", " 49: 6,\n", " 50: 7,\n", " 51: 6,\n", " 52: 7,\n", " 53: 7,\n", " 54: 6,\n", " 55: 2,\n", " 56: 6,\n", " 57: 4,\n", " 58: 4,\n", " 59: 3,\n", " 60: 6,\n", " 61: 6,\n", " 62: 7,\n", " 63: 5,\n", " 64: 6,\n", " 65: 4,\n", " 66: 9,\n", " 67: 5,\n", " 68: 5,\n", " 69: 7,\n", " 70: 6,\n", " 71: 5,\n", " 72: 3,\n", " 73: 6,\n", " 74: 4,\n", " 75: 3,\n", " 76: 2,\n", " 77: 5,\n", " 78: 7,\n", " 79: 2,\n", " 80: 3,\n", " 81: 6,\n", " 82: 5,\n", " 83: 4,\n", " 84: 4,\n", " 85: 5,\n", " 86: 4,\n", " 87: 6,\n", " 88: 5,\n", " 89: 2,\n", " 90: 6,\n", " 91: 4,\n", " 92: 4,\n", " 93: 5,\n", " 94: 3,\n", " 95: 5,\n", " 96: 7,\n", " 97: 4,\n", " 98: 2,\n", " 99: 3}" ] }, "execution_count": 27, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen.cardinalities" ] }, { "cell_type": "code", "execution_count": 28, "id": "358a43aa-7859-4574-9150-780ef3b4c2c6", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "focal\n", "0 3\n", "1 3\n", "2 5\n", "3 2\n", "4 4\n", " ..\n", "95 5\n", "96 7\n", "97 4\n", "98 2\n", "99 3\n", "Name: cardinalities, Length: 100, dtype: int64" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.cardinalities" ] }, { "cell_type": "markdown", "id": "5bbcf980-4658-4e43-80f3-3955151b39a9", "metadata": {}, "source": [ "Here we see that, although the attribute name is common to both classes, the data types are different. (Note: other cases of common names but different types that we won't cover here are listed [here](../../migration.rst)).\n" ] }, { "cell_type": "code", "execution_count": 29, "id": "0ff66bd7-c78f-4ac3-8eae-27fbd09646d6", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(dict, pandas.core.series.Series)" ] }, "execution_count": 29, "metadata": {}, "output_type": "execute_result" } ], "source": [ "type(w_queen.cardinalities), type(g_queen.cardinalities)" ] }, { "cell_type": "markdown", "id": "88aaf817-8b72-4da4-93e3-41976bda415f", "metadata": {}, "source": [ "Summaries of the cardinality distribution can be obtained for the `W` as:" ] }, { "cell_type": "code", "execution_count": 30, "id": "18a20a7e-9377-4f66-bd77-48e89866a874", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[(2, 8), (3, 15), (4, 17), (5, 23), (6, 19), (7, 14), (8, 2), (9, 2)]" ] }, "execution_count": 30, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen.histogram" ] }, { "cell_type": "markdown", "id": "8f2f0877-4c46-4502-9391-820719b77fcc", "metadata": {}, "source": [ "which indicates that 8 units have 2 neighbors, 15 have 3 neighbors and so on.\n", "\n", "For the `Graph` we can more easily visualize this distribution with:" ] }, { "cell_type": "code", "execution_count": 31, "id": "2364e921-4e95-4441-a8a2-37a8b2ea351c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 31, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "g_queen.cardinalities.hist(bins=range(2, 10))" ] }, { "cell_type": "markdown", "id": "072e73ce-f9ba-4f5e-9d7a-904ce50f12af", "metadata": {}, "source": [ "While to get a similar visualization for `W` requires other packages:" ] }, { "cell_type": "code", "execution_count": 32, "id": "519487e9-8a23-4a15-8731-4cc1796ddda3", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "sns.displot(pd.Series(w_queen.cardinalities), bins=range(2, 10));" ] }, { "cell_type": "markdown", "id": "b8fe49b2-8bb8-4e93-bfb3-dc9e93410df2", "metadata": {}, "source": [ "## Geovisualization of the Weights\n", "\n", "Both `W` and `Graph` afford the ability to visualize the connectivity structure as a graph embedded in the geographic space.\n", "For the `W`, the `plot` method can be used" ] }, { "cell_type": "code", "execution_count": 33, "id": "0cb29997-4555-4883-b118-3802d2adda9e", "metadata": {}, "outputs": [], "source": [ "gdf = gdf.to_crs(gdf.estimate_utm_crs())" ] }, { "cell_type": "code", "execution_count": 34, "id": "72a287f3-c85e-411b-939f-a1ba5fdf5ff1", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjMAAAG+CAYAAABmlIVZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOy9eZwNZ9r//znn9L7Zd60tse9rrLGLLYjlmyCCCWMZhAwePMlXPDGMDMMQhhCRMJahjcQwfsKIEGMiRgzxiBEhHaSJSKfTaUd39fX743yv2111aj3ntBap9+tVL62Wu+6qU1X3dV+rh4gILi4uLi4uLi4/UbxF3QEXFxcXFxcXl3BwhRkXFxcXFxeXnzSuMOPi4uLi4uLyk8YVZlxcXFxcXFx+0rjCjIuLi4uLi8tPGleYcXFxcXFxcflJ4wozLi4uLi4uLj9pXGHGxcXFxcXF5SeNK8y4uLi4uLi4/KRxhRkXFxcXFxeXnzQPnTDzwQcf4IknnkDFihXh8Xiwa9cux20QERYvXoxatWohNjYWqampWLBgQeQ76+Li4uLi4hI2UUXdgUiTk5ODxo0bY/To0Rg0aFBIbTz//PPYv38/Fi9ejIYNGyIrKwvffPNNhHvq4uLi4uLiEgk8D3OhSY/Hg7/85S8YMGCAWHf37l28+OKL+NOf/oTvvvsODRo0wKJFi9CpUycAwP/+7/+iUaNGOHv2LGrXrl00HXdxcXFxcXGxzUNnZrJi9OjR+PDDD7F161b8+9//xpAhQ9CzZ0/85z//AQDs3r0b1atXx1//+ldUq1YNVatWxZgxY/Dtt98Wcc9dXFxcXFxc9PhZCTOff/45tmzZgu3bt6NDhw6oUaMGpk+fjvbt2+PNN98EAFy6dAlXrlzB9u3b8fbbb2PDhg04efIkBg8eXMS9d3FxcXFxcdHjofOZMeNf//oXiAi1atVSrff7/ShVqhQAoKCgAH6/H2+//bbY74033kDz5s3x2WefuaYnFxcXFxeXB4yflTBTUFAAn8+HkydPwufzqbYlJSUBACpUqICoqCiVwFO3bl0AwJdffukKMy4uLi4uLg8YPythpmnTplAUBTdu3ECHDh1092nXrh3y8/Px+eefo0aNGgCACxcuAADS0tLuW19dXFxcXFxc7PHQRTP98MMPuHjxIoCA8PL73/8enTt3RsmSJVGlShU888wz+PDDD7FkyRI0bdoU33zzDf7+97+jYcOG6N27NwoKCtCyZUskJSVh2bJlKCgowK9+9SukpKRg//79RXx1Li4uLi4uLloeOmHm/fffR+fOnYPWjxw5Ehs2bEBeXh7mz5+Pt99+G1evXkWpUqXQpk0bzJs3Dw0bNgQAXLt2DZMnT8b+/fuRmJiIXr16YcmSJShZsuT9vhwXFxcXFxcXCx46YcbFxcXFxcXl58XPKjTbxcXFxcXF5eHDFWZcXFxcXFxcftI8NNFMBQUFuHbtGpKTk+HxeIq6Oy4uLi4uLi42ICJkZ2ejYsWK8HpD07E8NMLMtWvXkJqaWtTdcHFxcXFxcQmBjIwMVK5cOaRjHxphJjk5GUDgZqSkpBRxb1xcXFxcXFzs8P333yM1NVWM46EQljCzcOFCzJkzB88//zyWLVtmuN/hw4fxwgsv4NNPP0XFihUxc+ZMjB8/XmzfsGEDRo8eHXRcbm4u4uLibPWFTUspKSmuMOPi4uLi4vITIxwXkZCFmRMnTuD1119Ho0aNTPf74osv0Lt3b4wdOxabNm3Chx9+iIkTJ6JMmTIYNGiQ2C8lJQWfffaZ6li7goyLi4uLi4vLz5eQhJkffvgBw4cPx9q1azF//nzTfVevXo0qVaoIzU3dunXx8ccfY/HixSphxuPxoHz58qF0x8XFxcXFxeVnTEhuw7/61a/Qp08fdOvWzXLff/zjH+jRo4dq3eOPP46PP/4YeXl5Yt0PP/yAtLQ0VK5cGX379sWpU6dM2/X7/fj+++9Vi4uLi4uLi8vPD8fCzNatW/Gvf/0LCxcutLX/119/jXLlyqnWlStXDvn5+fjmm28AAHXq1MGGDRvw7rvvYsuWLYiLi0O7du3wn//8x7DdhQsXolixYmJxI5lcXFxcXFx+njgSZjIyMvD8889j06ZNjvxZtE49XEGB17du3RrPPPMMGjdujA4dOuDPf/4zatWqhRUrVhi2OXv2bGRlZYklIyPDyaW4uLi4uLi4PCQ48pk5efIkbty4gebNm4t1iqLggw8+wGuvvQa/3w+fz6c6pnz58vj6669V627cuIGoqCiUKlVK9zxerxctW7Y01czExsYiNjbWSfddXFxcXFxcHkIcCTNdu3bFmTNnVOtGjx6NOnXq4L/+67+CBBkAaNOmDXbv3q1at3//frRo0QLR0dG65yEifPLJJ6KKtYuLi4uLi4uLEY6EmeTkZDRo0EC1LjExEaVKlRLrZ8+ejatXr+Ltt98GAIwfPx6vvfYaXnjhBYwdOxb/+Mc/8MYbb2DLli2ijXnz5qF169aoWbMmvv/+eyxfvhyffPIJVq5cGe71ubi4uLi4uDzkRDwD8PXr1/Hll1+K/1erVg179+7FtGnTsHLlSlSsWBHLly9XhWV/9913+OUvf4mvv/4axYoVQ9OmTfHBBx+gVatWke6ei4uLi4tLoaMoCo4cOYLr16+jQoUK6NChg671wiUyeIi9cX/ifP/99yhWrBiysrLcDMAuLi4uLkXGzp078fzzz+Orr74S6ypXrow//OEPGDhwYBH27MEkEuN3aOUpXVxcXFxcXILYuXMnBg8erBJkAODq1asYPHgwdu7cWUQ9C2iL3n//fWzZsgXvv/8+FEUpsr5EGleYcXFxcXFxiQCKouD555+HnsGD102dOrVIhIidO3eiatWq6Ny5M4YNG4bOnTujatWqRSpcRRJXmHFxcXFxcYkAR44cCdLIyBARMjIy8Le//S1i57SjbXmQtUWRIuIOwC4uLpHFdSR0cflpcP36dVv7PfHEE/D5fEhKSkLZsmVRtWpV1K1bFy1btkSHDh2QlpZmqx07vjlW2iKPx4OpU6eif//+P+nviusA7OLyAOM6Erq4/HR4//330blz57Db8Xg8SEhIQKlSpZCamoo6deqgSZMmaNu2LZo0aQKv1yu0LdohnDPrr1ixAn6/H3/5y19w9OhRy3MeOnQInTp1CrvvoRCJ8dvVzLi4PKAYfaxYNbxjxw5XoHH52fMgaS7z8/Mt9/F4PNi8eTO+//57nDp1CufPn8eXX36JmzdvIicnBwUFBSAi5OTkICcnB19++SU+/PBDVRuxsbHIy8sz9c2ZNGmSo77b1So9qLiaGReXBxBFUVC1alVD+7vH40HlypXxxRdf/KRVwy4u4fAgaS5XrFhhaM7R4vF48Jvf/AazZ88O2nbt2jUcPXoUH3/8MT799FN88cUXyMzMxPfff29LWJLx+XwoVqwYvv32W8t9f+qaGdBDQlZWFgGgrKysou6Ki0vY7N69mwBYLocOHSrqrrq4FAnp6enk8XiC3gmPx0Mej4fS09PvW19GjRolzp+QkEC/+93vqHLlykF969q1K8XHx4v/P/bYY+T3+22fJzs7m/bv309Dhgyx9X3YvHkz5efnU+XKlXXvFd+v1NRUys/PL8Q7ZE4kxm83msnF5QHivffeQ8uWLdGvXz9b+7vV4l1+jjwoIdB3795Fy5YtsWHDBgBAamoqrly5gunTp+Py5cvYvHmzav/Tp0/j2rVrqFu3LgDggw8+QPny5XH27Flb50tKSkL37t0xceJEW/tXqFABPp8Pf/jDHwDc86dh+P/Lli376Wt4IyZaFTGuZsblp0pOTg5Nnz6dSpYsaWu2JS/R0dE0atQo97l3eai5ffs2bdu2jSZNmkTt27enUqVK2Xo/3nvvvULr09WrV6lcuXLiXB07diRFUYL202pE3nnnHSIiGj9+vFjn9Xpp+fLlts/N2haj69bTtqSnpwcdk5qael81WEZEYvx2hRkXlyLi2LFj1L59e/J6vaoPTJkyZWjOnDlUsWJFQ9WwdvF6vdSzZ0+6fPlyUV+Wi0vIZGZm0saNG2ncuHHUunVrKl++PEVHRzsW8uVBvWbNmjRz5ky6evVqxPp59OhRiouLE+eZPHmy4b6JiYkEQPxbp04dse2dd96h2NhY0U7Pnj0pLy/PVh/S09MNr9nIzJafn0//9//+X7FfUZqWZFxhRsIVZlweBPLz8+nQoUO0efNmOnToUNDHwu/300svvURly5YNEkbatm1LR44cEfta+QSsXr2a+vTpEyQMNW/enD766KP7fekuDzlWz7YTrly5QmvXrqVRo0ZR8+bNqUyZMhQVFWUpmCQnJ1OdOnWoXbt2IQk3JUuWpAEDBtD+/ftDvuY1a9aI99Lr9dK6detM20hLSyMAVL9+fdGPc+fOie2ZmZlUrVo1sa1s2bJ08eJFW/1jAcmJtuXmzZti3/Pnz9s6T2HjCjMSrjDjUtToqXErV65M6enpdOrUKeratSv5fD7V9hIlStC0adMoOzvbsE2tQKP9WGVnZ9OYMWNUMzwAVL16ddqxY8f9unyXhxizZ9uMixcv0sqVK2n48OHUpEkTKlmyZJDwrSe0FCtWjOrXr0+DBw+mxYsX0+nTp1UmHCszCwCqVKkSLViwgJo2baqr3YmOjqamTZvSokWLdN8/vWuWhYe4uDg6fvy45b1r37690Mjw8Z07d1btoygKjRgxQrTt8/lo/fr1pu1euHBB7L906VJHQiZ/h1atWmW57/3AFWYkXGHGpSgx0qIYfaybN29O+/bts9U2q7Offvpp049VXl4ezZ07l4oVK6Y6X5kyZWjJkiW69vxIzrZdHk6sNITbt2+nM2fO0JIlS2jIkCHUsGFDKl68uKXQ4vV6qUSJEtSoUSMaOnQoLV++XKWxsGLZsmWm7WsFrX379lH//v2pRIkSuvtXrFiRRo0aRadPn7Z8n4sXL07Xr1+31c8xY8aIicvkyZPFvbt9+3bQvlu3blVpqAYNGqT73srtxsXF2b5nDPscjRw50vGxhYErzEi4woxLUWFnlgiAkpOTafz48bofMTP447Z27Vrbx6xfv54qVaqkOn9CQgJNnTqVcnNziSj02bZLePyUBEi7z7bZ4vP5qHTp0tS8eXMaOXIkrVmzJiK+XU2aNBGCgfac8fHxhkIAEVFGRgbNmDGDatasqSt0WU1MKlWqZPt3W7lyJQGgqKgoysnJEVqRUaNG6e5/5coV1btbuXJlysjICNqvfPnyBAQcj53SqFEjAkAtW7Z0fGxh4AozEq4w41JUHDp0yNZHPdScMPxh3blzp+NjDxw4oLLV80e1bdu2hlqjSOfouB+Dd2GeI5Jt/9QESLvPNj9XZcuWpVatWtGYMWNow4YNEXW6lTl27Jg4729+8xvx+6xZs0asf/XVV221lZeXR2+99RZ16tRJ1wcl3Pf59OnT4pjs7Gzq06cPAaDY2FhDgUtRFHryySfFcdHR0bRt2zYiCjyPO3fuFNu2bNliqx8ynKemfPnyjo8tDFxhRsIVZlyKis2bN9v6+G3evDmk9vl42TnYKefOnaPHHnvMlikskkm07sfgXZjniGTbD1KSNysURaHNmzerHFPNltWrV9/X/tWqVYuAgLlHKxA0bNiQAFBKSoqpdsaIRYsWRfR9VhRFHLNv3z66fPmy+P/ChQtNj127dq3Kz65Tp05UsWJFVT8qVqzo+Nnha4yOjnZ0XGHhCjMSrjDjUlQUpmbG7/eL4yMReZCZmUkdO3a01d9nn32WVqxYQdu2baMjR47QpUuXhInKDvdj8C7Mc0Sy7VDygkQaOxqmI0eOUPfu3SkmJsa2hgIo3HwuWvbu3WsqRH300Udi+5IlSxy3XxjvM/u9vfzyy0RE1LhxYwJApUuXtjz2woULVKZMGdNnx+nzKN+jnJwc28cVFpEYv93aTC4uYcJ1lK5evWpYlyU1NTWkOkpfffUVUlNTAQC3b99G8eLFw+0utmzZgmHDhoXVhtfrRVRUFKKjoxEXF4f4+HgkJCQgOTkZKSkpSElJwXvvvYcff/zRsI2SJUti1apV8Pl8IhOp1+uF1xtITO7xeMTfeuuICCNGjMA333xjeI4SJUpg/vz5UBQFd+/eVS15eXnIy8tT/Z2Xl4f8/HzcvXsXf/vb33Dnzh3DtmNiYtCkSRMAEMUBedH+/4cffsDly5ct72th1ccxq2HUsGFDzJ07F3v27MH333+vOq5y5crIyspCdna2afulSpXC4cOHUb9+/Yj3XUuVKlWQkZGBsmXLIjMzU3efRo0a4cyZMyhWrBi+++47R+1bvc+h1EWrVKkSrl27hoEDByI9PV1VXXvXrl3o37+/6fF+vx/FihWD3+/X3e60TwUFBWK/vXv3olevXrauo7BwazNJuJoZl6LEKIEVL6FqCE6ePCnaCEVlrofdmWdcXBxFR0dbRqW4S+SWUE2RZjiJtAMCUTejR4+mK1euODre6/XS4sWLI95/ma1bt4rzbd++3XA/WfOwdOlSx+eJtMavZcuWBIAaN24s1rG5qHbt2pbHF4a2iGtEzZkzx9G1FAaRGL+j4OLiEjYDBw5EnTp1cP78ed3tn3/+eUjt3rx5U/zNGolw6dChA8qWLYsbN27objea5f3www+4evUqrl69iuvXr+PmzZu4ceMGvvnmG3z77bfIysoSs/ibN2/i1q1bln3xeDxCK0MOlMRO9vX5fEL7w5od/ldefD6f+PfOnTu2+l+1alWUKlUqqG3tebKysnD69GnL9ipUqGD7uuxgVsNIJj4+Hl27dsXcuXPRokUL1baBAwdix44deOqpp1RVmytUqIDXXnsNHo8HQ4cOhd/vx/Tp0/HOO+9g//79iIuLi+i1AMCUKVMAAGlpaRg8eLDhfi1btkT9+vXx6aefYt68eZg6daqj8/A1T5gwQfWeVK5cGcuWLXNckbtWrVo4ceIErl27Jta9+OKLmDhxIj777DN8+umnplqt69ev2zqP3f0AoGzZsrhy5QpOnTpl+5gHmkhJVkWNq5lxKUry8vKEo97AgQOFXwI7I8bFxTmqjsts2rSJgEB4a6S4deuWKhW73hKuP0thR3gV9jki3bZV5WIg4Ix57do1x329X9eRkJBg+IzcvHlTOOUCAefbSGehXrVqlWjfTgZfOeLJSd0jmS+++EK0sXv37pB9ml599VUCQDExMWKdoigieqpTp06mxxfGs96pUycCQDVr1gzpmiKJ6wAs4QozLkWJnMDr1q1bYv2FCxfEADZixAjH7S5fvjzoIxgOfr9fFQ2hV7CP82GEg9XgHQmH1/z8fEpKSjL8sIdzjsLov5Upku99KGYRIyIVaZednS32ZefguXPnBu03adIk1T166aWXInIdiqJQSkoKAeraRlbUrVuXgIDpLBR4XAEgzG6hcPToUdGOPKmxSqLHFIYDOZ87OTk55OuKFK4wI+EKMy5FSc2aNQ0/tAMGDBAfHL3kV2a8/PLLBAQS3kWCZs2aqWar+fn5IvlW06ZNxeA9YMCAsM9l5WsRrvbnzJkzpu0XVjRTOG3rhf2mpqbS0KFDVSG4DRo0iEiOlkGDBkVkRs8aQo/HI56XwYMH6+574MAB4Y8BBGqFGZXrsMv8+fNFe3ZKCDCyELFy5UrH55XDqk+fPu34eEaOSjx69KhYbyeJHjNnzhzDZzGU55H9jzweT0jXFElcYUbCFWZcigq5cJveBzM7O1vUhnn00UcdtT116lQCAvk0wmXgwIGin3KV39q1axMA6tatGz377LMR+Xgz6enpumG+PXr0CKtdRVHEoBodHa0aPCMhyMj914bFRkdHh9z2woULVW1NnTpVzKYzMjKoTp06YpvP56NFixaFdJ6rV6/SI488YinE2J3R83NRokQJatq0qRBSjMjKyhLhx0CgplGoJkVFUcTv26xZM8fH8z0tWbJkSOfnawgnzxMRifdAm8zPThI9onvvqdYh36qwpBGZmZmijQsXLjg+PpK4woyEK8y4FBUTJ04kIGAiMPoYvfjii7ZnwTKjRo0iIFBJNxxmzZolzv/444+rtnFq83bt2lFeXp6w49eoUSOsczLNmzcX7bVq1UoM1E7LOsiMHTtWXM+2bdto2LBhBEBldtqzZ09E+r9jxw7V4BEbGxtyW6wpYY3PpEmTgvZ59dVXVfV56tSp40ijt27duqDj5XOGIvTVq1ePAFCrVq1EZtpKlSpZHjd79mzVeadOnWr7Opjp06eL453UbmKOHDkijg+lsCL3f/fu3Y6PlWGhePjw4ar1chK9BQsW6B576tQpsc8f//jHiGWkZq3Q/U56qMUVZiRcYcalqGC/E7MaKYqiUMmSJW0PAgxrU6pUqRJy/9avX68a2LQCFwsYPNOW91+zZk3I52VYWGrfvj3dunVLfED79OkTUnvHjx8X/evevTsREQ0dOpSAQDbUcuXKEaAOgw0HdjyVzUBnzpwJqa0GDRoIgQgA9ezZU3e/69evC+dxPvf8+fPFdr0EeH6/n7p166Y6hh1f9TIZly5d2vaMngXcqVOnCnOHXdPnsWPHKDk5WfUMyn5lZvj9fqHR6NChg61j9GCtRijaGdaEhBs2z5oqvXpIXGeqVKlSusc++uijBAQcqyMJf5OsTFyFjSvMSLjCjEtRIOeBOXjwoOm+8gzfbnRF9+7dCbCXi0KPQ4cOiZllmTJldDP4PvbYYwSA6tevL9axD1B8fHxIUVgy7ITZpUsXIrpX7dfj8dDFixcdtSULhYmJieJ6nnrqKSEoyv4VdisbmzF37lwxeHMU2JQpU0Jqq3jx4kLTBgR8Y8xYunSpSstSs2ZN+uMf/xgkmJQpU0ZlaktNTQ0q5sgCEF/DxIkTbfU5JydHtHvgwAHavn27+P3skpubq6oHFhsbS++8847lcfKzEo4D7uHDh0MW0Pn+h6u94IlJhQoVgrbJ0Uq7du1SbcvIyBDbXnzxxbD6oIUF5latWkW0Xae4woyEK8y4FAVs77YbEcADe3x8POXl5Vnu36ZNGwICzrlOuXjxopjVxsfHGzqUPv744wSAHnnkEbHu7Nmz4gM6bNgwx+eWYcGItRB+v18MqGZ+F3qwBkZrRmLzTWpqKuXl5QkfJa1KPxTYjCj7i9StWzektniW37p1ayHUWJGZmSlm7naWcePGmfpecOFRuwOYnKhOURTV4OrUoX3RokUqn49Ro0YZ9jUrK0tow3r37u3oPHrwc2ik/TCCtWjhJgRkZ/64uDjd7UZJ9PgbExMTY+ub4YTBgwcTUPQFJyMxfkcmC5eLy8+QgoICHDhwAADQr18/W8ds374dAJCbm4vx48db7p+TkwMASExMdNS37777Ds2aNcPdu3fh8/lw9OhRVKxYUXffhIQEAMDdu3fFuvr162PQoEEAAuUP/vOf/zg6v0xeXh4AIDY2FkCgDMCLL74IADh58iQ++OADW+38/e9/x5YtWwAAAwYMQO/evcW2goICAPfKLHB69vT0dLEtVDh5Xnx8vGj34sWLjtv58ssvRV/atWsHAEHlA/QoW7YsTp06heXLl1vuW6ZMGaxcudI0wWLjxo0B2E/k+Le//Q0AULx4cXi9XlSuXFkkOvznP/9pqw1m5syZOHXqFEqWLAkA2LBhA2rUqCGSySmKgvfffx9btmxBv379oCgKvF4v3nrrLUfn0eP1118HEPg9161bZ/s4Thz5ww8/hHX+Vq1aAQDu3Lmj+0zyO8FJ9PicfP+HDx+OqKjI5rlt3rw5AODbb7+NaLtFQgSFqyLF1cy43G/kGatWpW9G7969CQhEJViZQbRaDTvk5eVRlSpVRN927txpuv/w4cN1Z2e5ublCgyKboJzCJpEhQ4ao1rO5KC0tzbINv98v8owUK1YsaIbav39/AkBVq1YlIqJLly6J6w/F6VOGTX01a9ZUOWueOHHCUTvr1q0Tv/s777zj2FQTqcRpcj/swJqcFi1aiHXhpsLPy8ujHj16iD5HR0fT1KlTdXOptGvXLqRz6MHvk50Cjww/d9OnTw/r3HLOGr1IQTmJXseOHenQoUMisZ3X6y2UsU32PyvKgpOuZsbFpQh59dVXAQSKSKalpdk+bsuWLYiKikJBQQGGDBlium9ubi4AIDk52Xb7HTp0wJdffin6+OSTT5ruz5oZ1qAwcXFx+O1vfwsA+PTTT/GnP/3Jdh9kOAW+Nr39a6+9BgC4cuUKNm3aZNrGkCFDhBbj3XffDZqhKooC4F7Jh2rVqqFmzZoAgEWLFoXUbyYrKwtA4DdIS0sT9+uNN95w1M7x48cBAMWKFRMlA4hIVfzRjEiltH/iiScABLRZZ8+etWyPC2S2adNGrCtRogQAGJbvsCIqKgr/3//3/4lCo3l5eVi2bJnuvfjwww+xc+fOkM6jZfXq1QCAb775Bm+++abtvgL3tKShkpKSIrQ8R48eDdru9Xrx3HPPAQAOHz6Mzp074/333wcQ0GayFjiSsGYGAI4cORLx9u8nrjDj4hICd+7cETVNfvGLXzg6NiUlBdOmTQMQ+Kj94x//MNyXq+QWK1bMVtvPPPOMGDTHjBmDGTNmWB6TlJQEIFiYAYDnn39eCGrjx49X1eaxCwsaMTExqvVDhw5F9erVAQCTJ082NAf99a9/xbvvvgsgoGp/7LHHDM/B5g8AmDt3LoCAsGRn0DaChSj+DbiGzsGDBx21w32oVKkSKlasKPp68uRJW8fbrdtktV/ZsmURHR0NANi9e7fpvnfu3BGDOAtB8jnsVAI3Y8KECTh37pypWczj8WDq1KniNw6HLl26oEaNGgCA//qv/7J1DN+rcIUZAKIitFE9JFlglLlz5w4GDx4cMaGOiYqKQnx8PAB9AeunhCvMuLiEwG9/+1sQETweD2bOnBnS8Tw4Dh061HA/FmaKFy9u2eb//M//CO1Jp06dsHbtWlt9YWHGaLDYsWMHgID9fuLEibbalGEBiH1mZN5++20AAR+fV155JWj7nTt38PTTTwMASpUqJfbXIvvMMMOHDxe+Riw8hgIPYqyN6Nu3LwDgiy++cNQOD/y1a9cGcO9+2ClACQQ0brK/ih6pqano0KGDZVvly5cHYD2A/fWvfxV/d+7cWfxdrVo1AM4KGxpx7do1U78mIkJGRkbENAd//OMfAQSKuNrxxWFh5scffwz73Hzf9TRaiqJYTj4iJdTJlClTBgDwr3/9K6Lt3m9cYcbFJQTWr18PAGjSpIkwOzjB6/Vi5cqVAAKaA3ZO1MLaEith5k9/+pPQRDzyyCOOtAZswjLSurRo0QI9e/YEEDCt2DWLMPzx1aui3K5dO7Rs2RIAsHDhQpUTMhDQBuTk5MDj8WDfvn2GM3g9YQYIaKoA4NChQ7hz546jfjM8iLHT6pgxYwAE7peT2Sw7WbKJiYXZzz77zNbxPp8Pf/jDH0wrYP/+979XVTo3ok6dOgCAc+fOme63d+9eAIHnTzbt1a1bF0BACA2XwqgIbUb37t2FRtDORIQ1ipEQZlgIzMjICNp25MgR03cr0kKdtk+hOLU/SLjCjIuLQ65cuSI+RnbMOEYMHz5c+HW88MILusIEr+OBVI9//OMfePbZZwEEtAenTp0yVdtrYdW32Yxv27ZtiI6ORkFBAQYMGGC7bbldPWGG2/Z4PPD7/Rg3bpxqPfsJjBs3TggBerAwo9Va/Pa3v4XH44GiKJg3b56jfjOsHeMZbMWKFYU2i4VaK3788UchqHXt2hVAwNwDODPVDBw4UERC6cHmOCvatm0LwFpA+PjjjwFAmGaYpk2bAkDIAqKMXfNZuXLlwj4Xw74zN27cMNT2MSzMROJaGzRoACDgs6Plfgt1TMOGDQEAX3/9dUTbvd+4woyLi0NeeuklAAEzwVNPPRVWW3/+858BBEwZzz//fNB2FgRKlSqle/yVK1fQuXNnFBQUIDY2Fp988okYaO3C+5up+lNSUoTm5+TJk3jnnXdst8/tsm1eS7Vq1URo+9tvv41vvvkG33//PUaNGgUgMNixFssIrQMwU7x4cRESa6T9soKFEBY+AKBRo0YAAo6admBHTuCe02XlypUBQIQl20Vropg/fz769OkDANi4caOtUHfe3+/3m2pX2JT26KOPqtazbwcR4cqVK7b7rse2bdts7Tds2DCcOHEirHMx3bt3FxoJK+0MmwMjIcywFlJPyxMpnyintG7dGgCQnZ0d0XbvN64w4+LiEB7Iu3bt6kgDokeTJk3QrVs3AMCaNWuCZmwsCJQuXTro2B9++AFNmzaF3++H1+vF4cOHUaVKFcd9YHOHmfkCAP77v/9bfEhHjx5tO3+LlTADBIQYjvAaOnQoevXqhTt37sDj8eC9996zvM98Dj0Ty+LFiwEEzDycs8MJrB2TB5H+/fsDCAiTdu4DCz0JCQniWh555BHRL7tcu3ZN5L1hLVSdOnWwc+dOoWHr16+fpaN28+bNxfF79uzR3efOnTsitwr7CTHly5cXx7PDuVN+/PFHNGnSRGhJrMjMzESrVq0wdOjQkBzRtbCAnJmZaRqpxxrFSAgz7LxOREF5fqyECY/HY9snygmsKdTr008JV5hxcXHA3//+dxHdouewGgrbtm2Dz+eDoigqTY/8wdaq2AsKCtC4cWPcvn0bALBp06ag2bNd2B/HSpgBgK1btwIAbt++bdvEZkeYSUlJEUkEDxw4gGPHjgEApk+fLqKH7JxDzzm2ffv2QqsyZ84cW33Wa1sWZjiCTVEUW/5Jn3zyCQD178jX5SRKZtmyZQACUSh8rXfv3kVMTIwQsrOyskTCQyO8Xq/43Y36Lws53bt3D9rOvydfmxM++eQTVKhQQTg/t2/f3tCM5PP5sHXrVqFB3Lp1K0qVKhWSYCrTq1cvEak3ffp0w/1YmGFzYziULVtWCLOyv9Xf/vY3U/Mt/9bLli2z5RPlhPLly4s2nUboPVCEne3mAcFNmudSmHBdGy5H4DQluhWTJ08OSsaWmZkp1t28eVO1f8eOHcW2efPmhXXuM2fOqNLVW8Hn9vl8lJmZabk/p6Rfv3696X55eXmq+kKcAM8OXPahSZMmuttfeeUV0a6dPjPZ2dniuEuXLqm2FStWzHbJhOrVqxMA6tatm1j30Ucfibbt1r+qUaMGAaBmzZqJ+7phwwaxnRMgAtZVnps1a0aAcUFOrotUrFgx3e2VKlUiADRgwABbfWdWr16tKmkwe/ZsunDhgvh/ixYtCAjUb+K6YoMHDya/3y/qG/HSvXv3sJK97dmzR7RlVEiyZ8+eBKjLfYQDJ+Hr0aMHHTp0iP7617+K+xEXF0fLly8PSh6YmppquyhoKJQoUYIA0HPPPVdo5zDDrc0k4QozLoWFXsXhxMTEiH5cFEWhpKQkAkA1atQgImMh47nnnhPrI1F76MqVK6K927dvW+5/69YtUXzPTnZW/lBv3brVcl+uTwOA3nrrLTvdJ6J7VYWNaj3l5eWJPo8YMcJ2u+fOnTMUOFios1PRnDO7yllk/X6/aPujjz6ybMPv94vBfcmSJeJ61q5dK/ZRFIXKlClDQKAwptlAP3bsWFNhhaudN2vWTHd7y5YtTQVILYqi0JAhQ8Q1x8bG0r59+4iIqG/fvgQEaoitWLGCAFBUVBQ9/fTTBKgLTR45ckRUqmcBwMmzoiUtLY0A/QKQRERPPvmk7d/ZivT0dFX1dXmJi4uj8+fPE5F+VfTChAtOPvroo4V6HiNcYUbCFWZcCoP09HQxgGgXj8cTUYFm/fr1qoH84MGD4v/MokWLxLrWrVtH5Lxm2gcjZs2aJY45cOCA6b58/6yqJC9dulR1fytVqmT7GnhglVPua+EBMy4uzpYGioho//79Qb8Bs2zZMgICqebN2lMURbShvQdOKjKvXr1aPHc5OTmimObKlStV+508eVLc886dOxu2t23bNtGeXv+Tk5MJAE2YMEH3eK5UbqdI4a1bt4R2ijUNXMpDK2ju3btX7JeTkyOKpcrPu6IoNH78eNW72bx58yANph12794t2ti2bVvQdqNyH04x+5YAoGXLloXVfjhwoVYjga6wcYUZCVeYcYk0+fn5urViZGEmNTU1orOmatWqEQBKSUkRtZ+4hs6OHTvEudPS0iJWQVcebE+dOmX7GJ4dW1V+tiP0XL16VQxosnZG1jqY0bx5cwLMK0FfvHhRtGtHeCAieuuttwgImNS0yLV2zAS1kydPiv203yc2VU2ePNmyL6x94lpWPMjrDYKTJk0S59y4caNue7IQe/ToUdU2WWtkZK6yqgLNHD58WGU+7Nu3r0p40poA5d8pKyuL5syZI/5/7NgxVdtnz54VmhXW5rz66qum/dEjNTVVPHtaWIMVjmnZ6lvCAl5ha2CMWLBgAQGBytxFgSvMSLjCjEuksVvY729/+1vEzin7UXTp0kUMpKtXrxbq6WLFikX8OedzWhUplJFn0C+99JJl29qBSKZOnTpiMLp69SrVqlWLAFBycrItLUrTpk1taau40KCd4pZE9zRhsbGxutvZ10BbRFPm1VdfJSBQTFELC6+9evWy7EtsbCwBoEmTJhERiSKgeoO3oihi8IyNjTU0H3Ib2oKRu3btEr+bkT+PrNEw+o3mz58vtBEej4eWLFkStE+5cuUIuGeuysvLE+2eOHGCFEURQl+1atV0zzN37lyV+aZmzZq2tYza692+fbtq29SpU8V7FyqRKhJaWBw7dkz0ITc3976f3xVmJFxhxiXSbN682dYHiAWO5ORkSktLozZt2tDw4cNp/vz5tGfPHseqb9m5V7tERUXRxYsXI36tPODs2rXL0XGtWrUS/TIaMLnvepWCidQz86VLlxIR0YkTJ8S6WbNmWfajcePGBFj78GzcuFG0e/bsWct2X3jhBQICmjI9unXrRoC5SYz9RPTMFG3btiUA1LBhQ9N+yCZHHqRZ2zF//nzdY86fPy/8lYzMb2z66dKli2o9ayOMrpuI6ObNm6JP2mcyLy9P3Bsg4GN2/PjxoDaOHDki9tm7d69Yr/WzYg0ZANq0aZNufzIyMoTvBxDQaL7wwgu2TYqsndH+li+++KK4hlCx+y0xckIubGQBcv/+/ff9/K4wI+EKMy6Rxu5sys7i8XgoISGBKlWqRM2bN6dBgwbRnDlzaNu2bXT58mXVed944w3TtgojqoFntU4dKa9evSoGHjlSh5HNFRcuXAjafunSJXG81pG0Xbt2QqNhFbHCg1iHDh0s+8zOuN27d7fcd+TIkQQYm9JkPxYjsx/3Tc8ENmzYMAJA5cqVM+3HgAEDCAAVL15crEtISCAA9PLLLxsexwMxAFqxYkXQ9n79+hEQbF5h4dDKuZeFYFnAuHLlCpUvX16ct06dOobf5datWxMAKlGihGo9a4wWLlwo1rE5qXjx4qYCyqpVq4QJDgj4gZw8edL0OoiIdu7cKY7ZsWOHWL9w4UICrM1pZjzomhmie/fcTMtaWLjCjIQrzLhEGjs+M5UqVaJ//etftG7dOpoyZQr16tWLGjRoQGXLlhVmAbtLXFwclS1bVviOGJ2zMGzr7EyqN+BZMXHiRNE/rSnp1q1bYtvVq1eDjmUzS3R0dJAGKyMjQwyWQ4cONe1DgwYNCAB17NjRsr+sdfD5fJYh0ew0bBQmnpOTI65Pz3mU6J4pavTo0UHb5s2bZ2ug5DbkMGiOftOaiLSwaS0qKko43TJsAouKilKt5/DhcePGmbbNguGMGTOIiOidd94RzxIAevbZZw2PzcrKEr/vzJkzda934sSJYp2sxZk7d65pv27fvk3t27dXvTvDhw+39DPjcPPKlSuLdStXrhTPaKjwt8QsmKAofWaI7mmm+vTpc9/P7QozEq4w41IYPPPMM4YfHzvRTIqi0KVLl2jr1q00e/ZsGjhwIDVv3pwqVqxICQkJptEN93MGpzcTtouiKGLwkwcBInXYt/bdnDFjhthm5Og7ePBgcb/1hCGG8/+YRe8wt2/fFvd99uzZpvt26NCBAFCDBg0M9yldunSQoCHDmqc1a9YEbWNfDY/HY9j+5cuXxX06ePCgWM/RRixIGJGRkSE0b3Xr1lVtO3/+fJCwKWvTrCLQeADs168fTZs2TRzn8/lU+W/0mDBhAgEBc5DWT4PbffLJJ1XrOWotJibGlm/H9u3bhcDFWh0zM4rsZM8mVzMncCdwNJP2nbf7LSlsHnvsMQJAtWvXvu/ndoUZCVeYcYk02dnZQl0tzzaByCaxyszMpN27d9Mrr7wiTCtWS0xMDDVp0oRmzJhhy/fDCv7gW83yjdi+fbvom+yQKudpkWfF586dEx/1Nm3aGLYrhyA/9thjhvuxA7GeqUsP9vWxilDhXCtt27Y13KdXr14E6PvEXL16VVy/nkNqRkaGqeaK6F5CRW2kCTvFTps2zfQaiIiWLFkizvPKK6+otrGwxRFesmOvleaKI6zkaKUSJUrQuXPnLPvEmiWtvw7RPdOc1gfq8uXL4rkZNmyY5TmIAsJZ//79Ve9Pz549DU2XHE2XmppKRPYETrvo5awq7IR4dmENq5mfVGHhCjMSrjDjEml69+4tPmKff/65yv/g7t27hXLOUP10YmJiqE6dOjR+/PigMFsr8vPzhTAzcODAkFXdPADFxMSIgUJ25GUURREDRlxcnOU7O336dNGGUeg4Rz/16NHDVl8PHz4s2uTEbXqwg2zv3r0N99mwYYPh4M+zeg6v18PK+Zp9RbQJzdgUw9FNVrAfjNfrVQlWnGTvqaeeIiKi8ePHExCIJLOC3xFeWrVqZUtjIjvE6jmGd+7cmYCAv40WzgLs9XpNtXVaDh8+TCVLlhTnjY+P13Um5vw7rJnSy/cUDvc7IZ5d2Dk+EkKbU1xhRsIVZlwiiTwIs6+DnJfDKlFcqOTn54tBSm/xeDxUrlw5mjx5MjVo0ECYh7SLz+ejatWq0fDhw2nPnj2GDpN6M8XKlSuHNFO8dOmSGJj79+9PRGqhgZF9bOxkBZazI+sNbkREjzzyiJhx26Vs2bIEgJo2bWq4DzuymmVazs3NFdejdaAeN24cAeZhvfwb6pWlyMnJEfd01apVqm2c58fKr4W5efOm0HLJPkBcCoLvLYe5N2rUyLS99evXq0wmL7zwgq1+EN3TpLH2Qwtn/tVL4padnS2uw47Dt4yiKDRmzBhVv1u1akW3bt1S7VehQgUCAll/9QTyh5Hr16+bahELE1eYkXCFGZdIwrPhpKQklXmENRh2kpyFgqIoKpW9VpDRs61fvHiR5syZQy1atFD5B2iPrVixIg0YMIA2bdpEfr/fMCNpODZ8jv7hGTfnouHZnpydtmvXrrbbZSdMQD+JG2tQnDgvsvMtYFyviU05U6dONW2Lc6VoNTgcel2vXj3DY1mo0jObcFZkj8cTpPVhjYqTejrr1q0T18ylFTiPCoces//TmDFjDNsZMWJE0HNjNwRa9gFavny57j4cEm+kHZK1dVzLzAlnzpwRfjlAwIzMaQGISCSsBECvv/66+LsocrDcT9jkaDdZZaQocmGGswY+//zzpvu9//771KxZM4qNjaVq1arRH//4x6B9duzYQXXr1qWYmBiqW7cu7dy501FfXGHGJVLIafW1amiOmjGqARQu7BQJICiqqWLFirYEjOvXr9OiRYuoQ4cOVLx4cVMtj9m2UKIr8vLyhEBVvXp14VDJKfN5AE5MTHQ8MPBsWc83haOi+vXr56ivfI+Nom5Ya2KUy4Vhn4wyZcqo1rM5TevIKsPPlJ5fDheD5HpdMixAmUUM6cF+WR6Ph86cORNUPoD/1vsG3759m2rXri324T4AEHWFrOBaR7GxsYYCEJeKMIogUhRFOEDXrFnT/sVrmDNnjqroZe3atUWqBNbKyZpLbTTYwwZrhc0E2cKgSIWZjz76iKpWrUqNGjUyFWYuXbpECQkJ9Pzzz9O5c+do7dq1FB0drYrjP3bsGPl8PlqwYAH97//+Ly1YsICioqJ0kywZ4QozLpEgJydHhFTrzaa5yGM42UCNyMjIEB/WHj16CHU/L6FWx87KyqLVq1fT448/LoQJu0soUVNyjSnW1Hi9XurRo4epdsUKedDVzuhZk+a0gnOfPn0ICPhP6A2sHAW0bt0603bkmbzsWMoO5GbCEDsQa7PbKooizClygUqGhTursHUt2dnZQkgrX768ykz2P//zP4ZaiOPHj6s0fz169KC8vDzxzNrJUSRf09NPP2243zvvvCMELiPWrFkj+qLN2uuEK1euUL169URbXq+XZs6cSZs2bQp6HzZs2PDA+LgUBixYR6rum12KTJjJzs6mmjVr0nvvvUcdO3Y0FWZmzpwZZOceN26c6mb9n//zf4Js3Y8//rjpw67FFWZcIgEnJ/N4PLqZduXEWpGqjcS0aNFCzEazsrKEQy0PPI0bN47Iefx+v6p2j9ni8XioVKlS1Lx5cxo9ejRt2LDB0CQjww65eotZMUgr6tevLzQ7svDBJoOBAwc6au/ChQuiX3qh06y92rNnj2k7eXl5Yl9uR9ZymJVy4GglrYC8Z88ecbyeoytrDMxKKRghR58999xzQkjh6K2kpCTV/osXL1aVJViwYIHYxv5MdnxmOK+N0TUxchScWcJEvgfh1E1ili9froparFSpkrg2eQnVp+ynADtX3++Ck0UmzDz77LPChmwlzHTo0IGmTJmiWrdz506KiooSESGpqan0+9//XrXP73//e9OS63fu3KGsrCyxcIijK8y4hMrp06fFB2vEiBG6+8izWKsBzgnywMUZXTl3Cg/U4STt0hJuduOoqCgqW7YstW7dmiZMmEBbt25VlTNgM4HREupgcObMGdGG7McSzsDOzsNazYic4t2oFIMMm5Q4s/C+fftsCb6cRVibuI41NkYDdZUqVQgwN2GZ8fjjj4v+sWaLhRour6AoitBeAQEN1pEjR1Tt8LFmEV8Ma5PM8vYQqXPdmN17OdJIFrBC5fbt28Ih2kzAfxDywhQG8+fPJ8C4FllhUSTCzJYtW6hBgwZCBWklzNSsWZN+85vfqNZ9+OGHBICuXbtGRIHso3/6059U+/zpT38yreA5d+5c3QfNFWZcQoWdSBMSEkzza7Ctfvz48RE5r6IoImRUTpvPmVs5mRUQuWR5djKSVqhQgVauXEkjRoygxo0bU4kSJVT+BXpLdHQ0VaxYUZVOXq/tcLKdcthuVFQUZWdnE9G9zK1OtLmMXPdHztkjJ/zTRrvowQn+SpYsSUREs2fPFgKAGXJxUfm5Y0dcDpnWEoqfkIzf7xeaB20epdGjR9PVq1dV/iI1atTQvQ88+FsJKPJ12qkBxs+m7JKgB5tj4+LiLPPi2GXLli2F4lP2oCNnWb6fzs6REGa8cEBGRgaef/55bNq0CXFxcbaP83g8qv8TUdB6vX2062Rmz56NrKwssWRkZNjuj4uLllWrVuHSpUsAgNdeew0xMTGG+9aoUQMA8I9//CMi5549eza+/fZbAMCWLVvE+vz8fABAcnIyUlJSAABr1qyJyDl9Ph/+8Ic/mO7z2muvYeLEiXj77bfxySef4Ntvv4WiKDh37hyWLl2KIUOGoF69ekhJSRHval5eHq5du4a7d+8atktEyMjIwJEjR0Lq++bNm+H1epGfn4+RI0cCAAoKCgAAXq+jTxoA4Nlnn0ViYiIA4Ne//rVYf/XqVfF38eLFLdt55plnAADffvstvv/+e5w6dQoAULZsWdPjGjduLP7+97//DQD47LPP8P333wMAnn/+ed3jfD4fgMA9D4WYmBjs3LlTt40///nPSEtLw1dffQUAePrpp3HhwgWULFkyqJ1HHnkEAJCZmWl6Pr63xYoVQ//+/S37Fx0dDQC4fPmy6X7btm2Dx+PBnTt3MH78eMt27VC+fHkxTukR7jP8oNK6dWvx94cffliEPXGOozf/5MmTuHHjBpo3b46oqChERUXh8OHDWL58OaKioqAoStAx5cuXx9dff61ad+PGDURFRaFUqVKm+5QrV86wL7GxsUhJSVEtLi6hcOfOHfGhrV27NkaPHm26P7/wLPyEwzfffIMlS5YAANq3b48uXbqIbSzMxMTE4NFHHwUAvP/++2Gfkxk4cCB27NghBg0mISEBO3bswMCBA3WPq1u3LqZOnYo///nP+PTTT5GVlYX8/HycPHkSCxcuRPPmzW2df/r06UhPTxeCiF3Kly+PoUOHAgD+8pe/4MqVK2Lg4QHeKdzegQMHhCDGwozH47ElJD3xxBNCqHvrrbdw8eJFAPeEXyNiYmJEv//1r38BgHgm4uLi0KZNG93juE/8nIRCdna27vqcnBzk5+fD4/FgzZo12LJli+E9qFevHgAI4UuPH374QQyOVu8XEx8fDwD48ssvTferWbMmevfuDSBw32/cuGGrfTOuX78e0f1+KkRFRQlFxU9NUHMkzHTt2hVnzpzBJ598IpYWLVpg+PDh+OSTT3Q/JG3atMF7772nWrd//360aNFCfESN9mnbtq3T63FxccyIESNw584deDwe/OUvf7Hc/4knngAQGAju3LkT1rkHDx4MRVHg8/mCzs2Tg5iYGDEAfP3116aDhlMGDhwo3kPWPlSqVMlQkDHC6/WiWbNmmDVrFhYvXmzrmJMnT2Lw4MGIjo5G1apV8Ytf/AL//Oc/bR27bt06xMTEgIjw9NNPC4EoVGFm0aJF8Hg8UBQF8+bNAwAxwYqKirLVhtfrReXKlQEAO3fuFAOdrHkxgjVDn376KQBg7969AICWLVsaHsPXqjeJtIOiKIZaH6Zs2bJ47rnnTPdh4dXv9xsKpv/93/+NgoICeL1e/OY3v7HVv6SkJADAtWvXLPfdtGkToqKiUFBQgGHDhtlq34wKFSpEdL+fEqVLlwYQeD9/UoRr69L6zMyaNUvlPMmh2dOmTaNz587RG2+8ERSa/eGHH5LP56Pf/va39L//+7/029/+1g3Ndik05HTichZTI98ELbJjqJU93wzZCVcvEkTOIyKHwC5evDjkc2rJzMwUfeAw6nCd//Lz84UzrNESFxcnCjRql+joaGrQoAHNnDlT5PzQQy4vwb4fvXv3DtmPgYsYssMtt8/J5OwwbNgwAiCS7dl9RqpWrSr6z98yAKbFGjnazaxmlRl2ncCt/LTk/hrVCeP74aSvHC5t9xg5Qu/MmTO2z6OHlU8ZgIfSZ4boXnFVo2zbhUGRJ80jChZmRo4cSR07dlTt8/7771PTpk0pJiaGqlatqps0b/v27VS7dm2Kjo6mOnXqOPYUd4UZFzvope8HAvWEnDgPcjI6LnUQCpyUq0SJEro5Tniw5wRW7KDcsmXLkM+phRMEer1eOnnypLgfTmre6PHEE08YOk7KkSDZ2dm0bNkyat++vW4YLAsTbdu2pSVLlqjeb0VRKCEhIWj/UENn5cF9//79okYRO/TaQY5g4uWbb76xPI6daBs2bCiSkXq9XtMoKK61ZFYE0wy5PpLZsnnzZsu2WNBev3590Da5EvVHH31ku388qNavX9/W/nLCRrOMy3YxypDNi1UixZ8qnLizMHJpGfFACDMPCq4w42KF1cfJyQDIOWGsIjiMeOWVV8R533nnHd19OBvnxIkTiejezDMuLi6kc+rRvXt3McskupckbtmyZSG3mZ2dLTLrarMYW1UIvnLlCs2ePZsaNmxoGBFVqlQp6tOnD02bNs2WwOQETirYvHlzEZ1UqVIl28fL+VucCFdDhw4lIJBRlzUuVjPj5s2bExBcgNIukdLMEN2L8NOm4SC6lxuoYsWKjvo3aNAgx/d/+fLllu+VE/S+Gfz/5OTkh7K8gZ3iqJHGFWYkXGHGxQxWGxt9sJ2GWk6ZMkVoDZwiF8ozK4vAobnTpk0jIqLz58+HNMM1g+sCca4SNg/ZyRliBKer5+rM3Of169c7Vst/9NFHNGbMGKpWrZplWHg4vyfz8ssvi+NZM1CrVi1bx4ZT64rPGxsbKwTKF1980fR8rVq1snyGzLB6J3j5+OOPLdtiM5k2+Snn/wKcm0eNkglawRpPOc1BqCiKIn7TSZMm0aFDh+jvf/+7uCanhS5/Cly9elVcn5mZN5K4woyEK8y4mBHJWSiROlkX5zqxCydD83q9lJGRYbgfm11mz54t1rEafeTIkY7OqYectZZ9M3r27Cm0CaEgV85m3zm+T+FWGlcUhXbt2kVPPvmkaWVxeXFaMM/v9wttEguTtWrVshSKwhWWZVMMLzdv3jQ9J5ummjRp4ugaZdLT0y3vYVRUlOVv1759ewKCzTtPPfUUAQEzrt1ClAxnC3bqwyWXvViyZImjY7UcPXpU9z3nyQwAVYHKhwWeOFiV8YgUrjAj4QozLmZE0j+ASD1js3sMEdGJEyfEucaOHWu6L/uDcEZgooCPGuBM9W7E7t27RV84ZfySJUsIAPl8vpDaZPNbbGysUMHzh3Hjxo1h95mx+3uy0FihQgXq3r07LViwQLdMhQwXd5QXK1NROMJyenq60CbIfbYyTbHmiLP1hoLf7xeaIHlJTU2ll19+WQh2Ho8nqOiqDDuPly5dWqxTFEXUOXNaZoLonsnOrD6TEWyqi4+PD6vsCCc+1DPtclJLr9dLFy5cCPkcDyLsE2j1jYoUrjAj4QozLmZEWjNDRCJr7/Dhw20fw6nfk5KSLD+yXJNp4cKFYt3atWtFX8O11/MAJKvxZbOA04gQOXuoLICx70u4s2SZSJRjqFKlCvXt25eWL18uqiEbaSq0pqJbt27Rhg0baMSIEdSgQQPxW1ktDRo0oFdffVVo5Mz8uKxMU506ddLVhjhh5syZ4lwsuEyYMEFokM6cOUPx8fGiT0amItaiyFnb5ZIWV65ccdy3U6dOieOdCiRybacJEyY4PjfDk4dHHnkkaFtmZqZ4tsuXL+9Y8/Qgw5Fkbdq0uS/nc4UZCVeYcTHiypUrVKNGDcuBJjk52ZGPBav57YYwys6JZrNchj+Ushrb7/eLwW/VqlW2+6oHzyy10TA8m37ppZcctceCmjY6i81ls2bNCqu/MnbNOl999RWtWLGC+vXrR2lpaUFp++VF9lcxWnw+n2mpBidLbGxskIO03jUYPZPdunUjAFS7du2Q7qGiKMJs2bZtW8MK4VeuXBEzdUA/jYBsduXfnmuKhRriKxfqPH/+vOPjuUq7z+ezVY5CD9aYGdW/kjWEdlM7/BRgvzenTtuh4gozEq4w46LHggULLAcoeRk0aJDttmfMmEGAde0dokCBSp69251J80CnFVq4wGC7du1s91UPHpTnzp2rWs+CnxPnxg0bNoh7qDW7lSpVioDIq6z/+Mc/GgoBZlqNS5cu0cKFC6lHjx5UoUIFR8+H9jwlS5akli1b0oQJE8R1Gi3x8fFUpUoVUwFGbzHSFrJ/U82aNUO6f7JwfebMGSEk79y5M2jf27dvq3IHDR06VLU9OztbbDt9+rRKq7J169aQ+kd0rz7T7t27HR9769Yt8dtqHZPtwscvX77ccJ8BAwaIaw0n79SDBEdb3q+Ck64wI+EKMz9P5AR4hw4dErPYq1eviqrTQCAR26pVqyg9PV2EkcqDDBcqBEBNmza1lXPm2LFj4hi5YrQeHObr8Xhs29f5Q6pNnDZmzBgCAgUxQ0WOjNL6kLDTpuz/YIaiKCIpWvXq1YO2swZl8ODBIfdXDw5N1i5W4d96nD59WsxGrZYhQ4bQqVOngswKTqKZTp8+Tf369bN1PiOfLK5mrXfP7cCRbHXr1iWie47ahw8f1t3f7/er3qlOnTqp7gE/r2vWrBHFQJOTk0PqG8OaNDNhwoyxY8eK/jrV7ly4cEEca2YmUxRFhPTHxsZaOm7/FDh8+LC49kgV7zTDFWYkXGHm54deArzKlSvTqFGjVLPfRo0aUWZmpjiOBy3++KalpZGiKGKmCwRs4PIxeshOwHrJwphz586J/bQzWjPYcVY7s5VnvUYZV63g7LZ6lek5z4TH47HlBzBnzhzRH72s3bVq1SIA1K1bt5D6qoecnG7evHli0PvVr34VclbWSPhVpaenB/nPGAlX4Z6PNQJpaWmOr3Xnzp2i/QMHDqiyWps9U4qiqKq4N2jQQAx2HP31y1/+Ujy7nCMpVHjiMX369JCOz8vLEz4/jRo1cnTs4sWLCQj4V1lx9uxZcc0sHP6U8fv94jc+ePBgoZ/PFWYkXGHm54VVAjwWVvScTps2bSoGGSAQjcCDthxymZiYaOkEyzMyM3t57dq1CQhogJzMcvj6du3aFbSNP9Djx4+33Z5M69atCdDPoSKbDKw+ZDk5OcJcZeQsyNFBrVq1CqmverBWoUyZMkR0bwYfjh9RpHIRNWrUiABQjRo1VNpCo/OZOQCbnY+1fZzw0AmcTZqj4uSyFuwMbcaQIUNUE4isrCzRJr8THo8n7O8x+6w4mQRoYedkALRv3z7bx/Xt25cA+34jixYtEucJVfh6kGChXHbmLyxcYUbCFWZ+PthJ9hUVFUVffvml7vEVKlQgACpbt2yTX7VqlRhgoqKiTDOJcniskd/C+vXrxTmcDrRmH2AWRkKZlRPdq5UzatQo3e3sGDpp0iTTdp5++mkxcBmp4jnqJlIzVk73D4D27t1LRPf8i8w0ZHawG81kBg/mdqLczPK8WJ2P771TJ03ZPMomzLNnz4p1diOHpk6dqhL8tc7VMTExIWVhluGJQJcuXcJqh38TJ/eqWrVqBIC6du1q+5hHH31U/HbHjh0LpasPDGx679+/f6GfyxVmJFxh5udDuOp5ecbByde0mpUDBw6oPs5GIalsYtHLQyHXignFr4HPfeTIkaBt7Lzp8Xgc27RlzQsLA1pYu9C0aVPDdjIyMoTQN2TIEMP9+vfvH5bgJZObmyuireS+scnQTpSYGVlZWbqakkqVKtkemFlT9eqrr9ran2ssaRetY7aWESNGEBAwiTqBNWXFixcX6+SweifI2ohQBDIrOGLQqYlIy65du0SfVq5caeuYUKL6cnJyxDufkpLyky53wIkQ70fByUiM3164uPzEuH79elj73blzBwDQtGlTNG/eHABw9OhR1T5du3bFmTNnkJycDACYPn06xo4dG9TWoEGDRJs3btxQbRs7dixycnIAAOnp6bb6rEdcXFzQuueeew4AQETYsmWLo/a2b98OAPB4PHj88cd192nTpg0A4OLFi4btPP300yAiREdHY8OGDYb7lShRAgCQm5vrqJ96jBgxAn6/Hx6PR3VPiQgAEBsbG1b7kydPBhHB5/Nh9+7dYv3SpUsxcOBAy+Pv3r2Lu3fvAgDatm1ruf+dO3dw9uxZAECfPn2wefNmFCtWDABw8OBB02OjoqIAAAUFBZbnYT7//HP861//AgD8+te/Futv3boFIPBMOOHXv/61+H2NmDp1KhRFcdQuU7ZsWQDAd999F9LxTP/+/VGnTh0AwMyZMy3v2bfffgu/3w8A6Nevn+3zJCQk4N133wUAfP/99+jZs2eIPS56GjRoAMD+97bIiZBgVeS4mpmfD+FoZuRaQTdv3qQ1a9aIGaSeev327dsiFBoAdezYMcgplh3/ZDPS5cuXxQy/X79+jq9RURRxTiOHTDaXOVXBs3nNrHaNnBKeswPLfPTRR2K7Vf6YF154QcxUw+HixYvinj799NOqbbw+nOKCfr9faOP4N2MHVCtzGyOnv7ejMeN7I/uXsPnG5/OZtsGROqVKlbLVN6J7uWni4uJUzzGbQ51mfi6MZJQyXLm8RIkSIR0vIzvOT5061XRfTjVg1wley8SJE8W5Qo3EKmr4HtyPgpOumUnCFWZ+PoTjqMlJrjhFupyYyygfRl5enlB34/+ZjOQ6LXqJtdh0EBMToysMWCGbgi5duqS7z/DhwwlwHv7KtvDHH3/ccB85ukXvvnAumpSUFMuPPfu4hFvtm01fcqkEhvvqxMFTC+cN8ng8dPXqVSIiEYqsTSxoBEfAREdH29qfI4Dat28v1t2+fVsIZ3L2Zy08YJYsWdLWuW7duiXa1WbF5TIWetFtZtgtK1GlShVasGCB4+/z/PnzI/LsMOy/FRUVZdoXNuHZvbd68Dvi8/ksS2g8iMjZwEPJ4OwE18zk8rPE5/PhD3/4g+42VpMvW7YMPp8vaPupU6cABNTB/G+ZMmUAAJs3b9ZtMyoqCseOHcOzzz4LALh06RJSU1PxxRdfAADq1asHAPjkk08ABExKp0+fBgC88sor4lxO+PHHH8XfiYmJuvv86le/AgBkZ2fj888/t9VuQUGBUBv36dPHcL+oqCiULFkSAFTmFgDYtm2bON/vf/97eL3mnxE2FeTl5dnqox7vvvsu/v3vfwMI3FM90xsQupmpoKAAq1atAhAwsVWsWBEAUL9+fQCB39wO/LsXL17cct9t27bh+++/BxAwYzHFixcXKv7Vq1cbHh8TEyP6bodJkyYJE9rixYtV27gfbLqyyzfffGNrvy+//BJz5sxBsWLFULFiRYwcOVKYu8yoUqUKAMDv9+P9998P2VzFbNu2DV6vF/n5+eJ91oP7VqNGjZDP9cEHHyA6OhqKoqBDhw6OzIEPApUrVxbvtpXJ84EgcrJV0eJqZn5ecC4U7WKVMI0TlcnhrJx8zMzswvBMEf9vFnv48GGaN2+emO1t2rRJOACGUwzy8uXL4jxmVbnZSdFKbc7IZhCr5F5t27YlQB2+rSiKqElVpUoVW+fkgpahFAzkc3J23QoVKujuw9ekl+fGDnIdITkcnzUWdnKNEBG1atWKAHsOq1xOQi+0Wn6+L1++rHv89OnTbZvv5GrgekUf2bQl1+myYvbs2ZbpEYBAHbJatWoJc6y8xMfHU/v27Wnt2rVBJrX09HQRhcSLVcFPO3BNMo/HY6j15BIcds2LRmzcuFH0PZzw8qKCy1iMGzeuUM/jmpkkXGHm5wV/5KpXr07jxo0TA45VDhCuptu6dWuxbtOmTeKDYyf6YPv27SJ6xuPx0NChQ3U/4vPnzw/5+uRCeWZmHM6CazelPZsmEhMTLffVqxjMac4B40yxWk6ePGnrWoyYO3euON4o7w1vP3XqlOP2ie49T9pSE/LvYJVEkeieCc+olg8jZ2BevXp10Ha54vSwYcN02+DfJykpybJf06ZNE8+r3nWMHj2agHt5e8zIzs5WVRbnfmoXbTh7Xl4ebdq0ibp06SIEfu3+1atXpylTptAf//hH29mUneL3+0WfmzdvrrudzxdKGQUtcqZnvZxRDzJOzayh4gozEq4w8/NBrhx94MABVeptqxwZnGxNHiDk4o3aIntGnDx5UlVN2OhjHupHVxYAzFi4cCEB6sR/ZrAwZxZyrdeHzMxMy0HAiOvXrzsSCGSys7OFU65R0j3ZWfrcuXOO2icKCKfy86SFn4233nrLsi1+JqzCqh9//HECAiUpjH43HgSNhJWXXnrJlmCqKAolJCQQAHrsscd09+EEfFbaxMOHD6sEkebNm1N2drZuIVcrLenZs2dpzJgxljmj9N4rO8kLzZCFcq1zsuz8HolU/nl5eUKzGBsbG3LRy6KAgwXC0TLbwRVmJFxh5ucDfxhYG5Gbmys+PidOnDA9lgfjBQsWqNZzZFCPHj1s9+Orr77SVZ1H4qNrN+/H7du3xX7bt2+3bJcH22nTptnqB1/fyJEjxQBspp7XQxY2nGpOeED3er2UkZGhu4/sxO2kX0zVqlUJCJgw9OBcRGPGjDFtR77OPXv2GO6Xm5srNHtmbZ45c8ZUQ8ADslWhU3ZKNhP2evXqRUAga7ERM2fOVD3bc+bMIaLAdbMJi53hy5Ur5+i5z8nJoaVLl1KLFi1MK5vLS6gRUtxnNpdqzXyTJ0+2JSQ6QS7k2aBBg4i1W9iwCb2wC066woyEK8z8PFi5cqX4mMlmDv6YrlixwvBYebDRRr0MGjSIAGfRC4UZlrp//35bwgzRPW2TVWXgq1evijZPnjxp2W56erqusNayZUvb18Hwh9woSZ8esoln5MiRhvvdunVL7MdRSHaRfYg2btyouw9HUbVo0cK0LTmLrtl3iH1dPB6PZYFSFg70tFKcsM4q0oeF/4YNGxruw/WW6tevH7QtKytLldgvKSmJjh49KrbLCel+8YtfEBBeGL7dCKlWrVqFVTdo69atoi1ZI8tZfO1Wt7eL7G9nlc7gQUH+xhVmwUlXmJFwhZmfB+yQps1KybOs5557zvBY2U9BO4jIH2S7z5Ddj27p0qVp8ODBtGnTJtsZQf/yl7+ohCGzWS6bCOSMrnpwjRo7uUSsal85NZ+xJsJJuQG211vVtJKFNCvhQAsP0mZ5TDgE3qqKOOcssrq/XEqiXbt2lv2bNWuW0Expnx12TjabNW/bts2WUM2+V1qB7dChQ8JExYKsNtVA9+7dCQg40K9atYoA+w7TetidJPASHR1NTZs2pYULFzr+/j/yyCNC+GJzHwt/dspROKVFixZCkA3VWf1+IvsPhaMJs8IVZiRcYebhhz/eAILqnvDA16FDB8PjzZJAyRWw7Sa5cvrR5aV48eLUpk0beumll+j8+fNB7aanpwvTBi9mURwHDhywpZno3LkzAdZlBSJVcFGGS0gsWrTI1v6yH4uZto1InQjRSfr4CxcuiOPM8rmsXr3alpDCCd7MhEpZuLAyiRIFfIb4udQW/FuxYgUB5rlh0tLSCLCOPKtXrx4BgaSQDCf049/cyA+IfWhGjBih0nTZrfGkxU7ttbi4ODGx0S7ly5en4cOH2xIWtMkf5e9AuHW+9MjOzhbCYbFixejHH3+kQ4cO0ebNmy0nLXbIz8+PaHtE90zz8+bNC7stI1xhRsIVZh5O+OWUQ5711OVs869ataphWxzRYZRkjqtoyx90q75ZfXSLFStGHTt2FLM9o5lljRo1aPjw4cIEoSdAmDkUs5lt9uzZhv3lPpjVUSIqHPMZZ9K1U01YURQxUBn5scjYjfzSwgnU4uPjTY+7cuWKaN8oTJqIqGPHjgSAateubbhPrVq1CHBW6ZqrvGudMFkTZKQFkf2uNm/ebHoOrnjdu3dvun37tnAU5/fFSDA4ceKE2O/06dMq/yWrivNmyCHNeu8CvwdXrlyh6dOnU82aNXVNonFxcfToo4/S8uXLDZNXtmvXTryHXPMMAN24cSPk/pshm5C1kWDhhJ6np6cHfY8iEcrOEXoDBgwIqx0zXGFGwhVmHj70Xk6j2T0LAWZhqlYCD5sTnNj7nVQ9zsvLo127dtGIESOoZs2aoiCh3cVMI8KDj1Flajmjr9XAZtd8ZtWODIc+jx492nJf2dFUr8imFjlNvV0yMzPFDHzy5MmW+7OZTC+MmuEqy0aZlWVNkJMK6rKW6sKFC2K9VQkC9vWx4wfGA1bHjh1VUXqPPvqoaQZrrtwtTxBYqDDyQbID537SLlFRUYaDc15eHm3evJm6desmhGftkpqaSmPGjFEJWnLB1EgLAkawaU7vHQ8lCtLILByJUHYW9iJV9V4PV5iRcIWZhwsznw29l1NbpkCP+vXrE2DsqyCba6wSyslwgi1tH+1EF124cIHmzp1L7dq1021Hb9HTiHAuFp/Pp6tl2LFjhzjeyhRTGJoZrm9llX8lKytLaJns+JQQER07dsyxMMN+RlFRUbZMU6VLlybA3I+Cf78XXnhBd3vPnj0JsNYE6cECxuDBg8U6zo+kJ8zI/mF2THt8ffLza8eswNq+Xr16iXWsQQ3VyVV2pH7uuefo0KFD9NRTTxFgPwUBUeDdmjRpElWrVk33W5KQkECPPfYYTZgwIaKChRV2NLqVKlWi77//3paprjDMwjKcx8vKJy8cXGFGwhVmHh5CeTllU8D169d12+UPtlFkjKIoYgZu5kMhI2fq/d3vfqcKg7Xy9dASjkYkMzNTbNcLCx42bBgB9gr2fffdd+I+GC0VKlRw9HGsU6cOAaDOnTub7scDvtfrNfwdtcjClx1yc3OFwPTUU0/ZOoYdN80y+/KAqff7yOHYZk7qRnC0nRyGzRobPR8wuyY0okB4v2yiSUlJseXPI79zcnRgxYoVgwQvJ7D/W0JCghjM+fsOgDZt2uS4Tb/fT2vXrqUOHTqoHJrtLMWKFaPZs2fTlClTaMyYMTR06FAaMGAA9ejRgzp06EAtW7akhg0bUu3atalq1apUqVIlKlOmDBUvXpySkpIoPj6eYmJiyOfzmaZycLqwsGUnCzMQugMvawALs+CkK8xIuMLMwwM7XDp9OfmlNtKIsFnn1VdfNTw3+w7IGYLN4Iy6ckQJD9xGqfeNCFcjwhFderZtvi4zB2miwMDGod5mS1xcnKMw6JYtWxJgnmzv9OnTov2xY8fabnvfvn3i424H/s08Ho/tBGZjxowhwHh2KkdU6eXDkYtYOo24IlKbqPj55gg87XXfvHlTvAtTpkwxbXffvn3CORsImGHsOlFzPhatAzJHiD366KMOrjCAbFJbuXKlahuHqTutEq/H6dOn6bnnngsql/AwL07MwjKy0GqU6ylcIjF+O6sq5uJigKIoOHLkCK5fv44KFSqgQ4cOuoUetXzzzTf485//jL179+KTTz7B119/bbuYHBdMZBISEpCTk4OTJ09i8ODBqm0FBQW4e/cuAKBly5aGbXbp0gWXLl3CmTNnbPXh3XffBQA0b95crPvd736HJ554AtevX8ff//53dOnSxVZbHTp0QOXKlXH16lUQke4+SUlJ6NChg+62tm3b4q9//SuOHDkStO3LL78EAPTo0cPw/Ddu3EDdunXx7bffAgCeeuopfPjhh/jqq6/EPiVLlsS3336LO3fuoG7duvjPf/4jCkmakZKSAgD44YcfDPfh3ywxMVEUfbSD3++3vW9BQQHWr18PIPBbczFNKzp06IB169YhKytLdzvfc4/Hg8qVKwdtf/311wEEfiM7RSi11KxZE5UqVcLVq1excOFCDB48WBSa1DJhwgQQEaKiorBo0SLDNidPnozXXntNte6FF14wLOKpZdeuXQCAZs2aqdZXrlwZp0+fxtdff22rHaagoAC//OUvAQCpqamYOHGianuPHj3w9ttv45///KejdvVo1KgR1q1bhy1btmDYsGGW+8fGxiI+Ph5RUVGIjo5GTEwMYmJiEBsbK7bFxcUhISEB8fHxSEhIQGJiIpKSksSSkpKC5ORkXL58GTNmzLA856ZNm9CiRQvk5eXh7t27yMvLC/pbURScOnUK8+bNs2yvQoUKtu6NlipVqsDr9aKgoAAHDx7EyJEjQ2qn0ImYaFXEuJqZosOuF72iKLR//34aN24cNWjQwLIcgNWi1VBwNFK/fv2C+ijP+s0cGmX/C6uy97LqWxvGyTO+xo0bm988DVb5XQDjom/vvPOO2Ef2+ZEzyRpF42RkZFBKSorYb+nSpUQUMPlxmHiPHj0oPz9fhLjj/2kq7PgXsZnEKIpHro+1Zs0ay/ZkzMwtWuQ09hcvXrR9DtmMd/bs2aDtHMas54Auaxs++ugj2+fUwn5RHo+HcnJy6ODBg6Jdxo4J7datW0J7CATMKGz+2LBhg62+yM++tgQIa2ycFK0kuldrCtB3/JbLa+j9BqFQmIkvjYi0jwu3Z+ZjGG75B86NNH78+JDbMMM1M0m4wkzRYOZFDwRqIHXu3JnKlClj+LJ5vV6qUKEC9erVi5YtW0ZfffVVSC9nmzZtCNBPF243oRkRiXTqL730kul+nOrb6/UGOepxzSQzAcKI9PT0oJTulSpVEmp2QN8RVfb5kZ03OfGaUXK1ixcvqvwItMIEf3hlHwi5PlbJkiUtTSdsptFLPKcoihCkrHLg6MFhvHZ+Ww75tlObSgsLCUuWLAnaxr4+epFytWvXNhXk7JKbmyuEjtmzZ+uWvGBBwuPx6AqZe/fuVZmVOnToQLm5ueJde+edd2z1hbPZ6j37oSTOkx2/zYoa8gTIqrSEXQrbedYIJ1GQdtsrrGgmonum81q1akUsf42MK8xIuMLM/ceOV77ekpycTM2bN6dp06bRsWPHdB0UzV52QD8D7ciRIw0HzEmTJtmeLfKL26xZM9P92FFRL+25XPW4T58+lufUwvlIGjRoID4eeXl5qvwf/fv3N+y7rBFi51Vt1mSiQOQID24ej0fXuZL9bbTXwYMW33Ozd4/DrfU0F1OnThXthJIVdd26dbYGT1mjFMp5WJgcNGhQ0Da+79pCjrKvi9YHJBTY96h8+fKqhG9EgWeOB3s9vxJO6se/tRzlxOvlMgVmcJI9vXDdUBLn9e3bVwhHZr5YHTp0IMBe/iG7RFqwsItRtXFtckS76E2CrIp92m1XFoD5/kfyvrjCjIQrzNx/7Kpoy5cvT4MHD6YNGzZQdna27faNcjEYObFyhuDo6GjDtswK6TE8uzWreZOXlydmyfPnz9fdZ9SoUUJj4CQzLdG9vB/aBHeKolCrVq3EvejatatKGGRHU/kecM4NrVPtiRMnhFO01+ulXbt26faFB2q9AXLZsmWiL2XLljX8fTnKS6sdunXrltAmderUyfymGMBCld7vLsP3tHr16iGdh/Nt6AmFrPHRmgA5t1Eo4dh67N69W9xvOdyeSK0NlPPR3Lx5UwjHQMA0KBf8lFPW26k6Lj/7r7zyStB2OXGeHXOQnPTQKg8RC66AubnYCYqi6Ba3jI6OLjRB5vjx4+I8b7zxBm3evFmYcriAbihwbqFIaVAKW+PDuMKMhCvM3H8KI7maDId4NmzYkDZv3ixmZdHR0bq1euQZoXY7D8h2svvKPibyoCAjf1SNnrnbt2/bjirRwpFJejZqRVGoS5cu4vytW7cWA6UceXDo0CGVb8OBAwdEG4cPHxZqfZ/PR/v37zfsC0enGKn/ueYTEIjg0htk3nrrLXEuGb4On8/nKLePDAtUZmn9OeIJAO3cuTOk87B2Ty+DtF5SvXDDsY3gPC7yM6AoivBtatKkidh39+7dKg1Ap06dgt4NORLLzm/Aobpmz76TxHms5YmPj7fU5OTl5Yl3Ss/cFwry9WzatIkGDBggnslICKB6DBkyJOhZYlM4oC6i6wTOpWU3i7kZ99ME5wozEq4wc/8pTOc5uRIymz7krK16icnkGaa2dpOdQpQyPAAYpd9v3bo1AdZ+EJzvIzEx0dGHkX1YzPx2WDUPBMxKPBDwDO+pp54SYe4ej0ecf9++fWKQjYqKsjQtcBVhs7BquSJw5cqVgzRRelXA5VT4drLwGsHClJkmjYXZsmXLhnweWRMi/5aywChnlmXTWqjh2EYMHTpUCG/y7J7/5t9z7NixYp3X66XFixfrtic7x9t5Rtk3zezZZ4HLrLwGkfqe2q2JVqNGDcvn0Qns01StWjUiCmiyuE87duyIyDm08Peod+/euuv1tH924Oe8a9euYffxfjpHu8KMhCvM3H8K04v+xRdf1J0dPfbYYwQEEmrpfXhZXczRONr1y5Yts3V+VtfqORMT3ctZM2nSJNN25GymZqnwtXB/rfwseGADAqplv99PPXr0IABUrlw5kRaec97s3LlTzJpjYmLo5MmTln3hmkP169c33e/ll18WfUlLS1NpAGRtFwtdnP4/OTk5rBkwC1JyQjkZebB2mshQJjs7W7QjRyXt3btXVxhg05OZQ2soyIkaeWETWrVq1SgzM1NUgwYCDtpmdZKcJh1kQd/s2beTOE/WJmnrTpnBNdbMNHF2kbVSsjDFuZZ69OgR9jm0yIVRtRrRlStXGk7I7FCzZk0CjEtqOKGwNe8yrjAj4QozRUNhOc/xS6nNuCrb1/XUzJxefdSoUWKdXJfIroMjz6r1PpiyycIqfJvonjOvkw82C4h2ZobyDDwtLU2lNmctTa9evWjjxo2i3bi4OFv+EUREjz/+OAH2bPlz5swR565evboQaGRNW0ZGhioa6q233rLVDyNeeuklAgLaLz1Yi5aUlBS22YAHctlXhEOmZc1QpMKxjeBq2NplxowZKrNS165ddU2yMqwdsZN0UH72zaL07CTO4wkL4MysIptSZdNpKHA9tpiYGNWzwRmzzWq9hQqXTzCKLmQh2GryoIeRs34ouJqZIsIVZooO2SFVXkIVZPx+vxh09VTjDRo0IEA/aoltxvJsWI76sOuIe/HiRXGMVnvRu3dvMeO1w86dOx19tBVFEfvbjbrh2SoA8TGUFzmnT2JiIl26dMlWu0RETz75JAGgKlWq2Npfrvxds2ZNysvLU13TsWPHhBnikUcesd0PIzj0XM+XJSMjQ5zXyuRhB/Yh6Nu3r1g3cOBAAkAVK1YU61jdH8moGxkebI0Wr9drWwvJgqWdUGq7zz5rBI2KumZnZwvto91s2zL8jFvV+rKCzbny70mk/mbYFfrtwlo0I43d0qVLQxaEWcjVi3R0Sn5+viodhN5k1fWZKQRcYabo4IGyY8eO9Nprr4Utsa9YsUK8LHrOpPKMYdu2bapt7EciD7ysunWS90K+rokTJ6rWc3SQ3bo+RPds4Vbh3kRqm72TkgGsoTBbEhISHKck54HTSXmGKVOmiHPWrVuX7t69KwRUDi8GoIqqCRVOWKcXds+Db0xMjO0wYTM6d+4cJISxFqJVq1ZEpBaEIxGOrcUqbUFSUpKjpHJGkWZ62H32rRLn9e/fXwhdoaTI5/fc7oRCD9mMcv78+aDtkc5pQxQICuBzmiUoZI1qw4YNHbXPiUP10gc4RVEUw3IPbjRTIeIKM0WDrE7nyB9OgDZs2LCQ2mzevDkB5iG0/NJqZ35sHpJNDlz11U6RRRnOz1KrVi2xTvb9cDJrkjPPWn28T506JfZ1YhbJz89XZfHVW5wWiCQyT3hnhlyNmCOn5MVpdmQjeODU/r5ZWVnCP0g2O4YDh74nJCSIdfzB50SGLEBFKhxbxm7FZSe/MZt7jMx0jJNn32wCIVf0Nir6asWePXtCEvhlWItr5MjM0ZNOzMNWsF+ZXrJBGTlC0I5fG8Nan6effjrsvrIJDgjW9kYif42MK8xIuMJM0cCCh6xi55DRUD4CiqKIgW/mzJmG+8np7+XU59u2bRMzB21/nOZv4Ay/8gf52WefDRrM7KAoinAatlIBcy4Ru4UTmcKycXNSO6fp6YmIunbtatqXSHwQORlcqVKlVOs5iaLX643Yd0H2GeHBSC5gKodjW+VMCYXC+I1ZGDQqosk4efbNEuexmTg+Pt7Sn8cIu98JI+TyFEZRXnKSRSf5scxg86NRYAGjKIrQgjnJVs1moREjRoTVz61bt4pr79WrF+Xn59OhQ4do8+bNbgbgwsYVZu4/cnr1OXPmiPVsg/d4PI4/VnKoZmZmpum+bLqRZ/hydAI757IzsdNwRbktFpg4yiGU0McRI0YI4cjMd4fzTTg1ixVW9AE79VrN3LXcrzwVo0ePJgBUpkwZsS4vL08IGZFwhmTk8P+DBw+q/n/06FFVOLbditx2OX36tMjJEsnfmIU+q7B1fva7detm2aZR4jzZf0wbcegUNu/JmlO78DMTFRVlqCGRkwPK2ZJDRfYFXLhwoeX+CxYsEPfq9OnTts7Bv1E4gnRGRobwZypTpkxEzLNWuMKMhCvM3H+0he8YudaLtgidFaxFKV++vOW+RhlP+dxbt24lonvOgqEUSUtKShIfh+vXr4vzhZJ/4tatW6Z5chjWCBmFGhtRWJoZvs9meVzuZ3+0sJAoPzPsFAxYm/Wcwr4Uc+bMCUrUyM9amzZtInIuRVFo5cqVVKVKFVv3MpR7alUElIhUz77dpIMsCHCeKEVRxARE1uSGCg/2ViYbPfi9tgph5kjESJhEOeeTXU2PoijCUd5uTh2O5tRm+7aLoijChO/z+QyThkaaSIzfXri4hMgbb7wBAGjcuDESEhLE+ri4OFFufsuWLY7a/Mc//gEA6NWrl+W+M2fORHx8PABg3LhxYj335eOPPwYA/PDDDwCARo0aOeoLELg2AHj//ffx+9//HgAQFRWFJ5980nFbJUuWRLt27QAAa9euNdzvm2++AQDExsY6ar9Dhw4oWbKk6T6pqano0KGDo3aTkpIAAIqiODru+vXrEd3PiLy8PACAz+cDABQUFOC1114DALRp0waVK1cOq30t5cuXBwCcOHECx48fBwBER0dj9+7d+O677wAAS5cuDescX3/9NZ555hkkJibiV7/6Fb788ksAgd8iMTHR8DiPx+P4N+b3Iy4uznCfJUuWAAg8+/3797fVLr+bn376KQDgf/7nf/Dtt98CADZu3Gi7f0b86le/AhD4vbdu3Wr7uPT0dHHNr776qum+AwcOBBC4hoKCghB7GoC/l2lpaeKdMsPr9WLGjBkAgJMnT4r7aAYRAQg8j6HwzDPPICMjAwCwcuVK1KxZM6R2ioTIyVZFi6uZub/IiZ9YAyLDydyc+FkcO3ZMtGk3HFKvSjCHJ/bu3Ztyc3NFmydOnLDdF4YjPbxeL5UuXZqA0CouM3ICNyOtFd87J9FDREQHDx40TGDIy7PPPuu4z+w7YKcqtYxdzUxSUpKjhIJatJqF5cuXO1bPO4ErZKelpQmtUJkyZYQ/RDgOo3v37hXmE3mpVauW0HAY1cvhxakfUvv27Qkwj5zh/CVOsu5WqFCBgEB9sZycHGG64KivSMDn6Ny5s+1jOCGmHe2QrJGyW1FcD7n+09SpUx0dx+HjZjl7GNYMOjkHI5upI2matYNrZpJwhZn7y9NPP21qepAHMrvRBjwoORGAcnNzhSMgF2Xkj3O9evXoyJEjoh+h2H5lJ0BeihUrFpbjKqdjN1Lrc5I6JzlYTp48qRsxpF08Hg/t27fPUX/Zz8GpQ3J+fr4QAO0spUuXpvXr1zs6B9G9MF+ObOPoolBTwlvBIfBxcXEixxKnxAecZxnOzc2lWbNmCRMML9HR0dSvXz/dnEByFKG8hBJB2KxZM1MhIycnRwhPTq6NhbJHHnlEvJMejyeiZj92SrbrzyWbeo0KxGrhZ7hXr14h9/Odd94Rv5HT65eTC+qFkMtwNOOMGTMcnSMjI0N8P8qVK1doNamMcIUZCVeYub+wLbdfv36G+/DLYVZfSIZTmztN+MRF29ixlp37SpUqJRJQWVVU1sNsBhxOjgWOuAL0U5Zzxlq7GqCLFy+KrK8cSQOowyk9Ho+Y4UVHRztKmnfgwAHRjhNu3rypW41Y7tP8+fNVggB/TPW0fUZwZeoaNWqoHMjNimeGgywgs2agXLlyQsCxOxCcOXOGunXrpvrNuK358+ebCt9yXavXX39dCMiyE7Rd6tata6rdYO2kE4f+9PR0iouLC/rNH3vsMcf9M+PkyZOibbOSDQxHvvl8PtuTG/6+pKSkhNzP7t27h/z7KIoi/LSsSmOwL5CTBJGKoghHfZ/PRxcvXnTcx3BxhRkJV5i5f8i1aMw+IPyRtOM8Jycac+oQevPmTSF0TJ48WVRRjoqKoueee04INk4o7EgcFtxatGgRtI0jVuxUvr1+/boQLL1erygKGR8fHxT5Mm/ePDFwlihRQjchoR5yNlS7KIoiai95PB4RZcGL1+tVCYMHDx4UAzIvFStWtOVsygNFrVq1hDkkkrlBtMjlMThiip8/q3w2iqLQ6tWrg8oReDweevTRR1VpBsxg7RAnupMFzr179zq6Hv6djEwLrGGxm9og0mYwK3igt1NEljUXTqIRZSfvUB1i+R3lXEROkR3azYQNnrDMnTvXdtssrAGgtWvXhtS/cHGFGQlXmLl/cNVcq1BOTjBmpyAcJ7ZzGjHDcGbW+Ph4On78uHg5Wb3t1ORQ2JE4crZerRmOI1cGDBhg2kZWVpYQijweD23btk2V44TNBzzgtmvXTqUVqlevni0twoULF8QxdmfmgwcPFscsXbqU8vPzg8xgeuHpe/fuDRroU1NTac+ePYbn4t++atWq4phw6z1ZwYOTViAxCsfOzMykESNGqMpKAAHzyLhx4xxX1eaoP670TERCO+Q0ayxrl/QSrSmKIp4pOf2CEfcrHF/GbnI7OdGeU/851jJNmDDBcf9kLVqoPlx5eXmiDx06dDDcj58vuya0jRs3ir5FogRCqLjCjIQrzNwf/H6/+LhNmzbNdF+5uq9VxlCushuqGloecH/zm9+oZveA8+q3clI+syXUirF5eXnCBDNw4EDVNvb5MJtp5ubmimsDQOvXrw/KccJCJ4drsqAoF4M0q2rMyAnGrHL/EBGtWrVK7C/XvNGaU8xmgbt27RLZTHmpXr16UGHB/Px8atiwoUpoc5rpORQ4ZFde9GoM7d+/n5o2bRq0b82aNWnjxo0hn58FENnMy+YgwLwIpBazcF45eZqdvDn3szghIxdWNQt55gSfdtI+aGnbtq0QrJ3C/oXhFq3ksh1mvy8LPK+++qple1euXBETjPLly993PxkZV5iRcIWZ+8OiRYvEgGlnNskzWLMkTnJF5XDKybM6vGTJkmJgY1+SyZMn22pDURRavHix7sxbb2nTpk3IDo38kYuKilJpPNjubZTZVFEUkQhQ/nCxjwwPqp06dSIg4IDJ+3LVcPYzsTOLkxPDWanZT58+LfKLpKamqj6QvJ61SXaSr23bti2o2F3NmjXpyJEjlJ6erqsFGDp0qGW74WDkD7JgwQIiCtyvOXPmCCGBl6ioKOrTp09EcnfwICRXjlcURTzv2sKJZrDpZfr06UHbOnbsSID9yLrCStxoRl5enmlhWqLA+MD72PXhk+FElgBsm2cZfg7CcSAmClwn/75G/k383bMqMqooipgs+Hw+Rz50hYErzEi4wsz9ge3r9erVs7U/q4DNKi6zt77P5wtrdiA7ZnIqcDtaAKLAy/3qq69a1jYyE2rsOCDKZGZmig+sLLjwB0keqOR+8gxTPk7OrMpOxRxC/Mgjj4jrYk2MoigqH5Xdu3eb9pX3M9Ow5eTkiPPExsbS9evXddt47LHHCLBOny/z1ltvGRa901si7ZfBWPmDNG7cOEgDVbZsWZo3b17EMqnKaRG0gxBn8/X5fLYrxLNpYt68eUHbWGizW2zRrmamS5cuIpVCJOBn2Sh0nFM4eL1e2/dFRs7eq/deGiFrp51GEerBBVw9Ho/IcC7D2l6rAqccOQogpAjCSOMKMxKuMFP4ZGRk2BYOGLkCttHHnLUMkciyyX4T2igaI0FDURRatGhRkPDTqFEjoYUyWjp27CgED17q1KnjKIqGTUFypARrMPRmr+zsCqi1XexsLefOePLJJwkI5EPp06cPAeoqw1lZWUL4iIqKMtUY8Dm1Zh4Z2Zyi54TK29hBG7BntpJZs2ZNUAizdikMvwwie0Ue5T60bNkyouYUZsmSJeI30yKHHtsNz+V3Zfny5ar18uTArjYpPz8/6F0yu0dt2rQJKf+TFjbBGEUtsjYwnGgqK4FJj0mTJhFgz2/QDn6/X3xz9DSbrLEz+z7L6SaefPLJiPQrXO67MLNq1Spq2LAhJScnU3JyMrVu3drSc/61116jOnXqUFxcHNWqVSvIMe/NN9/UfdCdSs+uMFP4jBo1SryYdjUoco0WvcE51BmPEbKDq7zcvXtXtZ+iKLRw4cKgD2/jxo1VVWplR13tMm7cOMrLy6NZs2YJ0xAvlSpVsuWEKjsH8v78/8OHD6v2ZbMUoHbWk2fqcr0bOfmeHIEmm8XOnTsnPoDFihUz9Dng38iojAN/tAF985iiKCrBkrUXdh0VZezO/j0eD3m9XvL5fBQVFUXR0dEUExNDMTExFBsbS3FxcRQfH0+JiYmUlJREycnJlJKSQsWLF6cSJUpQqVKlqEyZMlS2bFkqX758kNnIaOndu3fE6zLJ9O3bV/yuerRr1y5IQDaDhWetDw8Lw07yPskOpUZLq1atgpzBq1atGpYP0ZUrVwwFbjnSi82soTB9+nRTgUkPLg0QqfIWREQTJ04Uz7c2eIDfK6Nvz+XLl8W9r1ChQpH6ycjcd2Hm3XffpT179tBnn31Gn332Gc2ZM4eio6NVhcRkVq1aRcnJybR161b6/PPPacuWLZSUlETvvvuu2OfNN9+klJQUun79umpxiivMFD7FihUjwLkzLUdZ9OzZM2ibrLkJRf2rh97MsHLlypSenm4oxDRp0kQlxDBycUgW5DiSRBY4FEWhFStWBIUgFy9enObPn2/60WDTXVpamkr4k0MwZWGhffv2quP79etHQHCOEw5L59wWPAPXOm7v2rVLtF2zZk3dvvJHcsOGDUHb5Nwuek6wRPfeTyBQAJS1cXqh6VbY9csoqiWS/iB6sPbRKLxYzr3CWYPNYEFVGzHGPlhWUXXM0aNHhWCUlJQUFLkF3PNX8fv9NHPmzCCzbkpKCs2cOTOkatrcX622gdMVhJLjRUbWTNsxGcnPvNMadWb4/X7xLmtrS/H918vTpCiKcByPiopy5CRe2DwQZqYSJUoY/lBt2rQJcip7/vnnqV27duL/b775piPJ3whXmClcDh8+LF5Mq8gkLWyf1Ysy4fDh6tWrR6Sf6enppgON1nGzadOmdOrUKcP2OEy4QYMG4pjc3FwhDCUnJwcJYTt37hT5TniJj4+nSZMm6QpscuSUnNmV93355ZfFugYNGqiEDb/fLwStESNGqNplPwH2TWFfG717zcUtAX3nUf54ajPAXr58WWwrWbKk4SAkz5yzsrLE7NJpMU0i+5qZ//t//y9t2rSJNmzYQGvXrqVVq1bR8uXLaenSpfTqq6/SggULaN68efTSSy/R7NmzacaMGfTCCy/QlClTaOLEiTRu3Dh67rnnaOTIkTRixAiVEGu2FIZpSYadQM1yifBza+e94n7LCRzl6ECthlCPS5cuCfNHXFwcZWRkCCfzMmXKCDNP7dq1g47duHGjKqyeB9snn3zSkXM9C/WyKTU7O1vXLy1U2MRpx8Ga36lQCmFaMXbsWAICk0B58s/XqpefiTVtQOGnLnBKkQoz+fn5tGXLFoqJiaFPP/1Ud59mzZrRiy++qFo3a9Ysio6OFmr/N998k3w+H1WpUoUqVapEffr0oX/961+W579z5w5lZWWJhaVmV5gpHDgyRv5Q2EU2cchOf4qiiIE4Eh8aJz4NzZo1MxViuH8805HDvU+ePKkS7oxmyMeOHRPCmvyRHjx4cJDzI2u95Gy4hw4doj/84Q/i/2lpaUEfRXaelmtTMRyqnZycTEQkfFWMtGADBgwImkEzLARyxA5RILqCtW4+n8801frZs2dF24qi0JkzZ8T/rVK0a+Hf2Sw7c2H6zNzv88rcvHlT9RwaIYcrm+U2kWuXyX4xPFjayfuUlZUltCJer1dMdlq2bCkmDHLFaCMfmRMnTlDbtm2D7m+zZs1sJROUvzNsfmHTkNfrdRyFpMfAgQNVEwQz2I/NbrCEE+QyLnKyQ753Wod++XkYNGhQxPsTLkUizPz73/+mxMRE8vl8VKxYMdNkVrNnz6by5cvTxx9/TAUFBXTixAmhhr927RoREf3jH/+gjRs30ieffEIffPABDRo0iOLj4y0dzubOnav7QXGFmcgjCx3jx48P6Xg9HwlZC+HUEVQPuzN2uypf2dZ+69YtcQ/YUZIT/Vm1efHiReratavqI+3xeKhTp07iOZ89e7Zpn8uWLav7MeaZop65hmeGrP3Izc01rbGjKIoolgio/WM4VF1OnNatWzfbMz25iCjDAlIoRfE4qkhPsAin1ITd8xr9ToV1XmbdunXiGq38HdiPyyyTtKwxk1MtsDnCymFWURSVVmXbtm1iG6cE6N69OxHdE9itHGhv3rxJw4YNC3Kur1SpEq1atcrwuuXvFDs/s59TpHxW5G+MWThzXl6emAiF4hdmBy7bIk9kuG9yEMKlS5fE97dSpUoPjJ+MTJEIM36/n/7zn//QiRMnaNasWVS6dGlDzcyPP/5Io0ePpqioKPL5fFSxYkUxWzQavBRFocaNG1vmBXE1M/cP9msJR+hgHwn5Q8aZW0NJYqXHa6+9ZkuYsevTwEXs2AzKH0ZOWa8oinDwi46OtvT1unXrFj399NNBzo+NGzdWmXn0ljfffDOoPTmjqZ4pgKNeOOU90T1/CyNflZycHDHL9vl8wh+OByIWPObPny/OPXLkSKtbSfv27QsSZlhrFWpByPT0dFXiQKfCaqikp6cHhV9bCQ2Rgp267ZTnkMN4jb6Lp06dEv3nQc5J3id2NgbUWjsiEhNXNn/Kz7id+j95eXk0b968oOi1xMREmjRpkq5w36RJEwLu5SLiYyJp+rOTu2rt2rXi3GaJ/MIhJydHfEs4IED7PVAUReRpioqK0g3nfhB4IHxmunbtSr/85S9N97l79y5lZGRQfn6+cAo2kw7HjBmj6yxqhuszU3hwtlO7tVn04A+rrLZmB0GzhHpWXL16lSZNmiRmknYWux82Hvh5dsp+M/IsT5711KpVy1a7ubm5NGXKFF0HSaNFz3zBmW+NhEFO9CWH8E6dOpUA81DRixcvCj+Y5ORk+uabb4Qw07t3b3r//feFdsKuIMJaOK/XK9axL5BeiLFdPvzwQ3GPeCY8ZcqUkNuzi16Fci50Wpiw5szI0VomJydH3BOjPDF6RUTt5n1iYV8W8GVYmzdr1iwiCgysrI3r0qWLZf9ldu7cGZR12ev10uOPP64SjBYuXCgEON4/FNO4GVwpPS0tzXAfdjo2y68VCfg38Hq9KiH0+PHjRHTPjwiw5wxeVDwQwkyXLl1szcyYxx57zDRDZ0FBAbVo0cLxAOcKM4WDbKO3yippxvnz50U7p0+fVhVvO3funKO2QhFg+ANn16dB9pfhrKL9+/cnIBAZJcOVuQFnvj+c48Zuoj5ZCJMjK4xSl7NjsSxAyGYFs4gM2f9ATwsBgBISEmzXFOLcFj6fT6y7fv162DNn9gXw+Xyi2nhhFpkkUj/L2nxGdjNNhwoLCHZNc48//jgBAVOjnmDCqQzkZ4S1qI0aNTJs95VXXhHXbFQriAW+VatWiXXslO7xeELS8p49e5a6dOki3k1e6tevT3v37tUNDU9MTIyo+W/lypXiGoyEVzaRFbZgnZ2dLd5P9ucBQKdOnVJph5566qlC7Ue43HdhZvbs2fTBBx/QF198Qf/+979pzpw55PV6hX1u1qxZqoiKzz77jDZu3EgXLlygf/7zn/TUU09RyZIl6YsvvhD7vPzyy7Rv3z76/PPP6dSpU8Is9c9//tPRhbjCTOEwYcIEMesM1yOfNRHjx48XL57dSDYzASY6Opratm1L27Zto+3bt+v6UvA6ux81ecbK9miesepF4PBMzOPxmDpm6hFKCni+f7GxsYaz53feeUf0ScZuRMYzzzxj2h+tWcEMNlVqc3SwIKeNxLIL+xolJiaqosIi4YNlBGuUYmJixPPI2Ym14fGRRA7bN0teKCNHJWmT4hGRcMxl7Zic98moNIBcr6l69eqG16snNMuOq9qaZE7IysqiMWPG6JaVMJrIREqgkX3P9O6pbP4NtdSJE9j0KH/zfvvb3wohR1tW5EHkvgszv/jFLygtLY1iYmKoTJky1LVrV5Wj0ciRI1V243PnzlGTJk0oPj6eUlJSqH///kGRC1OnTqUqVaqINnv06KEKEbSLK8wUDjzwRcIfgGfOFSpUEKXq5UJ5WuwKMNoXVa9mT4UKFRx9zFh9Kyce2717t2hPe87s7GwhrJUsWdKR4Oe0OJ9cpFKv0rFeuzJ2BEk7kWFOIndY/a+NjuEouVAK+BHd+5BzoTweKF944YWQ2rMD1yuqXr26yOAs18rSRnBGCjmfj5Pni01TcnZohrNc8+8iax30fFKOHTumqrFl5A8i51jROsoOGzaMgIA2LdwII0VRaMmSJSKqzkyYiWSkGeeGatmyZdA21oaVLl06IueyIisrK0hTxYvP57svAlW4PBBmpgcFV5iJPB999FHQQBoOso2dl9KlS6uEjFAFGC35+fl06NAhMVt5/fXXHfWV/WVkFfrt27dFX/QSRcozMruJxrivZoKD9kMsq/jNnI7l7MIycli5kSNmpKsfcyblxMRE1XqusG1W7sIMrvNUt25dIroXDhyqcGSH0qVLEwAaMmQIjRkzRgzsrJ1LTEwslJkwh0vbzezLyEkRtSHOsmaL6F4uomrVqgW1c+XKFeH8Ghsba5p0Tf52aO/F7du3LX15nHK/q3WzD6Ce7xmbAocNGxaRc1lhlVursCPsIoErzEi4wkzk4RmG04+nHkYvHAsbPXv2DFuA0YM1S88++6ztY/T8ZRhW3a5evVr3WLnkwPbt222f0+z+aFXkbNJo2rSpaZuyiUFr2+dBacKECbrHRrr6MdfO0T5LVuUurGCtQ6dOnYhIXXcmkoUMGUVRxDPLyfiAgM+JnDsnEqU5tHDtq1BqmPF7oA2L5uSFJUqUUGm2tMlOs7OzReI7r9drqT3n50f2xZHhhHoxMTERSShn93lt1KiRYzOwHnIBSdnkJ0eHReI8VjidCD2ouMKMhCvMRBZFUYQTmxMHbz2cJLOLhAAjwzNNJwOAnr8Mwx/0sWPH6h6rKIoISY2NjbXtICtrdeQlNTVVJcjs379f9yOqR2ZmpthX60PStm1b0b4ekZ7pjh8/ngD9yBK+X71797bVlgybF3gWLOc0ikQiRi3ys5Gdna2qi3X16lURYRbpCBqie+Hxzz33nONj5bxcsjaPy3WUK1dOpcGR91EURZXR2o7QuWDBAvEO6CE7sHO0UzjYfV55KVu2LM2YMSOssGm98glsQtNqIAsL2fQdife0qHCFGQlXmIkscsbIcG2udj809evXj4gAI8OJpfRKKRih5y/DcFZPowgOokDEBc/emzRpYuucnHOFfWGqVq1Khw4dCppRcR4NO3Vm5OyuWnOSVR4Mq2y3LAjZnfGNHDlSDJpaOPOwndwpWtj3Sk7mxwKsWehsqLBDfFJSkljH92jjxo2qKL3169dH7LxyoU45MZ1dZD+rIUOGiPWc4r5KlSoiCWLZsmVVx7IpDwDNmzfP1vn4PpllymWBOhJmud69e1t+XxISEoIiBz0eDzVr1kw3/b8VHN0of1vYBKmtmRQpMjMzaeHChdS+fXsh3NpZCrteWLi4woyEK8xEFs6pEs6AkJGRQbNnz7atlSmMF+6tt94SHy27sHOfnsDSp08fW/dFrqdkFfVz69YtMSByEjg9x0I5lNluVlHeX6vyzsvLE+dctGiR7rFmWXYBdRp1K4YMGUKAftj0zp07RZtO3182B8oFMGVBza5mzC6NGzcmQK3p40GFTXac+daoqnUoyBOCUJ1mBw8eLARmNu10796dgIADM/t6yJFlPBnQrreCBVQz3yW5xEU4Zjk2WZktsql237591Lp16yCn2cTERBo6dKjtAoyyljQjI0OlbTLLjO+ES5cu0UsvvUQtW7YUv08oi6uZ+QnhCjORIysrSwxgTsJv/X4/rV+/nrp27Wo7d0phv3BynpwzZ85Y7i+bKfRCU2fNmiU+fFawycHr9ZrWH+KyCD6fTwyWeunXn3rqqaDByAqze8sROGYmOL3IMHmx6xfUt29fAvQdS2UfpaVLl9pqjyggkHE/5OKneXl54jecPXu27fbswAPKxIkTxTr222HhVzYZhjLj14NrDIVSmJORheGXX36ZiO5pR1gAA+7VcuIINADUtm1bR+fidhs2bGi6H78joWjliO759QEBvymOLtMKKXpOsLm5uTR37lzdLNJVq1alxYsXW75nbIqfNm2aqVOwXc6ePUsvvPACNWrUyDDs3OfzUbVq1WjEiBH017/+lSpWrFik9cIigSvMSLjCTPhwBBCrbL1er2EVZOb48eM0cuRISk1N1X2hoqOjqUGDBqLKtNFSrly5QnvhWL1upIGQOXjwoOiTngMpaxHsaHpu3bolPnYVKlQwVKXzvenWrZv4uGu1QrIPk5NCcUaF54juRbJERUWZqvnz8/OFwzC3x2HoUVFRtlLTswlDr2oy0b3osXbt2tm+tnPnzonfSqutYHOcnvAUKtnZ2eJ8sr8SC2qyFqJSpUpiUIwELBwY3T+7sAmOhQe+T+wgzEVJ5bppaWlpjs1AWsdsI+SSAxs3bnR0DtYqAfcyCrMPVatWrYSAVqNGDcu2zp49S/379xfPOS9RUVHUuXNnQ4dnvp8VKlQQ5U5atWpl+xqOHz9OEyZMoDp16gTVopK/obVr16Zx48bR0aNHg9ow0qA6za1VlLjCjIQrzISH3gw8JiYm6EXIzMykl19+mZo0aWL48lWsWJGGDx+uqhdkFT4YFRVlqzJuKPDMy064NPt2GEVwyZoeO4O4nGBML6JKzh1y6tQp4ZPTuXNn1X6cDwRw5sPEGg89E57sIGylQWCBkAfpypUri8iX0qVLWwq9HTp0MJ2pP/fcc2IWbRcecPUESw75juQ3gSOltEUe2blWzqEjJ/CLhMaRI9jCzeQqh+W/8847Qujg37JXr1700UcfieemWLFiId0/jky0018262oza5vRtWtXcR3dunUjIrWwuXnzZqFZsirLIKMoCm3YsIHq1asX9I0qWbIkTZ48WWW6lM1wvBQvXlxXgFAUhQ4cOEAjR46k6tWr65bE4OeoYcOGNHXqVDp16pStfut9v7XBAw8yrjAj4QozoWNVCXjq1KnUs2dPEcmjXZKSkqhjx460Zs0a09o0skqYl9KlS4sPp9frDcm50Qp2YLRTW4o/rO3btzfch/tr18FTdk6Uk0wS3VOzsy8J15Pp0aOHaj8uFmelttfCH8w1a9bobudIIq5sbARfMwsdfP38t5UZgvO/GFVMPn78uGjLrs8CF7vUJuIjUlctfumll2y1ZwX7nGidr+UIJ9kswc6gnAMnHPhajH5HJ7CgUbduXUpLS1O9j5s2bRLmjZiYGNPK0GawmXnatGmW+8o+U2YlNhguUKt9T5YvXy6Ezby8PNXE4+DBg46v4ebNmzRhwoSg757H46GGDRvS888/b/jN9Hg8tG3bNtq5cyc99dRTlJqaapjYLjExkZo3b05z5syxNUEygjXrmzdv1g0eeJBxhRkJV5gJDadh0zzTqVOnDk2fPt32wENEKrs8APrVr35F+fn5dOLECZV9ONI5OjjHidWsX/aXMap3RHTP4VP2mzAjLy9PhHEmJiYKk4isGWETWI0aNQhQhynLzp979+61dU6G1eZG95RDSeXoHD3kwUYWEjhPCRCcm0SmUaNGBJibkVjTZ9fPhQUrI38LPqcdM4MdODxZmw1bjhqTtYucTRcIL+fI6dOnRTuRKNPAgz5PJvjvmJgYYW7yer1haUr5t7TrA8XmISszGmdfBoIjhtq3b0+A2jmf39VQy2Uwhw8fpscee8ywTpmTJSUlhdq1a0cLFiygq1evhtWvhwVXmJFwhZnQsBs2XaJECRo8eDDt27cvpDBK2cmTTRZyfonLly+rQg3tFtKzg1w00cyhz8pfhmHtidYUZMbx48eF9ou1PmzSkutese9I//79xbFaPwcnsG/LK6+8ortdzhJs5CAthwWfPn1aOClXr16diO6FlQMB04UebM7o2rWrYV85gs6u9qlHjx6mwoo8Uw8nnwjDz61eeDILjdooM9ZQGGmk7DBv3jwhbEQCRVHEcyEPzvKEwqn/ihZ+1nft2mVrf67wDoBOnDih22c5RLxXr15B+7BztpyHh8tlRKp6td/vp4ULFwqNpp2lZMmS1KVLF1q2bBndunUrIv142HCFGQlXmHFGVlYWzZs3T/hAWC3hhk3LyZ3YUW7UqFGqfW7fvq3qjxNHVzP8fr9o0yzRnJW/DMPmMh7M7TJ16lTRj5UrV4qPr6yFSU1NVV17ZmamGBjmzp3r6HxERElJSUGCoxbO1aL9PZhbt26Jfl+/fl2lccjKyqLc3FyheYqOjqYrV64EtcFaDbPEeDNnznQ0aLNQ1bp1a93tfr9fCNB286MYISfHu3DhQtB2Nt1oiyfKpSf0jrMDO7pGMm+O7HOiXcI1y8maKr2yH0bwZEYr+CmKIrQugH5KgIsXL+oKQ1zgVOvnFC52Mw5HMtfQw4wrzEi4wow158+fp7Fjx9oWYOQlXCfGQYMGERDQ8LD2Qa9qs9/vF34kPFBF4iPEM1GzQd2OvwzRPbOVlWlGDzYjyfbzc+fOie3srMwFJIcPH05A6FXLWciYPHmy4T48ey1fvrzudjlqKC8vT9cf5dy5c2KWX7Zs2aC+aoU0PWSBQQ61NoLNo3IGVi2s7bHjL2UGZ7Tl6tJaOCS4fv36qvWyFuSxxx4L6dz8voaSIVmPwq7lI5d1MPOh0yILfps2baLNmzfTwYMHVeHWRoVpp02bRkBwxmHZKThSYfJE978W1MOOK8xIuMKMPvv376e+ffvq5n3xer1Ur149y7DpSOQpYJt4z549hbBi5D+hKIoq7LJ69ephV9dlzQCHcOqd046/DNG9CCUnifiYjIwMVRSDVv3Njr7PPvssKYoizBey2ckJHAVjVtBPnmXqqcHlCBiG/VFk7dTGjRvFflq/Er4uK98F1lbZKUDIz7SZSXLJkiXitwrnGerSpYupdkQuOKmFhV+PxxOSjwQ/L3ZSC1hxP2r5OElfICOnH9BbzKIR2YypV6+MNcFarVk4WGXI/qnkd3lQiMT47YXLQ8Xdu3fxxz/+EY8++ihiY2PRo0cP/PWvf8X3338PAIiLi0OHDh2wceNG5OXl4dNPP8WGDRvg8XgM2yxfvjx8Pl/Iffrxxx+RmZkJAHjmmWdQokQJAMB3332nu7/X68X+/fsxevRoAMClS5dQtWpV3LhxI+Q+1K1bFwBw4cIF3e0ffPABFEUBAHFeIzp06AAAICJcuXLFUT8qV66MV155Rfw/LS1NtZ37EB0djeXLl8Pv9wMA/vCHPzg6DxMVFQUg8BsY8dRTT4nfd/ny5UHbb968CSDwuzBjx44FEPht+Nl65plnMGbMGADA4cOH8d///d9i/7y8PACB58+Mxo0bAwD+/ve/m+4H3LumGjVqGO4zceJEeDweEFHI9xAAzp49CwBo1qyZ7vZHH30UAJCVlRW0bdGiRYiJiQERiftmly+//BL5+fkAgCeffNLRsXocOXIEX331leF2IkJGRgaOHDkS8jm++OILAIFn2Am7du3C3bt3DbePGDFCd31BQQH+85//AAAGDBgQtL1ly5YAgGPHjjnqjxk+n8/weeJv6bJly8L6bro4JEKCVZHzMGpm7IbaZWZm0qxZs+iRRx7RnSmULFmSBg8erJtwiZEjHHiRZ0kvvPBCyNfBOT84ZJJTndtxypML5CUmJqpMMk7gMF4jf4xRo0YRcC9pmBVsZgnFUXLo0KGq+ywn5OJZ5Pjx44XJqV69eo7PwVSpUsXSFENEVL9+fQL0w4jZOVM2scimJq0vj2wm5Ogr9oewcuxmTYrX6zU1L8pOyVYFNzl3T506dUz3MzsXv1dr167V3UdbcFILa268Xq+jEgvLli0jIOCoGwkiXQ1dD/YNs/I9kwlHY7Rv3z6xn160l1xJ3SofklP0vps/pfwuDwqumUniYRNm9JIgVa5cWbwkp06douHDh+t61Xs8HkpLS6MpU6boOmPqwWn6ZRPIm2++qRqYVq1aFdK1cDglZ0jlD7vd6Jx169aJwSQ6OjokO/TJkyfFdegNJmyGspuBlk1zU6ZMcdQPRVFE1AibtYoVKyY+spzTgnOaAPYjQvTgcHi96A8ZjpjREyI4+Zg2n4tR6HNOTo4wAcXExNDVq1eF+UguCKnH7du3xXUbRUYREV29elXsJ1d41oMTDno8Hkc+HIycpdZMEJELTmrJyckRv7eTxHcs+Bv5MznljTfesCXMhOPrwXW4nNSmCscHhd8Vo6KWskPyW2+9Fepl6fLiiy+KZ4v74JqWnOMKMxIPkzBjlcROm3KbhZCmTZvSkiVLQvINaNq0qdACcK6JESNGkN/vF/4OHo/HVlIrLTzwcwZc/gA4yfa6d+9eMRh4vV7HM0d5dr1169agbdy2Xb8EduTl7KN2WbdunerDyn/37NmTiO75gfA9d1LtWw+jjMJaZCFCOxjPmDGDgGCtFUeK6L13p0+fFpqbChUqCAHOTq0vfv7MfBzk2bgVubm54re38ofSg2vuJCQkmO6nLTiphQd5n89n+x1lIduqLIAdNmzYYCtPSmJiYlgDMjuUO9GEhaMx4nfFrFI172Ml1DuldevWBED4pkUqfP7nhivMSDwswoyTJHaJiYnUvXt32rlzZ9gRPxxxMXnyZHr00UcJuGdyuHnzpphZR0VFOQq3lCvJchIuDu2Njo521MeTJ0+qcmE4dYjkwWbcuHGq9fKs0G5SMg5tfeSRRxz1gTUl7DjL5i0A9Pbbb4vfgZdwCyVy7R29wpVaOLxYG3UzduxYAoI1aWamJiJ19WoWJpYtW2bZD66CXK5cOcN92Pxi9xmqXbu2ENadwnl0rPLf8Dlq166taxa+deuWuF9jx461dW5+3q00WlbIafdZ+2o2YXKqcZSxcvDXI1TNTE5Ojti2adMmw/affPJJ3Wc4XPibIlfuDiXq8OeOK8xIPCzCjN2X2sh2HwpXrlwR7Z48eVLYvGXNydmzZ8VHMDEx0fagP2fOnKBBZ9euXeJjGkpfOdyYhS+78Ee2ZcuWqvVO/WWIiCZPnkyAM7+Ay5cvi36vXr2aiAJaIbNQea0WySksmDZr1sxyX86oq63MzNW6ueSCDN9To8R1zz77rOp61q1bZ9kPua6RkVmI73+xYsUs2yO65zPl8Xgc+02wZtFMAElPTw+qciybhRke9GJiYiwHPVlbdvz4cUd9ZrKzs0V4OhDQ+GVkZOiasStVqqSqID1z5syQzslh+FZ+WjKyUK+3GPnMyHlkzO6nXAMtEgkUidSClFySQS/pn4s5rjAj8bAIM/fDQU/Lyy+/rBI45Ey4sjp83759YjZXvnx5W4MC+1XI+TdOnToV1iwmKytLfDABewUkie6l7i9durRqvVN/GSISJiKv12v7GM61ExcXp9Kkvfbaa6Yf8XCcCTlraoMGDSz3PXv2rO7gyQOwnsBiJ8sum7oAewnZ8vLyxHNmVIuof//+BKgrVZuRk5Mj2rSbYp+P474blZIwMgvrVS3OyMgQ+86YMcP03PyMhZrw7fTp06q0C126dFG9b/n5+WKCMnnyZMrPzye/3y/eByC0RI082bBb7kNOJmn0Dhi9B5xMzyqYQH6mVq5c6fia9JALj+bl5Yl7uXz58oi0/3PCFWYkHhZhpiiSMfHsnROLyZEi2sKPq1evFtsaNGhg+ZHliCj5wy3POO06KGvJy8sTghIQ0LZYCUYcVSVHhoTiL0Okjl6xckDl87Cv05AhQ8T6ws77wSn/7SaN48GPk/YRmVe8lrPsvvzyy7pt8rvJArOd+8X3xCgvUKtWrQjQzytiRM2aNQ2vwwieXBjN/EP5/fh+xsfHm74/I0aMICAQjeiUtWvXqhIzGgmRerW7cnNzVZOFhQsXOjo3t2nnuJdeekmcp06dOrR9+3ZH1Z/Z/D169GjLc/E1OSlDYgabrsqWLUtE93y95HIKLvZwhRmJh0WY0Ybt6i2RTsbEKe9lNTrPrvQ+EtOnTxd9MctK+tFHH4n9tAUpeX04xewURaGePXuKtqpVq2aqQpbNaVwNOBR/GYZnenYqfcslAGQBrrCFV9Zg2E2Fz6UaZO0V+4w8+uijusewqcnMf0j7/FoJwaxFMzIjsfaAHaftwBpIr9dr29TE76ORQBHK73f+/HmxXlvLSaZevXoEgFq1amX7GonUpr3Y2FhTp302jWkF+ZycHOFDBdjzdWJYiLIykb766qui/bS0NPGb5OfnizpY48ePN/zWySUM7GSMtnqmnMLCEVecZw1khw4dItL+zwk3ad5DRt++fbFlyxbL/WbPnh2xZEw3btzADz/8AAB49tlnxXpORPbxxx8HHfO73/1OJKfau3cvpkyZotv2mjVrAACJiYlByeE4AZvTpHPaNv72t7+JRG1ffPEFqlatimvXrunuX6VKFXHf9uzZAwB46623AADJyckoW7aso/MnJiYCAE6cOGG57+9+9zsAQK1atVClShWx/vr167bOZXc/LfHx8QAgEq9ZMXHiRADAN998g6+//hrAveR0SUlJusdwIrjPP/9cPEsyBQUFqv9nZGTgiSeeMO3Hc889ByCQhI77IcMJF1NTU03bkfn1r38t+vP666/bOoaf/3r16uluD+X3q127tki+x8+FHvxucEI+K3744QfUr18fb7/9NgCgYsWKuHTpEh5//HHDYzjBGydrZBISEnD+/HmUKVMGADB16lTxPpuRn58vfu/q1asb7rdmzRrMnDkTAFChQgWcO3cOMTExAAIJ6fi5TUhIMPzWrVq1CgAQGxsrEuOZ8Ytf/AJA4Jn65ptvLPe3gr8zvXv3BgBUrVoVQOD5drn/uMLMA0BBQQEee+wxMcC2bt0a27dvR+XKlXX3//3vfx+xc2/YsAFA4APSvn17sZ4/oJzNU8tf/vIXNGnSBACwYsUKrFixImifAwcOAIDYT4azgxoJHk5Yu3Yt5s2bBwC4desWatasiU8//VR331KlSgEAjh49CiCQ+RcAGjZs6Pi8/KHn7LBGfPbZZ7h8+TIAYM6cOaptFSpUsHUuu/tp4Yy7nIHXir59+4qswUuXLgUA5ObmAggIfHqMGzcOXq8XRIQlS5YEbb9z5474u0+fPgACQvCrr75q2I9OnTqJfugJHjk5OQDMB0wtSUlJYv+1a9faOubLL78U/dEj1N+PBYOsrCwxKMvcvXtXXGOvXr0s2//kk09QsWJFnDt3DgDQvXt3ZGRkoGLFirb6p/d8pKSk4MKFCyhZsiQAYPz48UL4N0KenNSuXVt3nz/96U8YP348gMD7eP78eSQkJKj24f9z9mk99u7dC+Bedm8rOnfuLASj9evX2zrGiH/+859CABw2bJiqH5EQlFxCIHKKoqLlp2pmUhRF5HgBQD169BDb8vPzRWXgtLQ0laki3CrADNvvq1Wrplq/d+9ecS4jlbzf7xfRDx6Ph3bv3i22yQ53eipqzqcSTmZhLevXrxfnjIqKooMHDwbtw3kh6tevr/KXceoXQHQvn0bt2rVN9+vbt6+hj0R+fr4qOktvCcesOHHiRAKc5avh55HNRlxXi/ME6WFmapIT3N28eVP4rng8Hjp8+LBhmxzqrK2iTHTPxOe0eCD7aHi9XksfKzmtgFE6Aju/HwD67W9/G3Qs1xMqU6ZM0LZ33nlHHGvVz9WrV6v8Y4x8l/RgnxMzx+ybN2+KEGSPx2NqPpK/G3q888474rdLSUmhmzdv6u7HeZyMzIjyu+vkerldo0rrdtHLPcRReE6CAlwCuD4zEj9FYcbv94sPNhDI+qqFncw4aoP9F7xeb8jOszL8kdIOVHl5eaJfZhlob926JXxufD4fnT59moiItm/fLo7X82PhJFbDhw8P+xpk9u3bp8qjoU0AN2HCBPEhlQsoOvWXkdsyyjxKFPjosv1f71r3799vmu8DCK+KMSe8cxJCvnjxYnH//H6/rcrbnPdFL6pJW3X79u3blJCQIAQ8owGNBwxt5mHZofjChQu2r0t7LIfHG8H3wahSNlGglIidRHRAcOZfOWpQKyDws2WVLoD9QPD//GOsSjto4XfXKp/R1atXxb4ej8dQiOQJl949O3jwoBC6EhISKCMjw/B8/J0z8hfav39/SO8u50xykrBTD55kNmrUSKyTffJC+Z78nHGFGYmfmjCTk5Mj6uYAxh7w2oRlN2/eFIOjlUbACjmqSM9JkLUn2iRzWs6fPy/6FB8fT9evXxfaCL1ZJ5H1zCscTp8+rUo+J2tdZCErlPwyMpzN16xuDg+IQHDNngsXLoj7FhsbKwQ87RKK1ojh2lZW2Wtl5HDk1atX25q9y1FNWq2h7AjOHD9+XAhxVatW1XUIloUgWTNy7NgxsT6UkOWqVasSYB0JxZFgZuHf7PTp9XqDfr/U1FR66623VKHO9evXV6U7SEtLIyCQk0amefPmBBhHXmVlZQnNDhDIEWMnSkwLR6/ZySlz5coVIYR6PB7as2dP0D5cFkUrLBw/flwIfbGxsZZCKCelNMoizNmUzSYSehw/flzcs3Amg3wftAJ+qBrDnzuuMCPxUxJmbt++LVT3gLmphVP/y4ORnEbeTnp4IzhHiFFRP84eayf89cCBA+JFLlOmjAhT7N+/v+7+bMowipAJl4yMDFHnCLiXYj47O1usYxNZ27ZtQzrHhQsXRFu3bt3S3YcjHrSZZ2/fvi2ERZ/PR2fOnKH8/HwRfZSUlCRMNz6fL6SBiuhexEhsbKyj41jQbt26tQivt8rPwsnZtGHgBw4cCBJmiO5pcwDjBGscbSMPGnaESDNmz54tjjcz4XA0T79+/XS3y2HFixcvNiwMqyiKKkNsiRIl6OLFi0SkTuYm57HhZ3fkyJFB5z1x4oQQMHlCEGoGcKfm3osXL4qJgtfrDTLlcjg5hysTEZ05c0YI7dHR0UJ7awYLK3qJGonuaXZls7xduC8vvvii42OJAto4vvdykViie0JOuBmbf264wozET0WYuX79umqQtbL3cm4UrdqWc6z4fD7dKr124NmP0cyTtUJ2TRRy3SFejGYoXbp0MZ15RYKsrCyV9qtv375EREG1rX7zm9+E1L5c70nPFHfmzBlxDjl8W1EUMSMH1AUVWZMTGxtL169fF7NZWZ3tBH5+nJaOYF+b2NhY0QerIn2yqUnWPnB2VL2Mz2xGNRKWWENRq1Ytsc5o9m+XW7duiXMaZSRWFEVomvSSrJ09e1b89no+PXpwNmz+PdjHjAdmOc8Tt61N0b9q1SrRL4/HYxrabQc2ITopX3Du3DnxDvl8Pjp69KjY1r17d9W1XLx4UVVYVTv4GzFp0iRDzYusOQylaj1r05zkKJLh51xvEsiTF7uJPF0CuMKMxE9BmLl06ZKwOwP2cjewI6B2IMjMzBS+IaHUmyG6l+RJTuImI6fotpupl2e9vBjNGLnSrdHMK1Lk5eWpHKyrV6+ucpYEAtmMQ/VL4ZmYns8B52zRDrqctRQILnyo/b0XLFgg9g2lajlnkXWqxZCTAvKiZ1aQkU1Nr7zyiq0+KIpC1apVE9esTdv/yiuviGP5WXr66afF7xYqLOQaCSKyOULr06MoitCsxsXFmVbS1rJjxw5VZfpXXnlFNQk4duyYqko3f88URRHXzefVc3B3Ck+sJk2a5Oi406dPCw1HVFQUffTRR5Sfny9qj9WpU4cuX74sNEher9eRPw//7trSGkT3/HKsShgYwRmHtb5YdmHzo9Y0SHQvwCDUycfPFVeYkXjQhZmzZ88K9azH46H169fbOs4s9f/SpUvFtsWLFzvqj7auiB65ubliHyfVsllIAowrCLODY7hVoe2gKAr16dMnaHDmxSxduhU8KLLWh5HTm8v+UHLBP72EhHIiMH6WOfonJibG0cBJZK4VsUIbpfPHP/7RMqpKz9RkpR26efOmeDcSEhJU1yir9Hnw5hINoQrxRPcco2UhSYYTQ+oNpmxKAUDbt293fO5z584J8w7P4lmoaNSokRBaYmJiKD8/n27fvi2eASCgSY2Ugym/q+PHj3d87IkTJ8Qz7vP5qGzZskHvFf9rFkSgx/r16w0FYI7AtCphYIRctuPcuXOOj+fr1NO+6JnZXKxxhRmJB0WY0bObf/TRR8LvwOv10o4dO2y3Z5X6nwePqKgoR34VPBu0mt2wJsnJzE225wPqVOnMvHnzDAeLwiA/P19oUYwEmlBCoFnLoh1YubChPLOXnYGN6kDJpSRYJX/p0iWh8XAaUmrkr2KHli1bBt0nveKJMixgy6YmO347R44cEYPfI488ohIwOOJu2LBhRHQvpLlTp06Or4m5efOmuCY98xlfO1eOZ+T72adPn5DPn52dTbVq1RJtlSlTRve5LFOmjMos2rt375D9Y/QoXbo0AaAxY8aEdPzRo0eDNJ3a5fnnn3fcrtlzy9+XUaNGhdRnonu+WNOmTXN0nGwC1JuQ8rMeExMTct9+jrjCjMSDIMzoVaItXbq0eNl9Pp8jDQfDbenl5JD9KpzUnGGHxAoVKpjux8JSixYtbLUraxZkJ2et9mft2rVCCLsfFFbZgDFjxhAQrGFi51FWN8t5Q6pUqWIqQPLvKX8sZ86cKY7X+lGYoRdJZIf09HRDoc9Mi6VnarIbUbVo0SJxHjmMmX272CTJM+Nww/r5XdUL/2UBStae5ebmioG0ePHiIZk4ZBRFUfkMWS3hRLUZwffSTm0jPfLz84VAZLSEMkmQnetl/yvZ/BlqJXGie8ENTrV7ch4dvZQTcrqHcJ+PnxOuMCNR1MKMUeVcXrSOck7gwcFoEJNrnNitocIfMaNoI4bDl+2ag1544QUxC9dG7Jw8eVLsZ5VcKxL4/X7asGED9ejRI0hbZLQ4rUauV7zyxIkTor1du3bRmTNnhDo+OTnZ0lTEfdVGRPDgGx8fr/rAm6HN8WKHcItf1q9fn4B7piYnuW5kcyA73rLgy3lvWMMWajQKM23aNCFQy9oOv9+v+v2Yzp07R2Qg1cJaSrOlTJkyEa3HxthJiGhGYU0S5N9ADstn81+4mg92xnbqGM9mYqNnWTbNy987F3NcYUaiKIUZq48/ENBShPoxYjWz1llUhlXvUVFRlvZ0+UNhNcvfunWrGEjsqLc5UqBJkyZEFNDUsIktLi5ORF7Jg6zdon9WKIpCBw4coKeffpoqVqxomYxObxk8eLCjGZUcscTPHg96nOGUhZPo6Gg6f/68ZZs8wGg1D2fOnBHX1K1bN1v9k7PvGoWPawl3gFqyZIl4ZnJychxlIVYURUSEeL1eOnHihGqA2LhxoxDuraKrrJD9ceT3QM5FxM8m+3AAoIkTJ4Z1Xi2FXWzUDNYgsgnPKVxVPNKTBKJ7OVtkx3P+zvH3JVQuX74s+manSCXDSU7NUkrwxGXFihVh9fHnhFto8gHhyJEj+Oqrr0z3yczMxJEjR0JqPzY2FgB0C+4x7733Hnw+H/Lz800LywFAenq6+HvIkCGm+3JBQCISdYzM+M9//qM6rkaNGnjvvffg9Xpx584dNGrUCD/++KOq8GQ4xSY//fRTTJkyBbVr10ZMTAy6deuGrVu34tq1ayAiAIG6OEOGDBHF68zYsWMHkpKSMHnyZNy9e9dyf7kA4dGjR3H37l1xn4YOHYrGjRsjJycHHo8He/bsMaxXI1OiRAkAwXWrGjRogHHjxgEI1L169913LdsqXry4+Ds7O9tyfyD84peTJk2Cx+MBEWHp0qWiUCXX4zLD6/Xi448/RlxcHAoKCtC5c2fcvXsX5cuXBwBs3LhRFDI0Kv5ol7Jly6JSpUoAgJUrV4r1u3fvBhD4HWJiYnDjxg1x36tUqaLaN1w+++wzzJ8/39a+oRYbNYMLvmqLgdqlMGuLcR0l/rYWFBTg4sWLAID+/fs7bk8mLS1NFIp94403bB/HNda6du1quE9KSgoA4PTp06F30MUxrjATAe5X5WOzomuVK1cWxRY/+eQT3eJ1zJ///GcAgUKJXKnWiISEBFH0zWrwfP/990V1Zi4kBwCPPfYY3nzzTQCBQpBNmjRRDWzvvvtuUNVeI27cuIH58+ejRYsWiI+PR4MGDbBixQpcuHBBtFG8eHF069YN69atQ25uLq5du4Znn31WFEzUwpWD69evD4/HA7/fj9deew3JycmYMGGCqlCiFq/XK36f48ePY/78+VAUBR6PBx999JEQSJYvX47u3bvbukYuYHnjxo2gbStXrhTbhw8fbilwcaFJIFDU0A7hDlAxMTFC0Hj77bfFfbcjzAABIWPPnj3weDz44Ycf0Lp1a7Rr1w5AoMAfE64wAwADBw4EEKiOzQM6n4MFz06dOiEvLw8+nw/vv/9+WOcrKCjAli1b0L17dyQnJ6NOnTo4ePCgrWNDLTZqBj/7dquqa+nQoYNhQVwmKioKjRs3dtw2f5v4HTp06JDu9yVUGjRoAOBeQVwrvvjiC/j9fgDAM888Y7hf2bJlAQAXLlwIs4cujoiUmqioKUozU2GriZ2k/ucQzujoaMOaN6xatltKgFW7VpE0w4cPN7UnyxlTtYnrjKJkcnNzae3atdS1a1fhlKld4uPjqUWLFjR//nxdE9vFixeF6jc+Pl5k/uUlNTVVnPvKlSvUrVs3lYkqOjqaxowZQ7m5ubrXValSJQICoZrsiyQnRnRqlnjqqacIMHbOlnOg2EnOxfseOXLE1vnz8/OpVKlSls9z9+7dDf1/ZFMT+8GkpaXZOj/DuUYAiKKevIQSaq6HbIbjxIYc6TJr1izVM2tm5rU6x6xZs6h27dq6kT9xcXHCFKu3hBppZwdOLWCUgdkORs7i8pKYmEgnTpxw1C6HjXM5FX4vnJYwMIJzOBmF52th3yYrP5uePXsSEFy818UY12dG4kHwmTHz0QjnY+Qk9f/ly5fFB1MvIZhczXrNmjW2zs9CCteHMoIH9c6dOxvuwzki9D7YHo+Htm/fTnv37qUhQ4YIoUu7+Hw+qlWrFk2ePNmwmjGTk5MjBAufz0fnzp0zTDsvc/XqVXr88cdVv2lUVBSNHDkyyPn20UcfJUA/vLZr166m/dODHWaTkpIM9xk6dKhtIVnP98CM69evqxK7mS1RUVE0Z86coMHA7/erQq2B4DIHduCMsvyM8N+cgyUS8HPWvn17un79ujjHn//8Z3HOZs2aOWpz//79NGDAAFXOJXkpX748DR8+XPhrmAUQhJoDyQ5cp8qoZIPTdrTfvKFDh4rr8nq9utmUjWDfqYEDBxJReCUM9JDD8+0k9OM0DHqV4WXYsTzUmm8/R1xhRuJBiGYy++hbVeg1gx1K7ab+f/nll8V5tQLLrl27xDa90EI9OIOrmROwPHgZCUl2HKWNPuaVKlWioUOH0oEDBxzl2eD6RgAc5fdhrl+/Tr1791YNND6fj4YPH07Z2dmUnp5umL9Gmy/FLqtXrxaCghF5eXlCU1WiRAnT87BwK5dUMEJ2wPV4PEGJ81JTU2n79u00YcIElZahVKlSQcJSvXr1CIBIiNegQQP7N0Hqj1aTxotVzhu7cALHmJgYVap6Hjw5Ms+MrKwsWrhwITVp0kRkxtUKfY0bN6YFCxYYtpWenq57rU5zoTiBi2CGkzMnNzdXvB+smUxLSxPC5qFDh4S2CwhkHLfzXvDz07FjR1WSz3Adv2X4HbIT5s+TohEjRpjut3HjRvEMudjDFWYkilqYIaKgDJjyUqlSpZCTXQ0aNEh8vO3CH6mYmBhVFAtnF3WSeVdO3GdUW4UFHgCG5hi75jju3+OPP05vvfVWyNFOcvp3vXIDTsjMzKR+/fqpBnCrZGFO8sHIyPfJ7JmRE4uZhdaylsWoDpGMnPdk9erV9PHHH4v/a7VYV69eFenbeWnevDllZGQQkTpJIBCorxSKNoVDs/WE3EhoLa5cuSLaZDOtbAY1EgJPnDhBI0aMMNQglihRgvr160d79+61/e7n5+cLwYAHWiuNaDiw1iyc6vU8efJ4POL50ZYpyczMVNVJq1mzpqWA2LZtWwICuZo4BQKH50cKNl+aVUYnUodcW2UzliOljEz9LmpcYUaiqIUZOTnZ7NmzhQlDrrsyaNCgkNoeP368YwFEzhorJwXjWXeXLl0c9YFn17NmzdLdzvVKzOrl2A3jfP311x31TQ/22QAip5YmChQpHDhwoKUgwx/IUAZvOWSYBQMj+vbtK/Y18klgfwyrHEQ8YAD3VPvnz58X64xC1vfv368S5L1eLz333HNihiovTrUp4ea8sYuRP5ZsqvX7/bRmzRpq166deB/kxev1Us2aNWnGjBmWv5sZLHyyuQIonIR5RCSyEHfv3j3kNlhIady4sah7VKxYsaD9FEVRPa8JCQmm+XrY16patWqihIWV0OGUFStWiOfITOCUv112UjeYFaB1CcYVZiSKWphp1aqV4Us8cOBA8SLYrckkwzMfp6n/5Sq969ato4MHD4qX7Pe//72jtnjG2r59e93trII1KlpJZF8zww5/oSJfZ1paWkTTvzO7d++2dS2hOn3z8VYZo/1+v8hjU65cOd1r5YF3wYIFhu2cOXNGCGipqaminYyMDNEXq3fr5Zdf1jWxaIUPI21KdnY2HTp0iJYuXUpjx46lrl27qmbzZkuZMmWoSZMm1KtXLxo3bhwtWrSIdu3aRRcvXrT8/a1MxD179qSqVavq+rQkJiZSly5daNOmTRHL+Mpmy7lz54pMtfHx8YWSUZad+0Px7SJSPx9vvfWWmESYJbVbtGiRqm6TXrV0onsJO8uUKSPKqoRTwkCP7Oxs0X+jGnVEREOGDCEgkNHdDpFK7PhzwRVmJIpSmJEjIvQeXkVRhHOsz+ejixcvOmp/zZo1BISW+l/PMQ8AVaxY0dEMmV/mMmXKBG2THen27t1r2EZ+fn6QD4bRUrx4cdsOyjIZGRlCE5GYmGg7UZxT7GqZypUrR+PGjXMcycFCwbPPPmvopMzIflB6NbSMMgozubm54neJjY1V1fi6deuWaJsTHpqRnZ0tSmWYLfHx8dS8eXOqXLkyJSYm2tJ0hbP4fD5KSkqiihUrUsOGDalbt240atQomjt3riryzGphTdC4ceNCKlJoB7n4o6wZC6UYpBXslxJqnSvOiBsTE0OKotgubHrkyBGVduvJJ58MEjpnzZpFAFQ+aZHMvMxw5B5rI/VIS0sjwL5Gm7WJ4USJ/ZxwhRmJohRmBgwYIF5oo9nTlStXhPrYaAZtxJ49e2x9IPR47bXXDD/KTvwN2G9Bz6mNiypahTjKJje9/gAQHw1eypYtS1u3brXVR7/fLyKKvF4vnTp1ytZxoeDE/4eXmJgYql+/Pk2fPp0uXLhg2LZeZIuVeYadxD0eT9AgyyaUqVOn6h4rF5TUOvHK2aLtCuGh3Bt5iY6OppIlS1LNmjWFZsJqad++PbVq1YqqV69OJUqUoNjY2JAyQBstderUoeXLlxv6g0USnvgMHjyYiO6ZcH0+n+Oq6Vawg3yHDh1COp6FYDZTyRWpre7VzZs3VZOt6tWrqyYfsjM2vz+FAYdSm5nI+dttNzyfIxwbN24coV4+3LjCjERRCTO5ubmiMKBVSvBNmzaJF9dJ9ICcMt+J81sk/Q1k7Yu25kiLFi0IMA+9la89MTFRRIrwIud6OXbsmEgbLg/m2oFWG2LdvHlzsX8kIx70sBOOX6xYMWrVqpVhXSjWTsybN09oQ4xCdK2Ez+zsbBExUqVKFdU2nnmOHTs26DgOAwdAL7zwgm7bvP306dO27o1drVW1atXoySefpJkzZ9LGjRvp3LlzQcKw1X22eoZv375NR44coTVr1tCMGTNo8ODB1LZtW6pZsyaVLl3a0izGSyjp+EOFTT+sBbh586YY0Pv27RvRczVu3JgA4yruZsj5jrjunCz8njlzxrINbbHN+Ph4kQ/pz3/+s+o3KCzBYMOGDabf15MnT4rtdrSTRPfSWZQtWzbS3X0ocYUZiaISZtjhzePx2DJpyPlBVq1aZescsl3XiYkq0sn8OMLjpZdeUq3nQVTPxEFEtGPHDjEYJScn0/Xr123letm/f3+QpqZGjRp06NAh3QrlvEyZMsX2PQoHFjy0A62e4HH58mWaPXs2NW7cOChhIC/JyclhJU+TP8qySYkFR21IqVzsUy8nEcP72FXxR/q5K8wcLEVZF8kI9r+TfxP2H/F4PHTp0qWInYtzWFklxNTj8ccfJyDYT5B/KzMfFC1Lly5V+dGMGDEiKDo0OTm5UPLtyJFKepMgLp7rxGeRK8DHxsZGsqsPLa4wI1EUwoyiKGLW3bFjR9vH8ADt9Xpt2935ZTt48KDt/kW6CByHe8v2dVlrpJfAbs+ePeIjlZiYGFKUx44dOwzDX/WWwkowpoeeUCVrmYw4ffo0TZo0iWrXrm07QZ2dgZVNRl6vly5fvkxE9yLY5Gi6zMxMIVSlpKSYVuF2+uyFq03RQ0+gSUpKCvu3tpP7qFKlSoWSfdcINivJmk6/3y9+r5YtW0bsXKzNdNqmoiiiPyNHjlRtY4F8yZIljto8duyYYc6mSAmvRrDA36tXr6BtzZo1I8BZniS76RVcArjCjERRCDMrV640HciNuHr1qnjhS5UqZStKgU1ZTswnkZ51sjpYti1PmjTJcNZy4MABoR6Pj48Pe0a5YcMGW2n2Cyv1uxF2tExWHD582DA7shPh89atW8J0Urt2bSK6J4T27t2biIIFaivfIqcZhInMI4RCGZBk8wU7yEZqULeKZqpVq9Z9HZBYe6stZyEnwzTK9+QUPS2QHWSz8ZUrV1Tb2EcrFA3pjRs3xLcuUoKwHfjbppfThyesEyZMsN2enOSvMH33HhZcYUaiKIQZlubtZuaV2bFjh3jY7eR4YFOOk3wT+fn5hv4a8tKxY0db2YBZePP5fGIdh2y3aNFCte/Ro0fFRyk2NpbOnz9vu99mPIhmgUhh99pKly5NM2bMMHQGXb58udj31VdfDQq/HTx4sNi+YsUKy36xQLp9+3ZH1zNlypSgvpcsWTKkmbX8vrA/gpO8S2bI2avlRfanadas2X0TaPi+aWsQKYoiHG6rV68ekXNx0sOmTZs6Oo7NU3p5X7QOzE4oqvdbfr7kb6EczXf48GFHbbLG1UkJh58rkRi/3arZIfLee+/h66+/BgAsWbLE8fGDBg3C6NGjRVvLli0z3T82NhYAkJmZafsc69atQ05OjuV+hw8fRpkyZbBp0ybT/QYMGAAAUBQFn376KQoKCnDp0iUA96oPA8CJEyfQqVMnKIqCmJgYfPzxx6ICcbgUdoXyooQrEHMlYyO++eYb/O53v0OJEiVQs2ZNLFy4UFXZe/LkyaIi8Jw5c+D1Bl7zjIwM/PrXv8aOHTsAAP3798ekSZMs+8X9Masersfx48cBBKqAc+X1gQMHqp4Vu/zlL38BAJQoUQJ9+vQBAHz33Xei0nU4vPbaayAi8X++3tatW+Ppp58GAPzrX/9CkyZNQq4u7QSuuqy9316vV3wnLl26hG3btoV9Lp/PBwC2q9Zzvz755BMAwKhRo4K2Fy9eHICzbxVTVO93//79xe++YcMGsX7r1q0AAve+ffv2jtpMTk4GAHGvXAoZJ5LPqlWrqGHDhpScnEzJycnUunVr07wiRIHQ4Dp16lBcXBzVqlVL10yyY8cOqlu3LsXExFDdunUdOY4x91szU79+fQLC91bndOJer9fU+5+1QEOHDrXVrpw4rnjx4rqVordv305TpkxRzUpbt25tGv7J5rFXXnlF5UDKzs+nT58WM9qoqCjH+VWseJg1M0TmTsUA6Ne//jW1atUqSBXv8XioUaNGtGbNGlIUha5evWqqri9ZsqRtTQPPMJ3k/VEURRz3wgsviBBgrQbPLqwBbNeuncoh3okPmRH8LqekpBAAoQXl3CkjR44U56tdu3ahJK+TWb9+fZAGVIZ9oCJR5oAz6zZs2ND2MXL5Aj1fK26zXr16jvtj9/2Oj4+nMWPGRLRcAN9X2SeQcyZVrFjRcXusEbXrT/lz5r6bmd599/9v787jY7r6P4B/ZsmKkCCEBBGVxBr7GvterVpqqS3atEUXy4OH1lP18FCtklJbKWpvJbRaqlQjYqulxBLEEgQRS5NILFlmzu+P/M7pTDLLvTN3QpLv+/WaV3XmzL3nJpOZ75zl+93Bdu7cyS5dusQuXbrEPvroI+bk5GR2vcjSpUtZmTJl2JYtW9jVq1fZ5s2bWenSpdmOHTtEm8OHDzONRsPmzJnDLly4wObMmcO0Wq3s5EiFGcwkJCSIPypz2Sulun//vlhI5+npaXbrNX8zl5Ka/9q1ayLocHd3ZykpKRbXdcTHxxtlWnV2djZbGJOvtejatauYruDD/fHx8eJaNBqNYvP6hhyxuPRFI2VRcU5ODluyZAmrW7dugZ+FVqtlrVq1ElV+zd2kTvfw19KiRYskX4Phzqr79++LqSGpGVTN9YHvpOPTp/buXMvKyhLTaHyhp2Gla16c9N133xX3BQQEKFofKL/du3eLc5myb98+8bi9ZQ54bSI5gQd/r2jQoIHJx3mCzfz1maSQkvLAVBBvSxHZ/N544w0GGO/O4l8ibdkSz3d7KTUlWJy9EGtmPD09zRawa9WqFZs0aZLRfePGjTPKaTBw4MACRc66d+/OBg8eLKsfhRnMdO7cWXw7UGIe3TA1frt27Uy2kbrrICMjQ8yrazSaAjlhLJk2bZpRJtZGjRqxlJQUoza8tkqVKlXEH3r37t3ZlStXxDdatVqtyDdmc+RsiS6q5Cwqfvz4MZs5cybz9/eX9AEgN+jjv9d58+ZJ7j/PPcTXVPBK4OZGGywxLNzHd//xzLX2LgJeuHCh+HnwbLbly5cXr23D8h0ffPCB6Ef16tUdlkAvPj5enMdc0KRUmQP+XiZ13Z9h+YK1a9eabMPTVXh4eNjUJ2sLsufPn8+6du1aYBdg6dKl2VtvvWXzaM2ePXuMAnCdTifeY2zJRs5/DmXKlLGpPyXJcw1mcnNz2ebNm5mzszM7f/68yTaNGzcukN5/6tSpzMnJiWVnZzPG8ob28tcJWrBgQYHEX/k9e/aMpaenixv/I3N0MJOamipe4GPHjlXsuGPHjhV/SKZq6PA3Hb5DxRSdTiemrQD5CzYZyxvV4aNA/Fu+4RZLXnvFMOhZsGCB2FKpVqutTj0qwdYt0cXd/fv32fjx441GFyzdpEzH8bTzM2fOlNQHnU4nprgmT57MGDOuTC13VxvPMG1YzoNP/di7CLhevXoMyNsGPWbMGDEyw/8enZycjL6wTJo0SVxH1apVJS2cl8twJ4y5TNFKlTkwtQ3ckvzlC0yRUp/JmlatWpl8vTo5OYkgMicnh82ZM0csODa81a9fX/ZojeHrdu7cuUZV6W35XPnuu+/EeyKx7LkEM2fOnGGlSpViGo2GlS1b1uJ2zWnTprHKlSuzEydOML1ez44fPy4SId25c4cxlleDZuPGjUbP27hxo9U/hBkzZph8sTs6mOFvohqNxmJuDlvwOVaVSlVgrQkfurU0d9u1a1fxc5D6wWPO7NmzjdZc1KlThyUlJbFly5aZ/JbP/1uYVWKV2BJdXCmZY4gX+TNX2yk/vuYDgFEiSf56kru7g0+F1KhRQ9y3ZcsWcQ5bR0cNp5hmzpwpRl48PT2N6q19//33Rs+bPn26eKxSpUoOec/hx7c0wqlEmQO+JqRWrVqS2vNR3y5duphtw2uF2VJ+heOvubZt27JNmzaxDRs2iPcZU+sG4+LiWLdu3UyO1owaNarACLM5AQEBDMhbO/j222/bNbJy7do10Y/ly5fTe5QFzyWYycrKYpcvX2bHjx9nU6dOZRUqVDA7MvPkyRM2atQoptVqmUajYVWqVGFTpkxhAMSLy8nJqcAb6oYNG6xmTnweIzOGiaLklCOQKjU1VXwL9vDwMBrG5vlc8m/X5Ay3wQ4aNEiR/iQlJYnFkflHY0zdJkyYoMh5if2UXCjNF8aaK3eQH58SzT+6yutmyZ1CrlChAgOMK7Ibjl78/vvvso7H8do/KpWKZWRkiGkB/jfGEzW2bt26wHNnzZolzl+hQgXFC5pKySulRJmDl19+mQHS1nUYli/gJQdMkVOfyRTD165hjha+Rk+lUomEkPnl5OSwuXPnmh2tyR+Y5scDmFKlSrHg4GAG2L5o3dR0mbUaayXVC7FmpnPnzuydd96x2CY7O5slJSWx3NxcsSiYf5uydZopv8JYMzNz5kzxosyfKEophkObhinG+ZunqeR0vKo2ID9fhBQLFy6UlKW2qC+8LU6UXCjNq0qbK1dhyHCofsqUKUaP8eJ7cjKp5uTkiGtYv3690WN8EfAHH3wg+XiG+A4rPirB09bzBaB8pCb/VBP3+eefi5+nl5eXojtrpK5TsrfMQZ8+fRiQtwbIGnPlC/KTW58pvy5duoifqaHHjx+LheBSkvydPXuW9ejRo8B7V6lSpVhYWJjJ0RrDgI0/j0+VymFrjbWS6oUIZjp16lQgnbUl7dq1MxomHDhwYIEU0j169HghFwDzN3W52TLlmjBhgnjxz5gxgzH2T8Xp/Asoo6OjxR9N5cqVHbbLYtu2bYp90yeFw9xCaf6mKvUNlY+MhIeHW23Lq6sDKDBaMXr0aDHqKJXh9v/807p81NCWb845OTliVOPTTz9ljP1TdJP3Lzk5WZzb3HScYYLCsmXLioKh9uKjYdZGO+0tc9CvXz+To2j5WSpfYIot9Zn4eXjAMmrUqAKP8/VTANju3bslHTMnJ4fNmzfPZNmKunXrGo3WmBpN8fb2lhV8KFngt6Qo9GBm2rRp7MCBAywxMZGdOXOGffTRR0ytVrM9e/YwxvIW9xoWs7t06RJbv349S0hIYH/++ScbNGgQ8/LyYomJiaLNoUOHmEajYZ999hm7cOEC++yzz17IrdmGc/SO2HKcX4MGDcQL//Dhwya3axpuwXZzc5M8L2wLpes8kcJhriCnnMXhfGfPiBEjrLbl25tNfdP//vvvxWta6jqXt956y2wAxEclzE29WrJ48WLxs+CLeKdOncoA4zUSfLrCUiFGw5HRMmXKKDJqy3/mQ4cOtdrWcP2g3PcmPnXj6+trsZ2l8gWm8Pel+fPny+qP4fusufPw6cpKlSrJOjZjeaM1PXv2NDlaw9dmyRlNefr0KTt8+DBbtGgRe/fdd1mnTp2s1vqiL34FFXow8+abb7Lq1aszZ2dnVrFiRda5c2cRyDCWtzjWMEFQfHw8CwkJYW5ubszDw4P16dPHZFr7rVu3ssDAQObk5MSCgoJsGoJzRDBjuMCUz5/Lnf6yVXp6uhhGL126NDt+/Lj4I3j69CnLyMgQI0Vyt2DborgnqyvO+OuY7zIB8socSMU/0K2txcrJyRFTTNOmTSvweGpqqji/1GSKfN2CqVEHHhwBkL09mX9ZMFwr8tFHH4m/N46vo9FqtRYDsLVr1xoVVLW3Dhmvp2Wq8GF+9pQ5GDx4MAOsJ4WzVL7AFFvrM/Et/ZZy1OzcuVP83m3Ns5OTk8Pmz58vEuVJubm7u7N27dqxgIAAVq5cOdkFYumLn3kvxDTTi0LpYMbcN1pb5+dtERMTI94gGzZsKPqwdu1ao+3T1ha1KaEkJKsrCfg6ETnZY3mixL59+1psZzhCYW53Df/GPnv2bEnn5mtHTK1bMFwEbPilyhrDKSaehI8xxv7zn/+IYIRLSUkR58i/Zie/LVu2iL8PNzc3u+qR8TwylkaEDBkmKdyyZYvk8/BkhvmLWhp6+vSpyZ+XJTwANqzUbo3h7rL8663y4z8fFxcXu/P9nDt3TqznsvWm1WqZp6cne+mll0TgR1/8pKNgxoCSwYy5xVv8Q7swF2/xb4vmbny+vzCUhGR1xV1MTIzsIDh/1W1z+Ju44Rbq/PgXBCm7bwzXrJgbeeSjl1IWJ3OG1e4N3y94mn53d3eTfW7evLnVY2/btk38fbi6uprNjm4Nn/IIDg6W/BxbyhyMGDGCAZanbPjGB3PlC0zh65lCQ0Ml94UnMJTyPn7lyhXxc5a7vtIUqdPoVapUYV27dmVjx45ly5cvZ8ePHy8wKkhf/OSjYMaAUsHMi7Z4y1o2zMIOIChZXdHHP/T8/f0ltQ8MDGSA5eruhqMdlvLRdOzYkQF5JQGs4dNilrIG85EmOYvy+Shn/ikZczsG+S4na1NN3M6dO8XPwtnZ2Wh7sVR9+/YVf1tS2VLmgCfBq1ixotk2fGTOXPkCU2ypz8RfZ7Vr15bUnufesnUnlyGlp9Hpi588Snx+a0GMxMbG4tatW2YfZ4whKSkJderUEVVROcNqx6b+LfdxxhhOnDhhti8qlQrjx49Hnz59RPVbR+vXrx/69OmD2NhYJCcnw8fHB6GhoYV2fmK/2bNnY+TIkUhMTMSff/6JFi1aWGyv1ea9TWRnZ5tts3LlSlHBevLkyWbbNWvWDNHR0ZKqHv/2228AgCpVqpht07RpU5w9exZXrlyxejwAyM3NxdmzZwEAb7zxhtFj/DqZQQVtAJg2bRoWLFiA3NxcrF+/HiNHjrR4jl69emH37t3o2bMnsrOz0bx5cxw6dAjNmjWT1EcAqFChAgDgyZMnkp/TqVMnhISE4PTp0/jvf/+LSZMmiWsyx9w1c3fu3MGNGzcAABMnTpTcl0qVKgHIq2wuxaNHj5CQkAAAePfddyU9Z926ddixYweysrIwYMAAnDx5UnL/8gsNDUXFihVx//59k4+rVCr4+voiNDRU0vH69euHyMhIjBs3zujzxNfXFxERETZVjidWKBRYPXdKjcxIHW58kW4090rk4gtGpYxo8JEMU8nj8rextgDVsP6NtTQCfEdPnz59zLbZunWrOJ6URcC8RpSp9wqeN8bV1bXA8/holpxt4DExMWJBtFartZhoLj+eZdhw/Y4Ucssc8O3y+XO6cHw3maXyBabIrc/Ed5Kp1WpZ6SXmzJkjrteeMir3798X67nM3WwZTaEs5dLQyIwD+Pj4SGoXHBwMDw8P8Y2GGXyzMfdvS4+bui81NRU3b9602hcp33IJMTRhwgTMmDEDJ0+exI0bN1C9enWzbZ2cnACYH5kxHO0YOnSoxfO2b99e/DsmJgZdu3Y12U6v1yMlJQVA3kiHOb179xb/3rt3L3r27Gnx/CtWrAAA+Pv7w8PDw+gxPrpo6m928ODB+OKLL3Dq1Cno9Xqo1WqL5wGAdu3aITY2Fu3atUNubi46dOiAPXv2oFOnTlafy0c2LI2GmRIYGIhu3bphz549WLlyJebOnYty5cqZbW9tZGbbtm3iWqRcM8dfT8+ePZPUfv369QCARo0awdnZWfJ5pk2bhoiICNy7dw9hYWHiNSOHXq9H48aNkZ2dDZVKhQoVKhQYoXFxccGrr74q+9gajQYdOnSQ/TxiA7tDqheE0mtmrJWg37x5s0I9N+327duSqyDTyAyRyzAJWv6q9fnxon8hISEmHze3oNYcXpTUUmZVwzUM1uoO8To+1gq/Gm4dN7Wuh5c3MFUX7uHDh6I/q1atsnie/I4fPy6+9UstxMqTVNpS30hOmQM+gmIqq++xY8fENcsZVWJMXn0mw0rctmxXNkysaKpIrzU8szEAtmjRIqPRlG+//VY89u6778o+NpGGFgAbcMRuJmsBTY8ePRyScTd/kUdzN1oVT+wRHh4uPmAt/d20b9+eAXnZUk0xlbPFEl7Mr2PHjmbb8KrVUqZZpC4CtrZ1nCfSM1fkli+EtSUDeFxcnAge1Wo1++mnnyy2NwwkbCmkKbXMAV/cbGo6qEePHmYDHWvk1GeydSrLEN9JJ3ertmHCQVMFLBn75+egVqsVLVlB/kHBjIHCyDPj5+fHIiIixFw+fxOIiYlR5JxXrlwRb/T4/3n2YcOG0ap44hAZGRniG3xYWJjZdrxWTmBgYIHHDHOD8NIb1vTu3ZsBlhOj8QDJ3GiQIf5haO1D11J2YsYYW7p0KQPyajGZwtd1aDQa2Un6GMtLIsrz5qhUKotZmA3z29hSIkFqmQNe+Dd/ZWjDkTspmZ9NnZ/3Py4uzmJbLy8vBoB17txZ9nm4a9euifdIqYV2DZPvWdoCn5qaKr5cdurUyeY+EvMomDHg6AzAhou3dDqd2NLIbyNHjrT5WwVjeXVhDKtSh4SEiDcx2g5NHKVXr17iW7G5D2jextTIi2FtIl4WwBpeX8fcCAhj8opIGtYNM3cNhlNMU6dONdmGj9xotVqTjxtmMF6+fLnVfply5coVMc2mUqnYhg0bTLbT6XTiXHJLu3CGow4HDx402Ybnsco/Ama4EcJchWpreHARGRlptk1cXJw4z759+2w6Dzdo0CBJo1GMMXb9+nXm5OQkvpBae+1++OGHop+OzrZeElEwY6AwCk3mFxMTIwrCAXmFHuUmyTp79qxRuXpnZ2eTb5S0Kp44wo0bN8Rrz1x219dee83siEb+ytNSnDx50uJ0j+H6FClrNZ4+fSra79y502QbSwUwOV7M1VwwwxhjNWrUkDxiZM7169fFOh/A/BocKcGAJVLKHPCAJ38ww0exrNVsskRKfSaeT8ewhIStnj59KkaTGjVqZLZdVlaWKJ6q0WgkvWfrdDrxO3vppZfs7isxRsGMgecRzDCW94fRs2dPo+kfw7o0lkZ3Ro8ebTR91Lp1a6uLHQlRGq+HY26ahn/jzT8tZDjFNHPmTMnn0+l04nVvKvU+n+5Rq9WSRzvLlCnDALAxY8aYfLxJkyYWp5gYY+y7774TH3Dm8JEMtVpt01QTd/v2baMvQkuWLCnQho8cLFq0yObzWCtzwLP7GiYKNCxfMH36dJvPzeszWRpd4yNwcsoeWDJ37lyrgW3z5s1FG3MjY6bwYFfu84h1FMwYeF7BDLd161YxH86j95UrVxaYHvL19WVz584VlV/5GwkVHSPPi+Fi09WrVxd4fPjw4WLk0RBPP69SqSRPMXH8g9xULhS+psZSvaD8+BobU9/IdTqdmGKyVPOHV4a2FMzw9xlzAYgcKSkpYuQEAFu4cKHR4/yD3lJGZSkslTngeVoMc+vYUr7AFGv1mQxzDtla9sEUb29vBoB5e3sXeIwvLJc6hZkfXwRerlw5u5YVEGMUzBh43sEM70OzZs2Mghdrt65du9r1hkGIEvjCc1OLcvmup/wp7+vVq2fzsDt/rql6R/xD0NqWcVN9NLUrZ/Xq1VanmBjLKxTJR10s4bWq5KT3N+f+/fusfPnyon+GZQj4/eHh4Xadw1KZA1OJAvkHdv369e06r7X6TLz+VIUKFew6T36GW7X/+9//ipHxadOmifulFvDM7/Dhw+IYpirDE9tQMGPgRQhmOMOCaeZuKpXK6vZMQgqL4SLa/Asx+bdZT09PcV9WVpaYKpo1a5bs8w0ePNhkgMQYE6MoERERko/H85oABTML82m0atWqWTwGzyZsLZjh60zkZqs1JzU1VYwmAP8Uj61WrRoD8rIO27tOjmdodnNzM5oe4/WvXFxcGGN501+WRunk4Fv6Te0U0ul0YhrNEflb+JofU7eyZcva9Xtr3bo1A/J2vckdkSSmKfH5LT2lI5EsJCTEahvGWIEMpIQ8L3379kXFihUBAOPGjTN6zNXVFQBE7SUA+Prrr8EYg0qlklWzh+M1bv7++2+j+48fPw6dTgcAGDJkiOTjGWb+3bNnj/i3Xq/HqVOnAACDBg2yeAxLGYANTZo0SRybZxS2R7ly5XD16lWRffzTTz9Fv379cOfOHQDAiRMn0LFjR9SoUUNk5JVr8+bNAICnT5/igw8+EPfz7M78mj/55BNxv7UaVNZ4e3sDANLT0ws8tmHDBuTk5BidU0nh4eFmH0tPT8cvv/xi87G3bt0KtVqNnJwcDB8+3ObjEGVRMOMAUssLUBkC8iKZOnUqAODcuXO4cOGCuN/FxQUARJABAN9++y0A4KWXXoK7u7vsc/HgQ6fTGZXs2LJlCwDAzc1NfBhK4ezsLAq/7ty5U9y/fv160e8pU6ZYPIbUYqmlS5dGrVq1AOQV2FRC6dKlceXKFfj5+QEAtm/fjtzcXKM2t2/fxoABA2wKaIKDg9GtWzcAeX3mBSDzlzOwtXyBKVWrVgUAZGZmFngsIiICAODn52exkKgtdDod5syZY/ZxXqDX8PUsR5UqVTBs2DAAwE8//YSrV6/adByiLApmHEBqfSep7QgpDOPHjxeByZgxY8T9+Udmnj17JoKdsLAwm87l7+8vggfD4CMmJgYAULNmTdnH5M85evSouG/JkiUA8j40eSVqc6xVmDbEv5GfP39edv0kc9zd3XHhwgWzQRUPOGz9IN64cSPUajV0Op3ov+Fo1PHjx5Gamgogb3TIXubqM2VnZyMuLg4A7B79MSU2NtaoUnV+jDEkJSUhNjbW5nOsXLkSrq6uYIyhf//+Nh+HKIeCGQcIDQ2Fr68vVCqVycdVKhX8/Pwkl5MnpDCo1WoxPH/gwAE8ePAAQMFgZvHixWKKKf+UlByenp4AgIMHD4r7EhISAACtW7eWfbyWLVsCAK5duyb6y6eYXn/9davPlzoyAwATJ06ESqWCXq/H0qVLZffVHMNpNlPs+SCuUKECRowYASAvgExMTDSaZuLTPR4eHmjbtq0NvTfm7+8PAGI6iYuIiIBer4dKpcK///1vu89jKC0tTfK0lT0j487Ozpg5cyYAIC4uDr/++qvNxyLKoGDGATQaDb766isAKBDQ8P+PiIiQ9eZJSGGYN28etFotGGN4//33AeRN+QD/BDNr1qwBANSuXdumKSauRo0aACAqbmdmZiIjIwMAbPq2y6trP3r0CNnZ2diwYYOYqpk2bZrV51urIG3IcKpp1apVsvtq6M6dO5g7dy5CQ0MtVgg3tH79ejx58kT2uVasWAEXFxcwxjBo0CCjYCY6OhoA0KdPH9nHNYX/fBhjRn3lU3NBQUEoXbq0Iue6dOkSOnXqBC8vL8mBnr0j41OmTBFToaNGjbLrWEQBtq8/frG8SLuZOCpDQIqiAQMGMCAvE25WVpZRmv+nT5+KXUz5t/nK9fbbb4vdJYz9k+BNpVLZlMPDsB7QTz/9JJKjSc1iGxMTI54vBS/LoFKp2Nq1ayXvOLp48SKbOnUqa9KkiShtYOutcuXKrH///uznn3+W/DMzLHPA/22YvNPW8gX5marPZJjdefHixXafY/fu3axOnTpGPxONRmOU88vcbfDgwXbnijGs7/T555/bfT0lFW3NNvAiBjOMURkCUvSkpKSID7eJEycaZcadN2+e+PCTU53YFJ6kjgcv/fv3N7tdWyqejC88PJxptVoGgE2YMEHScw8ePCgrmNm4cWOBD0hfX98CX1aOHj3Kxo4dy+rUqSPS7ee/abVaVrNmTTZ8+HCjvDOmbmq1ukDhWX5/QEAAGz16NDt16pTZfhuWOTDcEi4n8JMqf0kGXqXb1mKdjOX1f/HixUYFfwEwd3d39v7777PHjx+zqKgokz+j/DdPT0+zdauk4gkbXV1dFdmqXxJRMGPgRQ1mCCmKQkNDGZBXs+f7778XH5ZBQUEMAAsKCrL7HIbf0k+dOiWStdlTmTgkJIQBMAoIUlJSJD336NGjkoMZax+W7du3ZzVr1hQBVf6bi4sLCw4OZmPHji1QSDIqKsrkc1QqFVOpVCwqKoplZWWx9evXsx49eoiq06bOERISwmbMmMFu375tdA7DMgeGtzJlyig6cszrM/FRCx8fHwaAtWjRQvaxnj59ysaPHy8yI/Obt7c3i4iIKDDKYurn6Ofnx77//ntRb4zf3njjDZtHaRISEsRrYeTIkTYdo6SjYMYABTOEKOfcuXPijZ5PBxlORyg1pM4Tp82dO1d88Ns6fRUVFVVg2katVkv+cD5+/LikYCY3N5dVqVLF6rf+/KMGjRs3ZlOnTmUXL1602hee6C//B7G5a3n48CGbN28ea9GihdmpKw8PD9ahQwe2fPlyk6NKhr9jpQIaw/pM165dE+eQUzzz9u3b7LXXXhPJFPktKCjIbP0lxvKyK/O2ixYtKjAyvmfPHlHTCwDz8vKyeZTm1VdfFa+3/IEjsY6CGQMUzBCirODgYAZAfCAZftjZO8XE8dIFPLU9AHbjxg3Zx7E0UiLlwzklJYV9++234jnh4eGsV69erFmzZqxWrVrM29ublSpVyuxIi6lbvXr12Jw5c2z6cOPTP4aZbKVUEOcSEhLYxIkTWVBQkKw+85+Xn5+fIlPihvWZRowYIUaMpDh+/LhRUUjet/bt20uq5fT777+L55kbdcnKyiowSjN06FDZozQZGRkiMG/Tpo2s5xIKZoxQMEOIsgwLARre6tSpo9g5eMp7T09PWR90hnJzcwsstDc1MtKjRw/WpEkTVrNmTVaxYkXm7u5e4Nu+kjdbi8c+ffpUHGP9+vViWsWeGk3R0dFs+PDhzM/PT9JaEgAsOjra5vNxvD5T27ZtxTqd7t27W3zO999/L+pf8ZuTkxN74403LNbWyi8iIkI81xpTozSHDx+WfC7GGJsyZYp4vtznlnQUzBigYIYQ5fEgw/Dm4eGh2DQEXxDKP2B9fX1ljwjwApFK3ry8vFiNGjVYo0aNWNeuXdmIESPYtGnT2PLly9kXX3zh0GCAF8ZUqVQsKyuLtWzZkgFgNWvWtOl4+a1bt05S/ytWrMjCwsJYTEyMzefiwaq/v784rqmpHJ1Ox2bNmlXg9Va2bFk2ffp0mxYL85pi5cqVk9Q+KyuL9enTx+j8w4cPlzxKo9PpxChmjRo1ZPe3JKNgxgAFM4Qoy9xCVP5Bq0RA8/HHHxc4tqkdQdz9+/fZ8uXLWd++fZm/v79YYCr1w7lJkyase/fuLCwsjE2fPp2tXLmS7du3jyUlJbGLFy+KtpZ2pVgbCbJ3mqZ79+4MyNt2zRhjc+fOZUDeDiB7txIzljdKIze402g0rFatWuztt98usGDZEsNt/jwQNpSens7CwsIK7PKqVq0a++677+y6zl69etkUWOzevdtolKZ8+fKSR1r4Dj3A/kKdJQkFMwYomCFEOY7+wGbM+q6dDRs2sA0bNrA33niD1a5dW1LuEHtGSq5fvy7aPn782GLb5cuXW+y7PYEe34nVr18/xpjxQtbff//d5uNy/HdrabqpbNmyrFWrVkYf6oY3rVbLateuzcaMGcNOnjxp8jxRUVGsdOnSRs9zc3NjUVFRLCEhgXXp0oWp1Wqjxxs3bmz3VmmuUaNGDABr3ry57OdmZWWJRb1yR2kCAgIYkLczTIngsySgYMYABTOEKEfqt3e1Ws1cXFxYmTJlWPny5VnVqlXZSy+9xBo2bMjatGnDevbsyQYPHsxGjx7Npk2bxubPn8/Wrl3LfvnllwI5TqTe1Go18/b2Zh07dmQzZ85kZ86csfrhLCXwSkpKEu2tvY+0adPG5HmqVKliVyCTkZEhjmW444dPXwwdOtTmYxuSs2A6OTmZzZ49m7Vo0aLAtmh+c3JyYsHBwezDDz9kZ8+elZznhf8+e/XqpViyPo4H43379rX5GLt27TIKyKSM0hjuipOa46iko2DGAAUzhChn06ZNdo2CKHnz8vJirVq1YlOmTGHHjh0z+W2Xf3ia+wBdtGiR1WtOTk4W7S0tND116pRoN2XKFLZv3z7x/z/88INdP/clS5aIgMJwnQhfe+Ln52fX8Q1t3rzZ5M9qy5YtFp+XlJTEZsyYYXcGYzc3NzZ69GiWkZGh2DUZ4qNKEydOtOs4WVlZrHfv3kZ9HzFihMVRF747T6vVstTUVLvOXxJQMGOAghlClCN1ZGbSpEls3rx5bMqUKeztt99mAwYMYN26dWOtWrVi9evXZwEBAczHx4d5eXmxUqVKMRcXF1k7iDZs2CC5z6bKh/CblCDAcDrHUqI9nj6/dOnS4gONj1jY+8HZqVMnBhTMxLto0SIR5Cg1dfHjjz+K6122bJkIBD/66CNZx7l+/Tr76KOPWKNGjWRNBe7bt0+R6zCHv85WrFihyPFMjdKYWz+UkpIiptB69uypyPmLMwpmDFAwQ4hyrK2rsHfNjNRgSe6OoPzlQ9asWWMUeFmSmpoq2prLDWO4Xd1wtIdvJe7SpYus/ubHty8PGTLE6H7+/gaA/fjjj3adg+O7fcqUKcMYYyKnS/ny5e067pdffinpd2vr1nUpcnJyxHmU3CadlZXFXn75ZaPrGDlypMkA0zDZpJS8OCUZBTMGKJghRFnmpm6UWOTq6GDJUOvWrcUxLWXfNVyvYi5xn5+fHwMK1o/q2rUrA+zbPm1Y3mHXrl0FHudlC/r372/zOQw1a9aMAWB169ZljBkHmPYsNHZUoCpHfHy8OI8jprF27txpNEpToUIFduzYMaM2OTk5zM3NjQHK5mYqjiiYMUDBDCHKc2Tld0cGS4bS09PF1l9L002GyequXbtW4HFecBMA27p1q9FjkyZNYkBeLStbzZ8/nwF5C2JNfdPnW7Z9fHxsPochvmvKcFFxxYoVGQDWpEkTm48rZbeUp6enQ4vu8txDKpXKYefIysoS27/5LSwszOh3x6cHlRxRK44omDFAwQwhjuHIyu+ODJYMGQYi5qabDKcmTI3g8IRupkZftm3bJp5r65qWtm3bMsB8XpRVq1aJD2glqjPzYGPVqlXivhkzZohz3L9/3+ZjW1uQDYDNmDHD7mswh1+Hm5ubw87B/fzzz0a7vPKP0vA6XvZO3xVnFMwYoGCGkKLJkcGSIWvTTTqdzuwaB564DjBdI8lwvU1cXJxN/eO7b8xVXjYcOZKzMNqUuLg4cSzDoCUrK0skuBs1apRd5zAVqPr4+DAPDw/x/2+//bZd5zBn6NChDMirqF0Ynj59ynr27Gl0raNGjWI6nc6oRtSsWbMKpT9FDQUzBiiYIYRYYjjdVK1aNZNtTAUkhmsfQkJCzB6fBwERERGy+2a4LdxS+QCem+fll1+WfQ5Ds2bNYgCYs7Nzgcd69OjBgLx6VvbKzc0Vu4rGjx/PcnNz2ePHj41qL9l7Laa0a9eOAXmVtQtT/lGaihUrsuPHj7MmTZowIK/2mFJFWosTJT6/1SCEkBLAw8MD33zzDQDg5s2bmDJlitm2ubm54t//+te/8PTpUwDApk2bzD7H09MTAHD8+HHZfVu1ahUAQKPRoF27dmbbtWrVCgBw7Ngx2ecwdOjQIQBApUqVCjy2YMECAMCTJ0/w3Xff2XUejUYDrVYLAKhSpQo0Gg3c3d1x6dIlNG3aFACwc+dONG/eHHq93q5zGUpJSQEA+Pj4KHZMKXr37o0HDx6gR48eAID79++jWbNmqFmzJlQqFbKystCrVy9s3rwZ+/fvh06nK9T+FWcUzBBCSowRI0agdevWAID58+fj0qVLJtvxD5nMzEwsXboUANCpUycEBwebPbavry8A4OLFi7L79fPPPwMAatasabHdkCFDAOR9SD558kT2ebj4+HgAMHk9wcHB8Pf3BwDMnj3b5nNwGo0GAPD48WNxn1arxfHjx9GrVy8AeQFg7dq17bomQ3///TcAoFq1aoocTw5XV1f8+uuv+Omnn1CqVCkAwNatW8XPITo6Gm+88QY6duyIGjVqYNu2bYXex+KIghlCSIny66+/wsXFBYwxdOvWzWQbPjITHh6O3NxcqNVqbNy40eJxg4KCAABJSUmy+3T+/HkAeQGTJf3794dKpQIAbNiwQfZ5uLt37wIA2rRpY/LxqVOnAgCuXLmCy5cv23weAGJkxjCY4Xbu3Im33noLAHD16lX4+/vjwYMHdp0PADIyMgAAtWrVsvtYtnr11Vfx4MEDdO/eHYDxaB93+/ZtDBgwgAIaBVAwQwgpUaRMN+l0Oty9exdbt24FkBdEVK5c2eJx+bRJamqqrP7cvHlTfNDzD3ZztFotqlSpAgCib3L9/fffyM7OBpA3LWJKeHg43NzcAAATJ0606TycpWAGyJtimz59OgDg3r17qFmzJhITE+06Z1ZWFgDTI0+FydXVFTt37kT58uVNPs4YAwCMHz+eppzsRMEMIaTEGTFihFh/Ymq6KTc3F8OGDYNer4dWq8Xq1autHrNDhw4AgJycHGRmZkruy8qVKwEATk5OaNasmdX2bdu2BQCcPHlS8jkM8SktlUqFkJAQk23UajUGDBgAANi9e7fJUQWpnJycAMDiFNKsWbOwZMkSqFQqZGRkIDg4GCdOnLDpfGlpaSJIaNCggU3HUFJsbCwePnxo9nHGGJKSkhAbG1uIvSp+KJghhJRIu3fvNjvdlJiYiH379gEAxowZg9KlS1s9nmFgsH//fsn92LlzJwAgICBAUvthw4YByBsBSktLk3wejl+Xp6cn1GrzHwHz588HkBfYzZs3T/Z5OCnBDACMHTsWW7duhVqtRlZWFlq2bIlff/1V9vnOnDkj/i31Z+oIFy5cwMCBA8ViYGuSk5Md3KPijYIZQkiJlH+6afLkyeKxjz/+GADg5uYmdvdYo1ar4e7uDgCyvmVfuHABAMyu38mvV69eIgiRMmKU3+nTpwFYX2zs7e2Nhg0bAgC+/vpr2efhnJ2dAUDsCLOkf//+iImJgZOTE3Q6HV5++WXZO6rOnTsHIG96y1Kw5gj37t3DBx98AG9vb9SpUwdbt24VU17WFPbOq+KGghlCSImVf7qJ49+S+/btK9Z8SOHt7Q0AiIuLk9T+0qVLePbsGQDg7bfflvQctVoNPz8/AMCPP/4ouW/c9evXAfyzxseSOXPmAMhbMMy3c8slJ5gB8qbR4uLi4O7uDsYYwsLC8Nlnn0k+H58y5IGloz179gz/+9//ULNmTVSqVAlff/017t+/L/rw6quvwtvbWyzczk+lUsHPzw+hoaGF0t/iioIZQkiJNnbsWLOPbdq0SdZOEz7acfXqVUnteX4ZFxcX1KtXT/J5+PocPsoiVW5urtjpI2UkqFevXiJ/jqW8PJbwYIYHbVIEBwfj8uXL8PLyAgBMmzYNH374oaTn8mCtbNmy8joqg16vx3fffYeQkBCUKlUK06dPF4uWtVot2rZti507d+Lx48f46aefsGzZMgAoENDw/4+IiBBbt4ltKJghhJRYOp0O06ZNM/u4SqWStdOELzjlW5+t2b17NwCgdu3aktpzb775JoC8LchSzwUABw4cEP/mW4at4Tusjhw5gkePHsnoZR4XFxcA8oIZIC/JXmJiohiFWrx4sViUbMnt27cB/DNKpqQ//vgDnTp1gqurK8LCwhAXFwe9Xg+VSoV69ephxYoVyMrKQmxsrMihAwD9+vVDZGQkqlatanQ8X19fREZGol+/for3taShYIYQUmLFxsbi1q1bZh+Xu9OET1mZ24acX0JCAgCgZ8+ektpz7dq1E9/k+eiOFDx4cnNzkzwNM3PmTKjVajDGxFoiOVxdXQFAbAeXw8PDA9euXUP9+vUBAFFRUQgNDbWYLZhP8eQPHGx14cIFvP766yhdujQ6d+6M6Oho5OTkAMhLyjdjxgxkZmbi7NmzeOedd8yu0+nXrx+uX7+O6OhobNq0CdHR0UhMTKRARiEUzBBCSiypO0iktuNJ7xhjZrMLc2fOnBEf8O+8846k4xviU1p8q7UUR48eBfBPtmIp3N3dRYmFdevWyehhHh7MSF0Im59Wq8Xp06fFz/bgwYOoW7eu2eCI7/CqUaOGTecD8hbyvvfee2Ihb2RkpAhQy5cvj7fffhu3b9/GjRs38Omnn0oODDUaDTp06IAhQ4agQ4cONLWkIApmCCElltQdJFLbVahQQXxA/fHHHxbb8vwybm5uNm0h7ty5M4B/du9IwUeC+EiHVF9++SUA4NGjR4iMjJT1XJ58z5aRGU6tVmPfvn2inMPFixfh7+9vcms63wIeGBgo6xzPnj3D7Nmz4e/vj0qVKmHp0qVGC3lfe+01xMXF4cGDB/jmm29E8kLyYqBghhBSYoWGhsLX19fsThMAsnea8IWn1opB/v777wCAOnXqSD62Ib6W5cmTJ5Iz5vJSAZaKWZrSuHFjMW0zY8YMWc/loxZ8asYemzZtwqRJkwAAd+7cgb+/P27evCke1+v1IsGflJ+rXq/HmjVrEBISAnd3d/znP/8RC4i1Wi1CQ0Oxa9cuPH78GNu3b38hkvAR0yiYIYSUWBqNBl999RWAgjtNuPDwcFnTAfxDn9dbMufKlSsAYLRQVI6mTZuKhHR8lMeSq1evioXMr732muzz8bIG8fHxFtcZ5cdHZpQIZgDgiy++ECNFaWlpCAoKwpkzZ6DT6bB582bRztLo0759+9CxY0e4urrizTffRFxcHBhjUKlUqF+/PlauXImsrCwcOHBA9nom8pwwGZYuXcrq16/PypQpw8qUKcNatmzJdu3aZfE5GzZsYA0aNGBubm6scuXKLCwsjD148EA8vmbNGgagwO3p06dyusbS09MZAJaeni7reYQQEhUVxXx9fY3eg1QqFQPA3NzcWGpqquRjDRgwgAFglStXNtvm8OHD4jxJSUk297tOnToMAAsJCbHaduHChQwA02g0Np1Lp9MxZ2dnBoC9/vrrkp/3/vvvMwDM09PTpvOas2nTJvE70mg0rEKFCka/P19fXxYVFSXanzt3jvXv35+VKlWqwOdN9erV2aeffsoeP36saB+JNEp8fssamfH19cVnn32GEydO4MSJE+jUqRP69Olj9hvIwYMHMWLECLz11ls4f/48tm7diuPHjyM8PNyonYeHB5KTk41ufNEYIYQ4mqmdJocPH4ZKpcLTp0/RsmVLycfi9ZX+/vtvs2145t5SpUrJWoybH99effHiRattY2JiAAAVK1a06VxqtRqvvPIKAGDHjh0WdxQZ4tNM9tR3MmXIkCHYu3cv1Go1dDpdgWrbt2/fRv/+/dGjRw9UrFgR9erVQ1RUlNFC3tGjRyM5ORnXr1/HjBkzCi3RHnEAeyMqT09PtmrVKpOPffHFF6xmzZpG9y1atIj5+vqK/1+zZg0rW7asvd2gkRlCiOLmzZsnvr2PGjVK0nOOHj0qnmPum35AQAADwFq2bGlX/+Lj48W5zp07Z7FtrVq1GADWvn17m89348YNcb6IiAhJz5k5c6YY4VJabm4u8/b2Njm6b+pWqlQp1q9fP3b27FnF+0JsV+gjM4Z0Oh22bNmCx48fi9wK+bVu3Rq3bt3Crl27wBhDSkoKIiMj8fLLLxu1y8zMRPXq1eHr64vevXvj1KlTVs+flZWFR48eGd0IIURJU6ZMEbuG1qxZg++//97qc5o0aSL+bSo/jV6vF4tMX331Vbv6FxwcLEaxreWb4cnkWrRoYfP5qlWrhuDgYAD/7HCyhhfplDqSI0dsbCzu3btntV3Dhg2xe/duZGZmIioqSla2ZVI0yA5mzp49i9KlS8PFxQWjR4/G9u3bza4ab926NTZu3IhBgwbB2dkZlStXRrly5bB48WLRJigoCGvXrsWOHTuwefNmuLq6ok2bNrh8+bLFfsydOxdly5YVN54lkhBClLR7924xNTNs2DDcuHHDYnutVisWvR48eLDA4/v37xcLcfmOJHsEBQUBAPbs2WO2TWZmpqiNlP/LpFyffvopACApKUlSOQUezEjNoiyH1Pw///73vyVnPCZFk+xgJjAwEKdPn8bRo0cxZswYjBw5EvHx8SbbxsfH48MPP8Qnn3yCkydPYvfu3UhMTMTo0aNFm5YtW2LYsGFo2LAhQkND8cMPP6B27dpGAY8p06ZNQ3p6urglJSXJvRRCCLFKq9XiyJEj0Gg0yM3NRcuWLa2OMlSoUAEATI4y8yrQHh4eiqTc57uhLH0B5Jl/gbxCjvYYOHAgypQpAwD417/+ZbU9b+uIkRml8wSRokt2MOPs7IxatWqhadOmmDt3Lho2bCi2NuY3d+5ctGnTBpMnT0aDBg3QvXt3LF26FKtXrzYbUavVajRr1szqyIyLiws8PDyMboQQ4ggBAQFi0e7du3fRu3dvi+39/f0B/LP92hBfiBsSEqJI33j24JycHBw/ftxkG57TxsPDw2y6fTmGDx8OIG+UyVrNJR7MMMbsPm9+oaGhFpPXUUXqksPuVzVjzGya6idPnhT4w+H5Gsy9sBljOH36NEXShJAXyogRIzB06FAAwK+//oqIiAizbfmajPxf2vR6vRhF7tu3ryL9ql69OkqVKgUA+Pbbb022OXHihGirhLlz50KlUkGv11tNoleuXDkAjglmNBoNtFqtyceoInXJIiuY+eijjxAbG4vr16/j7Nmz+Pjjj7F//37xBz5t2jSMGDFCtH/llVewbds2LFu2DNeuXcOhQ4fw4Ycfonnz5iKanjlzJn777Tdcu3YNp0+fxltvvYXTp08bTUURQsiLYN26daIm0r/+9S+za0b4poiMjAyj+3ft2iWmW8LCwhTrV926dQHkJYMz5dq1awDyMvkqwcPDQ2xXt5awz3DUXOmppmHDhokMwHwEiKOK1CWLrGAmJSUFw4cPR2BgIDp37ow///wTu3fvRteuXQHkfQsxTC0dFhaGBQsW4Ouvv0a9evXw+uuvIzAwENu2bRNt0tLS8M477yA4OBjdunXD7du3ceDAATRv3lyhSySEEGWo1Wr8+eefcHV1hV6vR/v27UUtIENdunQBkDcaYVhqYMOGDQAAT09PMWKhBL4rKjExsUDAoNfrRQ0jXqxRCV988QUAIDU1Fbt27TLbjpd3AGB1SkqOxYsXY+PGjQCArl27IjU1lSpSl2RK7BF/EVCeGUJIYdmzZ4/IXdK0aVOTbdRqNQPAVq5cKe6rWrUqA8A6d+6saH9SUlJEf6Kjo40eO378uHjs4cOHip63UqVKVjMQp6amivPfvn1bkfMePHhQZP/18/NjOp1OkeOS5+O55pkhhJCSqmvXrpg8eTKAvPUoU6ZMKdCGj0j8+eefAPIy4N65cwcA8PrrryvaH29vbzGds2bNGqPHfvnlFwB5mya8vLwUPe/7778PADh9+rTZfC+G00ypqal2n/Pu3bvo0qULGGNwc3PDX3/9pciiZlK00SuAEEJs8Pnnn6Np06YA8qZc9u7da/Q4XxfIy71s27ZNLILlu4GU1LBhQwD/7JbiDh8+DACoXLmy4uecOnWqWIDLg7v8DAMNe5Ob5ubmomnTpnj27BlUKhWio6PFNnhSslEwQwghNoqNjRUjD6+++qpRfaDatWsDgEiyxys6V6hQwSE1gPj6kKSkJKN1MxcuXAAAs8lN7aHVakUyuq1bt1pd4GtvMMPXVQLAsmXL7MpmTIoXCmYIIcRGrq6uiImJgVqtxrNnz9CiRQvxgc53Dj18+BAAcPToUQD/FKJUGt8dpdfrjRbkpqSkAIDDcq0sXLgQAPD06VORiyc/PjqTf3eXHFOmTEF0dDSAvGt99913bT4WKX4omCGEEDuEhIRgwYIFAPK2QPP0FO3atQOQV0cuMzMTd+/eBQAMHjzYIf0oV64cPD09AeRtIQeAe/fuIScnBwBExWulvfTSSwgICACQl3/GFJ7zxdZg5ocffhC7p0JCQgqsCyKEghlCCLHTuHHj0LNnTwDAxo0bsW7dOpGHBQD+97//iX87KpgB/ilyeejQIQDAjh07AOSNjDiyuOLHH38MIC+Yu3TpUoHHedK6zMxM2ce+cOEC3njjDQCAl5cXjhw5YkdPSXFFwQwhhCjgl19+EZnL33zzTSQlJcHFxQUAEBkZCQCoVKkSnJ2dHdYHvksqOTkZubm5+OOPPwBAjNg4ysiRI8U6oAkTJhR43NZgJjMzE61atYJOp4OTkxOOHTsmqoQTYoiCGUIIUYBarcaRI0eg1Wqh0+nQqlUrlC9fHgBE4jzD0RpHGDZsGIC8ZH1RUVGIi4sDANSqVcuh51Wr1Rg4cCAAYO/evcjOzjZ63NZgpkWLFkhPTweQtxuMT2cRkh8FM4QQopDq1auLrLT3798XtZl0Oh0A4MCBA0YZ0JXm7u6OihUrAgA2bdokMrI7atGxoS+++AIqlQq5ubkF1s7w7dumsiWbM2TIEMTHxwMAZsyYYbW4JynZKJghhBAFDRw4UJQNYPmKK6ampmLAgAEODWj4duXDhw+LkRC+fdqRKlSogEaNGgEAli5davQYD2akjswsWLAAW7ZsAQD06NEDn376qXIdJcUSBTOEEKIgnU6HhIQEi23Gjx8vRmuUxhcYG+a86dixo0POld+cOXMA5O2iOnDggLjfyckJgLSRmQMHDmDSpEkA8ka6du7c6YCekuKGghlCCFFQbGwsbt26ZfZxxhiSkpIwZcoUHDhwwCjoUAIfBTEUFBTk0NEgrnv37qJkgmGJBx7MPH361OLz79y5g27duoExBnd3dypVQCQr+KonhBBiM75OxpoFCxaI/DRA3ge+m5sbypQpAy8vL3h7e6Nq1aqoVq0aatWqhdq1a6Nu3bpGtY7y27ZtG4YMGVLg/tu3b2PAgAGIjIx0eCXpd955B5999hmOHTuGtLQ0lCtXTuzgshTM8FIFWVlZUKvViImJUbyWFCm+KJghhBAF8e3Z1mg0GqOpppycHOTk5ODRo0ciZb8pKpUKTk5OcHd3R5kyZVC+fHl4e3ujSpUqiIqKKrBOB8gbDVKpVBg/fjz69Okjdhc5wowZM/DFF19Ap9Nh2rRpWLZsmQhmnj17ZvZ5nTp1EoHgihUrRN0rQqSgYIYQQhQUGhoKX19f3L5922RgoVKp4Ovri8TERKhUKty5cwfnz59HQkICEhMTcfPmTdy9exf3799HWloaMjMzkZWVJQIfxhiys7ORnZ2NtLQ0JCUlSeoXn96KjY1Fhw4dlLxkI66urujYsSN+//13bNiwAcuWLRP5dswFMxMmTEBsbCwAIDw8HOHh4Q7rHymeKJghhBAFaTQafPXVVxgwYABUKlWBgIYxhoiICDE64uvrC19fX6s7jvR6PRITE3HhwgUR+Ny6dQvJycl4+PAhUlJSJJULGDVqlKht5IhK2gDw5ZdfomHDhsjMzMTmzZtForusrKwCbTdv3oyIiAgAeRmMV65c6ZA+keJNxUx9dSiCHj16hLJlyyI9Pd3inDIhhBSGbdu2Ydy4cQUWA7/88sv45ZdfFD/f/v37Ze9a8vDwQOPGjfH6668jLCxM0Wre1atXx82bNxEUFAQfHx9ER0ejTp06OH/+vGhz7tw5hISEQKfToXz58rhz545DMySTF5MSn9+0TJwQQhygX79+uH79OqKjo7Fp0yaRg+X3339Hbm6u4ufj01u8qKMppUuXRu3atcWOp0ePHmH//v147733UKpUKVSqVAmvvfYaoqKiRPVvW02ePBkAcPHiRXEsw8zAmZmZaN26tShVcPLkSQpkiM0omCGEEAfRaDTo0KEDhgwZgu3btwPIm2qZNm2aQ8711VdfmXxMpVJBpVLhu+++w6VLl5CTk4Po6GgMHz4cfn5+IgC6d+8efvrpJwwYMABOTk6oUaMGRo0aJQpXyjF27FixVuby5csA/glm9Ho9mjRpIqbFfvrpJ1SvXl32OQgRWDGRnp7OALD09PTn3RVCCDGpXbt2DABzdXVlOTk5DjlHVFQUc3NzYwDEzc/Pj0VFRZl9Tk5ODtuyZQt7+eWXWfny5Y2ey29OTk4sODiYTZw4kSUkJEjqy+DBgxkAplarGQBWuXJlxhhj/fv3F8edOXOmItdNii4lPr9pzQwhhBSSq1eviqKPU6dOLVDDSClt27bFoUOH4Ovri/Xr1yM0NFTWduxHjx5hzZo1iIqKwqlTp0yWIXBzc0P9+vXRt29fhIeHo0KFCgXa3LlzB1WrVhX/X7p0aQwdOhQrVqwAAPTu3Rs///yzDVdIihMlPr8pmCGEkELEAw03NzdkZmY6JMNtYGAgEhIS0LlzZ/z+++92H+/WrVtYsWIFfv75Z1y8eNHkrqRy5cqhWbNmGDRoEIYOHSp2MPGFwPlVqlQJd+7coQy/hBYAE0JIUbNmzRoAedlwHVVAMTU1FUDetm8l+Pr6YtasWTh9+jSePXuGkydPYvTo0QgICBAjPmlpadi7dy/Cw8Ph5uYGHx8ftG7d2mQgAwApKSn48ccfFekfIRTMEEJIIXrppZdEZesvv/zS7l1DpvBpIX9/f8WPDQCNGzfGsmXLcOXKFWRnZ2PXrl0YNGiQUfbju3fv4siRI2aPwTMSO6rgJilZKJghhJBCxkdnnjx5gtmzZyt+fD4NxNfnOJJarUbPnj2xZcsW3LlzB1lZWVi7dq3VcgTMICMxIfaiYIYQQgpZcHCw+LD//PPPFR2dyc7OFserW7euYseVytnZGSNHjsTEiRMltZdamJMQSyiYIYSQ54CPzjx+/Biff/65Yse9ePGi+HedOnUUO65cpUqVktROamFOQiyhYIYQQp6DevXqiazAc+bMUWx0hpcLUKvVzy2j7r179zBy5EiLbVQqFfz8/BAaGlpIvSLFGQUzhBDynKxatQoAkJGRgYULFypyzISEBAAQ2XcL261bt1CrVi2kpaWZbcMzDhsW3CTEHhTMEELIc9K4cWM0aNAAADBr1ixFjpmYmAggL0FdYbt69SqCgoKQkZEBlUqFVatWoXz58gXa+fr6IjIyEv369Sv0PpLiiYIZQgh5jvjoTHp6utnaSnLwKt2enp52H0uO8+fPo169enj8+DFUKhU2btyI4cOH4++//zZq9/XXXyMxMZECGaIoCmYIIeQ5atasmdh1NHPmTLuPd+/ePQCAt7e33ceS6q+//kLjxo3x7NkzqNVqbN++HUOGDMH69euRP8l89erVaWqJKI6CGUIIec5WrlwJIC9z79KlS+061sOHDwHAqCaSIx05cgQtW7ZEdnY2NBoNdu3ahT59+gAANm7cKPrCAxjeP0KURMEMIYQ8Z61atUJQUBAA4JNPPrHrWI8ePQIAVKtWze5+WfPHH38gNDQUOTk50Gq1+OOPP9C9e3fx+IkTJwAAnTp1omCGOBQFM4QQ8gLgozMPHz4U62hs8ezZMwCOz/67a9cudO3aFTqdDk5OTjh8+DDatWsnHr9x4wYyMjIAAGPGjIGTkxMAWNzlRIitKJghhJAXQNu2bfHSSy8BAD766CObjqHX65GbmwvAsQnzoqKi0Lt3b+j1eri4uODEiRNo1qyZUZuvv/4aQF5G4FatWlEwQxyKghlCCHlBLF++HABw//59fPfdd7Kff+PGDfHvevXqKdYvQxs2bMDrr78Oxhjc3NwQFxcntpcb+uWXXwD8E1TxBH7p6ekO6Rcp2SiYIYSQF0SnTp0QEBAAAPj3v/8t+/lnz54FkJeUrly5ckp2DQDwzTffYPjw4WCMoVSpUjh//jwCAwMLtNPr9bh8+TIAoG/fvgAAV1dXABBTT4QoiYIZQgh5gSxZsgQAkJKSInYDSXXp0iUAcEgZg4iICLz77rsAgLJlyyIhIQH+/v4m2+7evRs6nQ4AMHbsWAAUzBDHomCGEEJeIN27d0eNGjUAAJMnT5b13GvXrgEA3N3dFe3T//73P0yYMAEA4OXlhStXrqBKlSpm2/PFzF5eXqhQoYJRnx4/fqxo3wgBKJghhJAXzuLFiwEAycnJ+OGHHyQ/LykpCQAUnWL6+OOPMX36dAB5ifiuXr0qAhRzDh48CCBvyznHq2hTMEMcgYIZQgh5wfTu3VvkiZk4caLk5929excATNZDssXEiRMxZ84cAHmJ765evWo1UEpLS8ODBw8AAKNGjRL382Dm6dOnivSNEEMUzBBCyAto0aJFAIDbt29j+/btkp7DgwhLU0BSjR49WlTyrlGjBq5cuSKpeCXfkaVWq8XiXwAoU6YMgH/y4BCiJApmCCHkBdSnTx9RkmDcuHGSnsO3Pdub/XfYsGFYsWIFAKB27dq4dOmSWMBrTWRkJACgZs2aUKv/+Yjx8PAAQMEMcQwKZggh5AUVEREBIG8tDM/bYsmTJ08AQGzvtkX//v3FLqr69evj/PnzsnZHnTt3DgDQs2dPo/vLli0LAMjOzra5b4SYQ8EMIYS8oAYMGAAfHx8AwAcffGC1PQ8UTOV+kaJnz57Ytm0bgLxq3qdPn4ZWq5X8/BMnTiArKwtAwf7ytTY5OTk29Y0QSyiYIYSQF9iXX34JALh+/Tp+++03s+34ehkgb0RFDr1ej/bt22P37t0AgHbt2uHo0aNG00RS8IrfpUqVEqUZOB7M8HILhCiJghlCCHmBDRkyBJUqVQLwTwI6U86cOSP+LWcBsF6vR8uWLXHgwAEAeXluYmJiZAcyAPD7778DABo1alTgMb7DioIZ4ggUzBBCyAvu888/B5CXFG/fvn0m2/Dsv1qtVnIgkpubi5CQEBw/fhwA0K9fPzE6I1d2djZu3boFIC8Ay48HM4wxm45PiCUUzBBCyAtuxIgRqFixItv/kQAAGq9JREFUIgBgzJgxJttcuXIFAODm5ibpmNnZ2ahXr56o5zR06FBERUXZ3MeNGzeKQCUsLKzA4zzRHgUzxBEomCGEkCKAJ6+7fPmymBIyxCtm8y3Qljx79gyBgYFiNOftt9/Ghg0b7Ooff76Pj4/Jcgre3t7i35mZmXadi5D8KJghhJAiIDw8XEzV8IKPhu7cuQPAevbfzMxM1KpVC9evXwcATJgwAd98843d/eNTVR07djT5OB9ZAvKKaBKiJFnBzLJly9CgQQN4eHjAw8MDrVq1wq+//mrxORs3bkTDhg3h7u4OHx8fjBo1Cg8fPjRqExUVhTp16sDFxQV16tSRnO2SEEJKklmzZgEALl68iEOHDhk9dv/+fQBA5cqVzT4/LS0NAQEBuH37NgDgo48+woIFC+zu182bN0U1bHPTYIajNffu3bP7nIQYkhXM+Pr64rPPPsOJEydw4sQJdOrUCX369MH58+dNtj948CBGjBiBt956C+fPn8fWrVtx/PhxhIeHizZHjhzBoEGDMHz4cMTFxWH48OEYOHAg/vzzT/uujBBCipkxY8bA09MTQMHRmdTUVACAn5+fyefeu3cPAQEBIpCYPXs2/ve//ynSr6+//hoA4OTkhLZt25ptp1KpAKDAF1pC7Mbs5OnpyVatWmXysS+++ILVrFnT6L5FixYxX19f8f8DBw5kPXr0MGrTvXt3NnjwYFn9SE9PZwBYenq6rOcRQkhRsmjRIgaAAWDHjh0T97u6ujIAbPbs2QWec/v2bebh4SGet3DhQkX7FBQUxACwBg0aWGyn0WgYALZ27VpFz0+KNiU+v21eM6PT6bBlyxY8fvzYqMy7odatW+PWrVvYtWsXGGNISUlBZGQkXn75ZdHmyJEj6Natm9HzunfvjsOHD1s8f1ZWFh49emR0I4SQ4u6DDz4QpQHeeustcT/PvJs/+++NGzdQu3Zt8R65fPlyjB8/XrH+6PV6XL58GQCMCkuaotFoANDIDFGe7GDm7NmzKF26NFxcXDB69Ghs374dderUMdm2devW2LhxIwYNGgRnZ2dUrlwZ5cqVw+LFi0Wbu3fvioRQXKVKlUQpe3Pmzp2LsmXLipu5oVVCCCluPvnkEwB578d//fUXnjx5IrY8161bV7S7dOkSgoOD8fjxY6hUKqxfv97k4mF7/Pbbb9DpdAAsJ/XT6XQi/01cXJx4DiFKkB3MBAYG4vTp0zh69CjGjBmDkSNHIj4+3mTb+Ph4fPjhh/jkk09w8uRJ7N69G4mJiRg9erRROz6PyjHGCtyX37Rp05Ceni5uSUlJci+FEEKKpPHjx4st2OHh4UbvwbyMwJkzZ9CwYUM8ffoUKpUKP/zwA4YNG6Z4X1atWgUA8PT0NNp+bWjbtm2oUaOGqJi9bt061KhRQ9SBIsRu9s51de7cmb3zzjsmHxs2bBgbMGCA0X2xsbEMALtz5w5jjDE/Pz+2YMECozYLFixg1apVk9UPWjNDCClJ5s6dK9bAzJo1iwFgarWaMcbYsWPHmJOTk7hv586dDutHxYoVGQDWs2dPk49HRUUxlUol+spvKpWKqVQqFhUV5bC+kaLhua6ZMQiGxFxtfk+ePCmQVpvPmbL/HxJt1aoV9u7da9Rmz549aN26tb1dI4SQYmvKlCkoXbo0AGDhwoUAAGdnZ0RHR6N169bIycmBRqPB3r170atXL4f04dGjR2JLuKmsvzqdDuPGjTOZ9ZffN378eJpyInaTFcx89NFHiI2NxfXr13H27Fl8/PHH2L9/P4YOHQogb+pnxIgRov0rr7yCbdu2YdmyZbh27RoOHTqEDz/8EM2bNxeF0MaNG4c9e/Zg3rx5uHjxIubNm4fff/9d0QVqhBBS3KjVarzyyisAgL///htAXmbfTp06ITc3F1qtFrGxsejUqZPD+rBixQrRlwEDBhR4PDY2VtRrMoUxhqSkJMTGxjqsj6Rk0MppnJKSguHDhyM5ORlly5ZFgwYNsHv3bnTt2hUAkJycjJs3b4r2YWFhyMjIwNdff41//etfKFeuHDp16oR58+aJNq1bt8aWLVswffp0/Oc//0FAQAC+//57tGjRQqFLJISQ4mfbtm3YvHmz2cfnzp1rdqepUrZu3QoA8Pf3N1ncMjk5WdJxpLYjxBwVMzX+VwQ9evQIZcuWRXp6uqTaJIQQUlTpdDrUqFHD4qiHn58fEhMTxdS+I7i5ueHZs2d4//33jXapcvv37zdb3sBQdHQ0OnTo4IAekqJAic9vqs1ECCFFjLXpGwAOn745ffq02J30/vvvm2wTGhoKX19fs8dQqVTw8/NDaGioQ/pISg4KZgghpIh5EaZvlixZAiCv5lL+RH2cRqMxm3uGp9+IiIhw6OgRKRkomCGEkCLGx8dH0Xa24LtQGzVqZLZNdnY2PvvsMwAF84n5+voiMjIS/fr1c1gfSckhawEwIYSQ549P39y+fdvktmcgb5u2o6ZvcnNzxWaPwYMHm23Xq1cvUUZhz5490Gq1SE5Oho+PD0JDQ2lEhiiGghlCCCliNBoNvvrqKwwYMAAqlcpkQJOdnY3+/fvjxx9/VPz8mzZtEud88803Tbb55ptvsG/fPgDA6NGj0aVLF8X7QQhH00yEEFIE9evXD5GRkahatarR/X5+fmjYsCEA4KeffipQPkYJ69evB5A3jeXu7l7g8Zs3b+K9994DANSoUQPLli1TvA+EGKKRGUIIKaL69euHPn36IDY21mj6RqVSoWnTpjh16hRWrFiBypUr49NPP1XsvMeOHQMAk9up9Xo92rVrJxL3xcTEKHZeQsyhYIYQQoowjUZjMqg4duwYateujcTERMycOROVKlXCmDFj7D7frVu3xDoYU6M+Y8eOxY0bNwDk7XiqVq2a3eckxBqaZiKEkGJIq9Xi3LlzqFixIgDgvffeQ1RUlN3H5VuynZyc0K5dO6PH/vjjD1HioHPnznjnnXfsPh8hUlAwQwghxZS7uzvi4+NRpkwZMMYwcOBAHDhwwK5j7tixAwAQFBRkdP+TJ0/w6quvAgDKli2LXbt22XUeQuSgYIYQQoqxChUqIC4uDi4uLtDr9ejSpQvOnTtn07H0ej0SEhIAAK+99prRY126dMHjx4+hUqnw22+/wdnZ2d6uEyIZBTOEEFLM+fv74/Dhw9BqtcjJyUHz5s2tlkMwZe/evcjNzQUAo8y+CxYswJEjRwAAEyZMoELBpNBRMEMIISVA48aNsWvXLqhUKjx9+hT169dHWlqarGOsXLkSAFCuXDlUrlwZAHD58mVMnjwZABAYGIgvv/xS0X4TIgUFM4QQUkJ07dpV5IhJS0tDnTp1RLFIKXjhypYtWwLIm3Zq37499Ho9nJyc7F6PQ4itKJghhJASZOjQoWL0JDk5GSEhIdDr9Vaf9+jRI9y7dw8AEBYWBgAYPny4KGa5Zs0aeHt7O6bThFhBwQwhhJQwEydOxJQpUwAAly5dQtu2ba0+h08xqdVq9O/fHzt27MCmTZsAAK+++iqGDh3quA4TYgUFM4QQUgLNmzcPI0aMAAAcOXKkwO6k/H744QcAeeUJMjMzMWjQIABA+fLlFclfQ4g9KJghhJAS6rvvvkO3bt0A5NVxspTk7syZMwCA7t27o1OnTnj27BlUKhX++OMPaLWUTJ48XxTMEEJICfbrr7+icePGAPKmkkzVcDpz5oxYKKxWq3Hq1CkAwCeffIIGDRoUWl8JMUfFTNWOL4IePXqEsmXLIj09HR4eHs+7O4QQUmTk5uYiMDAQ165dAwAsXbrUqI7TO++8g5UrV8LFxQXZ2dlgjKFBgwaIi4t7Xl0mxYgSn980MkMIISWcVqvF2bNnxW4kXsdJp9Nh//792L59OwBAp9OBMQZXV1eqhk1eKDTRSQghBO7u7jh//jxq1qyJjIwMvP766yhfvjwePHgg2vDsv1u2bEG5cuWeU08JKYhGZgghhAD4p46Tk5MTGGNGgYwhnU5XyD0jxDIKZgghhAjVqlWDp6enxTbjx4+ngIa8UCiYIYQQIsTGxopMv+YkJSWJ0gaEvAgomCGEECLw8gRKtSOkMFAwQwghRPDx8VG0HSGFgYIZQgghQmhoKHx9faFSqUw+rlKp4Ofnh9DQ0ELuGSHmUTBDCCFE0Gg0+OqrrwCgQEDD/z8iIgIajabQ+0aIORTMEEIIMdKvXz9ERkaiatWqRvf7+voiMjIS/fr1e049I8Q0KmdACCHEJJ1Oh9jYWCQnJ8PHxwehoaE0IkMUp8TnN2UAJoQQYpJGo0GHDh2edzcIsYqmmQghhBBSpFEwQwghhJAijYIZQgghhBRpFMwQQgghpEijYIYQQgghRRoFM4QQQggp0iiYIYQQQkiRRsEMIYQQQoo0CmYIIYQQUqQVmwzAvCrDo0ePnnNPCCGEECIV/9y2p7pSsQlmMjIyAAB+fn7PuSeEEEIIkSsjIwNly5a16bnFptCkXq/HnTt3UKZMmQJl6w09evQIfn5+SEpKKhEFKel6i7eSdr1Aybtmut7iraRdL1DwmhljyMjIQJUqVaBW27b6pdiMzKjVavj6+kpu7+HhUWJeOABdb3FX0q4XKHnXTNdbvJW06wWMr9nWERmOFgATQgghpEijYIYQQgghRVqJC2ZcXFwwY8YMuLi4PO+uFAq63uKtpF0vUPKuma63eCtp1ws45pqLzQJgQgghhJRMJW5khhBCCCHFCwUzhBBCCCnSKJghhBBCSJFGwQwhhBBCirRiF8wsXboU/v7+cHV1RZMmTRAbG2u27bZt29C1a1dUrFgRHh4eaNWqFX777bdC7K0y5FzzwYMH0aZNG5QvXx5ubm4ICgrCwoULC7G39pNzvYYOHToErVaLkJAQx3ZQYXKud//+/VCpVAVuFy9eLMQe20fu7zcrKwsff/wxqlevDhcXFwQEBGD16tWF1FtlyLnmsLAwk7/junXrFmKP7SP3d7xx40Y0bNgQ7u7u8PHxwahRo/Dw4cNC6q395F7vkiVLEBwcDDc3NwQGBmLdunWF1FP7HThwAK+88gqqVKkClUqFH3/80epzYmJi0KRJE7i6uqJmzZpYvny5/BOzYmTLli3MycmJrVy5ksXHx7Nx48axUqVKsRs3bphsP27cODZv3jx27NgxlpCQwKZNm8acnJzYX3/9Vcg9t53ca/7rr7/Ypk2b2Llz51hiYiJbv349c3d3ZytWrCjknttG7vVyaWlprGbNmqxbt26sYcOGhdNZBci93ujoaAaAXbp0iSUnJ4tbbm5uIffcNrb8fl999VXWokULtnfvXpaYmMj+/PNPdujQoULstX3kXnNaWprR7zYpKYl5eXmxGTNmFG7HbST3emNjY5larWZfffUVu3btGouNjWV169Zlr732WiH33DZyr3fp0qWsTJkybMuWLezq1ats8+bNrHTp0mzHjh2F3HPb7Nq1i3388ccsKiqKAWDbt2+32P7atWvM3d2djRs3jsXHx7OVK1cyJycnFhkZKeu8xSqYad68ORs9erTRfUFBQWzq1KmSj1GnTh02c+ZMpbvmMEpcc9++fdmwYcOU7ppD2Hq9gwYNYtOnT2czZswoUsGM3OvlwUxqamoh9E55cq/3119/ZWXLlmUPHz4sjO45hL1/w9u3b2cqlYpdv37dEd1TnNzr/eKLL1jNmjWN7lu0aBHz9fV1WB+VJPd6W7VqxSZNmmR037hx41ibNm0c1kdHkRLMTJkyhQUFBRnd9+6777KWLVvKOlexmWbKzs7GyZMn0a1bN6P7u3XrhsOHD0s6hl6vR0ZGBry8vBzRRcUpcc2nTp3C4cOH0b59e0d0UVG2Xu+aNWtw9epVzJgxw9FdVJQ9v99GjRrBx8cHnTt3RnR0tCO7qRhbrnfHjh1o2rQpPv/8c1StWhW1a9fGpEmT8PTp08Lost2U+Bv+9ttv0aVLF1SvXt0RXVSULdfbunVr3Lp1C7t27QJjDCkpKYiMjMTLL79cGF22iy3Xm5WVBVdXV6P73NzccOzYMeTk5Disr8/LkSNHCvx8unfvjhMnTsi63mITzDx48AA6nQ6VKlUyur9SpUq4e/eupGN8+eWXePz4MQYOHOiILirOnmv29fWFi4sLmjZtivfeew/h4eGO7KoibLney5cvY+rUqdi4cSO02qJVV9WW6/Xx8cE333yDqKgobNu2DYGBgejcuTMOHDhQGF22iy3Xe+3aNRw8eBDnzp3D9u3bERERgcjISLz33nuF0WW72fu+lZycjF9//bVI/P0Ctl1v69atsXHjRgwaNAjOzs6oXLkyypUrh8WLFxdGl+1iy/V2794dq1atwsmTJ8EYw4kTJ7B69Wrk5OTgwYMHhdHtQnX37l2TP5/c3FxZ11u03t0lUKlURv/PGCtwnymbN2/Gp59+ip9++gne3t6O6p5D2HLNsbGxyMzMxNGjRzF16lTUqlULQ4YMcWQ3FSP1enU6Hd544w3MnDkTtWvXLqzuKU7O7zcwMBCBgYHi/1u1aoWkpCTMnz8f7dq1c2g/lSLnevV6PVQqFTZu3Ciq7i5YsAADBgzAkiVL4Obm5vD+KsHW9621a9eiXLlyeO211xzUM8eQc73x8fH48MMP8cknn6B79+5ITk7G5MmTMXr0aHz77beF0V27ybne//znP7h79y5atmwJxhgqVaqEsLAwfP7559BoNIXR3UJn6udj6n5Lis3ITIUKFaDRaApEu/fu3SsQ9eX3/fff46233sIPP/yALl26OLKbirLnmv39/VG/fn28/fbbmDBhAj799FMH9lQZcq83IyMDJ06cwPvvvw+tVgutVov//ve/iIuLg1arxR9//FFYXbeJPb9fQy1btsTly5eV7p7ibLleHx8fVK1aVQQyABAcHAzGGG7duuXQ/irBnt8xYwyrV6/G8OHD4ezs7MhuKsaW6507dy7atGmDyZMno0GDBujevTuWLl2K1atXIzk5uTC6bTNbrtfNzQ2rV6/GkydPcP36ddy8eRM1atRAmTJlUKFChcLodqGqXLmyyZ+PVqtF+fLlJR+n2AQzzs7OaNKkCfbu3Wt0/969e9G6dWuzz9u8eTPCwsKwadOmIjEHa8jWa86PMYasrCylu6c4udfr4eGBs2fP4vTp0+I2evRoBAYG4vTp02jRokVhdd0mSv1+T506BR8fH6W7pzhbrrdNmza4c+cOMjMzxX0JCQlQq9Xw9fV1aH+VYM/vOCYmBleuXMFbb73lyC4qypbrffLkCdRq448qPkLBXvDSgvb8fp2cnODr6wuNRoMtW7agd+/eBX4OxUGrVq0K/Hz27NmDpk2bwsnJSfqBZC0XfsHxLXDffvsti4+PZ+PHj2elSpUSq/ynTp3Khg8fLtpv2rSJabVatmTJEqOtjmlpac/rEmSTe81ff/0127FjB0tISGAJCQls9erVzMPDg3388cfP6xJkkXu9+RW13Uxyr3fhwoVs+/btLCEhgZ07d45NnTqVAWBRUVHP6xJkkXu9GRkZzNfXlw0YMICdP3+excTEsJdeeomFh4c/r0uQzdbX9LBhw1iLFi0Ku7t2k3u9a9asYVqtli1dupRdvXqVHTx4kDVt2pQ1b978eV2CLHKv99KlS2z9+vUsISGB/fnnn2zQoEHMy8uLJSYmPqcrkCcjI4OdOnWKnTp1igFgCxYsYKdOnRJb0fNfL9+aPWHCBBYfH8++/fZb2prNGGNLlixh1atXZ87Ozqxx48YsJiZGPDZy5EjWvn178f/t27dnAArcRo4cWfgdt4Oca160aBGrW7cuc3d3Zx4eHqxRo0Zs6dKlTKfTPYee20bO9eZX1IIZxuRd77x581hAQABzdXVlnp6erG3btmznzp3Pode2k/v7vXDhAuvSpQtzc3Njvr6+bOLEiezJkyeF3Gv7yL3mtLQ05ubmxr755ptC7qky5F7vokWLWJ06dZibmxvz8fFhQ4cOZbdu3SrkXttOzvXGx8ezkJAQ5ubmxjw8PFifPn3YxYsXn0OvbcPTQ5j7XDX1+92/fz9r1KgRc3Z2ZjVq1GDLli2TfV4VYy/4OB0hhBBCiAXFbwKOEEIIISUKBTOEEEIIKdIomCGEEEJIkUbBDCGEEEKKNApmCCGEEFKkUTBDCCGEkCKNghlCCCGEFGkUzBBCCCGkSKNghhBCCCFmHThwAK+88gqqVKkClUqFH3/8UfYxGGOYP38+ateuDRcXF/j5+WHOnDmK9VGr2JEIIYQQUuw8fvwYDRs2xKhRo9C/f3+bjjFu3Djs2bMH8+fPR/369ZGeno4HDx4o1kcqZ0AIIYQQSVQqFbZv347XXntN3JednY3p06dj48aNSEtLQ7169TBv3jx06NABAHDhwgU0aNAA586dQ2BgoEP6RdNMhBBCCLHZqFGjcOjQIWzZsgVnzpzB66+/jh49euDy5csAgJ9//hk1a9bEL7/8An9/f9SoUQPh4eH4+++/FesDBTOEEEIIscnVq1exefNmbN26FaGhoQgICMCkSZPQtm1brFmzBgBw7do13LhxA1u3bsW6deuwdu1anDx5EgMGDFCsH7RmhhBCCCE2+euvv8AYQ+3atY3uz8rKQvny5QEAer0eWVlZWLdunWj37bffokmTJrh06ZIiU08UzBBCCCHEJnq9HhqNBidPnoRGozF6rHTp0gAAHx8faLVao4AnODgYAHDz5k0KZgghhBDy/DRq1Ag6nQ737t1DaGioyTZt2rRBbm4url69ioCAAABAQkICAKB69eqK9IN2MxFCCCHErMzMTFy5cgVAXvCyYMECdOzYEV5eXqhWrRqGDRuGQ4cO4csvv0SjRo3w4MED/PHHH6hfvz569eoFvV6PZs2aoXTp0oiIiIBer8d7770HDw8P7NmzR5E+UjBDCCGEELP279+Pjh07Frh/5MiRWLt2LXJycjB79mysW7cOt2/fRvny5dGqVSvMnDkT9evXBwDcuXMHH3zwAfbs2YNSpUqhZ8+e+PLLL+Hl5aVIHymYIYQQQkiRRluzCSGEEFKkUTBDCCGEkCKNghlCCCGEFGkUzBBCCCGkSKNghhBCCCFFGgUzhBBCCCnSKJghhBBCSJFGwQwhhBBCijQKZgghhBBSpFEwQwghhJAijYIZQgghhBRp/wc7RghWNr8O9wAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "w_queen.plot(gdf);" ] }, { "cell_type": "markdown", "id": "068e0437-0e1e-47cd-a0e5-8f477bc9d2d9", "metadata": {}, "source": [ "and the same can be done with the `Graph`:" ] }, { "cell_type": "code", "execution_count": 35, "id": "b560cc2e-9438-4e65-a3a2-d70bec954f8b", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "g_queen.plot(gdf)" ] }, { "cell_type": "markdown", "id": "ff87e52b-5035-44c3-9e25-adff9d1ffcfe", "metadata": {}, "source": [ "However, the `Graph` adds an `explore` function allowing for a richer visualization:" ] }, { "cell_type": "code", "execution_count": 36, "id": "e372a3c5-cc3c-45b1-b1be-b73756ff1279", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
Make this Notebook Trusted to load map: File -> Trust Notebook
" ], "text/plain": [ "" ] }, "execution_count": 36, "metadata": {}, "output_type": "execute_result" } ], "source": [ "m = gdf.explore(tiles=\"CartoDB Positron\")\n", "g_queen.explore(gdf, m=m)" ] }, { "cell_type": "markdown", "id": "b44a171d-d458-4773-859f-00eef92aea6e", "metadata": {}, "source": [ "This can be leveraged to look at the spatial distribution of the cardinalities, for example:" ] }, { "cell_type": "code", "execution_count": 37, "id": "8cfb4407-c8a3-4e87-a634-e5b993436a74", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "focal\n", "0 3\n", "1 3\n", "2 5\n", "3 2\n", "4 4\n", " ..\n", "95 5\n", "96 7\n", "97 4\n", "98 2\n", "99 3\n", "Name: cardinalities, Length: 100, dtype: int64" ] }, "execution_count": 37, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.cardinalities" ] }, { "cell_type": "code", "execution_count": 38, "id": "5ebf7c0d-f74d-4345-b6ab-d71a2cab197b", "metadata": {}, "outputs": [], "source": [ "gdf[\"cardinalities\"] = g_queen.cardinalities" ] }, { "cell_type": "code", "execution_count": 39, "id": "2359c40d-bb0d-4823-aca5-5a46fba40192", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
Make this Notebook Trusted to load map: File -> Trust Notebook
" ], "text/plain": [ "" ] }, "execution_count": 39, "metadata": {}, "output_type": "execute_result" } ], "source": [ "m = gdf.explore(tiles=\"CartoDB Positron\", column=\"cardinalities\", legend=True)\n", "g_queen.explore(gdf, m=m)" ] }, { "cell_type": "markdown", "id": "ca807531-72df-4027-a946-0f65f847c0b2", "metadata": {}, "source": [ "## Transformations\n", "\n", "Transformation of the weight values is often required in various spatial statistics and operations. How these are handled is a major change between the `W` and `Graph` classes.\n", "\n", "PySAL currently supports the following transformations:\n", "\n", "- O: original, returning the object to the initial state.\n", "- B: binary, with every neighbor having assigned a weight of one.\n", "- R: row, with all the neighbors of a given observation adding up to one.\n", "- V: variance stabilizing, with the sum of all the weights being constrained to the number of observations.\n", "- D: double, with all the weights across all observations adding up to one.\n", "\n", "\n", "For the `W`, the `transform` property stores the type of transformation that is associated with the weight values:" ] }, { "cell_type": "code", "execution_count": 40, "id": "c4a0cb0a-76f2-41e5-b439-d99a24e7c280", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'O'" ] }, "execution_count": 40, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen.transform" ] }, { "cell_type": "markdown", "id": "394d8a97-1496-449b-bc46-3b3c815c1ff8", "metadata": {}, "source": [ "An 'O' here means the weight values are set to the original transformation upon construction. In this case they are binary:" ] }, { "cell_type": "code", "execution_count": 41, "id": "7fdba8fd-8520-40e2-bc3c-b839f0e14bad", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{96: 1.0, 99: 1.0}" ] }, "execution_count": 41, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen[98]" ] }, { "cell_type": "markdown", "id": "66ae843c-9563-45c3-8239-40d802e1ddd9", "metadata": {}, "source": [ "In some cases, rather than using the binary weights, we may need to row-standardize the weights, so that in the full $n \\times n$ weights matrix, the row sums would all be equal to unity. The relevant transform in this case is 'r':" ] }, { "cell_type": "code", "execution_count": 42, "id": "8827dd89-b78f-46bc-b3bf-cc7762af0f4b", "metadata": {}, "outputs": [], "source": [ "w_queen.transform = \"r\"" ] }, { "cell_type": "code", "execution_count": 43, "id": "bb868961-6809-4812-99f1-e5c225e9cc7c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{96: 0.5, 99: 0.5}" ] }, "execution_count": 43, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w_queen[98]" ] }, { "cell_type": "markdown", "id": "126d3420-97f9-4682-babf-ff0e61268821", "metadata": {}, "source": [ "Since `transform` is a property of `W`, setting the `transform` will update the values of the weights.\n", "\n", "For the `Graph` class, things have changed in terms of these standardizations." ] }, { "cell_type": "code", "execution_count": 44, "id": "c6bb10a2-e4d2-49f2-a67e-6d57e923d058", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "neighbor\n", "96 1\n", "99 1\n", "Name: weight, dtype: int64" ] }, "execution_count": 44, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen[98]" ] }, { "cell_type": "code", "execution_count": 45, "id": "9c161b95-9cdc-479c-a449-e33d9a584ffc", "metadata": {}, "outputs": [ { "data": { "text/plain": [ ">" ] }, "execution_count": 45, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.transform" ] }, { "cell_type": "markdown", "id": "35ad4391-6829-48d2-a7c0-403b1b1688cc", "metadata": {}, "source": [ "This tells us that the `transform` member is now a method of the `Graph`.\n", "\n", "The related change is the addition of a `transformation` property:" ] }, { "cell_type": "code", "execution_count": 46, "id": "ed78f5f3-d306-4b32-b159-3b2fb3aad0b8", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'O'" ] }, "execution_count": 46, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.transformation" ] }, { "cell_type": "markdown", "id": "c94993a4-8c24-497d-a803-25b9a7343e2e", "metadata": {}, "source": [ "which plays a similar role to the `W.transform` property, in the sense it informs us as to what standardization is associated with the value of the spatial weights.\n", "\n", "However, `Graph.transformation` is not a setter in the sense that if the user changes its value, the weight values will not be affected. It is an information-only property.\n", "\n", "This is because, by design, the weights for the `Graph` are **immutable**.\n", "In other words, once the `Graph` instance is created, its state cannot be changed.\n", "\n", "If transformed spatial weights are required, the `Graph` method `transform` can be called with the type of transformation required:" ] }, { "cell_type": "code", "execution_count": 47, "id": "a18cb028-c9a3-4bd3-86c7-dd28f8bf52eb", "metadata": {}, "outputs": [], "source": [ "g1 = g_queen.transform(\"r\")" ] }, { "cell_type": "code", "execution_count": 48, "id": "86f2ed73-49c0-493e-a41a-55915ac402cf", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "neighbor\n", "96 0.5\n", "99 0.5\n", "Name: weight, dtype: float64" ] }, "execution_count": 48, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g1[98]" ] }, { "cell_type": "markdown", "id": "3f2f5116-4ba1-4619-aaea-0d262d3d11cd", "metadata": {}, "source": [ "This will return a new `Graph` instance with the weight values suitably transformed:\n" ] }, { "cell_type": "code", "execution_count": 49, "id": "30a3a8eb-c917-4093-9e8d-d88477fe55b5", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "'R'" ] }, "execution_count": 49, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g1.transformation" ] }, { "cell_type": "markdown", "id": "a8e3b89a-5e0b-4ffc-a0bd-68b0fd9b9358", "metadata": {}, "source": [ "## Spatial Lag" ] }, { "cell_type": "markdown", "id": "00295e4f-283d-4b34-be43-087749b7ba5e", "metadata": {}, "source": [ "The spatial lag of a variable is a weight sum, or weighted average, of the neighboring values for that variable:\n", "\n", "$$[Wy]_i = \\sum_j w_{i,j} y_j$$" ] }, { "cell_type": "markdown", "id": "5766e374-e416-4595-a71d-456dcb468076", "metadata": {}, "source": [ "The implementation of the spatial lag has changed between the `W` and `Graph` classes.\n", "\n", "To illustrate, we will pull out the Sudden Infant Death Rate in 1979 for the counties into the variable `y`:" ] }, { "cell_type": "code", "execution_count": 50, "id": "34f5d317-00bb-4d79-9f49-1ec6b02b45a6", "metadata": {}, "outputs": [], "source": [ "y = gdf.SID79" ] }, { "cell_type": "markdown", "id": "ec106df5-4899-4113-b8aa-7f6013c6e8b1", "metadata": {}, "source": [ "To calculate the spatial lag as a weighted average, we need to use the weights that have been row-standardized. For the `W` class, the calculation of the spatial lag is done with a function that takes the `W` as an argument together with the variable of interest:" ] }, { "cell_type": "code", "execution_count": 51, "id": "7b499f38-3e54-4aa6-b70a-ac237b4d3cf9", "metadata": {}, "outputs": [], "source": [ "from libpysal.weights import lag_spatial" ] }, { "cell_type": "code", "execution_count": 52, "id": "47159a15-996d-455d-bf65-74cb2a6234fe", "metadata": {}, "outputs": [], "source": [ "w_queen.transform = \"r\"\n", "wlag = lag_spatial(w_queen, y)" ] }, { "cell_type": "code", "execution_count": 53, "id": "3f20b6ee-702c-46e0-9d6e-dacba84b0831", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([ 3.66666667, 4.33333333, 6.8 , 1.5 , 7.25 ,\n", " 3.33333333, 2.66666667, 2.4 , 6.6 , 16.75 ,\n", " 6.5 , 14.8 , 12.6 , 8.5 , 2. ,\n", " 3.85714286, 1.33333333, 3.375 , 4. , 2.33333333,\n", " 1. , 6.4 , 7.8 , 11.42857143, 9.42857143,\n", " 9.83333333, 11. , 5.2 , 8.4 , 9.6 ,\n", " 12.14285714, 2. , 9.8 , 7.66666667, 6.75 ,\n", " 7.66666667, 8.42857143, 9. , 11.55555556, 8. ,\n", " 10.5 , 13.42857143, 10.14285714, 2. , 0. ,\n", " 7.33333333, 12.16666667, 12.875 , 11.16666667, 8.5 ,\n", " 9. , 9.83333333, 5.14285714, 12.57142857, 6.5 ,\n", " 1. , 5.16666667, 4.25 , 15.25 , 6. ,\n", " 11.16666667, 9.16666667, 17. , 15.4 , 20.5 ,\n", " 4.25 , 13.88888889, 13.4 , 12.8 , 7.28571429,\n", " 9.5 , 7.6 , 2. , 10.83333333, 9.75 ,\n", " 21. , 8. , 1.8 , 16.85714286, 11. ,\n", " 1.33333333, 9.33333333, 13.2 , 16.5 , 7.75 ,\n", " 22.2 , 1.25 , 11.5 , 7.8 , 2. ,\n", " 6. , 11. , 4. , 20.2 , 14.33333333,\n", " 21.4 , 10.14285714, 10. , 4.5 , 9.66666667])" ] }, "execution_count": 53, "metadata": {}, "output_type": "execute_result" } ], "source": [ "wlag" ] }, { "cell_type": "markdown", "id": "4874f241-ae96-4296-a941-e21c311a807b", "metadata": {}, "source": [ "For the `Graph`, the `lag` is now a method:" ] }, { "cell_type": "code", "execution_count": 54, "id": "25c264fd-55fd-440c-b4f4-bb3c6b1aa1d2", "metadata": {}, "outputs": [], "source": [ "glag = g1.lag(y)" ] }, { "cell_type": "code", "execution_count": 55, "id": "d4aec77e-b453-444f-ae8e-ba5c18aa2c63", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([ 3.66666667, 4.33333333, 6.8 , 1.5 , 7.25 ,\n", " 3.33333333, 2.66666667, 2.4 , 6.6 , 16.75 ,\n", " 6.5 , 14.8 , 12.6 , 8.5 , 2. ,\n", " 3.85714286, 1.33333333, 3.375 , 4. , 2.33333333,\n", " 1. , 6.4 , 7.8 , 11.42857143, 9.42857143,\n", " 9.83333333, 11. , 5.2 , 8.4 , 9.6 ,\n", " 12.14285714, 2. , 9.8 , 7.66666667, 6.75 ,\n", " 7.66666667, 8.42857143, 9. , 11.55555556, 8. ,\n", " 10.5 , 13.42857143, 10.14285714, 2. , 0. ,\n", " 7.33333333, 12.16666667, 12.875 , 11.16666667, 8.5 ,\n", " 9. , 9.83333333, 5.14285714, 12.57142857, 6.5 ,\n", " 1. , 5.16666667, 4.25 , 15.25 , 6. ,\n", " 11.16666667, 9.16666667, 17. , 15.4 , 20.5 ,\n", " 4.25 , 13.88888889, 13.4 , 12.8 , 7.28571429,\n", " 9.5 , 7.6 , 2. , 10.83333333, 9.75 ,\n", " 21. , 8. , 1.8 , 16.85714286, 11. ,\n", " 1.33333333, 9.33333333, 13.2 , 16.5 , 7.75 ,\n", " 22.2 , 1.25 , 11.5 , 7.8 , 2. ,\n", " 6. , 11. , 4. , 20.2 , 14.33333333,\n", " 21.4 , 10.14285714, 10. , 4.5 , 9.66666667])" ] }, "execution_count": 55, "metadata": {}, "output_type": "execute_result" } ], "source": [ "glag" ] }, { "cell_type": "markdown", "id": "891ae30f-34ff-4240-8205-aaf8288aaf81", "metadata": {}, "source": [ "## Pandas operations\n", "\n", "As mentioned above, gains in efficiency and scope have been the main motivating forces behind the development of the new `Graph` class. Here we highlight a few additional gains." ] }, { "cell_type": "markdown", "id": "4e649b10-e9e7-490f-b42f-e30418a7bca6", "metadata": {}, "source": [ "The `adjacency` attribute for the `Graph` lets users leverage the power of pandas series. For example, the operation:" ] }, { "cell_type": "code", "execution_count": 56, "id": "9921c8c7-e2f8-4bd8-8c94-1b1cdd11b49d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "neighbor\n", "96 1\n", "97 1\n", "98 1\n", "Name: weight, dtype: int64" ] }, "execution_count": 56, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen[99]" ] }, { "cell_type": "markdown", "id": "4798679f-ad54-4488-9b45-afd2fd978b4a", "metadata": {}, "source": [ "is actually a [slice](https://pandas.pydata.org/docs/user_guide/indexing.html) of the `Graph` that returns another pandas series where the index is the id of the neighboring unit and the value is the weight of that neighbor relationship.\n" ] }, { "cell_type": "code", "execution_count": 57, "id": "ea8dca41-4b87-40aa-a6eb-caa425c18a96", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "MultiIndex([( 0, 1),\n", " ( 0, 17),\n", " ( 0, 18),\n", " ( 1, 0),\n", " ( 1, 2),\n", " ( 1, 17),\n", " ( 2, 1),\n", " ( 2, 9),\n", " ( 2, 17),\n", " ( 2, 22),\n", " ...\n", " (96, 99),\n", " (97, 93),\n", " (97, 95),\n", " (97, 96),\n", " (97, 99),\n", " (98, 96),\n", " (98, 99),\n", " (99, 96),\n", " (99, 97),\n", " (99, 98)],\n", " names=['focal', 'neighbor'], length=490)" ] }, "execution_count": 57, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.adjacency.index" ] }, { "cell_type": "code", "execution_count": 58, "id": "c2128226-0976-49ea-8adf-133b8dabc9ae", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Index([96, 97, 98], dtype='int64', name='neighbor')" ] }, "execution_count": 58, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen[99].index" ] }, { "cell_type": "code", "execution_count": 59, "id": "406c3323-2d58-42ca-9ebc-a362de2046cb", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "focal neighbor\n", "96 78 1\n", " 87 1\n", " 92 1\n", " 95 1\n", " 97 1\n", " 98 1\n", " 99 1\n", "97 93 1\n", " 95 1\n", " 96 1\n", " 99 1\n", "98 96 1\n", " 99 1\n", "Name: weight, dtype: int64" ] }, "execution_count": 59, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g_queen.adjacency.loc[[96, 97, 98]]" ] }, { "cell_type": "markdown", "id": "4cfe70cb-18f2-432d-95e0-cbe2c7fe7b4f", "metadata": {}, "source": [ "Another way the `Graph` makes use of the powerful indexing afforded by pandas is seen it the method `subgraph`:" ] }, { "cell_type": "code", "execution_count": 60, "id": "760dc3be-6c4d-4e58-bbff-a2907f99f502", "metadata": {}, "outputs": [], "source": [ "g1 = g_queen.subgraph([96, 97, 98])" ] }, { "cell_type": "code", "execution_count": 61, "id": "fc216581-a49a-4205-b5c0-688bd200eb3d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "focal neighbor\n", "96 97 1\n", " 98 1\n", "97 96 1\n", "98 96 1\n", "Name: weight, dtype: int64" ] }, "execution_count": 61, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g1.adjacency" ] }, { "cell_type": "code", "execution_count": 62, "id": "1ecf948a-3b0c-4e03-8044-fde8621dae6e", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "3" ] }, "execution_count": 62, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g1.n" ] }, { "cell_type": "markdown", "id": "abf50253-917a-48ce-ba84-979f9fd04126", "metadata": {}, "source": [ "This allows more efficient extraction of subgraphs from the weights graph, relative to the way this is done with the `W` class:" ] }, { "cell_type": "code", "execution_count": 63, "id": "d145dc1b-aa21-4b67-b3d3-fe324e2bec2f", "metadata": {}, "outputs": [], "source": [ "from libpysal.weights import w_subset" ] }, { "cell_type": "code", "execution_count": 64, "id": "15d60171-3229-497e-9ed3-55812b0abdb3", "metadata": {}, "outputs": [], "source": [ "w1 = w_subset(w_queen, [96, 97, 98])" ] }, { "cell_type": "code", "execution_count": 65, "id": "9b7dabec-f7ce-49d5-9f41-91ea51245302", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{96: [1.0, 1.0], 97: [1.0], 98: [1.0]}" ] }, "execution_count": 65, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w1.weights" ] }, { "cell_type": "code", "execution_count": 66, "id": "50fda366-2a8e-494b-a1c7-9926fef979c3", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "3" ] }, "execution_count": 66, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w1.n" ] }, { "cell_type": "markdown", "id": "9a723115-c894-4bc2-8e12-d2668cfbec18", "metadata": {}, "source": [ "The index of the dataframe can also be set to a more informative attribute than simple integers:" ] }, { "cell_type": "code", "execution_count": 67, "id": "0a88d13f-0573-4680-ba61-e33735fa4d97", "metadata": {}, "outputs": [], "source": [ "ngdf = gdf.set_index(\"NAME\")" ] }, { "cell_type": "markdown", "id": "7fb4873d-ed96-46f7-8d89-26c6a6e9bd98", "metadata": {}, "source": [ "Once this is done, the new `NAME` based index will propagate to any `Graph` built on this dataframe:" ] }, { "cell_type": "code", "execution_count": 68, "id": "80c5aec6-9640-4793-80e4-6171d32e9c73", "metadata": {}, "outputs": [], "source": [ "g = graph.Graph.build_contiguity(ngdf, rook=False)" ] }, { "cell_type": "markdown", "id": "823c2b51-1dd0-4dee-90cf-7c3d7a1c8736", "metadata": {}, "source": [ "This facilities more user-friendly queries. For example, we can ask who are the neighbors for Ashe county:" ] }, { "cell_type": "code", "execution_count": 69, "id": "8466eb37-cb73-4bd4-aab0-86d83230dc18", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "neighbor\n", "Alleghany 1\n", "Wilkes 1\n", "Watauga 1\n", "Name: weight, dtype: int64" ] }, "execution_count": 69, "metadata": {}, "output_type": "execute_result" } ], "source": [ "g[\"Ashe\"]" ] }, { "cell_type": "markdown", "id": "c28a39d5-50d2-415b-98be-ec265663daa7", "metadata": {}, "source": [ "We encountered the `explore` method of the `Graph` above. This can also make handy use of the name-based indexing:" ] }, { "cell_type": "code", "execution_count": 70, "id": "726d4da2-52e0-4817-b91d-8782d7b4646e", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
Make this Notebook Trusted to load map: File -> Trust Notebook
" ], "text/plain": [ "" ] }, "execution_count": 70, "metadata": {}, "output_type": "execute_result" } ], "source": [ "m = ngdf.loc[g[\"Ashe\"].index].explore(color=\"#25b497\")\n", "ngdf.loc[[\"Ashe\"]].explore(m=m, color=\"#fa94a5\")\n", "g.explore(ngdf, m=m, focal=\"Ashe\")" ] }, { "cell_type": "markdown", "id": "4bd40414-eee6-4efa-8e8c-31648fea1eaf", "metadata": {}, "source": [ "## Conclusion\n", "\n", "This notebook highlights the main areas that users need to be aware of when considering porting their code from the `W` class to the new `Graph`.\n", "More details on the specifications and differences of the graphs can be found in the [W to Graph Member Comparisions](../../migration.rst).\n", "\n", "Downstream packages in the pysal family are in the process of supporting the new `Graph` while preserving backwards compatibility with the `W` class." ] }, { "cell_type": "markdown", "id": "10c4e285-977c-408a-9c9e-ffa80e8ffb9e", "metadata": {}, "source": [ "## Further Reading\n", "\n", "- Anselin, L. and S.J. Rey (2014) *[Modern Spatial Econometrics in Practice, Chs 3,4](https://www.amazon.com/Modern-Spatial-Econometrics-Practice-GeoDaSpace/dp/0986342106)*. GeoDa Press.\n", "- Arribas-Bel, D. (2019). *[Geographic Data Science, Lab 4](https://darribas.org/gds_course/content/bE/lab_E.html)*\n", "- Fleischmann, M. (2024). *[Spatial Data Science for Social Geography, Ch 4](https://martinfleischmann.net/sds/chapter_04/hands_on.html)*\n", "- Knaap, E. (2024). *[Urban Analysis & Spatial Science, Ch 9](https://knaaptime.com/urban_analysis/03_esda/spatial_graphs.html)*\n", "- Rey, S.J., D. Arribas-Bel, & L.J. Wolf. (2023) *[Geographic Data Science with Python](https://www.routledge.com/Geographic-Data-Science-with-Python/Rey-Arribas-Bel-Wolf/p/book/9781032445953?gad_source=1&gclid=CjwKCAjw1920BhA3EiwAJT3lScLq-TnzytfU0JRyROH9eOG97U4r7YX3G0ZCs3p5mVGUymqptvNe6hoCkqcQAvD_BwE)*. CRC/Taylor Francis. [Chapter 4](https://geographicdata.science/book/notebooks/04_spatial_weights.html).\n" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.12.2" } }, "nbformat": 4, "nbformat_minor": 5 }