diff --git a/load-keyword-extraction.ipynb b/load-keyword-extraction.ipynb
new file mode 100644
index 00000000..ff473315
--- /dev/null
+++ b/load-keyword-extraction.ipynb
@@ -0,0 +1,730 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Keyword Extraction"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\n",
+ "\n",
+ "This tutorial is available as an IPython notebook at [Malaya/example/keyword-extraction](https://github.com/huseinzol05/Malaya/tree/master/example/keyword-extraction).\n",
+ " \n",
+ "
"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import malaya"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# https://www.bharian.com.my/berita/nasional/2020/06/698386/isu-bersatu-tun-m-6-yang-lain-saman-muhyiddin\n",
+ "\n",
+ "string = \"\"\"\n",
+ "Dalam saman itu, plaintif memohon perisytiharan, antaranya mereka adalah ahli BERSATU yang sah, masih lagi memegang jawatan dalam parti (bagi pemegang jawatan) dan layak untuk bertanding pada pemilihan parti.\n",
+ "\n",
+ "Mereka memohon perisytiharan bahawa semua surat pemberhentian yang ditandatangani Muhammad Suhaimi bertarikh 28 Mei lalu dan pengesahan melalui mesyuarat Majlis Pimpinan Tertinggi (MPT) parti bertarikh 4 Jun lalu adalah tidak sah dan terbatal.\n",
+ "\n",
+ "Plaintif juga memohon perisytiharan bahawa keahlian Muhyiddin, Hamzah dan Muhammad Suhaimi di dalam BERSATU adalah terlucut, berkuat kuasa pada 28 Februari 2020 dan/atau 29 Februari 2020, menurut Fasal 10.2.3 perlembagaan parti.\n",
+ "\n",
+ "Yang turut dipohon, perisytiharan bahawa Seksyen 18C Akta Pertubuhan 1966 adalah tidak terpakai untuk menghalang pelupusan pertikaian berkenaan oleh mahkamah.\n",
+ "\n",
+ "Perisytiharan lain ialah Fasal 10.2.6 Perlembagaan BERSATU tidak terpakai di atas hal melucutkan/ memberhentikan keahlian semua plaintif.\n",
+ "\"\"\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import re\n",
+ "\n",
+ "# minimum cleaning, just simply to remove newlines.\n",
+ "def cleaning(string):\n",
+ " string = string.replace('\\n', ' ')\n",
+ " string = re.sub('[^A-Za-z\\-() ]+', ' ', string).strip()\n",
+ " string = re.sub(r'[ ]+', ' ', string).strip()\n",
+ " return string\n",
+ "\n",
+ "string = cleaning(string)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Use RAKE algorithm\n",
+ "\n",
+ "Original implementation from [https://github.com/aneesha/RAKE](https://github.com/aneesha/RAKE). Malaya added attention mechanism into RAKE algorithm.\n",
+ "\n",
+ "```python\n",
+ "def rake(\n",
+ " string: str,\n",
+ " model = None,\n",
+ " top_k: int = 5,\n",
+ " auto_ngram: bool = True,\n",
+ " ngram_method: str = 'bow',\n",
+ " ngram: Tuple[int, int] = (1, 1),\n",
+ " atleast: int = 1,\n",
+ " stop_words: List[str] = STOPWORDS,\n",
+ " **kwargs\n",
+ "):\n",
+ " \"\"\"\n",
+ " Extract keywords using Rake algorithm.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " string: str\n",
+ " model: Object, optional (default='None')\n",
+ " Transformer model or any model has `attention` method.\n",
+ " top_k: int, optional (default=5)\n",
+ " return top-k results.\n",
+ " auto_ngram: bool, optional (default=True)\n",
+ " If True, will generate keyword candidates using N suitable ngram. Else use `ngram_method`.\n",
+ " ngram_method: str, optional (default='bow')\n",
+ " Only usable if `auto_ngram` is False. supported ngram generator:\n",
+ "\n",
+ " * ``'bow'`` - bag-of-word.\n",
+ " * ``'skipgram'`` - bag-of-word with skip technique.\n",
+ " ngram: tuple, optional (default=(1,1))\n",
+ " n-grams size.\n",
+ " atleast: int, optional (default=1)\n",
+ " at least count appeared in the string to accept as candidate.\n",
+ " stop_words: list, (default=malaya.text.function.STOPWORDS)\n",
+ " list of stop words to remove. \n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " result: Tuple[float, str]\n",
+ " \"\"\"\n",
+ "```"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### auto-ngram\n",
+ "\n",
+ "This will auto generated N-size ngram for keyword candidates."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.11666666666666665, 'ditandatangani Muhammad Suhaimi bertarikh Mei'),\n",
+ " (0.08888888888888888, 'mesyuarat Majlis Pimpinan Tertinggi'),\n",
+ " (0.08888888888888888, 'Seksyen C Akta Pertubuhan'),\n",
+ " (0.05138888888888888, 'parti bertarikh Jun'),\n",
+ " (0.04999999999999999, 'keahlian Muhyiddin Hamzah')]"
+ ]
+ },
+ "execution_count": 4,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.rake(string)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### auto-gram with Attention\n",
+ "\n",
+ "This will use attention mechanism as the scores. I will use `small-electra` in this example."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:56: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/modeling.py:240: dense (from tensorflow.python.layers.core) is deprecated and will be removed in a future version.\n",
+ "Instructions for updating:\n",
+ "Use keras.layers.Dense instead.\n",
+ "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/tensorflow_core/python/layers/core.py:187: Layer.apply (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.\n",
+ "Instructions for updating:\n",
+ "Please use `layer.__call__` method instead.\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:79: The name tf.variable_scope is deprecated. Please use tf.compat.v1.variable_scope instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:93: The name tf.get_variable is deprecated. Please use tf.compat.v1.get_variable instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/sampling.py:26: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
+ "Instructions for updating:\n",
+ "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:115: multinomial (from tensorflow.python.ops.random_ops) is deprecated and will be removed in a future version.\n",
+ "Instructions for updating:\n",
+ "Use `tf.random.categorical` instead.\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:118: The name tf.InteractiveSession is deprecated. Please use tf.compat.v1.InteractiveSession instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:119: The name tf.global_variables_initializer is deprecated. Please use tf.compat.v1.global_variables_initializer instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:121: The name tf.get_collection is deprecated. Please use tf.compat.v1.get_collection instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:122: The name tf.GraphKeys is deprecated. Please use tf.compat.v1.GraphKeys instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:128: The name tf.train.Saver is deprecated. Please use tf.compat.v1.train.Saver instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/transformers/electra/__init__.py:130: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
+ "\n",
+ "INFO:tensorflow:Restoring parameters from /Users/huseinzolkepli/Malaya/electra-model/small/electra-small/model.ckpt\n"
+ ]
+ }
+ ],
+ "source": [
+ "electra = malaya.transformer.load(model = 'small-electra')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.2113546236771915, 'ditandatangani Muhammad Suhaimi bertarikh Mei'),\n",
+ " (0.1707678455680971, 'terlucut berkuat kuasa'),\n",
+ " (0.16650756665229807, 'Muhammad Suhaimi'),\n",
+ " (0.1620429894692799, 'mesyuarat Majlis Pimpinan Tertinggi'),\n",
+ " (0.08333952583953884, 'Seksyen C Akta Pertubuhan')]"
+ ]
+ },
+ "execution_count": 11,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.rake(string, model = electra)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### fixed-ngram"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.0010991603139160087, 'parti memohon perisytiharan'),\n",
+ " (0.0010989640254270869, 'memohon perisytiharan Muhammad'),\n",
+ " (0.0010985209375133323, 'perisytiharan Muhammad Suhaimi'),\n",
+ " (0.0010972572356757605, 'memohon perisytiharan BERSATU'),\n",
+ " (0.0010970435210070695, 'memohon perisytiharan sah')]"
+ ]
+ },
+ "execution_count": 12,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.rake(string, auto_ngram = False, ngram = (1, 3), \n",
+ " ngram_method = 'skipgram', skip = 3)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### fixed-ngram with Attention"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.007511555412415397, 'Suhaimi terlucut kuasa'),\n",
+ " (0.00726812348703141, 'Suhaimi terlucut Februari'),\n",
+ " (0.00725420955956774, 'Suhaimi terlucut berkuat'),\n",
+ " (0.007235384019369932, 'Muhyiddin Suhaimi terlucut'),\n",
+ " (0.00721164037502389, 'Hamzah Suhaimi terlucut')]"
+ ]
+ },
+ "execution_count": 13,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.rake(string, model = electra, auto_ngram = False, ngram = (1, 3), \n",
+ " ngram_method = 'skipgram', skip = 3)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Use Textrank algorithm\n",
+ "\n",
+ "Malaya simply use textrank algorithm from networkx library.\n",
+ "\n",
+ "```python\n",
+ "def textrank(\n",
+ " string: str,\n",
+ " vectorizer,\n",
+ " top_k: int = 5,\n",
+ " auto_ngram: bool = True,\n",
+ " ngram_method: str = 'bow',\n",
+ " ngram: Tuple[int, int] = (1, 1),\n",
+ " atleast: int = 1,\n",
+ " stop_words: List[str] = STOPWORDS,\n",
+ " **kwargs\n",
+ "):\n",
+ " \"\"\"\n",
+ " Extract keywords using Textrank algorithm.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " string: str\n",
+ " vectorizer: Object, optional (default='None')\n",
+ " model has `fit_transform` or `vectorize` method.\n",
+ " top_k: int, optional (default=5)\n",
+ " return top-k results.\n",
+ " auto_ngram: bool, optional (default=True)\n",
+ " If True, will generate keyword candidates using N suitable ngram. Else use `ngram_method`.\n",
+ " ngram_method: str, optional (default='bow')\n",
+ " Only usable if `auto_ngram` is False. supported ngram generator:\n",
+ "\n",
+ " * ``'bow'`` - bag-of-word.\n",
+ " * ``'skipgram'`` - bag-of-word with skip technique.\n",
+ " ngram: tuple, optional (default=(1,1))\n",
+ " n-grams size.\n",
+ " atleast: int, optional (default=1)\n",
+ " at least count appeared in the string to accept as candidate.\n",
+ " stop_words: list, (default=malaya.text.function.STOPWORDS)\n",
+ " list of stop words to remove. \n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " result: Tuple[float, str]\n",
+ " \"\"\"\n",
+ "```"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "from sklearn.feature_extraction.text import TfidfVectorizer\n",
+ "tfidf = TfidfVectorizer()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### auto-ngram with TFIDF\n",
+ "\n",
+ "This will auto generated N-size ngram for keyword candidates."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.00015733542115111895, 'plaintif memohon perisytiharan'),\n",
+ " (0.00012558589872969095, 'Fasal perlembagaan parti'),\n",
+ " (0.00011512878779574369, 'Fasal Perlembagaan BERSATU'),\n",
+ " (0.00011505807280697136, 'parti'),\n",
+ " (0.00010763518916902933, 'memohon perisytiharan')]"
+ ]
+ },
+ "execution_count": 15,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.textrank(string, vectorizer = tfidf)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### auto-ngram with Attention\n",
+ "\n",
+ "This will auto generated N-size ngram for keyword candidates."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "INFO:tensorflow:Restoring parameters from /Users/huseinzolkepli/Malaya/electra-model/small/electra-small/model.ckpt\n",
+ "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/albert/tokenization.py:240: The name tf.logging.info is deprecated. Please use tf.compat.v1.logging.info instead.\n",
+ "\n",
+ "INFO:tensorflow:loading sentence piece model\n",
+ "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/albert/modeling.py:116: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/albert/modeling.py:588: The name tf.assert_less_equal is deprecated. Please use tf.compat.v1.assert_less_equal instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/albert/modeling.py:1025: The name tf.AUTO_REUSE is deprecated. Please use tf.compat.v1.AUTO_REUSE instead.\n",
+ "\n",
+ "INFO:tensorflow:Restoring parameters from /Users/huseinzolkepli/Malaya/albert-model/base/albert-base/model.ckpt\n"
+ ]
+ }
+ ],
+ "source": [
+ "electra = malaya.transformer.load(model = 'small-electra')\n",
+ "albert = malaya.transformer.load(model = 'albert')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(6.3182663025223e-05, 'dipohon perisytiharan'),\n",
+ " (6.31674674645778e-05, 'pemegang jawatan'),\n",
+ " (6.316119389302752e-05, 'parti bertarikh Jun'),\n",
+ " (6.316104723812124e-05, 'Februari'),\n",
+ " (6.315819355276039e-05, 'plaintif')]"
+ ]
+ },
+ "execution_count": 17,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.textrank(string, vectorizer = electra)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(7.94645241452814e-05, 'Fasal Perlembagaan BERSATU'),\n",
+ " (7.728400390215039e-05, 'mesyuarat Majlis Pimpinan Tertinggi'),\n",
+ " (7.506390584039057e-05, 'Muhammad Suhaimi'),\n",
+ " (7.503252483650059e-05, 'pengesahan'),\n",
+ " (7.502407753712274e-05, 'terbatal Plaintif')]"
+ ]
+ },
+ "execution_count": 18,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.textrank(string, vectorizer = albert)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "**Or you can use any classification model to find keywords sensitive towards to specific domain**."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "WARNING:root:Load quantized model will cause accuracy drop.\n"
+ ]
+ }
+ ],
+ "source": [
+ "sentiment = malaya.sentiment.transformer(model = 'xlnet', quantized = True)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(6.621038516352373e-05, 'plaintif memohon perisytiharan'),\n",
+ " (6.61143060050603e-05, 'ditandatangani Muhammad Suhaimi bertarikh Mei'),\n",
+ " (6.517221024654814e-05, 'terbatal Plaintif'),\n",
+ " (6.469109066728589e-05, 'terlucut berkuat kuasa'),\n",
+ " (6.450719772460985e-05, 'pengesahan')]"
+ ]
+ },
+ "execution_count": 8,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.textrank(string, vectorizer = sentiment)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### fixed-ngram with Attention"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(1.7071539462023998e-09, 'perisytiharan ahli sah'),\n",
+ " (1.7071528386679705e-09, 'Fasal parti perisytiharan'),\n",
+ " (1.7071498274826471e-09, 'Plaintif perisytiharan keahlian'),\n",
+ " (1.7071355361007092e-09, 'Fasal dipohon perisytiharan'),\n",
+ " (1.707130673312775e-09, 'plaintif perisytiharan')]"
+ ]
+ },
+ "execution_count": 19,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.textrank(string, vectorizer = electra, auto_ngram = False,\n",
+ " ngram = (1, 3), ngram_method = 'skipgram', skip = 3)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(2.1995491577326747e-09, 'Perisytiharan Fasal melucutkan'),\n",
+ " (2.1990164283127147e-09, 'Pimpinan Tertinggi (MPT)'),\n",
+ " (2.1981574699825158e-09, 'Majlis Pimpinan (MPT)'),\n",
+ " (2.1980610020130363e-09, 'Perisytiharan Fasal BERSATU'),\n",
+ " (2.1973393621296214e-09, 'Perisytiharan Perlembagaan')]"
+ ]
+ },
+ "execution_count": 20,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.textrank(string, vectorizer = albert, auto_ngram = False,\n",
+ " ngram = (1, 3), ngram_method = 'skipgram', skip = 3)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Load Attention mechanism\n",
+ "\n",
+ "Use attention mechanism to get important keywords."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### auto-ngram\n",
+ "\n",
+ "This will auto generated N-size ngram for keyword candidates."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.9452064568002397, 'menghalang pelupusan pertikaian'),\n",
+ " (0.007486688404188947, 'Fasal Perlembagaan BERSATU'),\n",
+ " (0.005130747276971111, 'ahli BERSATU'),\n",
+ " (0.005036595631722718, 'melucutkan memberhentikan keahlian'),\n",
+ " (0.004883706288857347, 'BERSATU')]"
+ ]
+ },
+ "execution_count": 21,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.attention(string, model = electra)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.16196368022187793, 'plaintif memohon perisytiharan'),\n",
+ " (0.09294065744319371, 'memohon perisytiharan'),\n",
+ " (0.06902302277868422, 'plaintif'),\n",
+ " (0.05584840295920779, 'ditandatangani Muhammad Suhaimi bertarikh Mei'),\n",
+ " (0.05206225590337424, 'dipohon perisytiharan')]"
+ ]
+ },
+ "execution_count": 23,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.attention(string, model = albert)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### fixed-ngram"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.15667043125587973, 'pelupusan pertikaian mahkamah'),\n",
+ " (0.15665311872357476, 'pertikaian mahkamah Perisytiharan'),\n",
+ " (0.15657934237804905, 'pertikaian mahkamah'),\n",
+ " (0.1563242367855659, 'menghalang pelupusan pertikaian'),\n",
+ " (0.1562270516451705, 'pelupusan pertikaian')]"
+ ]
+ },
+ "execution_count": 25,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.attention(string, model = electra, auto_ngram = False,\n",
+ " ngram = (1, 3), ngram_method = 'bow')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 26,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(0.031264380566934015, 'saman plaintif memohon'),\n",
+ " (0.02621530292963218, 'plaintif memohon perisytiharan'),\n",
+ " (0.02573609954868083, 'Dalam saman plaintif'),\n",
+ " (0.022935623722179672, 'plaintif memohon'),\n",
+ " (0.019724791761830188, 'Mereka memohon perisytiharan')]"
+ ]
+ },
+ "execution_count": 26,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.keyword_extraction.attention(string, model = albert, auto_ngram = False,\n",
+ " ngram = (1, 3), ngram_method = 'bow')"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.7.7"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/load-similarity.ipynb b/load-similarity.ipynb
new file mode 100644
index 00000000..0e6a9587
--- /dev/null
+++ b/load-similarity.ipynb
@@ -0,0 +1,872 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Text Similarity"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\n",
+ "\n",
+ "This tutorial is available as an IPython notebook at [Malaya/example/similarity](https://github.com/huseinzol05/Malaya/tree/master/example/similarity).\n",
+ " \n",
+ "
"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\n",
+ "\n",
+ "This module trained on both standard and local (included social media) language structures, so it is save to use for both.\n",
+ " \n",
+ "
"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "CPU times: user 5.55 s, sys: 1.09 s, total: 6.64 s\n",
+ "Wall time: 7.7 s\n"
+ ]
+ }
+ ],
+ "source": [
+ "%%time\n",
+ "import malaya"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "string1 = 'Pemuda mogok lapar desak kerajaan prihatin isu iklim'\n",
+ "string2 = 'Perbincangan isu pembalakan perlu babit kerajaan negeri'\n",
+ "string3 = 'kerajaan perlu kisah isu iklim, pemuda mogok lapar'\n",
+ "string4 = 'Kerajaan dicadang tubuh jawatankuasa khas tangani isu alam sekitar'"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "news1 = 'Tun Dr Mahathir Mohamad mengakui pembubaran Parlimen bagi membolehkan pilihan raya diadakan tidak sesuai dilaksanakan pada masa ini berikutan isu COVID-19'\n",
+ "tweet1 = 'DrM sembang pilihan raya tak boleh buat sebab COVID 19'"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Calculate similarity using doc2vec\n",
+ "\n",
+ "We can use any word vector interface provided by Malaya to use doc2vec similarity interface.\n",
+ "\n",
+ "Important parameters,\n",
+ "1. `aggregation`, aggregation function to accumulate word vectors. Default is `mean`.\n",
+ "\n",
+ " * ``'mean'`` - mean.\n",
+ " * ``'min'`` - min.\n",
+ " * ``'max'`` - max.\n",
+ " * ``'sum'`` - sum.\n",
+ " * ``'sqrt'`` - square root.\n",
+ " \n",
+ "2. `similarity` distance function to calculate similarity. Default is `cosine`.\n",
+ "\n",
+ " * ``'cosine'`` - cosine similarity.\n",
+ " * ``'euclidean'`` - euclidean similarity.\n",
+ " * ``'manhattan'`` - manhattan similarity."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Using word2vec\n",
+ "\n",
+ "I will use `load_news`, word2vec from wikipedia took a very long time. wikipedia much more accurate."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "downloading frozen /Users/huseinzolkepli/Malaya/wordvector/news vocab\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "4.00MB [00:01, 2.03MB/s] \n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "downloading frozen /Users/huseinzolkepli/Malaya/wordvector/news model\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "191MB [01:01, 3.13MB/s] \n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/wordvector.py:114: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/wordvector.py:125: The name tf.InteractiveSession is deprecated. Please use tf.compat.v1.InteractiveSession instead.\n",
+ "\n"
+ ]
+ }
+ ],
+ "source": [
+ "vocab_news, embedded_news = malaya.wordvector.load_news()\n",
+ "w2v = malaya.wordvector.load(embedded_news, vocab_news)\n",
+ "doc2vec = malaya.similarity.doc2vec(w2v)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### predict for 2 strings"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "array([0.899711], dtype=float32)"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "doc2vec.predict_proba([string1], [string2], aggregation = 'mean', soft = False)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### predict batch of strings"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "array([0.9215344, 0.853461 ], dtype=float32)"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "doc2vec.predict_proba([string1, string2], [string3, string4])"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### visualize heatmap"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "doc2vec.heatmap([string1, string2, string3, string4])"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Different similarity function different percentage."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Calculate similarity using deep encoder\n",
+ "\n",
+ "We can use any encoder models provided by Malaya to use encoder similarity interface, example, BERT, XLNET, and skip-thought. Again, these encoder models not trained to do similarity classification, it just encode the strings into vector representation.\n",
+ "\n",
+ "Important parameters,\n",
+ " \n",
+ "1. `similarity` distance function to calculate similarity. Default is `cosine`.\n",
+ "\n",
+ " * ``'cosine'`` - cosine similarity.\n",
+ " * ``'euclidean'`` - euclidean similarity.\n",
+ " * ``'manhattan'`` - manhattan similarity."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### using xlnet"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "INFO:tensorflow:memory input None\n",
+ "INFO:tensorflow:Use float type \n",
+ "INFO:tensorflow:Restoring parameters from /Users/huseinzolkepli/Malaya/xlnet-model/base/xlnet-base/model.ckpt\n"
+ ]
+ }
+ ],
+ "source": [
+ "xlnet = malaya.transformer.load(model = 'xlnet')\n",
+ "encoder = malaya.similarity.encoder(xlnet)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### predict for 2 strings"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "array([0.8212017], dtype=float32)"
+ ]
+ },
+ "execution_count": 12,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "encoder.predict_proba([string1], [string2])"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### predict batch of strings"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "array([0.8097714 , 0.78071797, 0.8244793 , 0.5807183 ], dtype=float32)"
+ ]
+ },
+ "execution_count": 13,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "encoder.predict_proba([string1, string2, news1, news1], [string3, string4, husein, string1])"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### visualize heatmap"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "encoder.heatmap([string1, string2, string3, string4])"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### List available Transformer models"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "INFO:root:tested on 20% test set.\n"
+ ]
+ },
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " Size (MB) \n",
+ " Quantized Size (MB) \n",
+ " Accuracy \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " bert \n",
+ " 423.4 \n",
+ " 111.0 \n",
+ " 0.885 \n",
+ " \n",
+ " \n",
+ " tiny-bert \n",
+ " 56.6 \n",
+ " 15.0 \n",
+ " 0.873 \n",
+ " \n",
+ " \n",
+ " albert \n",
+ " 48.3 \n",
+ " 12.8 \n",
+ " 0.873 \n",
+ " \n",
+ " \n",
+ " tiny-albert \n",
+ " 21.9 \n",
+ " 6.0 \n",
+ " 0.824 \n",
+ " \n",
+ " \n",
+ " xlnet \n",
+ " 448.7 \n",
+ " 119.0 \n",
+ " 0.784 \n",
+ " \n",
+ " \n",
+ " alxlnet \n",
+ " 49.0 \n",
+ " 13.9 \n",
+ " 0.888 \n",
+ " \n",
+ " \n",
+ "
\n",
+ "
"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "model.heatmap([string1, string2, string3, string4])"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Vectorize\n",
+ "\n",
+ "Let say you want to visualize sentences in lower dimension, you can use `model.vectorize`,\n",
+ "\n",
+ "```python\n",
+ "def vectorize(self, strings: List[str]):\n",
+ " \"\"\"\n",
+ " Vectorize list of strings.\n",
+ "\n",
+ " Parameters\n",
+ " ----------\n",
+ " strings : List[str]\n",
+ "\n",
+ " Returns\n",
+ " -------\n",
+ " result: np.array\n",
+ " \"\"\"\n",
+ "```"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "texts = [string1, string2, string3, string4, news1, tweet1]\n",
+ "r = quantized_model.vectorize(texts)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "(6, 2)"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from sklearn.manifold import TSNE\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "tsne = TSNE().fit_transform(r)\n",
+ "tsne.shape"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "plt.figure(figsize = (7, 7))\n",
+ "plt.scatter(tsne[:, 0], tsne[:, 1])\n",
+ "labels = texts\n",
+ "for label, x, y in zip(\n",
+ " labels, tsne[:, 0], tsne[:, 1]\n",
+ "):\n",
+ " label = (\n",
+ " '%s, %.3f' % (label[0], label[1])\n",
+ " if isinstance(label, list)\n",
+ " else label\n",
+ " )\n",
+ " plt.annotate(\n",
+ " label,\n",
+ " xy = (x, y),\n",
+ " xytext = (0, 0),\n",
+ " textcoords = 'offset points',\n",
+ " )"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Stacking models\n",
+ "\n",
+ "More information, you can read at https://malaya.readthedocs.io/en/latest/Stack.html\n",
+ "\n",
+ "If you want to stack zero-shot classification models, you need to pass labels using keyword parameter,\n",
+ "\n",
+ "```python\n",
+ "malaya.stack.predict_stack([model1, model2], List[str], strings_right = List[str])\n",
+ "```\n",
+ "\n",
+ "We will passed `strings_right` as `**kwargs`."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/function/__init__.py:54: The name tf.gfile.GFile is deprecated. Please use tf.io.gfile.GFile instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/function/__init__.py:55: The name tf.GraphDef is deprecated. Please use tf.compat.v1.GraphDef instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /Users/huseinzolkepli/Documents/Malaya/malaya/function/__init__.py:49: The name tf.InteractiveSession is deprecated. Please use tf.compat.v1.InteractiveSession instead.\n",
+ "\n",
+ "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/albert/tokenization.py:240: The name tf.logging.info is deprecated. Please use tf.compat.v1.logging.info instead.\n",
+ "\n",
+ "INFO:tensorflow:loading sentence piece model\n"
+ ]
+ }
+ ],
+ "source": [
+ "alxlnet = malaya.similarity.transformer(model = 'alxlnet')\n",
+ "albert = malaya.similarity.transformer(model = 'albert')\n",
+ "tiny_bert = malaya.similarity.transformer(model = 'tiny-bert')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "array([0.99745977, 0.07261255, 0.16457608, 0.03985301], dtype=float32)"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "malaya.stack.predict_stack([alxlnet, albert, tiny_bert], [string1, string2, news1, news1], \n",
+ " strings_right = [string3, string4, tweet1, string1])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.7.7"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/load-zeroshot-classification.ipynb b/load-zeroshot-classification.ipynb
new file mode 100644
index 00000000..36e1c444
--- /dev/null
+++ b/load-zeroshot-classification.ipynb
@@ -0,0 +1,895 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Classification"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\n",
+ "\n",
+ "This tutorial is available as an IPython notebook at [Malaya/example/zeroshot-classification](https://github.com/huseinzol05/Malaya/tree/master/example/zeroshot-classification).\n",
+ " \n",
+ "
"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\n",
+ "\n",
+ "This module trained on both standard and local (included social media) language structures, so it is save to use for both.\n",
+ " \n",
+ "
"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "CPU times: user 5.61 s, sys: 1.08 s, total: 6.69 s\n",
+ "Wall time: 6.75 s\n"
+ ]
+ }
+ ],
+ "source": [
+ "%%time\n",
+ "import malaya"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### what is zero-shot classification\n",
+ "\n",
+ "Commonly we supervised a machine learning on specific labels, negative / positive for sentiment, anger / happy / sadness for emotion and etc. The model cannot give an output if we want to know how much percentage of 'jealous' in emotion analysis model because supported labels are only {anger, happy, sadness}. Imagine, for example, trying to identify a text without ever having seen one 'jealous' label before, impossible. **So, zero-shot trying to solve this problem.**\n",
+ "\n",
+ "zero-shot learning refers to the process by which a machine learns how to recognize objects (image, text, any features) without any labeled training data to help in the classification.\n",
+ "\n",
+ "[Yin et al. (2019)](https://arxiv.org/abs/1909.00161) stated in his paper, any pretrained language model finetuned on text similarity actually can acted as an out-of-the-box zero-shot text classifier."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "So, we are going to use transformer models from `malaya.similarity.transformer` with a little tweaks."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### List available Transformer models"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "\n",
+ "\n",
+ "
\n",
+ " \n",
+ " \n",
+ " Size (MB) \n",
+ " Quantized Size (MB) \n",
+ " Accuracy \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " bert \n",
+ " 423.4 \n",
+ " 111.0 \n",
+ " 0.885 \n",
+ " \n",
+ " \n",
+ " tiny-bert \n",
+ " 56.6 \n",
+ " 15.0 \n",
+ " 0.873 \n",
+ " \n",
+ " \n",
+ " albert \n",
+ " 48.3 \n",
+ " 12.8 \n",
+ " 0.873 \n",
+ " \n",
+ " \n",
+ " tiny-albert \n",
+ " 21.9 \n",
+ " 6.0 \n",
+ " 0.824 \n",
+ " \n",
+ " \n",
+ " xlnet \n",
+ " 448.7 \n",
+ " 119.0 \n",
+ " 0.784 \n",
+ " \n",
+ " \n",
+ " alxlnet \n",
+ " 49.0 \n",
+ " 13.9 \n",
+ " 0.888 \n",
+ " \n",
+ " \n",
+ "
\n",
+ "
"
+ ]
+ },
+ "execution_count": 9,
+ "metadata": {},
+ "output_type": "execute_result"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "plt.figure(figsize = (7, 7))\n",
+ "\n",
+ "for label in unique_labels:\n",
+ " indices = [i for i in range(len(r[0])) if r[0][i][1] == label]\n",
+ " plt.scatter(tsne[indices, 0], tsne[indices, 1], cmap = palette(unique_labels.index(label)),\n",
+ " label = label)\n",
+ " \n",
+ "labels = [i[0] for i in r[0]]\n",
+ "for label, x, y in zip(\n",
+ " labels, tsne[:, 0], tsne[:, 1]\n",
+ "):\n",
+ " label = (\n",
+ " '%s, %.3f' % (label[0], label[1])\n",
+ " if isinstance(label, list)\n",
+ " else label\n",
+ " )\n",
+ " plt.annotate(\n",
+ " label,\n",
+ " xy = (x, y),\n",
+ " xytext = (0, 0),\n",
+ " textcoords = 'offset points',\n",
+ " )\n",
+ "plt.legend()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Word level"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 28,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "texts = ['kerajaan sebenarnya sangat prihatin dengan rakyat, bagi duit bantuan',\n",
+ " 'gov macam bengong, kami nk pilihan raya, gov backdoor, sakai',\n",
+ " 'tolong order foodpanda jab, lapar',\n",
+ " 'Hapuskan vernacular school first, only then we can talk about UiTM']\n",
+ "labels = ['makan', 'makanan', 'novel', 'buku', 'kerajaan', 'food delivery',\n",
+ " 'kerajaan jahat', 'kerajaan prihatin', 'bantuan rakyat']\n",
+ "r = quantized_model.vectorize(texts, labels, method = 'word')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 29,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "x, y, labels = [], [], []\n",
+ "for no, row in enumerate(r[1]):\n",
+ " x.extend([i[0] for i in row])\n",
+ " y.extend([i[1] for i in row])\n",
+ " labels.extend([r[0][no][1]] * len(row))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 30,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "(315, 2)"
+ ]
+ },
+ "execution_count": 30,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "tsne = TSNE().fit_transform(y)\n",
+ "tsne.shape"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 31,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "unique_labels = list(set(labels))\n",
+ "palette = plt.cm.get_cmap('hsv', len(unique_labels))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 32,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ ""
+ ]
+ },
+ "execution_count": 32,
+ "metadata": {},
+ "output_type": "execute_result"
+ },
+ {
+ "data": {
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbEAAAGbCAYAAABK567hAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOydd3RUVdeHn+mZtEmvBEICJCQhJCT0Fgi9W0AElI6CooIgYAMVFQQUFFBBFBAEFJDeIfReE3pICCGNhPQyk0w53x+RkQj66auCyn3WmkXmzC3n3hlmz9n7t/eWCSGQkJCQkJD4NyJ/2BOQkJCQkJD4X5GMmISEhITEvxbJiElISEhI/GuRjJiEhISExL8WyYhJSEhISPxrUT7sCdyNm5ub8Pf3f9jTkJCQkJD4B3Hq1KnbQgj3+732jzJi/v7+nDx58mFPQ0JCQkLiH4RMJrvxa69J7kQJCQkJiX8tkhGTkJCQkPjXIhkxCQkJCYl/LZIRk5CQkJD41yIZMQkJCQmJfy2SEZOQkJCQ+NciGTEJCQkJiX8tkhGTkJCQkPjX8o9KdpaQkJCQ+POkpKTQrVs3zp8/bx2bMmUK9vb2FBUV0apVK+bNm8f169cpKSkhJyeHmjVrAjB//nxef/11kpOTuXHjBjKZDIBevXqxa9cuSkpKHso1/RqSEZOQkJB4hBgyZEgVA7d3715GjBjBgAEDKCoqoqysjPPnz1NaWoqfnx/FxcVUr16d5ORkLBYLMTEx/ygDJxkxCQkJiUeIcePGUVRUdN/X3n33XQDCwsLo0KEDJ06cwGg08vjjj5OTk8N7770HgJOTE4cOHaJFixYUFBSQmZn5wOb/S6SYmISEhMR/EKPRSFhYWJWxHTt2cOXKFQoLC9m1axePPfYYw4YN4/r167z55pu4uLjg7+9PUVERq1evZuPGjQghWLlyJU899RQGg4EDBw7Qt29fVq5cCcDatWt5/PHHH8YlApIRk5CQkPjPccfN92vodDratWvHjz/+yFdffYWnpyc1a9akR48ezJw5E0dHR2QyGba2tiQmJqLX63FyckIIAUBsbCz79+/HbDZbDdzDQjJiEhISD4WCggLmz58PVMZlunXr9pBn9O/m0oE4FrwwmFl9u7P+/TfIzcmxvpaZtZ4LFxZQWnaIsrIkbt/OZvXq1VX2v5/ha9myJUlJSfTp04e1a9eiVFZGoBQKBS1atGDlypXo9XoeZgstyYhJSEg8FO42YhJ/jksH4tixYC7Ft3NACIyFBaiFmfzc22Rmref48QkcOXIbDw8FcrkJo7GC/IIT1v3NZjNqtfqe43bt2hWVSkWfPn1YuXKl1YgB9O3bl5deeok+ffo8kGv8NSQjJiEh8VCYOHEiSUlJREREMH78eEpKSnjyyScJDg6mf//+VtfVqVOnaN26NVFRUXTs2NEqIoiJiWHChAk0atSIOnXqcODAgYd5OQ+VAyuXYqoorzLWLTyY27dv06plf14de51nn3XCwUGBXC5DpYIjh1cAUFRUREFBAc7OzvccV6FQ4OXlxa5du9Dr9VVWay1btmTSpEk8/fTTf+/F/T/I7nxQ/glER0cLqSmmhMSjwd25THv37qVnz55cuHABHx8fmjdvzowZM2jcuDGtW7dm/fr1uLu7s2rVKrZv387XX39NTEwMUVFRzJo1iy1btvDxxx+za9euh31ZD4VZfbvDL77Ly40mpm/dy9pNviCDrCwjI59PZ+RIVw4eLCU93UhpqY6CggIaN26MyWQiIyODoqIilEoleXl5uLq6kp+fjxAChUKByWRCCEHt2rVZvHgxPXr0IC8vD7PZjIODAyUlJX+L3F4mk50SQkTf7zVpJSYhIfHgiP8ePgmDKU6wqAMYCq0vNWrUiGrVqiGXy4mIiCAlJYUrV65w/vx52rdvT0REBFOnTiUtLc26zx1VXFRUFCkpKQ/4Yv45OLi63TOmUSnR2Wg4d7ByhVZSYkGvtxBWzwZ7eznDhgXSv39/XF1dmTp1Km5ubqSmpvLKK6/Qu3dvnJ2d2bp1K0ajkenTp1OnTh3s7e2xsbHh6tWrhISEUFFRAcChQ4coKSl5KHJ7KU9MQkLiwRD/PWx8CYz6yufFmVCsrxzHA41GY9307l/9oaGhHDly5L6HvLPPne0fVVr2fZZtn87AIpeTV1rGogMnmNChJZP9/Pj882SmzqlACIFaLSMz08jZswYuXsylqOgbiouL6dGjB0ajkcOHD7N48WJu376N0Whk9+7dREdHc/DgQS5evEidOnVISkoCKqX1bm5uGAwGVq5cSYsWLaxy+wsXLjywa5dWYhISEg+Egk2TmX+4cuW1N8XEiI16isstFGyazLp16yrHf6FSDAoKIicnx2rEjEbjA/2C/LdQt2UbIkpM2FQYQQjkQlDvZg7NKsysCGnK4CFP0q2bNx4eSpo3q8XZc6u5eTOb9u3bo1KpePzxx/n2229p1qwZUPnjQKfTERgYWGV15ebmRt26da3Seg8PDxQKxUOV20tGTEJC4oFQkJ3O/BMV1udqhYzmfgoazbrCl19+ed991Go1q1evZsKECdSvX5+IiAgOHz78oKb8r6LBqNHEXs+mzeVU7MqN+BaUILOxIfWxgXx/VsOGa825maVm9/onYbE3aRMPoL+cB/eRRdSqVYuysjKgajKzTCarIq23sbEBeKhye8mISUhIPBAm7pORlG8h4osSxu80UFIhqDBDVqkMo9FIWloa48ePJzMzk1WrVvH888/j7e2Nh4cH+/fvR6lUUlRUxCuvvIKdnR0vv/wy0dHRxMTEMGPGDDw8PB5plaKue3e833sXhYcnIEPp40P60DG8mO1BscGITKkBkwnnq4WYCyrjZBXlFVjMZkx5ButxTCYTQUFBlJeXY7FY7lld3U9a/zDl9pIRk5CQ+FP83qTlaR+8T6CLgr2D7Gjup+BMlpmnwm2JDq+LWq1m7ty5fPjhh5w7d45du3ZRVlaGg4MDw4cPx2g0cu3aNbp160ZpaSmjR4/mhRdesB7bZDJx/PhxZs+ezTvvvPNArvufxNVjWSx5/RDLNttxusGryKr5U3vPbt4urYbeaAZAmAwIs5HknOsA5OuLOJp6BrlMzoGDB4iPj6dr165kZGSwadMmTCYTL7zwAocPHyYjIwMhBMePH6dFixZWaf358+cxGAwPVW4vCTskJCT+FHeM2KhRo+77+qUDcRxYuZQbqankG+04WVST9VfiifKzZbNdLw7fOosJJav3nqa+rwMAAwcOBCA3NxeNRsOVK1coLS1l165dREREUF5eTkFBgfUcj7JK8eqxLOKWX8ZUYQFAZrAh+1Yue348wYlZg/EZOh+LoQSFgzsKmZxlZ9ez7Ox67NRalHIl5aYK0gqymDp1KhqNBrVaTfv27fnhhx/Izc1l9uzZ1liZ2WzGzs4OvV5PQUGBVZ0ok8kYN27cQ7l+aSUmISHxp/itpOVu7dux/cvPKL6dQ2ZBETlFJfRYdJEbhbD3hoUVW45QkZWIWV/CnMljee75kdjY2HD27FkWLVqEjY0NxcXFPPfcc9jY2LBixQqcnJxo3749ZrOZOnXqUFBQ8EirFI+sT7IaMACNSouj1oXvv9mEUiHHrC9Gn3wKtXdtZDIZrfwbcmb0eg4+t5J2tZpRw9mXVrUaERkZiV6vp0mTJrRt2xZnZ2c8PT0JCAgAQC6X4+3tbb2/a9eu5Y033sDOzu6eOT3IliySEZOQkPhTTJs2jcDAQM6ePcuMGTM4c+YMs2fP5uLFi1xMiOdaRhZmi4VdFxLRqtWMad8CrVqN3MYe5zaDEcZykMlw6fgiMq0jZWVlHDhwgNGjR9OhQweGDx/OqFGjMBgMxMfHA1BRUYGrqyuzZ8/mxo0bD/kOPFxK8srvGXu27QS2H/sBU34m2d+9hjCWo9J5YAbMQMdvBvPyxqlcuJVImVHP+dvXuHjxIrVq1eLgwYO89957FBYW0qNHD5577jmqV6+OEIIePXpgNpvp2bMnw4cPf6iFf+8gGTEJCYn/ifj4eD755BNmz55Nbm6u1cDcnbTsZW9LXlkZ2cWl3C7Rg0XBzK0H0ZdXoDT95IpSa1HYO1N4YBmmCgNKpZJXXnmFY8eO8e233zJr1iymTZuGra0tc+fO5eTJk2zZsoXy8nKioqIwGAy/Nc3/PPYumnvGvJ39eeGJd6nuV43RQwbgopXTRXUJhY0Wn4b1QVH51d85vA0vDh1FVKMoQkJC+Oqrr+jUqRPLli1Dp9MxZMgQCgsL0el02Nvb07FjRxQKhbWh5sMs/HsHyYhJSEj8YeLj49m4cSOFhZV5X2azmY0bN5KUlFQlaVmj1WKxCLxtahLsFsiFV7ayd8Qyarv5k/DSRqJQIJPJcX/sTaq9sAS30BY4OjqyePFiGjVqhMlkIi8vj4SEBIqKijh9+jTR0dGsXbuW3NxcFAoFnp6eREdXViRyc3N75GJiTXsGolRX/SpXYsBDvZ7i4iJM5ZVSeTd7FRgNaF3NqDxssYvyZEv6QdIrcu53WKDSPavT6ahbty7l5eX4+Pggk8nIysqqUgz4YSIZMQkJiT/M7t27MRqNQGUuV0VFBUajkdOnT1fZzs0/gKTb+XTw70lGcQ6PLRuFvdqWkvJSUnNS6MHPldO1KgX1fHWAlOT8R6jT2Is2/YOxtzcDFuzl2bRxnE+GrQ69/udVqlqtRqlUcubMGQDKy8vJyMggODj4N4/v7u7O9u3bgcofK2azmZdfflkyYhISEv9e7qzAAGxtbfHz82P+/PnWyht30Dq7klpchpPGlVdbDOZa7g2eWvEKxeWl9P/+VebHfVFZYSLlKB8+Xg+NsYiioiIpyfkPUqexFwN9x/CC1xMM9HiORPdT6NWu2Nvbc/PmTQD0ej0qlYqrV69y5coVdu7cSf369fHy8vrNY+t0OgYNGoSzszM7duwA4IMPPvh/G28+KP4ZplRCQuJfhU6nq2LInnjiCev4mDFjrOMVFRUUlZTS6dthKIWCYPdAXLQ6KsxG6nnVYd6AqVQ87UlsbCyfvNKPzMxMdDodmZmZREREIJfL6dSpE3FxccyYMYPg4GD27t1rPf7Dch8OGjSIbt268eSTTz7wc/8qhZWFkTfb2TLFzYUYQxmPPfYYmzZtIjc3lyVLltC2bVtatmzJggULiI2NtaoIJ06cyMyZM62HiomJoV69ekClfP6ZZ57hiSee4LHHHrtnBfYglYj34y9Ziclksq9lMlm2TCY7f9fYFJlMli6Tyc7+9OjyV5xLQkLi4RMbG4tKpaoyplKpiI2NrTJ2R7l4fPNB3mw3igu3EpkSO5o9w5aSWpjJBc9sXnvtNa5fv05ubi46nQ53d3eaNm1KcHAwly5dwmg0cvz4cV544QV69Ogh9RX7NXTVAJjj7IRBLue883mcPZwZPHgwnp6edOzYkaioKKKjozGbzVV+bMTExLBp06Yqh9u7d6811gg/9w9LT09/MNfzO/mr3ImLgU73Gf9ECBHx02PLX3QuCQmJh0x4eDjdu3dHp6uMYel0Orp37054eDgAm5M302F1Bzqt7kRKUQp7dSewb1WNiOqheDt6oHLWEtk0mlvqQoYOHYpMJkOpVJKfn8+VK1eoXr06Fy9eRK/XU6tWLYxGI8uWLcPJyYlTp04xZMgQ3njjDet8/oqKHR9//DFhYWGEhYUxe/ZsUlJSqFu3LsOHDyc0NJQOHTqg1+ur7LNnzx569eplfb5z504ee+yx/+n8f5rYt0GlJUupACDNIY1TrqcoVZTS67FexO2P49tvv2X06NFMnjyZwMDAP3T4OwnNbm73tn15mPwl7kQhxH6ZTOb/VxxLQkLi30F4eLjVaN3N5uTNTDk8BYPZgEBgspiYcngKj9s9jmOQO/YT6/Hdd99hQM6Kedu4kHQShVzF919uI0N/mffff5+dO3cCYG9vz+3bt7ly5QqXL1+mvLyciIgIzGYz3t7e1nP+2Yodp06d4ptvvuHYsWMIIazNOBMTE1mxYgULFy6kT58+rFmzhgEDBlj3a9OmDaNGjSInJwd3d3e++eYbhgwZ8ofP/5cQXlm30Ovku2RW2jHSHNJIc0iD6hDcJpgdT+6ossvixYurPI+JiSEmJqbK2N3u27t52G7EO/zdwo4XZTJZ/E/uxnt7XwMymWyETCY7KZPJTubk/LrUU0JC4t/BnNNzMJgrVXFyrRyLwYLBbGBN4hqgskzV7FmfknYlnwq9CbXSBovFzKKZa7h5MQ+1Wm1VIspkMsxmM0IIgoOD8fHx4ezZsyQkJFhFBvC/9xVbdyad5tP20GHClxR4RLDzagH29vY8/vjjHDhwgJo1azJ79mxWr159XwN5J160bNkyCgoKOHLkCJ07d/4zt+/PEd6Hl2OmY6OwqTJso7Dh5QYvP6RJ/b38ncKOz4H3qCz0/x4wC7jnJ4oQYgGwACA6Ovo+TQEkJCT+TWSVZln/Vtorsa1tS+IbicjVcrxCvJg4cSIpqdfJUGaiUdmis3VFJpPxyY+vIl8vR6mWc/jwYQwGA7dv32bhwoUcPnyYrKwsTCYTMTExREdHs337dsrLy7G1tf2f5rnuTDqT1iagN5oRQLHBxKS1CVW2+WWjzl+6EwEGDx5M9+7dsbGxoXfv3g9det41oCtQ+WMiqzQLLzsvXm7wsnX8v8bfdreFELfu/C2TyRYCm35jcwkJif8IXnZeZJb+3KLe73k/ALztvNn05CZSUlLYv+MYb/RZxNWMsyzY/jZ1/RqSU5BOXsktwsPDGTRoEK1bt2bAgAG0bNkSrVbLihUryMvL4+TJkyQkJDBt2jR8fX3p37//H57jxx9/zJsffYbJbME+vCMKZy/y9y7GVJRDvy8TkZflsnPnThYsWGDd59q1a+zdu5cpU6YAcO7cOZYvX86PP/6Ij48PU6dOZdeuXX/u5v1FdA3o+p81Wr/kbzNiMpnMWwhx55P8GHD+t7aXkJD4bzC01lBem/0ajm0cKblUQu62XJ56y58+bnns3lOL/DxnFJqfVzQ13IN4ruN7AKw5/hlDRj3BlStXOH/+PGVlZZw6dcoaA9u/fz8xMTG8//77NG/enFu3buHq6vqHKnbciX+595uJQJC19FXcur8KFhPlaReQKdWE1KrF0aNHq+wXGBjI6tWruRP2iIuLY8KECQD079+fnJwc6tat+1fdRonfyV9ixGQy2QogBnCTyWRpwGQgRiaTRVDpTkwBnvsrziUhIXEv9vb2VQLtixcv5uTJk8ydO/dvP/cvc6YaOzWGw+DdzZtrXMNVpeAplwpk5lIAyituobQpwKnmMcjQoFRUSvWVajk+gc6YTCaEEISGhlordvySPxoDKz2TTdH2FMwF5Wy58CNdm3bggIcz6QV6bOs0xXDzAkonL3xHLMTXSUsP2QlKS0s5f/48gwYNAmD8+PFUVFSwbNkyZs+eTWRkpDX+dfDgQYYPH/6n7qPE/8ZfIuwQQjwthPAWQqiEENWEEIuEEM8IIeoJIcKFED3uWpVJSEj8h5k4cSLZN7PJnpqN625XnEwW3p2cxuBBN/ngg2y0NjLKysyUOK3gx6PzScpM4Isdkwjt6IyTpy3Tpk1jyZIlnDlzBj8/Pw4cOPCnyk6VnsmmYG2itZuxRW/CcCmPD4N90KoU1u1kChValYLxHYN+1TgOHjyYZcuW0blzZ8LDw1EqlURFRREfH19FtSjx4JDKTklI/MfZuHEjjRs3JjIyknbt2nHrVmW4esqUKTzzzDM0bdqU2rVrs3DhQuDe7swvvviiVYo9ceJEQkJCCA8Pv28TxLfeeguTyURgYCBNmzalqKiIc+eKcHdXsOjramRmGBk2LB2tVsbIkQnkFKfTuFlD3vxwLF+vmWM9jhCCo0ePotPp6NKly58qO1W0PQVh/LnfVqNq4Wy/sh/f+Gwmdw6kIvkYNn6hKBVyPny8Hr0ifQFYt24dq1evrnIsHx8ffHx8OHfuHG3atAEq3ZP79++vIgKReHBIRkxC4j+AXq8nIiKCiIgIFAoFb7/9tvW11NRUoqOjOXPmDH379uWjjz6yvhYfH8+ePXs4cuQI7777LhkZGb96jtzcXH788UcuXLhAfHw8b775ZpXXX3r8cZbNmkXd/QeoSElhQsuWfPnllzRooONaUgUp1ysIrKXGbBY0bWaLVqvA1dWVhIQEpk6dSlpaGnPnzsXLy4vHH3+ciIgIdu/ejaenJxcuXLC66+6uJPF7YmB3VmB3qOcVRO+wznT5bBBTRzzGBxNf5vDkHjii581nOhIWFsaBAweoqKjglVde4fDhw4wZM8aa7Ny/f3/s7OyoVq3aPyvZ+RFFMmISEv8y7O3tASjcuJHEtrF84O2DXAj2vfceZ8+eRavV8u6771q3z8vL4/PPP6devXrMmDGDCxcusHjxYrZs2ULPnj3RarW4ubnRpk0bjh8//qvn1el05OTkEBsby9q1a6tI26eMH0/Wvv1EqFTIZCCMRha/NoGh/fpx7lwJ1xIruHHDiEJeWTQ2KsqJunVrsWbNGiIjI/+yvK/7oXD6eYW04PgqYhcNZFXCFvo37klwcDAzZsxgzJgxFBQU4OnpSVxcHNeuXSMwMJDMzEzGjRtHVFQUTk5OrFmzhhUrVmA2m4HKZOfLly9bxR4PNdn5EUUyYhIS/0IKN24k8623MWVkAAKEIPOttyncuPGebZcvX45MJiMhIYGBAwdy+PBh3n77bS5fvkxpaaXYYsqUKRw8eJAJEyYwYMAAa7fkvXv3smHDBgCUSiXu7u7Y2tqyadMmateuTUhICOvXr8eYk8OF0hIqhEAjl1NktvBJZgYeej2enl44OOhYtKiQPXtKKC8XxMZOpqDARJcuXbh27RqRkZHUrl2by5cv/+X3yrGjPzKVnPisK3yfsIWNz3zBxiFfsuLSVoqKUsnKykCljuPppz2wszOwfft2Hn/8cbKzs/Hw8GD48OFcvnyZoKAgXnvtNY4cOUL9+vWBf2Cy8yOIVMVeQuJfSPYnsxG/6GgsDAZWvvEmpaWlDB8+HJlMRlxcHMnJyQgheOaZZ9i6dStGo5HY2Fj27t3LokWL2Lt3LxkZGWRlZeHl5cWAAQP45JNPiI6O5tKlS5SVlTFq1CiKi4uxWCwolUquX7/OzZs3efbZZzl69CgV169TZhFsLS5mb3ExFYAZOFVQgMZkoqysDG9vb6pXd+LixYvk59UgMjKS1atXk5eXR1lZGe3bt2fmzJkcPXqUL7/8khEjRqDX6619y34vH3/8MV9//TUAw4YNw7GBI699NQ4fhTs5pXkM3fAGP367mk4V4WzbthEnJznl5YIjhzPp3KWchIStpKXB1atXcXBwsBqqY8eO8eyzz7Jq1Sr8/Pys5/unJTs/akh3W0LiX4Zer6f74UPW54VmM3ccbvX0euRqOW16tuHi4Yt06tSJ2NhYPvvsM5YvX45Op6OiooLFixejVqtxcHAgKSkJhUJB165dKSsrIyEhgVq1anH69GlsbGzw9PRk7NixfPjhh+Tl5ZGdnY1er8fW1haZTEZqair1bG2Z7eZOzLVEDECMnR02cjnbS0qQy+VUr16dzMxMcnJykMlkXLt2jS+//JKDBw/i7++PRqOhc+fOLF68GC8vL8LDw1m4cCHz58+/p9Hmb/HLGoihDUKxf9ae3Ox86ChHoVJzISuJaUc+Iz//KCBQKGR06mTP1i1FlJSUoy/bx+bNJdSpU4esrMrqI4MHD+bzzz/HbDZjZ2dHamqq9Zx3xB7/pGTnRwnJiElI/BuI/x52vwuFaWiVsDE6AtPtyn5ePxYWMPmnL9srGiMWo4W9u/aiqlCz48eDRNRojUwmY+TgV7iQdJqxY8cyffp0bty4gcViISQkhPT0dE6cOEFFRQXVqlUjOjqarKws3N3dcXBw4JVXXuH8+fPs2rULk8mETqcjMTGR3bt3s3btWi5XVDAk7abVmJ7V61kXFExC7m3qRUeza9cuYmNjrcdXq9V8//335OXl4eLiQnZ2Nqmpqdb4190FfdeuXfvbtyY+nt27d1NYWMi5c+do1qwZdnZ2ACjCFeRfzkftrsapsRNpX6Wha6hj7eG1iFNZeHmpyM01UaeODRGRWtb9WIRKXYxCrqNatWpWI+bj44OjoyP79++nXbt2HDlyhCNHjlhz46Rk54eHFBOTkPinE/89bHwJCm9SGf+y4BGUjkxdtZ+XQQlvFGUjU8gIa9+IQPdwLqScJO32NYQQLFu5mCNHjvDVV18BkJOTQ0FBAYcPHyY7Oxu1Wk3jxo3Jy8tjzZo1DBw4kNDQUBISEggPD6eiooLs7GyKiopIT09n5MiRdOnSBScnJ4rKytj29dfUsrPDTi5Hp9bQ+vIlMvPy2LdvH3K5nOTkZAoLC/n2229Zvnw5L7/8MhaLha+++oquXbtSXv6zivD3Cjvi4+PZuHGjtUGnXq/n6tWrxMfHA1BSUZkALlPK0PprcW7hTO7uXFJ3pNKjRzXutERb/UMBmRlGKioEzk4aevfuzZ49e6hTp461DcvEiRMJDQ1Fp9MRHByMvb29VZl48OBBmjVrJikTHwKSEZOQ+Kez+10wVi08q/MrwruFCaWPDwIwAe2zr5OVpUeYBMlbLpOdX9np1/2nZokOWhdqeoaybds2jh07hk6nQyaTIZPJ6NGjB7m5uezYsYNbt24RGBiIWq0mJSWF4OBgrly5wrp16wDo3bs3zs7OrFy5kuDgYG7evIm9vT2zjhyh3M0Nmb0942bNRC6v/Hrp1KkTGo2GrKwszpw5g0ajQaVSERwcjIODA7NmzWLr1q3/263ZvbtKzKxGjRpcvHiRrVu3UlpaStnZMmzr/KyidOvkhntnd2p0qsHESZ9ga6umZy8d27YX07GjAyqVjBs39OzevZvi4mKSkpJo1qwZx44dY8yYMQwcONB6rLCwME6fPo1OpyM+Pp709HQiIyMlQ/aAkYyYhMSvUFBQwPz5839zm5SUFMLCwv7eifzUdv6XZGgqiKtbnXg/DwCavVIfv5F+qNxVmIxGSgxFANTyDgdk5BXfIjEt3irycHJywsnJCTs7u8r+XgYDnp6e2NnZkZuby5w5c0hISODChQs4OjoyatQo5HI56enptGjRAiEEN2/eRKvVUlZWxrx587h58yYGg4EPPvgAf39/AHbt2oVerycgIICAgADKy8u5efMmN27cwGw2s2fPHpo3b868efNo3bo1mZmZtGrV6ncpFe+swO7g7e1NREQEs2bNonHjxvQb2A9bh6pV7pUyJRHUYuN76yjOUHMj2UzdujbExZUzbdpovL29SU9Px9nZmZs3b/LCCy9gb2+PwWDAxubnFicymYwRI0ag0Wj4+uuvOX78OBcvXpQk9g8YyYhJPJL8loG689r58+cZO3bsfbd5IMbrDj+tpO5Q8rojlwrd2ZFVh+LbOTSqWamUa57gSn1PL4JmBFE3MBqdnQsAs9aNRqu2Q6XUYKd1xMPDg2+++YbHHnuMoUOHUlBQQPfu3XF3d+eLL77giSeeoLS0lPHjx/P2229ja2vLlClT6NGjB+3atWPcuHG4uLjg6urKvHnz8PDwwMbGhvr165OTk8OtW7c4cOAA2dnZLFy4kOHDh+Pv78/o0aO5evUqPj4+xMXFkZubi6OjI2VlZcTFxWE2m3Fzc6O8vJwXX3yRmTNnEh0dfd+mjCkpKXz33XfWztJ307RpUyZNmsT58+dZ8N4CPuz5Ia1mt+L2tttcf+s6HDJhSc7ixo0bXEsp5eoZGbu3l1JeaMOlS+U0aNAAFxcXzGYzXl5eTJkyhTFjxtC0aVP27dvHnj17rOcaMmQIKpWKiRMn0r17d44dOyZJ7B8wkhGT+E/za8bq7vFfGqQ7r02fPp2KigoiIiIYP34848ePJywsjHr16rFp072dhQwGA4MHD6ZevXpERkYSFxcHVBbjffzxx+nUqRO1a9fmtddes+6zaNEi6tSpQ6NGjRg+fDgvvvjivRfxU9v5uzmQUxOTpep/X6VFTtQVZ+oZ62HRFmJSlqJSKrG3dUAhVyKTAXKzNVF39+7dfPjhh9ja2rJt2zby8vJo2LAhRUVF6PV6ZDIZcrmc8vJytmzZgl6v5+LFi4wbN45jx45ZRQ9yuRxbW1t0Oh21a9dm5MiR5OfnU15ezqhRo1i4cCFpaWnIZJWJzhkZGYSGhmJvb09aWhqenp7k5+ej0+nu6dD8a/GwO0YsNjYWlapqbFClUhEbG2t93jWgKx/W/BC3827cunSL8Z1iOZKYjL7CyO2SMprX9sdGpeJWfj67Nm4kOTnZKgxxcnLi8uXLlPyksjx16hS1a9e2HtvHx4eQkBA2bNiAk5OTJLF/CEh3W+I/zR2DNGrUqCrjEydOJCkpiYiICBo1akRWVhZhYWHIZDKr7NxgMCCXy2nfvj0rV64kKysLb29vdDodU6ZMwdXVldWrV5Oamoqvry+3bt1CpVKxZcsWvL29raWRLBYLBoOBevXqkZCQQFBQEO7u7nz++eekp6fTp08f7O3tOX/+vDWJtgo/tZ2/o05EV41i4711+j7ecYDMgiI8d3hRWFiIg4MDZosZnasDhXnFeHt5gdJiLdN07do1NBoNoaGh3Lhxg9zcXKCyIoi9vT2rVq2iZs2aVhdaTk4ON2/eJCQkBIDGjRsDlW61/v378+OPP2I2m9m6dStff/01Pj4+5OXlYTabsbGx4ebNm1y/fh2LxcJzzz1HdnY2X375JWlpaXzyySfk5eWh0WgYNGgQxcXFHD9+nNdee42ePXvy8ssvW8+1f/9+Jk6cyKVLl3j22Wdp3749Pj4+FBYWotPpiI2NJTw8vPKm/KTqPLg9ice8HbBL2oyxqJB6vl4k5+ThYqelQQ1f0vILOZiYQn5ODhNefpmpU6daK5LcyREzGo3cvHmTyMjIKhL7oUOHcvLkSRYtWiRJ7B8C0kpM4j/N3cbq7tXUmTNncHNz4+zZs0RGRmIwGDh37hybNm3i7KlTmA0GKlJTwWzm5M6daLVa7O3tycnJIS8vj8LCQsrKygAoKSmhuLgYe3t7hBB069YNtVpNeXk5jo6OTJ8+HQ8PD5KTk8nJySEwMJBZs2bxzjvv0KdPH1JSUlAoFPTu3fvXLyS8D4w5D1MKYMx53li7vcrLSrkcfzdnVCoVzz33HB06dKBGjRoA2NpqcfN0RqYS5Obm4urqCoDJZCI0NNTqAnN1dbWWnTKbzSQkJLBixQoAunTpQsuWLa3djWNiYkhPTwcqjdnOnTvRaDR8+umnKBQKtFotXl5euLq64uLiQlFREZcuXaJt27YAlJeX88477yCTyahVqxbOzs5VrufWrVtERkby8ccf8+6779KzZ0/Onj3LgQMH0Gq1TJs2jZYtW3L27Flr2ag7br8qBuxuVWd5IWx8CYefYmSnb6RjNFcWBu4REUKIhxtYLCxdupTg4GBr/cnBgwdbY2ShoaEsXbrUGu+DSmVinz598PPzkyT2DwHJiEn8p5k2bRqBgYGcPXuWJk2acGhvHC+2jKJzDQ8yMtLZ/+NqTp48iU6nI+lkDhOe+QAblRMOdm4MqNkQMxBzO5f87GwsFgs9evRgzpw5WCwWysvLKSwsRAjB0qVLad26NSqVCq1Wy4YNG9BqtTRp0gQHBwdq166Ni4sLN27coKioiPr16+Pg4IBcLv9t4/UryFVKXINKCemXSP0Rl5ApBDZORsR9tk1NTaVBgwZ88MEHyGQyMjIyaNeuHdWqVSM+Ph5vb2+OHz9Obm4uI0eOJDExEQcHBxo0aED16tWJiIjA2dmZWbNm0aZNGzIyMvjqq69IS0vjrbfe4uTJk2RnZ5OcnMykSZP4+OOPefHFFzl58iRvvPEGXbp0QaPRsGLFCrp27YqNrQ2fLfkMbz9vBILk68m8+eabVjcnQLt27azux8DAQObMmcOnn35KQUHB73fX3aXqbFlDybrLlZVDonRJnE/PwsHm59Ws3GKhmUZDPz8/zp8/T0hICB4eHpw/f96azHzs2DFresLixYt58sknrW1Y5HK51E/sISEZMYn/JJuTN9NhdQc6re5ESlEKm5M3s2HVCmooBaV5udipVagVCr6dM4uCrExMFRbill/m0s0LONt7IJAjQvoCkGPQoy0vR61Wo1Qq0ev1CCGQy+XcvHkTmUyGo6MjrVu3RqlUIoQgOzubiooKPD09rXOSyWRVYjwNGzZk3759lJWVYbFYWLNmze++PpkMqsdkoXYwVca6EDj4lmEymZgzZw6bNm3i5MmTQOWKy9vbm4MHD6LVatFqtcTHx2M0GgkKCqJdu3akpqbi5eXFzp07OX78OPn5+cyePRsPDw+OHTvGW2+9xf79++natSs9evRAp9Oh0+kYNGgQRqPRKvDIzc1l7Nix7NmzBw8PD2bMmMEXX3yBXl9pTObPn0+FuQJNHQ1KR2XlN5AWikqLsLGx4cKFC/z444/MmzePwsJCiouLiY+Px2w2M23aNOrVq/ebqsWPP/6YsLAwwsLCmL09iZQCC3XnlfD5iQqySwUeM4p5Zslp+sXGov7JSN5Iz2L17iPojBbsmjRh586dVcQbUJnMfL+V1qlTpygtLeXChQtSP7GHhBQTk/jPsTl5M1MOT8FgNiAQmCwmphyegt2lIjx/iu9olEosFoHZZMKmJJ/CwkIqDEZMZiPZhTdRyJVobV2RIWNRXh5eSiUmk4kffviBo0ePolQqrbEiuVzO8uXLmTdvHlOmTKGwsJD169ej0WgwGo2YzWaSk5OtAgQnJyfOnTuHra0tEyZMYOzYsTg6OtKpU6f7qu3uh8FgYMSIZOtzownAgkYjY9TIUZy/cJ7MzEyMRiMpKSl8//331pWjSqXC0dGRlJQUevXqxfbt2ykvL+ftt98mLCwMnU6HjY0NgwYN4tatW/Ts2ZMePXqwc+dOZs+eTX5+PqWlpVYD17x5c+RyOa1bt+bgwYOYzWZcXV1JTU1Fo9GgVCpRKpUYjUaEUmC5q7cXMrCYLLh2dMV0wsTMmTMJDg7GZDKxb98+bGxsGD16NCtWrGDTpk08+eSTXL58GT8/P4qLi62HSU9Pp1evXhgMBmvJqca13GhdQ5CYa2HFE1oW9tDS54cyekR4MmD5OvrFxjLkVjaxt0vpVl7Bc59+RuCAAfTr14958+bRvXt36/F/q3PzqVOnftd7JvH3IBkxif8cc07PwWCuLI4r18qxGCwYzAZCHO05mpRKtH81hBAIYFP8JQLcXVHIlLy5rHLl5e3sj6OdK3M2VMrr++ic2FZaSmlpKUIIHB0dCQ0NpaCggPr167NmzRr0ej3R0dGUlZXRrFkzDh48yKBBg1i9ejUJCQm0a9eOQ4cq6x3u2rWLBQsW0KhRI3Q6HX369MHHx4fz589X6U31S9adSWfG9itkFOhRq2V8ueBn6f32bcVcvVoOCGrlurD60iUKCgvR6/XY2dmRlpaGv78/bdu2ZdWqVdbGl/b29nTp0oXDhw9bv6Td3Nzo0KEDjz/+OD169MDJyQlHR0c0Gg1vv/02e/fu5dSpU/j6+uLj40NKSgpPPvkks2bNQq/XU1payokTJzAZTegLTZhNZqJqx5CWf5lb+iyEEBSfK0ZUVDo/5Ro5N3ffRFYmIzo6mqlTpyKTyWjcuDHDhg1j//797Nu3j/DwcEJDQ+ncuTNyuRyFQkFw7VAa1Iilmq4uacnZPN7rCauy8PEeXTmQvpOaznIivCo7OEdVsyHFuTkAaj8/vEeOJOTJJxn6/vtsyMlh8E+V6JcuXWq9t1FRUdjZ2TFr1qw/+cmU+DuQ3IkS/zmySrOsfyvtldjWtiXxjUSuFeTh7eTIrB37+WLfMZ5qGM6EzjH0aNYQBzsddjaOvNNvOS4OXmTkXcdW48DYDm8xqUYNJo8aSd++fQkKCuLMmTP4+vryzTffMGbMGGbNmsWJEyews7Ojf//+tG7dGoC5c+dSUFDAoUOHyMvLY8aMGcTExADQr18/EhMTadWqFevWrePbb7+lZs2av2rE1p1JZ9LaBNIL9AhAIPuVq5exYOsyJjUZwa1bt6wJyoBVaAKwZMmSyq1/ktFnZWVhMBjIzc0lLS2NmjVrEhgYiLOzMz179rT20QIwGo2Eh4eTk5PD22+/zblz57h+/ToXLlxg5syZLF26FFutPRqVLX2avVJZFcSsoLTYgLnIjKgQWMosIAeFnQIbPxtUGhVCCJRKJV9++SWBgYGUlJTQuXNnXn75ZRo2bMjEiRNZsWKFteLHm6On0zzgMZrX6kXKrUvU9AgjJeE2V4/99P57hkDoE2jUKkAGOj8UYb0wud0rvhg8eDDLli1jxYoV98jkpc7N/2yklZjEfw4vOy8ySzOtz/2er0wGDrztS9RZO0wVP9foU6o1dBnxAoeyTOyM28pHa0fhaOtMSLWGXLx5jLVH5hL6/FBGffQRXVJS6NatGykpKSQnJ7NkyRJGjx6NXC5nwYIFtGzZkujoaK5cucKpU6c4duwYKpUKFxcXOnToQG5uLnXq1MHJyQmDwUBGRgZubm4MHDiQOXPmWIUM92PG9ivojT8LH8wWBeVmFRrFXW1KhByZRUlxeSkecieg0g12h9atW7N06VJOnTpFs2bNMJlMvP/++8yYMQOLxUJMTAw5OTnY2Njg7OyMs7Mz3bt3Z9y4cfj5+VmFF0ajkdOnT2M0GgkJCeHjjz+mqKgIpVJJTk4OqamplJYVYzKZcPRQIZcpOHv9ACaLCZAhswHn5s4Uny1GrpVTeqGUjr07UpBSwJNPPklmZiaBgYHY2Nhw+fJlvL29yc3N5bvvvqNfv36kp6fz7LPP0tCrF8cu76JBzba0DuvFzZyrfLv3I/auvoBvmAM//vgj3377LQs2n4Ap5ytvwsyZUFJyz/2VKtH/e5GMmMR/jpcbvGyNid3BRmHDU4+NJqCRLQdWLqU49zYOrm607PssdVu2Yd7CmnSIvcyIjlN5f+kolCo5uzcdwiVQScOGDek0ZkyVc+Tl5VkbTQ4ePJiOHTvi7+9PjRo1SEpKYvv27Wg0GoKCgti5cycKhYJmzZpx+vRpHBwcaNu2La1atWLu3Lm/65oyCqrWTjQLBUsu9OW52lsQNrnIjfbY5nlxdcx0ticeYOSGybhFfU7btm2tq4rg4GCeeuopEhMT2b59O5999hk+Pj7odDqio6NJT08nMDDQ2jn6Tq6WVqslJyeHhQsXMnDgQPbu3cu0adN48skn+frrr/n2228ZOHAgLVu2ZPr06Tg6OhIQEMCNGzcod05DrdZQv0YLTiftwySMKIxyLGkWHCIcMFw0IMyCfev3YTAYCAkJqXT1CoFarWbLli0MHDiQa9euER8fj0qlwtnZmXr16rFq1+dkFaTy3qrBVHerzcDY12kc1JEpi4bx6VY7hg0bdo90/7eQKtH/O5GMmMR/jq4BXYHK2FhWaRZedl683ODlyvEAqNuyjXXb0jPZZE47TuaNFOSlRo4mLaPYkEehScPXP84GoKioiCZNmjBu3DjrfmVlZQwYMADDT40pVSoVcrmcBg0a4O/vz48//siGDRsoKyujadOmNGzYkNatW+Pi4sKiRYu4cuUKV69epby8HI1G8/8aMx8nLel3GbLqY1dzLAtqFLbmZZOaII0FKkM9dAppTd83h2EX6XHPccLDw6vEe+7QrVu3Ktd3hzt5Y3ezePFi699Dhgyx1gqMiYkhMjLS+trs2bNBLujQriN5103MGrqJV7/uxvwv5iKXyzl58iRjvhxD/fr1adasGfv376dp06aYzWbOnz9Ps2bNWL58OQUFBXz33Xc0bdqUw4cPM2vWLAICAji2Oo1NB5cTW78Pu899D0BseG96xgxg4AfNrfM4f/689e+7r/Hu64DfFm9I/HORjJjEf5KuAV2txuzXSN+fyJIPv+DZ+r0AEGbBuDoDiDu2n3feeQedTscnn3wCwHvvvcfEiROtMaU7zJs3D5lMRoMGDRg3bhxDhgyx5n2dPXuWqKgoxo4da42VZWRk8N577zFmzBhu3rxJQkLC/at0/ILxHYOYtDahiktRq1LQrEcdnFBRtD0Fc0E5CicNjh3972vAfo0pU6b87m1/i18W472DkFemFRgRmBGcTs0nWpEEZ1cx++mFlJVWcOjAPioqjMTHx9O4cWMKCws5dOgQrVu3JjAwEL1ez6lTp3jppZfYvXs3KpWK/Nwi7OVubDj2FVkFqXy4egRN63bk/Zlv/eG5S+KNfy+SEZN4ZEnffIElJ3/k2fq9sFfbUlpRhjBZ0AgVq1atIjg4mLZt23L8+HE6dOjAjz/+aO0ybGtry/Lly8nLy6NXr17s3r2bTp064erqSkFBAQCxsbFkZmaiVqsJCwvj0KFD7Nmzh5YtW7Jt2zbq1atH7969uXr16v87116RvgBWdaKPk5bxHYOs47/HaP1VxurX0Ol0VQzZK6+8AkCDeo1R+jRkm40RoXXEUpKGOXkZVBj4rIuWb85W8GlnG16L09CzVy9atmzJpUuXaN68OevWraO0tJS33nqLXbt2sXPnTvLz89m7dy/Xr19n2tSZNPV/hk0HlzO+/yya9gykTmOvPzx3SSb/70UyYhKPLFM3fMaNgnQ6fjOElv7RaJU2tF7QH7Ow4Ks3s3z5cpRKJZ6ennh5eWE2m8nNzaVTp04UFBSQnp7OkSNHOHz4MHXq1KFnz57WElJ3emilpqbyzDPPUFxcjJ+fH8899xy2trZ06dLld+eE3aFXpK/VaP0TiY2NZePGjVX6e5mRs8/iw3VdpZhGVJSx7eu5LC42Uc/TiUzDVwgGUGAYgKhYBEB2djYajYZevXpx8OBBOnfujKenJ9nZ2QwcOJCPPvqId955h0uXLqHT6eg0vB7nC91o94I/zz7bn927dz+U65d4OEgSe4lHljd7jKaGky/bB39NA58QvB3cOT5qNa52LsTHJzB9+nTq1KmDTqcjJyeHAwcO0L17d7Zt24afnx8XL17k9ddfJygoiPT0dN5//328vLwoLS0lNzeX0tJSioqKOHfuHM2aNUMIQe/evbG1tbXWBvwjVTr+6YSHh9O9e3ercdbpdByqqMF1i5t1G0uFnrV9NHzSKYK6nq0wU7mCFDgghIaKnDJ27NjB9evXefXVV5HJZDRt2pTNmzeTmJjIO++8g6enJ87Ozly5cgWDwcDQoUO5fv06mZmZUgX5RxDJiEk8cmRmrefQoZYk1n+ZCtssCr0OczwtgR4hsdjZOePgF4bRUMKmTZsIDg7m2rVrtG3bltq1a1sD/xqNhho1atC2bVssFgsFBQUMGzaMJUuWEBoaSm5uLrdu3bKKOeRyOSqVik2bNlFeXs7Ro0f56KOP8Pf3/8Mrsn8y4eHhVYrxljtWpjcUndxA6uynwGKm/TI9WWX1AQVfn1yNQqbg9e2zKDKUIL9VQXR0NA4ODjg7O5OXl8f+/fu5desWdnZ2+Pv7M2LECDw9PfHz82PSpEl8//33uLm5cfToUV544QXS09OrtGKR+G8jGTGJR4rMrPVcvvwGhvIMQIDcRFboYsptMyhCMB095h7jcKvbjMGDB/P6669Tq1YtEhIS8PWtdOX5+/tbFW8ajYZhw4bRr18/zpw5Q5s2bVAoFHTo0IERI0ZYz7tp0yYGDx5M3759uXbtGgUFBRw+fJi8vDxry5b/IuM7BqFVKSg+swWfIfNQOHrQ+/mXcLCpjFstPbOOVjUb0tw/Gi8Hd15tPIht27bRunVrzp49S3R0NMeOHbNWrW/Xrh2LFi1i6dKllJaW0rlzZ8LDw1EoFCxcuJCkpCRpRfaIIRkxiUeK5KSZWCyVUnWtVk5ZmQBFBR4tMplxaSc7LeWoKooh6xKNGjWqsm/Lli1Zvnw5AFevXiU1NZWgoKBfPdedAr/5+fmYTCar63DKlClEREQQFhb2m1U6/gv0ivTF+/AniIIMzN+NRmXI58LJeORqwY8XdnA97yZbr+4ntSCdMM86PL36Va5evcqGDRvw8/MjISGBq1ev0rFjR3bt2sXBgwc5cuQI7u7ulJSU0LhxY+bOncuePXs4d+4cY8aMsa7IJB4NpJ8rEo8UhvKfK3nodApCwzQMG3qThg1tUXtEkLP0ZdwdNHw2exZeXl7WBpIAo0aNYuTIkdSrVw+lUsnixYt/sxSRr68vr7/+Oo0aNcLFxYXg4GB0Oh3vv//+33mJ/yguHYijp489Z200jIxpwvSteylPS6K8bkPc7V2pMJtoH9gMZ1sn9iQf4ZtPFrJs32oSExPp3Lkzn332GRqNhg8//JDt27ezcuVKbGxsmDZtGjNnzrxvh+37dseW+M8iE+J+HYgeDtHR0eJO+wgJib+DQ4da/uRKrIqNxofmzQ/85ecrKSnB3t4ek8nEY489xpAhQ3jsscf+8vP8E8nMWs+5E5NQ2pbT/+lUJvWL4o3PTxPm60UdFycOX7pBvkXQ0r8hzk7OXCpNAa0Cg8HAoEGDGDNmDK1btyYpKclaqmvv3r1Uq1aNH374gZEjR3L79u2HfZkSDwCZTHZKCHFfv7vkTpR4pAgIHIdcrq0yJpdrCQi8t1rFX8Gj5Dq8mzuxR5VdOTIZyORQrdktZIrKH81GuQxh1ENFKaas0yhd9DRs1YSvvvqKzMxMxo8fT506dVAoFNYqI0IIAgICmDhxorXsl4SE5E6UeKTw9uoJVMbGDOWZ2Gi8CQgcZx3/q5k5c+bfctx/OnfHHu8gVwlGDHPl2FY5LX29KL+Swr6SEt5wdeNA8nXOu7hQr149unTpwuXLl5k4cSJ+fn6MHDmSBg0akJSURNeuXfnyyy9p1apVlYajd4oCSzlijx6SEZN45PD26vm3Ga1HmTuFe+VyeZXY490obCzIhCAoM4+7a8WbCwo4d+4cYWFhZGRkEBISYu0blpGRgY2NDS4/GTmtVsvTTz/NggULqF+/PoMGDaJly5a/qkg0mUySWvE/jPTOSkhIVOHOCuiOwm/KlCnY29sjhLB2iH7sscd45513SElJoWPHjjRu3JhTp04xf/58nn/+eWrXLiE+Pg83NyXvvufJ8u+qs3lzEZs2FGPOLOGjW3lM8/bhA285t00m4kxGZDIZCoWCTZs24ePjQ1RUFOfPnyctLY2ZM2dSUlKCm5sbRqMRpVLJnj17iIyMZO7cuXz00UeEh4cjhEAmkxETE0NERAQHDx7k6aef5tVXX33Id1Xi70KKiUlISFThqaee4vvvv7c+//7773F3dycxMZHjx49z9uxZa6NIgMTEREaNGsWFCxeoUaMGiYmJvDj6NRZ9XRs7ezkH9pcC0KqVK9vmTGBz7SAC1BrWFBYA8MHt27Tt2ZNz585x+vRpQkNDq8xn0KBBxMfHA5XKw+rVq1tfCw4OZvr06WRmZuLh4VFFrVhRUcHJkyerGDCTyfTX3iyJh88dF8A/4REVFSUkJCQePsHBwSI9PV0MHjxY1KxZU7z66quiRo0awtPTU3h7ewsXFxdRo0YNERQUJHQ6nRBCiOvXr4uaNWsKBwcHERISItasmSqcnFSidh2N8Pe3FfXrB4qmTZuKkBo1hJNSKZwVChFkZy/USqXIy8sTQgiRlZUlevXqJYKDg4VGoxGHDh0STzzxhNDpdGLy5MlCCCGioqJE7969hRBCxMTEiFq1aomwsDDh4OAgfH19RWhoqPD29hZxcXFCCCFat24tXn75ZREVFSVmzpz5oG+lxF8AcFL8it34S1ZiMpnsa5lMli2Tyc7fNeYik8l2ymSyxJ/+/f3d6SQkJB4YEydOZN68eVw9lsWS1w/RJXogcoM9z/QdwoYNG0hPT+fQoUNMmjQJBwcHLBYLLi4u2Nra8vbbb1NUVES3bt1o3749169fx9XVlQsXLnDlipyCAiO5t+1o0KATiYmZdOvWjT3HjlE9JAQcHdEEBiBTKJg1axZhYWG89NJLtG7dmieeeAJ7e3vrqkz8lAr07rvvcunSJbZv387QoUM5cuQIr776Kq6urtSpUweLxcLgwYMxm83WjgNw/1WZxH+Dv8qduBjo9IuxicBuIURtYPdPzyUkJP5hPPXUUyxZtIy45ZcpySvnTNJe6tZy5nT8bjSaQtzd5ZSU3GLatGnY29tz69YtioqKkMvlTJ48+Y4XhZ07d6LVaq1V7Dt16oSbmxujR4+mbt26GI1GioqKePHFF8nLy0OlUmGxWFAoFGzduhWAPXv20K9fPxwcHCgrK8NkMmE2m63NR1988UXq1q1Lx44dKSkpwWKx4OjoiNls5vLly8jlchYvXkxBQQHJyclVrlHiv8lfIuwQQuyXyWT+vxjuCcT89PcSYC8w4a84n4SExJ8nPj6e3bt3U1hYyPWbSaSXXcRYqMLeEWS6UxSXmjCUy9BqLRQXp+Lp6U16ejpyuRw7OzvWr19PZmYmsbGxVpGHnZ2d1YhduXKFwsJCPv/8c2xtbfH19eWLL76gtLQUrVZLTEwMGzZsYM6cOUyfPp28vDxMJhOXLl1CoVDQunVrGjVqRGFhoVVdGBcXx6VLl0hMTMTe3h5XV1fGjBmDXq/HaDTSu3dvPvnkE/z9/amoqLBeq52d3UO5xxJ/P3+nsMNTCHFHZ5sFeN5vI5lMNkImk52UyWQnc3Jy/sbpSEhI3CE+Pp6NGzdam1iGhNTlRNomjqdsol1HATIzo190Y9PmmsyY6UO1akr0+tts2rQJPz8/3N3d8fT0pFq1avj7+5OdnQ2AQqFg5MiRAIwbN47q1atz69YtJk+ejKurK6+88gpOTk6o1Wq++uorjEYjmzZtonnz5gQEBPDkk09y5swZDAYDDRs25PTp03Tp0gWFQsHw4cMZOXIkQUFBjBkzhuHDh1NeXs6AAQMIDw8HIDo62lolxcfH5+HcXIkHygOR2AshhEwmu299KyHEAmABVJadehDzkZB41Nm9e3eV5pWhoaFs3LgRfZmeEePtSEnRsvibfGLb2ePvr6a42IJOJ8fDwwN7e3v69etH06ZNMRqN3Lx5k+LiYhQKhfV4QghKS0uxt7enoqKCZcuWWV+LjY3FbDbTuHFj3NzciIyMxGAwkJ2dzeTJk3nttdfYvXs3jo6O1lyxJ598khYtWlBQUEDHjh2pqKhgw4YN1mMqlUqqV6/Oq6++yvz582nYsOF9r/vunDEpf+y/wd/5Dt6SyWTeQohMmUzmDWT/jeeSkJD4A9xZgd3Bw8ODiooKHBwc0GndiI7OI/WGkQH9U1EqZXh4KAkMdGbQoEHcvn2biooK8vLyqKioQC6X07dvX4qLiykuLubHH3/ku+++Ayqr/dva2mJra0tZWRn16tXjzJkzZGRkYDKZyMrKoqioCDs7OzQaDdHR0djY2GBjY4PZbGbChAkcPXoUi8XC/v37+e6775g/fz4uLi50796d7du3c/PmTQDkcjn29vYUFhaSkJBAQkICq1evJi8vj2+//Zbnn3+e/Px8unfvLuWP/Yf4O92JG4CBP/09EFj/N55LQkLiD3C/RpwjR45kyMDnyI5/DItZzeNP6Jg23RtfXxXz5tfihRdf48SJE0yfPp3ExEQcHByYNm0aLi4uzJw5k82bN1NWVsbzzz/Pjh07KCgoQC6Xc/LkSeLj47FYLNSqVYsrV67Qtm1bFi5ciF6vp1u3bgwePJj09HRq1KhB69atycvLIyYmhs6dO9OjRw+8vb0JDQ1l3LhxaDQa2rZtyzfffGN1Ge7du5cGDRowbdo0zp49S0BAAJMnT+b8+fOYTCYSExM5efIkfn5+klLxL2bp0qWEh4dTv359nnnmGTZu3Ejjxo2JjIykXbt23Lp1C6hMmh8yZAgxMTEEBATw6aef/jUT+DXt/R95ACuATMAIpAFDAVcqVYmJwC7A5f87jpQnJvEoMmfOHBEcHCz69ev3p45To0YNkZOT87u2PXfunJg6daqYPHmyaN68uejcubOYOnWqOHfunGjevLmIaREihg7yEHXqaIRchqhe20G4dnAVgJDJZEKpVAqZTCYcHBwEYH0OCE9PT/HUU08JuVwuZDKZCAsLE506dRK2trbC19dXhISECFtbW6FUKoWNjY1wdnYWarXaun/37t3FpEmThEwmEwaDoUqe2JgxY4RMJhOOjo5i+PDhwtvbWzg7OwtbW1tha2sr1Gq18PLyEkqlUtjb2wsbGxuhUCjE8OHDhRCVOWN79+79U/dZ4mfOnz8vateubf3c5ebmiry8PGGxWIQQQixcuFCMHTtWLFmyRHh4eAhbW1vx9NNPi2XLlgmFQiHq168vYmNjRVZWlhBCiMmTJ4vBgweL1q1bi5o1a4o5c+YIIR5AnpgQ4mkhhLcQQiWEqCaEWCSEyBVCxAohagsh2gkh8v6Kc0lI/NeYP38+O3futDbcfBCEh4fTvXt3dDodoaGhXLlyhe7duxMeHk5KcjIXz6Vyfr+G4fVjsVGpyL9Zjv5kCdhU/vDV2Gp48cUX6dy5MwCLFi0iOTkZe3t7cnJyePbZZ/H398fBwYEJEyYQGBiIXC5n2LBhhIWFUVFRQePGjSkpKaG0tJQaNWoQEhKCnZ0dWm3VLgMymcyaJzZ48GDUajUdO3ZEr9ejVqtp164dUVFR6PV6hg4dyvXr17FYLAQEBKDX66lWrVqV40lKxT/PujPpNJ+2h1avfEaZb0MO3iwHwMXFhbS0NDp27Ei9evWYMWMGx44dY+rUqQwcOJDXX3+duXPn0qVLF2rVqsWmTZvo27cvH330kfXYly9fZvv27Rw/fpx33nmnSuz2fkhRTQmJh8jzzz9PcnIynTt3ZtCgQRw4cIDk5GRsbW1ZsGAB4eHh5OXlMWTIkHvGc3Nzefrpp0lPT6dp06bWL/qUlBQ6depEVFSUtYzT0qVLsbW15d13360UcOj1NAv24stmmci807G7WcTcCRPZf/oKt3NzUSoUJGblM33rXvQVRgSAAfhJu1FaVMrRo0fJyKjszXby5ElatWqFSqVCp9MREhJCeXk5er2eV199FbVaTUVFBd988w3p6elYLBZrnldFRQWBgYGkpqZib2/P9evXCQwMRK1W8/HHH6PVaikvLyc/P59z585RXl7Otm3bcHBw4Pbt2+zevRuDwYAQgtzcXG7fvo3FYqFevXqUlJQgqZ7/WtadSWfS2gT0RjMCKC43MWltAlDZyXv06NGMHTuWHj16MGnSJD7++GOcnJzYvn07UVFRdO7cmcLCQhITE4mJiUGhUGA2m8nPz2fPnj0UFRXx5Zdf8tJLL+Hh4WF1R/4qv7ZEexgPyZ0o8Shyxw344osviilTpgghhNi9e7eoX7++EEL86vjo0aPFO++8I4QQYtOmTQIQOTk54vr16wIQBw8eFEIIMXjwYDFjxgwhRKW7RwghxLlVYkB9G7Ghr1aIyY7Cz1ErmlWvL15oMkD0r99D2KttRdeg1iLMu7qo7qITtmqVcLK1EcgQMoVMqD3Uom7duiIgIEAAQqvViho1agilUilcXFzE9evXhVKpFK6urmLy5MlixowZQiaTieeee064ubmJyMhIoVKpRMOGDYVKpRJarVbY2NiIIUOGCD8/P+Hq6ioUCoVo0aKFcHJyEjY2NsLe3l6oVCqr+9DPz0/IZDLRtWtX0bp1a+Hs7CwUCoUICgoSdnZ2wt7eXjRr1kx4enpWcSeeOHHiQbyt/1mafbhb1JiwSdSYsEl4D5kn5HZOQunqJ2y9AkTv3r1FzZo1RVCQn6heXSvkcoSHh1q89NKTolmzZiIiIkK0atVK2NjYCK1WK4YOHSri4uJEjRo1RNOmTcUbb7whpkyZIlxcXERFRYUIDQ2983n+e92JEhISf4zMrPUcOtSS3XtqUV6eRdatrRw8eJBnnnkGgLZt25Kbm0tRUdGvju/fv58BAwYA0LVrV5ydf67s5ufnR/PmzQEYMGAABw8eBCqThRs3bky9js+yJ7mcCzkWAFxsvcktK2LLlb0826AXaoWKuOTjJOXcolWdAHpFhmJjo0LlpULpoqTxu40ZNGgQcrkcmUyGnZ0d06ZNY/z48SgUCvz9/dHpdLz44ouYzWa2bduGWq3m3XffpU2bNpSWltKvXz+OHDlC9+7dee+997Czs+PDDz/k5MmTBAQEoFarWbduHd26dbNW1Xd2dsbZ2ZmhQ4diZ2eHra0tP/zwA2azGWdnZ8LCwpg6dSrR0dHExcVx6NAhsrKyWLBgAVApAImO/rlBsFQQ+I+TUVC1TxwCEAKjubL6ykvPdqSwMAu9vgIPDyUVFSbWrt1AeUWudVUlhECv13P16lWWLFkCVH6GlUoldnZ2v28F9hOSO1FC4gFzp+vxnaaRQpi5du19jKaK/2fP388vux7LZDIMBgOjRo2qVOktqseUvQKDqdIFqVU5UGDIwcfREy97N2xVWpRyJbll+WxJuIy9jQaHcAey9hcDkP5pOvNK5tGiRQuuX79OjRo1mDhxIqWlpZSVlQHwyiuvMHnyZDQaDVqtloCAAKpVq4ajoyNFRUUkJydz+vRpPvzwQ9544w2Kiorw8vLCwcGB7t274+rqSlRUFFlZWQgh+PTTT+nTpw/Lli3j888/Z8CAAbi7u1OtWjVKS0tp3769VW4PsHDhQkaOHElZWRnNmzfnyy+/RCaTYWNjw8CBAzl16pQks/8DLF26lJkzZ6LPzqa5RynPhph5YZsRjUyNcHAlrP+bfPVkK/q9EYbRaKGw0IzZDNX8VDzdV8eCBUmUlsqoW7cuq1at4oknnuD06dM0btwYAI1GY41/KRSK3/0DQ1qJSfwlxMTEcPLkyYc9jX8Un376KXXr1qV///5Vxu/X9dhiMVA32GAVd+zduxc3NzccHR1JSkpi4cKF94y3atXKmo+1detW8vPzrcdLTU3lyJEjAHz33Xe0aNHCWn/Qzc2NEhtvVl/8+UviYnYirzQbxPdPz2H2oSWUVJRy6sUfGddqMFobDW5NPcjXVIAF1Co17Vq0w8nJiWPHjuHq6sqZM2cwGo18//33yGQyunXrxtKlS7FYLGzZsoXc3FyeeeYZjEajNcere/fuvPbaa0RHR+Pr60vdunWpV68ejz32GLVq1eLmzZs0btyYadOm0atXL0JDQ5kzZw69evXCy8uLb775hnbt2mFra4ujoyPe3t7Ur1/fek13BCFDhgxBr9f/v21aJH6dCxcuMHXqVPZ89hJpLwgWdpbRqoaS15ooeDFS0LKeLz6pOynansLRU3nY2sr5bK4vLi4K0tOMtI21p0NHO9zd3fnoo4/o2bMnDg4OhIWFMWPGDAYNGgRUyvDHjRtnPe/58+fx9/f/zblJRkxC4m/i11SHv+x6XFZm4aXR6dzKLuLUqVOEh4czceJEq5tFp9Nx7ty5e8YnT57M/v37CQ0NZe3atSgUCnJzcwEICgpi3rx51K1bl/z8fEaOHImTkxPDhw8nLCyMjqssNKymvmsWFcio/BXcsU4LyowGZCo54z58i2KjhfRDOWj2lmCLDEeL4MqxY9jb23Pz5k2ys7NRqVQoFApeeOEFSktLiYqKYseOHdja2pKWlgZUFgSuVasWV69epW7duphMJq5evUpkZCTHjh3j2rVr5OXlsW3bNubNm8e1a9dYu3at1WVaUlLCU089xbZt2ygoKAAqCwK3atUKg8HAkiVL2Lt3r/WKjhw5wsWLF3njjTdYt24dFy5csL7Wq1evX33f/P39uX379u97k//jFG7cSGLbWL5rG0tsRQWqVR+jNBtw0cpIK7Kw8oKROUf0JB3eSXnODW5n5mCskBERqaV2bQ3e3kqEgBHD0zh9ykj79u3p3bs3UVFRVSq8/BkkIybxh0hJSaFu3boMHz6c0NBQOnTogF7/86rCYrEwaNAg3nzzzYc4y4fP3arDWbNm0atXL3x8fNBqtbz5RmW2SVGRmbffysJgsKDVyhk6pA7r1q0jLi4OR0dHnn76aYYNG4ZcLmfp0qVs2LCBgoICmjZtSp06dXjuueesX85+fn4oFAp69erFpEmTUCqVLFu2DE9PT2rVqkVMTAx16tShY8eOJCUlsevwWfQekay6BI+tKiPYU4VnAx05MsGmy3EYzSYafNGbNxZ/QF0fH7oolQizCSEDo9lM/Pnz2JWXM3HiRBwdHZkzZw6pqals2bKFWrVq8c477wBgb29PSkoKUFkQOCMjg3r16rF8+XJycnK4fPkyeXl52Nvbc/jwYc6fP29VFtaqVQtHR0er5D4rK4tVq1ZZP3M3b95kzZo1rFq1CovFglKpJC8vj+PHj6PX60lISODw4cPWjtBnz561vj+SzP7/p3DjRjLfehtTRgYgsBSXkLnPTGFK5fsxequBiS00LOhmg8VsIi4ujpYLnsZkULJ1SxHPDEglI6Nyte9gryS2XXvCwsJ49tlnqV+/Pnq9noyMDD799NP/aQVm5dcUHw/jIakT//lcv35dKBQKcebMGSGEEL179xbffvutaN26tThy5Ijo27evmDp16sOd5D+EX6oOg4KCxMqVK0VIqL/YExcqevZyFHXqqIVSifDyUglPT2fRs2dP4erqKnx9fcW5c+esqsNOnTqJoKAgAQhXV1eRk5Mj+vXrJ2rXri1CQkLEgAEDRPXq1UVOTo5o3769UCqVol+/fkKr1YratWuL0tJSsXnzZhEQECCio6OFp6enqFOnjrBYLOLcuXMCEI712gq1d22h0HkJua2TcG70mHD19BYBtrZCJ5eLII1G1LOxEbO8vUWEjVY4KBQiODhYqFQq4ePjI4qLi8X169dFUFCQuHXrlrh+/brw9PS0NrOsVq2aCAwMFEII8c0334ioqCjx4osvCrlcLmxsbMT69evFqVOnhE6nE08++aSoW7eueOKJJ0TdunWFo6Oj0Gg04uuvvxZdunQRgKhevbqQy+XCyclJ9OzZ09pI09bWVri4uAi1Wi3KysrE119/LZRKpdBqteK1114TGo1G7Ny5U3z33XciLCxMhIaGitdee+2e9+3vIj8/X8ybN+//3c7Ozk4IIURcXJzo2rXr3zafX+Nqm7biYlCwuBgULNb71xQ1VCpxuFZtcTWylsh9zUFU18lELReZCPeUi2o6pVCr1aJ+cD3hbuciqrm6iKVfhImdO2sKnaNS1Av2F9WqVRM1a9YUnTp1Ek2bNhUGg0Hk5ORYlYi/BZI6UeLPcMelcKluCNef7kcNDw8iIiIAiIqKsv7Sfu655wgLC+ONN954eJP9B3Lw4EGuXLlCcnIyU6dOJT0tn/en2rB1SzEmM+h0anbuWkRJSQUXLlygtLQUjUZDnz596Nq1KxqNhkuXLmE2m7GxsaGoqAiA27dvk5KSQkREBFu2bCE1NZUWLVoQHx+PyWTi1KlTVkHHvHnziIqKIiMjg7Zt21JeXs7NmzeZPn064eHhyNU2lKYkYCnXAwJLWQH5pzaReyuLAoMBJ4WC5PJyEgwG/NRqCs1m/FUqtm7dSp06ddBqtURGRtKxY0dSU1MpLi6+5z6UlpaiVCoxGo18/c1sSkouU1i0nBYtnAgKrkafPn3o3r07NjY2aLVasrOzadmyJSUlJej1euRyOdWqVWP//v14eHhQXl6OTCbDyckJk8lEUlISrq6utG7dmqKiIry9valbty7PP/88kZGRhIeHc+LECSwWC7dv32bChAns2bOHs2fPcuLECdatW/dAPg8FBQXMnz//gZzrz2DK/NntXVuj4TlXN55NvUH3izcZvF6P3iQnz6Amt8IeGxsnnHWOnLl4jmH9hxDhHcH6z3x5b6Q7JSUWrl7PpG7duhw/fpx9+/bRqVMnNBoNbm5uf0iJeD8kIybxm1RxKQiBOfsW8vx8CjduBKqqiJo1a0ZcXJxVQPAoUnomm8xpx0mbeABzYQWl8ZWJtlOnTsXHx4e4uDjMZjMBXnXwdHSmsVsIBfkmipId0Ov1xMbGEhQUxDvvvENycjJFRUVUVFTQsWNHdu7ciYODA0ajkeLiYi5fvozRaGTo0KEolUpUKhVPP/00ffv2xcHBgcuXL9OwYUO8vb1Zs2aNNal06dKlmM1mVCoVX331FQAWswmFoxsevaeAuTI2pnRwA4WSLt4+bA0IZKy7BzLgpfR0HBVy2np507VrV65du0ZhYSHvvfce27dvJyAggMDAQPz9/au4iN5//33Ky8tp1CgUV9eb1Kgho2tXB557zg4721u4uztSXFyMMENNeSvaBPXlnTenkpWZhYeHBzVq1MDLywutVku1atVYuXIlL7zwAmlpaVy8eJEaNWqg0Wh47rnnsLOzo7i4mPfee4/evXvj6OjIpEmT6N+/v7VxZ0xMDO7u7iiVSvr378/+/fsfyGdk4sSJJCUlERERwZgxY4iNjaVBgwbUq1eP9et/u8TsiRMniIyMJCkp6W+fp9Lbu8rzXjodn7WOYcjLL1NRsx3B9Rsxrl8fxrRvSZiPBzV19rRo1JCNh7dxtSKNPRnHeXXWG7z51lsMGDCAiooK3NzcsLOzq9Lr7Y8oEe87z/95T4lHguxPZiN+aZQsFrI/mY2ue/cqw0OHDmX//v306dOHtWvXPnJtLkrPZFOwNhFhrMy9wiIo2pxMk+Aoq7jj0KFDGI0VeJbkUcPJkbySMizCwhfT30cIQd++fVGr1aSkpGCxWFi/fj1CCLp06QJQpfpEYWEhDg4ONGjQAKiUKB88eJDU1FRMJhONGzfm/PnzqFQqfH19MRgMGI1G5s2bR0pKirV6xvnz58FsQlMtBFFRhlxjj7m0EOf2I7m99j3S3d144soVzGYLcuApnRPd3N0ZUZDPqdOncHaurG5vMBjw9/evPN5P3G3ERo4cyciRIzl0qCWGcjMAJ06UseDLPORysLcXzH7vB05uTWHp1o8xmsrR2boz/sm5nM3fyMETcaxdu5batWtbc+LmzJlDkyZNGD9+PHq9nmHDhtGzZ09atGhBZGQkb775JoWFhTz33HP07NmTRYsW/W3v/+9l2rRpnD9/nrNnz2IymSgrK8PR0ZHbt2/TpEkTevTocU+KBMDhw4cZPXo069evp3r16n/7PD3GvELmW29b///fqF6dE40aYv7p/7WQKyj3rgFALY9iNp07yDNt/fh650k2bNhAv3798PX15ZtvvmH79u3I5XKrQvXrr79m8+bNuLq6Wg3YlClTSE1NJTk5mdTUVF555RVeeuml/3ee0kpM4je526Xwy/GYmJgqeTkAY8eOJTIykmeeeQaLxfIgpviPoWh7ys8G7CeE0cJIp7bs/egj0lNTmfLqq6jlcj7ftZ9CvYG0/EKEgN0XrmCjUrJ+/XomT57M+vXrMZvN7Nq1C7lcztatW4HKJGaZTEbHjh0pKSkBwMnJiTZt2lBSUsLJkycJDQ1Fr9czadIkGjZsiJeXF76+vpSXV9a38/Dw4JlnnuHGjRsUFhby9ttvI1coUMnlqD0CULnXAGGm+OC3qJQKDl28yNZFi/i4fn0sgNneHpuRz+Pg4YFOp+PWrVvW+f0e7lZnNmxoy8KvqvHlgmrMnedJyVUHqjnXYdxjc5nUeyHDOkxBq9DRJWQkX375JceOHSMzM5PJkycTExNDcXExvXv3ZurUqTz99NNMnz7deuzY2FiOHj2KTqdj/PjxmM1mVqxYwdKlS2nXrh379u3j9u3b1vHWrVv/r2/97+LqsSyWvH6IpW8cpuBWGVePVea/vf7664SHh9OuXTvS09O5evXqPe7GS5cuMWLECDZu3Gg1YPb29kBl2kW3bt3+8vnqunfH+713Ufr4gExGfINIqwGrWbMmFy9epMxQToW7L442Guxt1Kw/cor69eszduxY/P396dWrFxs3biQyMpLo6GjmzJmDSqVi4MCBnDlzhr59+1ZRgv7RuokgrcQk/h8GZWUyTqcjzKZSkeSrUrOhZkClq8Fi5plnnqlSAQGwKtMeNcwF5VWeHxn5PQBC6JjrW412SdeYb6Ml1mSiQU0/ukeEcC07l4sZt6gwWbBVq9i2bRu7d+/GZDIRFBTEkiVL0Gg07Nu3j71793Lr1i2EEBw+fJh69eqRlZVFQEAAsbGxyGQyRowYQf/+/dmwYQOrVq0iMzOTzMxMnnjiCQIDA6levTp9+/alRo0adO3ala1bt7J27Vpq1aqFq7oQpZMWQ4OulF0+wHszZvPdrDdJTEyk07vvEhAQgAAM7dvR4qWXiDx9muDg4CrVQX4PNhpvDOUZ9x0vyat6DzPyrrPu6JfIZHJUK8w0bNiQVatWMXr0aPR6PVqtll27dv3quby9vZk2bRpt2rRBCEHXrl3p2bMn/v7+vPHGG/eM/11cPZZF3PLLmCoqf+RYzIK45ZdZu+kcOTk5nDp1CpVKhb+/Pzk5OfcYMW9vbwwGA2fOnHmgHat13btbPS6rpkyxjnt4eNCyZUsWL16MXC7HT6uiV2QomxKuoFQqqd+mPnGH47BpbYPDKQeCWwWz6MPKVbBOp+PgwYPUq1ePiooKoqOjrUrEOzFgjUZjjZX9soDzL5GMmMRvoqpWjZySErpdT6aBVssZvR5PtYY1U9+DuXOBSln9kCFDqFatGlOnTrXuGxMTw8yZM+8xcv9VFE6aewwZgND/3MBhemoq5UYTR5Nvcjo1A5PZglwmQ61U0K91MxLRkJiYSFZWFq6urjRp0oSPPvqIlJQUkpKSsFgsqFQqXnnlFeRyObVq1bIW+JXL5axZs4b4+HicnZ05fvw4RqMRtVrNwoULycnJoWbNmsycOZNx48bh5uaGXq+nTp06fPHFF3z11VecXTwKZUkJNjY2bFr4ESqVitatW3P16lXOnTuHQqFg3bp1+Pj4sHjx4v/pPgUEjqtSsQRALtcSEDiOay6aKoYsxK8hIX4NsXfRIOpU9gRr2LAhR48erXLMQYMGWRNmgSqJzU8//TRPP/30PfN4/PHHGTFixD3jd4RKfyVH1idRVFzEyWu7aRAYg8FYhqnCwqkDl8jhFiqViri4OG7cuEFMTAwqlYry8nICAgJwcnLizJkzBAcHM3ToUGrWrGlNZP8lJ06cYMSIEaxevZrAwMC/9Bp0Ol2VZqoRERFEREQgqyjHPikBpVrDax/OINm3jCmHp+DbwpfcnbmoaqmID41nc/JmugZ0pVq1atYCwXv37mXKXcZRo9FY//69sTLJnSgB/Hr+l9LVFef+/UmpqOC2yYxeJkMVGMCun/zkJpOJ/v37U7t27SoG7FHEsaM/MlXV/1LCVE75hR8B2BVYiw99fHBXq3mrV0fCq3nhaq8FGRiMJpYcPAHArVu3sLW1Zc2aNXzwwQeMGjUKW1tbCgsL8fLy4uTJkzz99NNkZGSQkZHBxYsXrUHzo0ePIpPJyM/Pp0mTJtjY2BATE0NKSgp6vR5fX19GjhzJiRMn2LdvHwaDgdmzZ9O7d29q1KhBWVkZxcXF7Nixg08//ZSTJ09SUVGBQqGgfv36REVFsWnTJvbt20efPn2Ijo4mNDSUyZMn/+775O3Vk+Dg97HR+AAyhMWT96fa0Knj27y/aihnU/ZyI/sys9aN5sMfhjPjxxeo184TgIyMDDp16kTt2rV57bXXrMdcsWIF1apVw9XVlQkTJlQZr1evHmFhYVXG/z/+6oTnkrxy9BUlHLiwAXsbHQFeYbz//VCu3bzE8ePHqVevHkuXLiU4OBi1Wk1gYCAajYa6devy3nvvoVAoOHDgAEIIzp8/z7Fjx+45x+HDh3n++edZv379X27AoNI9q1KpqozJhAV1TjoObu50GPEidVu2Yc7pORjMld8PdiF2FJ4opKSwhDmn55CXl0dhYSG+vr4A1sT9P4NkxCSsJCYm8sILL3DhwgUUFhteeepD0q8WcPiiLTZaO9pNmkjtJk2o17JlFVm9r68vy5Yte+QToO0iPXB6vDYKp8pfk5byAgxnvsWUfpwsj2gONXmXPa3nYlLrCGnVh9SCYp5pFoWTnS0bVyynzGAgJCQEPz8/pk2bxrPPPkvbtm1JSUkhJCSEatWq8eGHH/Lss8/StWtX1Go1tra27NixAzc3N6tMecCAAZjNZkaNGsWHH37Ivn37CAwMZM+ePQQGBnLixAkaNmyIj48PmzZtIioqCoPBgMlkokGDBkycOJF33nmH8PBwvL29sbGx4dKlS0RGRnL8+HH69+/PhQsX6NChg7Vr8759+4iPj//d98rbqyfNmx8gtu01CgtfpWbNSM6dO8fVpMs8P34Ai+Om8mSzF3j/uaWsXbGB8Fb+AJw9e5ZVq1aRkJDAqlWruHnzJhkZGUyYMIEJEybw1FNPWeXyd8Yfhoz+l9i7aFh/bCG3izL4cPUIHLXOOGiduZp+BoPBQEneLZySN+BTnojBoOda4hVrqsrWrVsxm834+fmRn59P48aNady4MRaLhcjISNLT0+8bM/urubsHHVSuzB574kkmLVzKiHnfULdlGwCySrOs+9j42uDe3Z3rH17nwNgDjB07lilTplirdri5uf3peUnuxEeY0jPZFG1PwVxQTrbIxd+nOhEREVw9loUl357lhz5FIVPw+frKX9ndW/Rj586dyOVyjEYjly9fxtfXlyNHjpCYmMiKFStYuHAhffr0Yc2aNcDPK7VHJX/MLtIDu0gP4Kf0hLh4sjyiuRzUD4ui0rgJ5CSedaRY4cLSav0pUVzkml0gFRUVPPXUU2zdupVmzZpZq9XfGd+yZUuVcbPZbBVU2NraVqlir9VqiYqKokaNGnz22WesXLmSiIgILl68SOPGjblw4QK2trZcuHDB2ovs8ccfZ8aMGdSuXduq4ispKaFz585cv36dlStXEh4ezqJFi5g7dy4HDhxg/vz5mEwmMjMzuXjxIuHh4b/rPm1O3syc03PIKs3CodCBa1uv4TLBhW7duuHk4UTtkJp8tHYUUJlf1rNnTxISEigpKWHbtm0EBARQVFREmzZtkMvlNG/eHAcHB7KyssjOzmbw4MHExMRYZfQrVqzg6tWrDBs2rIorbsWKFXzwwQfWuNjdwpC/kqY9A0nLeI7MvBQmPbkAs8WMRV5BeEcPRo3tiakom49aadGbbPGeVUx1R8Hnn36Mm7cfBoMBmUzGlStXaNSoERaLhcOHD1NeXs769etJTk5+YDGz8PDw//c9tpy0cG3dNWQyGZpqGnSNdMg1cpQWJWlpaTRp0oTk5GSrGhEgICDgHjXi3SrX30IyYo8ov5SDm4sqUBoqx4+sTwIho9RQiJ9bHer4RnI6KY45H1UWobVYLHz37VJyb+fQpoYXhYYS1CoVYWFhwL0J0H369HkkDNgvuRMQP7TOaDVgdxAmQV33MC5eiMNkEUz6fDUyuQIHBwdatmxpleQfOHAApVKJg4NDlYK/O3fuxGw2W4+XmpqKg4MDAOvXr0ej0Vjz9XQ6HWazme+//57U1FQuXbrEM888Q506dark9Gk0Glq1asW6deswmUycP3/eGgMpKirCzs7OWp9x06ZNKBQKLl++XEVi/3vYnLyZKYenWF1ORboifN70wVxm5s0336Rt27ZVtt+2bRs+Pj707t2bkydP0qlTJyIjIwkNDWXq1KlkZmayefNmoHKlNmbMGBITE1m8eDHu7u7WFdm4ceNISUnhxIkTlJWVkZWVxYQJEzh1qjJNoEOHDqxbt+6+dRX9/f05efLk/7RyKD2TjUNcKm20cpYoZPiqZOTaqolLXsEXkw5zIy0LEGSVCKYfKkdvhKQ8M5CP3gRpaWnY2NiQlJTErVu3UKvVjBgxAhsbG6pXr05ycjJOTk4sWrSI9u3bY2dnR0xMzB+e51/BhQsXyN+UT/DrwZhtzZhKTMhkMtynuPNO83fI3JPJRx99xKxZs4BKNWJcXBzFxcUEBQUxcuTIe1yW/x+SO/ER5b5ycAFFPxymJK/SFWirdkCj0tIgMAaVQsPmw8sQQrBqxXdQVoJSLgcEkd6uKISFnp06YLFYpATou9B1745B5XTf13pGPUvFrSTMJbnkHFiOY3Azli9fzpQpU9i+fTu3b9/mvffeo1evXixfvpzJkyezbt06bt++zbZt27C3t7eueH19fSkuLqZp06YUFhZSUFDApUuXGD58OE899RQJCQnWKitubm6YzWZ27959z5xGjhxJaWkp6enpvP322/j5+bF9+3bq16+Pv78/Z86c4c0336R+/fpoNJr/SWJ/d8wEwJhvpEJRQXz1eMaPH8+xY8dIvX6d15/szqy+3TmzaglbNm7k+++/JyMjg9TUVLy9vXFyqryvMTExHDhwgOLiYtq2bcv69euJjY3F19eXzMxMdu/eTevWrdm0aRNt2rShf//+1lXL35nwXFBQwCeTplOwNhFzQTkyQAY0dFKTbNrArvQtiMECg1lgby+n3xo9ay+ZsFXBkyFKLALkcrm1Q/bSpUuRyWR4eHhgNBrv+T/l6enJpk2beOGFF+4bM/s7uZM+MOW5L2jgE8skv2l423mjsleh0+uwLLAwsedEZsyYUaUQ8x014p+p3CEZsUeU+6noAMwWZ+zllQFtufznj4dKqcFgLK1M1jUYEBYL3joHPB0dEEJgslgQBblV9oHKBOguXbrQp0+fR7YBob1L1VXYu/2/w16rw6x1wOPxN1HYu+LZ5120rYdz6tQpYmJiMJlMHDt2jKNHjzJ37lxOnTpFmzZtUCqVHDt2jIULF5KSksKpU6fo0KEDpaWlVK9eHScnJ7744gtrFfs1a9bQvn17Ll68yAcffIBWqyUsLAyLxUKrVq2ASqPWpEkToNINuXDhQnx9fVm7di1XrlzBxsaGunXrolAoaNSoEZ9//jm7d++mbdu2BAcH069fvz8ksb87ZgJgSDOQ9G4SB8Ye4J133mHwYz3oUz+I5XEHmbVtH9/uPsDwppHYK+UcO3aMlStXcvbsWXbs2MHgwYPZv38/w4cPZ9y4cSxevJgLFy7Qtm1b1Go11atXZ8SIEXz33XcUFxdbZfRGo5HXX3+dDRs2oNPprCKOy5cvU69ePTIyMnj33Xf/5/ccKo3Yl98ssP5YtFfbUlpRhjBaKD9fiNHWiJALEHC7wMLpHAvqnwq7D2+gxiygfv361uocAwcORKVScfPmTT788EPs7OzYu3cvMTExVjVm9erVuXDhgrVH14PgTvrAHVVphd5MwQ4b5votJX5gPJqNGqaMn0JCQgJffvnlPav/O/yvlTskd+Ijyi/l4H46b3YPXYKCbJraLyOt7hMkpBzmlR4fk1ucRamhkD69BpCVn0rG5QvYa9RcycrBaDaz/mwiCrmc2Fr3DyiPHTuWwsJCnnnmGZYvX36Pofuv07RnYJUcIQAjgv02lf9hq438GgBfJy3r5qy7Z38XF5f7ChJcXV3ZsWMHKSkpdOvWDUdHRw4fPkxKSoq1iv0dUlJS6NChA56enveNNdzdwsTNzc3qDtZqtaxcufK+1/W/Suy97LzILP054dmhngMO9RzwtvNmx5M7WPDCYHwc7XipXaVhLNQbsFFAOx93nh3xJW+//TYWi4WePXvyww8/kJaWxqRJk5gwYQJ5eXlWYw2VbtCzZ8/Svn17MjIySElJYfHixdjb27N161a6dOlCdnY2mzZtYvHixVy+fJmLFy8SHR3NmTNnftW9+HuYOHEiKblpdPxmCE2rR3I5J4mS8jICZ8YS7RuGXqXnxsc3KjeWQZkBbpYLLAJ6/1DpDTly5Ig1neCFF16wFhBwcXGhYcOGvPDCC4wbN45PP/30b5HV/x6OrE+yfraDfCJZsONt2oY/yZH1SbjVVv/lasRfIhmxRxTHjv5VSyQBMgw4KpfgrTxARMAt1A5ZBPUeTvoNe3Sb7QiuH0DW3lR83d2wwULK7Xy+PniS1kEBDGnREAfXynjB3aWG7vCoJkAD1GnsBVT+Zy/JK0dhr2S7zMBlxc8xLa1KwfiOQf/T8e8u9WRvb8+xY8dISUmhQYMGGI1Gpk6dam0WeUdoc/r0aWt+ma2t7Z+8wj/Gyw1erhITA7BR2PByg5cpLS3lkx+3UFimxyIE7UNqU1xezpb4ywhkyBYuJywsDKPRyIYNG7C1tcVkMuHm5kZAQAB5eXls3ryZDz74gMTERHx9fQkKCmL69OkMGTKEOnXqWGNJYWFh6PV6jEYjY8eOJSYmhpo1a+Lu7o5MJuOJJ55g//79/7MRmzZtGmf3nmD7oK8xWUzojeU4aOzIKyug67IRBHwcQNrnaVRkVaD2UmPKN+Ess+Bip+JmESxcOJ9evXrRpEkTqwhHq9XSsGFDAMrLyzEajcyaNYstW7Y8kFJU9+PuvD5vF386RvZnzoaxyGVy4jJaWtWIzs7OtG3bluvXr/+l55eM2CPKHQXdHXWiQp6Lo/xr7JT7yHRXY1snl0VtfBg7NoMnn9Th5GzkuecjmTJlCuNHPU/CwX0422kxFVu4mHGLdvXq0rLvsw/5qv651GnsZTVmAF5n0pmx/QoZBXp8nLSM7xhEr0jfP3WOO2WIXnzxRQIDAzEajbz22mu8+uqrViFIYmIier2ehg0bYmtry/z58+/7o+PvpGtAVwCrOtHLzouXG7xM14CurFmzBjcnHcNaVn5R6yuMfLLzAMO7dcSjfjSGCiMXL14kNzeX27dvU6dOHbp168ZHH31EbGwsHTp0oEmTJpw6dYoNGzZY44h3Skp9//33WCwWRo0axbp167Czs8NsNvPUU0/RqVMna4wxJSXlL6mzKLdXIVPJEQaYvn8Bx26eQyaTkVmSg12hjvpd6rP/xH6e7/M8i5csRmNrj8rJiV7twnnhhReYO3cu6enp99RSPH/+PCdOnMDf358dO3Y80Coev8T+FwnqTYI60iSoI/YuGgZ+ULmavl81lLuTnOH3qxF/iWTEHmHuloMT/z1sPA5GSK5ph0XxywKkFpKTZuLt1RPPgFqU6fVc37EdRxsNdar5suVGNq80bfnAr+HfSq9I3z9ktKZMmYK9vf19DU7hxo1kfzIbS1kZyGQsGjaMj48cIS4uztpU8/bt2yQnJ+Ph4cGNGzcYMmQIarWagwcPPnAjBpWG7I4xAyD+e0qnhfDZF1c5etPCmeQbNAusgYeDHaXlFaw5cgrFiXM8++yzGAwGawJ2cnIyy5cvR6FQsG3bNtasWcOtW7eIiIhAqVRSp04d9u/fz4EDBzCZTIwZM4YuXbpgNBrZvn07QgjKyspo1qwZjRo14qWXXuL27ds4Ozvz4osvsmDBAvz9/asoQf8/MrPWk5w0k5QbqRgstzF2TWb9rERyywrY/vK3nLI9S/+Jr9DjSF0qPKqxX+xHhgyZkFFUVER2djYpKSlUVFSwbds2IiIiqhR+vnTpEhcuXKB+/fpYLJYHXorql9zPXa5Uy2na88G4Nh+t4ITEr5Li2Ii6ixQM3yan/6gMJryWSXn5XTEco6Bjx0NERkYyY8YMtu0/iGdgHTwDajFt0RJaxLajdu3aj6SU/q9GCPG7iidfPZbF4jG7iBw+kX3Vn0fIFCAE/Z97nm+/+cbaRcBkMlFeXm5NXIbKhOiEhIT7Vkt/4MR/DxtfYtvpFGq7yLk1zp632rqRdDuPtacv4KZz5PmRI3n22WdRqVQUFxdz69YtTCYTdnZ2lJWVERoayowZM9ixYwcymYw33niDoqIibt++TUlJCUuXLsXd3Z2vvvqKS5cuERUVxQ8//EBmZiYqlYrOnTtXqbNYv3591Go1nTt3/kOXkpm1nsuX38BQnoFWK6O0tILrhukUh2fg3y6U0lbl/LjmG/QVRipcvEBeqeRQq9XWAgFt27a1xvTi4uLuUex5e3ujVqsB2Lx5M5MmTaoS03zQ1GnsRZv+wVYBk72Lhjb9g6t4Hv5OpJWYhJXE1FusWHuSQaWjef2NMxzYXwqAk5MCv+pqMjLK+Oabb4iIiKBPnz706NGDr776CpPJRGJiIsOGDZOM2O/k448/5uuvKwUdw4YNo1evXnTs2JHGjRtz6tQptmzZwrJly1iyZAkeHh74+fkRFRUFQFJSEoMGDCM1KR2VQkP/1uMpt3GlQlQq3c6WlVLD1pZz587h7u5uVd7JZDJSU1Otyb7p6ek8/vjjAEyaNIlGjRrx2GOPPdD7ULhxI9nvTMZU4oSDXMOWG2k42pTTK8iMXXNH3oqzQ+3oxO3bt9mxYwf5+fnk5+cjl8uxWCyYTCZKS0s5ffo0rq6uaLVaLBYLn332GXq9nosXL5KWloa7uzu5ubmMHj2arKwsCgsLqVatGkajEaVSSatWrSguLsZgMODp6UnXrl1ZtWqVdZ5nzpz5XTliyUkzrTUhdToFoWEahgy+SkhIDnl5AbT7bhkGQ6XrbfkPq61xrCVLlmA0GjEajWRnZ+Pi4kJubi4LFy6kevXqpKWlAWAwGEhISGDw4MHMnTuXS5cusWnTJjp37szXX3/9QFWJd/NLd/kD5ddaPj+MR1RU1B9uoS3xv/Pj6TTR7MPdwn/CJtHgteXC289fCCFERuY6MXyEhxg02FmE17cRAQFq8XhvnfCp7mLdd9q0aeK9994TrVu3FuHh4WLq1KkP6zL+dZw8eVKEhYWJkpISUVxcLEJCQsTp06eFTCYTR44cEUIIMXz4cOHl5SVKS0tFYWGhCAwMFDNmzBBCCNG2bVsx7bkVYu5zu8W4XnOFTCYXNdyDBSAA4SiXC3+VWtSsWVMAQqlUCjs7OwEIrVYrnnrqKaHT6US9evVEaWmpmDhxooiKihKHDx9+oPehYMMGcal+hLgYFGx9zPb1Fj62cmGnQvg4yER4eLg4fvy4qF69utBqtcLFxUX4+fmJGjVqCI1GIxYuXCi0Wq2oW7eumDJlimjUqJFwd3cX3t7eQi6XC5VKJebPny+0Wq1wdnYWgYGBQi6XC3t7e7F69WqhVquFXC4X+/btE9WqVRMtWrQQq1evFm3atBHu7u4iJydH1KhRQ+Tk5Pyua9q1O1Ds2h1wn0egEEKImU91E693bSM8He3FlEkTxVtvvSVCQkKEi4uL0Gg0QqFQCC8vLzFo0CAhk8nE9evXRVxcnFAoFOLQoUNCJpOJtm3biri4ONG2bVsREhIijh49+ne+Tf8IgJPiV+yG5E58RFl3Jp1JaxNIL9AjgFtFBnINgnVn0vH26onFvj6lRkBAYF0bDsRXkGsqZXNyZWUEKaH5j3EnGXTe83v4eOJSWjWqrKxgb2/PY489xr59+6hRo4Y1Xys1NZWwsDBsbW1xdHSkR48eQGUZqMOHD/PZ6jf5cPUIVhz4BITg5R6fADJs1Q74qVSkmyq7PwcEBGAymXBxcSEoKAhvb2/27NlDkyZNMJlM2NrakpCQgEKhoGnTpg/0nvyy4er/sXfe4VFVWx9+z/T03hMIBEjvofcmvYhUpSNIUVDEglwQAUUFQbGLihQRBBGkK6F3AoSQQBIgCSU9Ib1N298fkZEI3qt+StF5n2ceJnv2nDbDrLP2Wuu38vQ62lvasN+/IesGWhDuZUVpYTE3Pk9gZbc3sFJaUq9+A3Jzc6mqqsLHx4fmzZuj0Wj46KOPmDBhApcvX6Zbt24sWLAAd3d3QkND8fDwQK/XExERwaJFi3B1dWXUqFEcOXIElUpFo0aNKCoqomPHjowePZojR47wxBNPmPqv/RE0ao/fHC/ZuhUL/S+xNVV+JsJQuyRaUVFBTU0NBoOB1q1bc+3aNSRJok+fPhw+fJj69eszYcIELCwsTEXqFhYWJCUlIZPJ7lm35wcR83Liv5RFu1Oo0tUNVgshWLQ7hf6RXhzMz6C4TEFFjYyKGBeQStHtKWTpqaV1A/KYOzr/L1JP5DBrxjyOJNWqWrjYeeGY50qD+o1o274VO3fuZMqUKVRWVtKkSRNcXV2prKzEyckJqF0+3LRpE3q9nk2bNmGlsWTfkyt4ZfsbyOUqvr2ZxvNf9AIE1boKLiPQA4sWLaJDhw60bt0aS0tLoqOjTTJWS5cuZdCgQUBt4e+9VniAOxuuptbUsDgvH0kCmzwDS56bT1GiLXO3LqFaX4OHhTMD3Trxdno2klJBXl4emzZtMr3fw8ODQYMGsWHDBvbu3Yubm5tJ2cPb25uTJ08ybdo0nJyciIyMrKMc8VfxW21mPEo6kT17Dk00CrIdrQBQld7kdOoVKsvLGT9+PBs3bkQmk/Hmm2/i6uqKp6cn+/btM6ml1NTU3BErvdfdnh9EzJ7Yv5Ss4qr/Ol6mLasz7tDeAUkpEbc07q5JB//mjs7/i28+2c6RC7uY8egHzHj0AzIL00jKOMHVa2mMGTMGNzc3GjduTElJCfHx8ezYsYP8/HwSExOpqqpi3LhxCCF49tlneWPibLRVNcQm7wMgvywflUyJwIilyprnwzuQu2UL3t7ejBkzhvLycgoKCkhJSeH48eOMHz8epVLJ6tWrTTVj165dY+DAgVRWVt7T66LwqOu1tLGyZnODBmwNacCpHV/TSNeWKLcAfhj5CasGLWLT8A/o1rAlszo8RY2VOxYWFuzevdskP3XL2xoyZAivv/46rVq1Mm27YcOGWFtbs2PHDkpLS1m1ahVt27YlIiKCoqIi/Pz82L9/v2n8m2++qaMm8Xv5dZsZjdqTgIDXMSw+jKiuxqu4nMjcErQ6fa3OW3ERNRoXvtdHUlQjyMrKMnV7rqyspEuXLhQUFGBhYcH27dvRarWmJI6/S7l+0aJFLFu2DIDnnnvOpGW5d+9ennjiCX788UdatmxJVFQUgwYNMnUYv1+Yb5n/pXjaW5B5myFT2LnhOe4jPO1rs6JCB4bWUVUACHw/EA8rD2Qymbmg+XdwK/X9fHUA4b6tUStrr23Txp3JupnOzfI8nnnmGZ588klSUlKwtbU1FR4PGjSIixcvEhoaSnp6Ora2tixZsgQHrHG0sGNdwnbisy6iUagwCiNKmQI3dzcS3JRETpvG9evXASgrK0Or1eLq6sqJEydo2LAhjz/+OHFxcaSkpDBy5Ei+/vprxo4de89rxlyfe5bs2XPqLClKGg2ur86DsD4Y1h4yjSfnp/H6/o+o1FaTV1GIzsYJD3dHysrKkMvl9O7dG19fX6qqqujSpcsd+9JoNIwfP55hw4YhhODChQvMnj2bXr16MXny5LuOX758+U+dl4d7Pzzc69ZFXcyeaXoeWFFNK4WKT7bsIURjwTVrI8b1czBUlqNy8uH1j1ZRlZ+PhYUF8fHxuLu7YzQacXNzQ6PRMGXKFKZMmfK3Kde3bduWd955h6lTpxIXF2cqqj506BBhYWEsWLCAPXv2YGVlxVtvvcWSJUuYM2fOX7b/P4rZiP1LeaGbPzM3na+zpHi7asR/U1Uw878p2brV9AOtsL3zLrmJZwSF5VmmAs93332XiRMn1pnTuXNn1q9fj7+/P9k/L73dePmXH/bntr9BF79WPL9zIUajoKK6nD37D6Np1BwLFwNV+deorq7G1taW/Px8UzZe27ZtSUtLQ6PR0LFjbQ+o4cOHs2zZsntqxG6p/OctfRd9djYKDw9cn3vWNH67NFqHhs3o0LAZADkYmWZv4MjLne6+4Z+5vdPzLVX3+fPn33Xu3To/32rL8ld0elZ4eKDPyjL9vehno5NrYc/Zbv/BUFVK9lfT8HjyYw4k7aC3rysVFRWmFPtb0lOSJJGUlMT+/fv/cuX6W+1xskuyuXLwChvObUCtVhMVFUVcXByHDh2ib9++XLhwwaSVqdVq73ks9deYlxP/pfSP9GLhgFC87C2QqNXtWzgg1FSA26thL+a2mouHlQcSEh5WHsxtNfeOeNjtdOjQgbi4uHt0Bg82tyctdNTnkpB+GK2umhpdFefSD9OkXjiWtirT/FstUKqqqigrK2Pr1q0A2Nra0qBBAzZs2ACAzE7FhbxfPIRybSX2GlskCSK9PbEM7ohz7+nYdq41iCt/2Evbtm0RQrBv3z58fHz44YcfTNJFP/zwA99/X9t5+n7UjNn16UPjvbEEXrxA472xJgMGtdJohl8V3Vch+FKm/dMSXfcL1+eeRfq5Ru8W1XIlXwXV1qHJLWxRewWR9cVkCjJSiYuLq9Pt+bf4q5Trb7XHya7IJv/HfAySgecXP8/1guvExsayb98+zp8/z6ZNmwgJCcHCwgKZTEZwcDDvvffen97vX4FUm734YBATEyPMP4IPLx06dGDx4sXExMTc70O571wMDKqNefzM+1pLthTlIyQ5HaL68uSU4UydPaaO1M7rr79uqgurV68eUVFRzJgxg/T0dLp27UplZSW2Ght6erZkeptxTNw8h92XDjOr40TeOvApL/fqwBunriG01ajcG1ORuAeFlR2PP9aPEydO4O3tTWxsLN26dWPp0qUEBQXRqlUrFi9ezBdffEFgYCDPP//8fasZuxsVZ/PI2XoFZaWOPATfWhhp1bfJ/1ui635wa3n5ltf5sV9XNjmF3jHPy97if3qZfxWLFi1CrVazzXMbZ5afofp6NW6PuXHt42sYygyo7FTIK+TodDo0Gg2TJk3ivffe4+TJk4SFhTF//nzy8/NNMbS/C0mSTgsh7vrDYl5ONPOHycjIoEePHrRp04ajR4/i5eXFli1bTK8bjUbGjh2Lt7c3CxYsuI9Hev/49fLRM6pKnnGzQuHpSeM9tXeuA88NZPHixaYlvFmzZqHT6bCysuL55583qf03aNCA4cOHm2Sntk6bT4WuhI/7vUalvpQj2bvQqBQYLR3xGFG73CuMBipSDoHahtTUVJRKJT4+PgQEBPDJJ58A4O/vT3p6OmPHjiUoKIhJkyYBcP78eVNK//3GKtIVv5+l0eoBD/PtkV2fPnU8zXZnM9n5X5b07wW34l85vXPgkhyHCnemZL3HzNKB6HV6dOU6xo4ey9atW6mqquL8+fOo1WpatGhhqtNq2/b+ys2ZjZiZP8WlS5f45ptvWL58OYMHDzYJp97KeAsJCflXq3f8OmlBCAEaDa7PPcvcuXM5ffo0x48fRyaToVAoTIodGo2GnJwcBg8efIdiR2RkJACqlg3oP30cZdXVZBWVYmuhQSmX88amrRil7cgsbNHUD0emtkIpl2jYsCFnzpyhrKyM06dPm5JHkpOT73rsOp3uvsc5/g3c8ib/aiHo/0VCQgKxsbGUlJRgbW3NsWPHaN+0N9XaE3i5NiCzMA2FpCCgQWuSbpxgx44dlJSUUFFRgU6nw93dHS8vL/bu3cvevXv54osv+PHHH3n11VepqanBz8+PFStWmASp/27My4lmfje3vvxXr17l66+/JjY2lrCwMN566y10Oh179uyhqKiIwYMH/ysN2K+lpDpZW9Nh0iT0RiM1QjBt8GCM3t4sXboUIQQ2NjaMHz+eVatWoVQqycrKwsbGBicnJxYvXszMmTMpLy+npqaGoqIiU4Ddw8ODyKBALiae51p+IY08XHl+7mJe+GAl2tICdDezQCbHPqwjhQfXcvjwYVq3bm3yuO6H4K+ZB4OEhAS2bt2KTqczja1evZqwhq2pvinwcmpIXvEN9pxbj0ymwMvFl4OndhMWFoZCoSA+Ph5vb28CAgKIi4tj7ty51NTUcPLkSXbu3GnKWKypqflLMxbNy4lm/t/8+ssvSZIp+UAul5vES2+pdzz//PMmsdl/OkuWLOHDDz8kMzOT1157jd69exMVFUWzZs0o0utZu3Yt3+87yeIv3kdIMjAaaeAfgkxfRWFhIcXFxUiShFKpxMHBgT59+jBmzBgAFAoF9vb2dO3albi4OC5fWUdeXg4HSnPx8LBAFIBkY0/CueMY0k9hlKkxVhYT/eLXjG/lzYL0Q6ZMsvuRgWjmwSI2NraOAQPw8fFh38kfGNF2Fp6ODdh07BM8HHxJz7tI3s0seiz6kUKtHMrKGDNtJq6uruTk5ODs7Ex5eTljx47l3LlzeHt74+joSHZ2tkkJ5l54ZH97dqIkSRmSJJ2XJClekiSzm/WQcrcvv06nM0ng3GLcuHH07NmTwYMH/6lW4w8TFw/tY9agPiyaN5cINweGP/Yoq1atoiQhB61Wi/5qObYaa8qzVPyweTMqn2DU3kEgSeTWKAhq0Ylt27YREhKC0Wis7TOVnc3GjRvRarWmIlK9Xs/evXvJyrrBG29MRaGQePppZz79zB21SmLipEfIz89nQP++fP35h9Tz8mDD+BgeCXa/I+PwgVCtN3PfKCkpuWOsXr16lJWX0cAtCFtLRxRyJeEN2/L+hJ94rNOLJH09H5lSjVGvY39cEig1fPjhhzg4OKBQKOjZsyf16tXj5ZdfxsvLi/z8fAoKCoiJiWHJkiV/+zndqxT7jkKIiN9yB808+Nzty/9b4/909Y7tadsZ+U53fvhoEUmX0wjxcsNQVcX18/F08I9mz1dbsVVbYa+xRSEpyNixn8DGrZAkOZJMjqSyoDr/KidTsygsLCQhIaFWyFQmw9ramunTp5uaOPbs2ZPXX38dOzs7goM1KJUGLCwkrqTd0vUTnDyxCVdXV2QyGYmJiVy9WtvyXq/X11GtX7t2LW3atAFqVetvpdab+fdgZ2d3x1jDhg1Z+PpbWFnVymG9OnQVncMGoZfghl8II0a9z9GxXzGt2VCssq9QlH+TiIgIcnJy8Pb2Rq1Wc+nSJcrLy7lw4QItW7YkMDCQlStXmr6Lfyd/e0xMkqQMIEYIUfC/5ppjYg8uS5cuvavBsrOz47nnnrsPR3R/uFVP0/snJ6yrFRxKTadCqyXEy531J8/Ry78bbpYuLDr0Oe/0fJnnd7zJjDZjuVFdytqUfYCEsboclUcTqjPOgjCiUChwdXXFYDBw8+ZNQkNDSU1NRaFQUF5ejiRJ2NnZobEooWdPGy5cqObqVR0F+XqqqgTOznK8vEK4evUqarWampoa6tevT3R0NEeOHCEmJobTp08TFBTE6tWrsbS0pHfv3syaNcucwPEv4/awwJEjR1AoFLRp04bk5GQuXUynvm04OXmZaEUl53NSqSrJIsDZl5sVRVTqqqjQVqGUK5HkElq9Dnt7e1QqFTdv3sTLy4vCwkLq1avHggUL/tIM1/8WE7sXnpgAfpQk6bQkSRN+/aIkSRMkSYqTJCnu9u6lZh4sOnfujFKprDOmVCrp3LnzfTqi+8N7Z96j2lCNVXVtM8MGLo4kZebiamNNRD1PVp3ewpr4LdhpbIj2CsFOY8PX57ayLm4j1hE9sGgYjbG6DG1eGnK1BT4+PvTr14/8/HxT9ld6ejqNGjViwoTa/y7u7u44OjpSUlx7w2lpIUMug0WLPXB1VfD66yE4OjpSVFRE9+7dad26NadPn+aLL74gOTmZNWvWcPHiRb777jtTZqI5A/HfSVhYGH369MHOzo769euTnZ1Nnz59yMjIwCjT8sy8YRSIS/Qf2ZmqshysFGq2jfiUSl01RiGI9AwmzM0fpaTA2dmZ4cOHo1Ao8PPzY+DAgdja2rJp0yb69u1LRUUFqampf/s53Qsj1kYIEQX0AKZIktTu9heFEJ8JIWKEEDEuLi734HDM/Blu//JDrQfWp08fwsLC7vOR3VtyKnIAqNDU1vZ4O9gR4+vNe3sOc+ZqJs+3G8M3Q5fiYGGHj50HZ57ezOGn1jGu1UjKz+1Gm5uGZWB7LDybMGDEkxw4cIDy8nIcHBywsrLC3t6emzdvEh4eTl5eHjNnzuTatWukpKQwbFhPJEnJM1OdyczU8dyzWRiN8N6yEpMsFdTqLsrl8v96Hrt37/77LpKZB5aSrVuxePY5un+2nOmXr1Cck4Ovry9qXTEtNVdIf28Il1KSObR2BUJfQ4W2gg6fD6dcW4mFQkNKfhoJOcmUVpeTn5/P6tWryc3NpaCggIKCAr766iuGDRtGWFgYLVu2/M0yjr+Svz07UQiR+fO/eZIkfQ80Aw7+3fs189cTFhb2rzNav8bdyp3simxO+xfR+rwTCqOM9v4Nae/fEIVKTc9ew7BMUhI7bqXpPZJSxtDpMziX/Dg+moMM8t9O7LZ0qiqz0Fh0ZNeuXcyePRsXFxc+++wzAL766iveffddbt68adqOlZUf1jaOqJTJ2Nhc54cfWtLQb8YdYrO3YhtmzNzO7XqeAFJODh4VFXw8biCtLNJxtLJmfYoVQkBBWSWuNtbYqpyoZ+dJUWUJg0K7szftBOU1FSzoP4Nd2ji++eabO/Zz6tSpe3pef6snJkmSlSRJNreeA48Aif/9XWbMPLhMi5qGRq4h3auSI6GFlGv0CAQKe2semfA0jYd2wH5AY+T2tW085PZq7Ac0pl3/ADaOLuOpiI3Yq28SFqrhwIFc4uNnciVtPd9///0dygd301O0tQnmkUeO4e/fjKysqXi490MIwblz5+7H5TDzEPHrJqQAUWo1723ZRzsfgV7VmCNXruNgZcHFnDzKa7R4O9iwP+0EjZzq09w7nEsF6TRw8qH9qB4cPnyYn376CeCeLR3ejb/bE3MDvv85rVcBrBVC7Pqb92nGzN/GLQHk9868R4ZXDtVN7JgW9UwdYWSrSFesfpZKup20K4tNzRIbN1HTrZs1kyddRpLGMnXq6zg4ONSZHxUVxZAhQwgPD8fV1dUk2gvw9ddfM2nSJBYsWIBOp2Po0KGEh4f/Hads5h/Cr5uQAkRbWPJZYSEtvS04d8UWhUxOIxcnTqRfZ3TraH6Iv4hBGLlRloNeGHC1cSatMpO+zw5FqVQyefJkLCxqWwwtWLCAJk2a3OvTMit2mPl3MHfuXKytrVGpVHz88cdERUXx9ddf33XusmXL/uccX19f4uLicHZ2/s19/npO7N5G1OY5/RqJzp3u7F31ewSVzaLLZn4vlzp1rqPneQuFNTTuncVnl5pSpr9ToMDG2YUJH64wiQVPnTqV5557jnPnztWRnho1atTfJj11v7MTzZi5pwghfrM+7aOPPuKnn376TeN0a87OnTv/65w/g0bt8YfGzZj5K7lbOxhJo8F1zABQWtDWNQOFZKjzukKlpu3QkUCtWPChQ7X97OLi4igvL79rs8wzZ87cs0JnMMtOmXlI+bVO4S0B3ebNm3P69Gl27Nhxh4BuRkYGaWlpdO7cGUmSyMnJQafT0ahRI1avXs2SJUs4duwYqampNGrUCC8vLxo2bEhhYSFKpRKVSkVpaSlRUVHk5eXRrl071Go1xcXFtGrVilOnTlFSUoKjoyOtW7emuLjY1EIlNzeXjh3DOHPmBiUlerp0sSYxsYbsbD0uLjVoNCE0a9aMyspKzp07R35+PiqVilGjRqHT6QgKCuLGjRtUVlYik8kwGAwEBASY5L4AJk2axKlTp6iqqmLgwIHmTttm6vBfm5AmhBMYOw+4xKGCRpRpFdg4uxA9qAk3ZXOI3ZuNQu7GyZNXKC0tfbCaZd6S038QHtHR0cKMmf9FXFycCAkJEeXl5aKsrEwEBQWJM2fOCEmSxLFjx+rMqaioECUlJcLPz08sWrRI1K9fX7Rp00YMHz5czJ07Vxw/flxERESI8PBwMWrUKFG/fn1hZ2cn8vPzRf/+/YWbm5sQQohHH33U9Lx3794CEPn5+eLrr78WgDh8+LB45plnRGRkpFi0aJHYtm2baU56eroAxJtvvimysjcLR0elCA3TiMOH24iPP35edO7cWQghRFRUlOl5dHS0kCRJnDp1Smzfvl20a9dOCCHEokWLhLu7uzh37pw4d+6ckMvl4tSpU0IIIQoLC4UQQuj1etG+fXtx7ty5e/ehmPnHkZW9WezdFyz2xDY0PSIjrcS8+U+K2bNniw0bNojXX39d1K9fX/zwww9i6NChf9uxAHHiN+yGeTnRzMNBwrewNATm2nN4QW8ebeWPlZUV1tbWDBgwgEOHDlG/fn1atGgBwKFDh3j00UexvLwN2y9a4VKdzv4VryFqKjh16hQbN25k3bp1PPXUU1RXV1NYWIhWq0UIYUpRz8jIQJIkSktLSUtLMz2/ceOGqV6uVatWyOVyQkNDOXjwINOnT+fw4cP06tULKysrunXrRvfu3ZHL5Qgh8HDvh49PEHa2MbRufQilMpDDhw8TGhrKhQsXaNCgAQDW1tY0btwYgOjoaC5cuEBUVBTz58+nqqqKCxcu3FHy8O233xIVFUVkZCRJSUlcuHDhXn5CZv5h3J6IdIuQUCUfvL+adu3a0bZtWz755BMiIyNp0aIFR44c4fLl2tjuvcxWNC8nmnnwSfgWtk4F3c//oaqKIHVX7XjYYNO0W8ZH3IqJ5V6ArZ/98j5dFVRVYmdljWe9hnz33Xc0bNgQABcPF/Ze28vNyptIRomdl3f+6cOVJInq6moqKyv58ssvsbOzIzw8nOrb0pvlcjnV1dW8/PLLuLi4cP78eQICAtBqtXW2A3Djxg2Ki4tJTU1lzJgxVFVV1dkWQHp6OosXL+bUqVM4ODgwevToO+aYMfNHqK65M5sxNNSCtV8X07JlS6ysrNBoNLRt2xYXFxdToXNNTa2u573KVjR7YmYefGLn/WKIgLb1FSyPqyTokZEEBQXx+eef4+fnx6VLlxg5ciQhISH4+/vz2drNNF6SR8svKkjMr030kBC4qqu5ceMGbdq0oW3btgx/bjgV6gpuXrqJMAq0RVqmz5qOhWtt6rCtra3J2Nna2uLj42PSkTx+/DgGg4GkpCTatWvHu+++S5s2bdi6dStCCBwdHamoqKC0tJTr168DkJubS0REhMnIyGQyysvLKS4u5vz580Dtneytu9qysjKTfmJkZCQHD9ZqBSQmJpKQkABAaWkpVlZW2NnZkZuby86df94ImzEDd084ioqy4MCBNqYbxtTUVKZPnw5Ap06dOHXqFAkJCSQkJNyz7uBmT8zMg0/JjTp/3qoKuVleQ1HRFWSy2nsxrVbL5MmTadGiBadPnwZhRIifuyrfhoWkx8/Pj4sXL3LixAkOHz5Mg5cbcOPLG8it5OhL9ZQll5GoSSQ6IJqwsDCUSiW+vr4EBwcTHR2NhYWFKbHD19eXDz/8kJMnT1JSUsLnn39OmzZtsLW1pV27djg5OSGXyzlw4ACBgYHo9XoGDhyIvb09I0aM4P3336dbt2507dqVs2fPEhgYSEFBAQEBAQCEhISgUqkICAjAy8sLJycnZs6cScuWLYmOjgYgPDycyMhIAgIC8PHxMQXYzZj5szT0m0Fy8qw6S4oymQUN/R6sfnTmOjEzDz5LQ6DkOgn4E0sbdh+/gLGqmEsXEzl0JplPP/0UFxcXli5dSnp6OkCtZNOON5jXqnZpY/ruajxtJCbGqHBZVI5/8C+xpHOJ5wh4N4CcdTlYBVrh0Ka26FhCImFUwn89tIyMDHr37k1i4m8L0WRkZBAcHMyBAwfM9VxmHiqyc7aQdmUx1TXZaNQed5U5uxeYOzubebjpPIeEzcvYamyPjlol/cRL18i7WUXHjh3x9fUlNzeX7OxsWrRoYdIfrPJoQf9vfyStUEd+pWBYiAKjXIOtrYSrqyuZmZm0bNmSlMwUAAzVBnI35lKeWE711Wrs6tlROagSS0tL5s2bx9atW6mqqqJVq1Z8+umnSJLE0KFDycnJoVmzZhQXF/PFF1/Qtm1bqqqqGDNmDOfOnaNevXp1vEFzKryZhwUP9373xWj9EcwxsQeUjIwMQkJC/tJtxsXFMXXq1L90m/eEsMHEqrubDFj9+vUxGAxYW9swdOhQzpw5Q6tWrWjUqBFvvPEGI0eOpF27dqzYfpSQNj058mwTZBKsuyCwHfQBMpUlDg4OJCUl0b9/f6oLqlHLa7UO9cV6qq5XEfp2KGE+YXz00UcAPP3005w6dYpt27bx7bffsm3bNgA0Gg2jRo3i5MmTvPvuuyaD9PHHH2NpacnFixdZtGhRnYSN119/nbi4OBISEjhw4IAprmXGjJk/jtmI/cPQ6/W/+VpMTAzLli27h0fz11FS9ct5eXh4EBERQVlZGR9//DEWFhY8/vjjQG1wubCwkEaNGqFSqfj6p7P02O5E+95DqZSsKPXtjoODA9nZ2YSHh/Piiy9iYWHBi01fRKPQILeW0/7d9sxtNZeZk2Zy+PBhAPbt20fz5s3p3r07FRUVJCUlmY5nwIABQG0qfEZGBgAHDx5k+PDhwJ3q/+ZUeDNm/jrMRuwB4re8r7S0NCIjIzlx4gTdu3cnOjqatm3bmnr1jB49mokTJ9K8eXNefPFFTp48ScuWLYmMjCQsLIwRI0YAsH//fnr37g1QZ06rVq1ISUkxHUPbtm2JiooiKiqKo0ePmt7boUMHBg4cSEBAAE888cQdCRN/J7dERm/RsmVLbGxsmDx5Mi4uLnh7e98Rl3JzcyM2NpbDhw+zdu1abGxsAFCpVHz11VecO3eOCxcuoNVqecT3EU6uO4mXoxdHRx+lvWt7nn/+efbu3YtarWbcuHFs3LiRXbt2YWdnx9dff01gYCBJSUmm6yCXy//rTQT8kgofGxtLQkICvXr1MqfCmzHz/8BsxB5wUlJSeOyxx/jqq6945ZVXeP/99zl9+jSLFy9m8uTJABiNRm7cuMHRo0dZsmQJAQEBHDp0iLNnz7JkyRIqKyvv2O7tc+bNm8crr7wCgKurKz/99BNnzpxh/fr1dZYfz549y7vvvsuFCxdIS0vjyJEj9+Yi/A/atm1r0jncv38/zs7O2Nra/uZ4u3btWLt2LQA7d+7EYPhFL+7atWsYDAY0Gg2hoaE888wz1K9fn/LycpycnKioqKCwsJCmTZty8eJFFAoFGzZsuOOYbt+HORXezF/NXyWsGx8fz44dO/6Sbd0vzIkdDxBpZ/Mpzq3kw4l7Scw+RFJSEt26dcPb25tRo0Zx/vx5+vbti1qt5tq1awghaN68OeXl5QwcOJA2bdpQXV2NTCbD0dGR7OxsKisrycvLA2DXrl389NNPREZG1pmj1+u5du0aUVFRGAwGPDw8yMzMpKqqivT0dAYOHMjJkyextLTEy8sLSZKIiIggIyODNm3a3JNrc7tG4O1UV1czd+5cxo4dS1hYGJaWlqxcWduQ8rfGX331VYYNG0ZwcDCtWrUCCfp/258zS84gU8qorq6u7XarViOXy02Gzc7ODmtraywtLalXrx5Q6+3Fx8ffcVyTJk1izJgxBAYGEhgYaE6FN/NAEh8fT1xcHD179rzfh/Ln+S09qvvx+DdrJ6YczxYLRn0jPBx8xewhXwlnGy8hl8mFna29mDdvnigpKRFOTk6iY8eOQgghRo0aJXr16iX0er0YNWqUWLlypdDpdEKn04muXbuKsLAwIYQQ33zzjbCwsBBCCLFt2zbRs2dPIYSoM+fChQuiXr16Qgghnn76aeHm5iYMBoPYs2ePAMT169dFbGyssLe3F4cOHRJCCDFlyhSxYsWKe3Z9lixZIl599dU7HkuWLPl/bXfblW1CppaJ4C+Chd8CP6H2UgtJJQk7JzsxaNAgkZqaKgDh5uYm0tPTxaBBg4S9vb3p/bGxsaJ///7/39MzY+Y36devn4iKihJBQUHi008/FUIIYWVlJZ599lkRFBQkOnXqJPLy8oQQQpw9e1Y0b95chIaGiv79+4ubN28KIYRo3769SWMzPz9f1K9fX9TU1AgfHx/h7OwswsPDxbp16+7PCf4OMGsnPthsT9vOlm+OYdAZKa0s5I0NT9IjZjgSMioqKnjjjTdwdnamvLyco0ePEh0dzc6dO2nYsCFyuZzDhw/z+eef4+rqiru7O0eOHOHixYtERkYydepU9Ho9U6dO5eTJk+zZs4eQkJA6c7p06UJ+fj6hoaF8/fXX5ObmEhMTY0qW8Pb2JiEhgaqqKqZPn05AQAA//vjjPY2Jde7cGaVSWWdMqVTSuXPnP76x23QY39v/Uu2YgIIdBWhztQidoLS4FGtra5RKJa6uruTm5gLQv39/iouLOXbsGABr1641eaMzZ87k+++///MnacbMXfjyyy85ffo0cXFxLFu2jMLCQioqKoiJiSEpKYn27dubsmJHjhzJW2+9RUJCAqGhof+1fEOlUjFv3jyGDBlCfHw8Q4YMuVen9JdiNmL3me1p25l7dC4W1bYAqFWWyGUKMgvTEIBaaUnjxo1p3Lgx9erVo2nTpuj1eioqKli/fr1pO0lJScyZM4eCggI+/fRTUz+tsLAwDAYDy5YtY+fOnahUKhITE+vMcXR0RKfTce7cOfbs2QOAwWCo00kYahUxhg4dyoULFygtLeXSpUv34AphOo8+ffqYhHft7Ozo06dPnay/30OrCP9aHcaS64Ag5+f/AcXHijFUGFC5qpCpZSjsFJw9e5Zu3bpRUVFhUs+A2njEhx9+SGBgIEVFRUyaNAmA8+fP4+7u/pecr5l/H78V51q2bBnh4eG0aNGC69evc+nSJWQymcnoDB8+nMOHD1NSUkJxcTF2dnbs2LGDUaNGmSTK/smYY2L3mffOvEe1oZpyVRE16gIkpR5btTXJeUeRy+VUVJfSqVMnEhMT2bt3Lzdu3KB+/fqoVCrKy8uZOHEiJSUlJkHZli1bcuXKFYQQVFRUUFJSgkqlonfv3pSWliKTye6Yc/PmTWxtbWnfvr0pRby8vJzk5GRkMhkdOnRACIFcLmfDhg0899xzPProowQFBd3Ta/XrVPU/w9FRSij5Jb7mrjeQBBiqDMit5GjztEgqCd1NHd+frvWqGjRowJdffomvry8LFixg7ty5PP/888ycOZNmzZphaWkJgE6nu3c9lMz8Y0lISCA2NpaSkhLy8/M5evQox44dw9LSkg4dOtw1m/WWWDT8Eufy9/c3jSkUCtNN6z8tG9bsid1ncipyAEjx2EeFTToCgSRJtG3fBq2uGlcXV5YvX87Vq1eRyWQMHz4cDw8P+vbtS9OmTblx4wY9e/Zk2rRpLF++nKqqKjw8PLC0tCQ0NJQ33ngDo9FIbm4uRUVF1NTU3DFnw4YNlJWVUVxcTL169VAoFNjb25u8tbNnz/L000/j5eVFfn4+R44c+V3p5A8i1i8lUa4VdF5VQdSn5aTOvowwCOxb2lN1tQptthaMoFAqmDRpElVVVfj7+/8uz2v37t3367TMPGT079+f6OhogoODTQozACNGjKBr1668//77pkzY4uJiWrRogUwm49ChQ5SVlWE0GgkLCyMuLo61a9cSHR1NeHg49vb2vPTSS6xfv57WrVubvp++vr61eqLAe++9Z0qUsrGxoays7N5fgL8QsxG7z7hb1X7JnPWW2DvYMWLECCorK9m9ezeOjo5MeXoyVVVVaLVa1Go1P/74I9nZ2axYsQJbW1sGDRqEJEn4+/uzf/9+GjRoQHp6ukmRYtOmTchkMtzc3Exf7F/POXDgAO7u7kRERJCSkoKLiwsXL17k2rVrADRr1gwXFxdCQkLo2rWryVt7KJFkaBTw/RBLzjxlzY+PqZGMAm8LGQ2ere3ltfCLhei0Ojw8PNi+fTvJycmsWbOGixcv8t1335k9LzP/b34rziWEYNKkSdSvX58DBw7QqFEjsrKyuH79OjKZDB8fH1auXImVlRWlpaUMGTKEvXv3MmNGrSjvqlWrsLa2RghBixYtTCUmM2bM4OOPPyYyMpLk5GSTEevYsSMXLlwgIiKiTnjiYcJsxO4z06KmoZGUWBosTWNqtRobGxt0Oh1VVVW4uLjg6urKG2+8QUBAACqViuDgYK5fv25qiQAwe/ZsOnbsyO7du5HL5Vy5coXY2FjTjy7Utk64fU5aWhp6vZ7S0lLc3Nw4cuQITk5OVFVVkZycjEajQa1Wm97/sHlgCQkJLF26lLlz57J06VKMMiVCYcErsdUEfVhO6CcVyIHVDaaxa+AufHx8eGHwC8AvsYbfwux5mfkjpJ7IYeUrR/hw4l7G9ptBQOPgO+Jct1r+XLt2jbNnz/LJJ58gk8mYNm0aGo2Gjh07sn37dpo3b069evVYv349S5YsYfjw4WRlZfHaa68xffp0hg4dSnFxMVeuXAHA2dmZ0tJSTpw4wblz51Cr1URERPDTTz9x6tQpc2KHmT9Pr/IK5hYUUiP/pSBZLpczfPhwrK2t+fjjj/Hz8+PFF19k2rRp7Ny5k1WrVnHhwgXCw8MB+Oqrrxg4cCAlJSV4eXnRsmVLRo0ahZeXF4WFhXz33XcAFBYWotPp2LVrl2mOp6cnCxYsYMyYMXh4eODv72/qD/TWW2+ZinE7dOhg0gsE+OCDDxg9evQ9ukp/joSEBLZu3Wrq/VVSUoLBKHjzZg/ydRacm2hNzaJgPN1dqG74CFA3tnC3v838wl+l7/nQanr+AVJP5LDv62TKb9aQmhXP+cunmNLlHTZ8tpvIyEg6duwI1ParA+jSpQvOzs4UFxej1WpRKBRUVFQQGhpKYGAg7du35/r16xiNRkaOHMlLL72Ep6cnLi4ufP7553fs//z581RVVf1jMhJvx2zE7jex8+hVWszjhgMo0ZmGVSoVAwYMoLq6miFDhvDFF18QHh5OcHAwW7ZsueumXnzxRWbOnElERMQd3lJubi6zZs1i3rx5XL58mcjIyDpzJk+ezMqVKwkPDyc5ObmOh/ewEhsbi06nqzMmhODM1RJc24xEOb+EfRHvczUr3/T6tWvXzOnz95iHWdPz93JsyxX02p8TK7QVWKqskQkV3325l+PHjwO1yjvfnryKXsj48ccfa2+6DAaEENTU1CCTySguLqZ9+/YMHz6cyspKDh8+THFxMZmZmQA0atSItLS0O+JciYmJvykY8LBjNmL3m58bPoaRQh9+or69jMmTJ2FHKb169aJRo0ZMmzaNjz76CJlMxooVK0y1YleuXDHdCY8ePZqVK1fi4OBAp06d6Nu3Lx4eHkRGRvLUU0/h5eVFcnIyTZo0oUmTJpw9e7bOnFGjRrFhwwbOnTvHpEmTiIyMJCoqiunTp5skqfbv309iYiLbtm27L/qJf5RbHtjtSJJE48aNiYuLIzQ0lFWrVtVJn/+9SRxmatHr9TzxxBMEBgYycOBAKisrmT59OhYWFoSEhDBhwgTTd+TUqVOEhYURERHBCy+8YPru3q7peTv/JA+t/GaN6fnR5B1czk5g2vJufPjdHFq0aIHBKACJUxs/4vV5r3L9+g3Gj5+AQqFAJpOxYMECjEYjhw8fZvny5fTr14/y8nJeeeUVbty4wauvvgrUhhTKy8vZsGEDx48fp0uXLmi1WhYuXEhFRQURERGcOHEC+GdIToHZiN1/7LxNT8NI4Tm+YC7v8pzdjwQGBgK/Tz8RuKt+4muvvUZlZSXFxcV37PqfpJ94N27VlN2isrISCwsLPDw8OHbsGOfPn2fFihVcvHgRX19ffH19zUkcf5CUlBQmT57MxYsXsbW15aOPPmLkyJH4+fmZ7v5vLUOPGTOGTz/9lPj4eORy+f/c9j/JQ7N2/CWuPKLDSyweu5VFY34Aych3332HTluD4yOTqf/iD1j4t8EgBJtoblot2bdvH4ApxpWSkkKrVq2orq7G2dkZnU5HRkYGNjY2+Pn5UVlZycCBA5HL5ahUKjp06ICVlRXx8fF06dKFsrKyf4wRM9eJ3W86z6ktvtXVuvolGRbknbdFX2kg59vHybtxg379+rFp0ybq1avH0aNHGTRokOntNTW/3OENGjTI9ONQUlLCqFGjuHTpEvb29uTk5Nyx69vnSJJkWnrT6XQ8/fTTph+b1NRU03uaNWuGt3et4b3X+ol/lM6dO7N161Z0Oh1lZWV89dVXtGnT5k+pfJiTOGqpOJtH6e4MDMU15IlCvN29TPqPw4cPZ9myZVhZWXHlyhXs7e2pqKggPj6e6Ohobty4wdSpU01lC7dITk7m4MGDRERE0LVrV3bu3EliYiL79+9n8eLFdWKxDysto/LZF6tBL1TsT/yehPTDIEFJdUGtaIAkQ19eQNaXT2PUa8FoIHtF7c2jhYUFo0ePRqlUkpmZidFoRAhBWloanTt35rXXXmPKlCk0adKEoqIidDodjz76KBMnTuTbb7/FxsaGiIgIAAoKCnj22Wdxc3Pjm2++QaPRcPjwYWbOnPnQxsfMntj9Jmww9FkGdj6UZFiQHeeAvkIOAgx5uVhWVeNpYcHhw4cxGo3Y29sTHx9vely8eNG0qbtlKiYmJrJ169a7Fjj+1pylS5fi5ubGuXPniIuLq9PQ8WHKVLxd5cPGxob//Oc/vPHGG//vgul/KxVn8yjedAlDce2Nk6FUi6jUU3E2zzRHkiRmz55NdXU127dvZ9asWSgUCr744gtsbGw4deoUiYmJ1NTUmOI2b731FqGhob/bQ3voSPiWJikT6WjzAdez95OSeZo5j73KO4ufRVlfweObHweM1Nw4ivuIxbgNnoek1OD4SO0qS1FRERcvXsTLywulUokQwtQ9oX379uzZswelUolGo8Hb25tHHnnEVP/p6elJdHQ0nTp1YuTIkUDt/9u4uDg+++wzRo4c+dAneJg9sQeBsMEQNpi8Tp0R+qw6LykRLLW1Y8rP9R8NGjRgw4YNDBo0CCEECQkJpizF27mVqQi12Yt347fmlJSU4O3tjUwmY+XKlXValTxs/BUqH2ZqKd2dgdAZ64xlluay97OtNHq8IXOenoiDSkF1RTluLi6Eh4fz1FNPER0dzenTp5EkieDgYGQyGRkZGVhZWVFcXExlZSUODg4APP744/8Iz6sOsfNAV0UTy0O0tNhLkq0Or8Yv8VK2NcWpxdhhBwKggCDpIEX7dpKtq6Zn8SFWUqtd6ubmxvXr12nQoAFpaWlYW1uzfv16pkyZwuLFi1GpVBQXF9O4cWOuX7/Om2++Sfv27VEoan/iLS0tH+76zv+C2RN7gNBnZ991PP/GDQoLC1m6dOkfzlT8dRYi1N4tx8XFoVAo7jrnn5ipaOb/zy0P7Hb8HOux4uB6OvTuR2lpGa386hPu7U5+QQHtWrYw6W9KkkR5eTlGoxG5XE5QUBAqlepen8L94efkLYDujRTojYIhS25ybUMuFn61zV4ltYSs1MCepe+TlJFBPSd7bhbUhgD0Wi0pKSlIkmRK4Fi+fDkff/wxgYGBaLVaU/z39ddfJyEhgaCgIMrLy4FaySk7OzuOHj1Ku3btqKiouJdn/7dj9sQeIBQeHuizfvHEvJQqfmjQkBwHB+RlpZw6dQqAadOm3fHeX3tbLVu25MKFC6Y7sQULFgC1tWKOjo7ExMTc0czx1pzGjRubmjhC7XIP1NaKdejQwTT+wQcf/MkzNfMwIrdX1zFkPnYe7B+/hkpjOVuvfmga7+DfkKOXr9K1SX3eWrGCJ598kjZt2nD48GHOnDmDwWDAz88PX19f7O3tcXFxYfbs2QCsW7funp/X346d98+C06BWSOx8woowXx/EbTWIwZ8GM3CvF9bVv/wkJ2bmIEkSbwzrR7tnXiIiIgJvb2+MRiN6vZ7KykoWLFhAbm4u77//PuHh4eh0OiorK5k7dy4lJSUsXbqUJ598kqNHj9KtWzcaNWrEu+++C/wzJKfA7Ik9ULg+9yySRlNnTNJocBo31vR3WloakZGRnDhxgu7duxMdHU3btm1JTk4GalPtJ06cSPPmzXnxxRc5efIkLVu2NDVifOGFF3jqqafqpDXfPqdVq1akpKQAtcWsbdu2JSoqiqioKI4ePQrUpkR36NCBgQMHPhSp9mb+Gmy7+SIpf/nJuF6STZcvR3GucN8dc11srNhz+lydUoW2bdtib2+Pq6srKpWKdu3aAbWSSL169SIiIoKKioo7skofejrPAaVFnSF3g/GOaRv2xPPCt9tZtOsAx69cI8DdBRA8u/xrmjVrRk1NDcXFxVhZWbFt2zYsLS158803uXLlCkVFRaxcudJUlnPgwAGCgoLw9fVl6tSprFu3jk8++YSFCxdSUlLCjh07/hGSU4C5KeaDRvEPP4jUjp3EhYBAkdqxkyj+4QeRnp4ugoODRXJysoiIiBDx8fGiU6dOIjU1VQghxPHjx+/aLFMIIUpKSoROpxNCCPHTTz+JAQMGCCGE2Ldvn+jVq9d/nVNRUSGqqqqEEEKkpqaKW5/Pvn37hK2trbh+/bowGAyiRYsWpmaZZv7ZlJ/JFVkLT4jrLx0UJ178XgT6+YtPJ48Wiwf3uuPx6eTRv2ubZWVlpucLFy4UU6dO/bsO//5xbr0QS4KFeNVOiCXBYtu+/4iY1TEi5KsQ0+OloZ2ESiEXCwd0F2621uK1fl0FIMZ2aSuEEEKlUokpU6YIIYQIDQ0V+/fvF0IIMXv2bDFt2jQhxC/NL1esWCHGjh0r6tevL4QQYsWKFab33v78YYH/0hTzvhuu2x9mI/YLWdmbxeHDbcSeWD+xYUOMcHa2E/7+/iIpKUmUlZUJjUYjwsPDTY+AgAAhRK0R++qrr0zbuXbtmujfv78IDg4WISEhwt/fXwhR14j91pzi4mIxfPhwERISIsLDw00dovft2ye6dOli2sfEiRPF6tWr78l1MfPgkJ6eLvz9/UWvLp2Fq62NCPV2F28M6C4eCW4ifBztRaMGvmL8+PHCaDQKIYQ4efKkCA0NFeHh4WLGjBkiODhYCFH7I2xjYyOCg4NFz549TV2K/+m0bdlSBLr7icZOvuI/vSaL4b36CQmElUolAKGUywQg5kx/Vpw9e1bIZDIhl8uFtbW1UKvVIi8vT7Rv314sWbJEWFpaiqCgICGTycSuXbuEj4+PUCgUQi6Xi3Xr1olly5YJGxubh6qb8+38NyNmXk58AMnO2UJy8iyqa7IAQY02F42mCjc3jTnV3swDRUpKCjPnvsb+bVuwsbZmz8XLJGTmsHvzJi6lpf+uYudOnTrRrl07EhMT2b59Oy4uLv8otY67UXE2j49azebHUSvYPmo5Xx3aRNLZSwhgSLvm1HeyJ6ZRbVeF73b/xOOPP47RaMTLy4tJkyYhhDB1bX7vvffw9PTkwIEDSJLEihUrmDdvHk5OTri4uNRJnzdrJ5q5J6RdWYzRWFfnTKEQzJ5tyapVq9i2bZsp1R5qvelz587ddVv/31R7Dw8PZDIZq1evfqhT7c38dWw+m0nrN/fS5q29qOxcybf0JbBtR+Z/+ClaJ3cKyisZPeNlQkND2bt3L0lJSRQXF1NUVMT48eOB2lT6P8vDauBKtm7lUqfOXAwMonDVCb48/i2PfDmGfqsnUlBVgM7uGjIZtB9Rzo3iUjKrasUH0tLSKCwsRJIkevfuzVNPPQXUFuDr9XoKCgpM8W21Wm2KXd+ulWhtbW1qivlPw2zEHkCqa+6eai/J8ti2bdtfmmr/3+aYU+3N/JrNZzOZuek8mcW1P5AGIZi56Tybz9YK0EqShF6vx8PDA71ej52dHWVlZbz11lvk5+dz+fLlO/QUx44dy8GDB+voKZaXl5OWlnbXY3gY5ahKtm4le/ac2uxjITiek8bhjNNsGfExG14ZT2N/BTqhw2iEt966gFoNzz3XF5lMhoODAwUFBQgh+M9//gOAl5cXWVlZxMfHo9frmTNnDtXV1VhbW5Ofn8+rr75aJ9kqJiYGnU5XRzvxn4I5xf4BRKP2+HkpsRZ3dyWff+GDRu2Bvb39H061v1026tep9v9tjjnV3syvWbQ7hSrdLx65oTSf4oxEFu1WEXh5LWFhYXW6D6enp3P8+HGaNWuGTCbDYDDwzTffsG7dOiRJYvDgwRQUFFBZWcmHH35ITU0NQgji4+M5evQoYWFhyGSyh16OKm/pu4jblvJLy3Kw09hgodRwTLOG5JQaFr7pzoznswkK0mBrKycp6QeGDRvGxo0b8fX1JS0tjQMHDpCcnIy7uzu9evWiqqqKXbt2kZiYyOnTp6mpqWHixIm0adOGl19+2dRWKTY2Fg8PD+Lj4/nuu+/44Ycf7tel+Mv52z0xSZK6S5KUIknSZUmSXv679/dPoKHfDGSyuim5MpkFDf1m/CXb/+GHH5g1a5ZpWcKMmd9LVvGvlrkdvSk7s51Ti0ZRVFRE//79gVo9RGdnZ1xdXTl//jz+/v74+Pig1+upqKjAy8sLlUrFtWvXmD17Nj169GDo0KEAJuNUWlr6hwSDH2R+LWTQsuI6eoOOjsuH8+nqNAKDamPMGo2EXi84d66K1auvU1hYSIsWLZg5cyYymYynn36aJUuWoFAomDNnDjNmzMDCwoLu3buzcOFCtFotc+bMoWPHjqjVaqZPn06DBg0oKCgw7fsfk1r/M9LtLudfvnFJkgOpQFfgBnAKGCaEuHC3+TExMSIuLu5vO56HieycLaRdWUx1TTYatQcN/Wbg4d7vfh+WmX85rd/ca1pKvEWH66cZl/IDTuUVxEoVTEu5ztbL2+jVsBfPP/88y5cvR6vVYmNjQ3FxMRYWFri6utKoUSN2797Nxo0befvtt8nOzub69essXLjQtHR2Kw6bkJDA448//tB6Ypc6da4jZACg8GqGOmwg6V3mo7covOM9GrUnrVsfuleH+EAjSdJpIUTM3V77uz2xZsBlIUSaEEILrAPMv8S/Aw/3frRufYjOnS7TuvWhB9aAzZ8/H39/f9q0acOwYcNYvHgx8fHxtGjRgrCwMB599FGKiopITk6mWbNmpvdlZGQQGhp6H4/czJ/hhW7+WCh/8Yo6XD/NtPhvcS6vQAIcKkAI+OrVZ1m5ZyVbt25Fo9FQU1NDYGAgRqORiooKioqKeOKJJ5AkifHjx7Nx40YeeeQR5HL5XTNoH3buJmRgKEzArhM0iZ71t668/NP5u42YF3D9tr9v/DxmQpKkCZIkxUmSFJefn4+Zh4dTp07x3Xffce7cOXbu3MktL3rkyJG89dZbJCQkEBoaymuvvUZAQABarZb09HQA1q9f/49I7/230T/Si4UDQvGyt0ACxiXvQnOXrNWrW7KY/vh08vLyCAsLQ6lUEh8fj9FoRKFQEBoayoEDB3BwcKC0tLYB7JkzZ0wZdBYWFkiSZEpCeNjlqOz69MFj/jwUnp4gSSg8PfGYP6923L0fAQGvo1F7AhIatScBAa8/sDeuDxr3PbFDCPEZ8BnULife58Mx8zuYP2UIazZsRqfTYalR8cGs8XQZ8Tzl5eUsWrSIsrIywsLCSE5OZsuWLaZ4RteuXWnZsiU5OTmsX7/+H7Ee/2+kf6QX/SNr70UvbnmhzmvuSiUNVCpcUZAuZLh5utG8eXO0Wi3nzp1DoVBgZWVFUlISTk5OrFmzhlGjRpGSkoJarUbzs7cSERFB8+bNGT9+PDKZjPbt2z/0clR2ffpg16fPXV/zcO9nNlp/kr/biGUCPrf97f3zmJmHlFNr3+C7TZs4N8GCd4/DoiM1cGEzIwcfpG3btgQHB7NkyRJee+013n33XXQ6nSltX5IkZDIZqampSJJE48aN7/PZmPn/cjfR6u0NGpKg1nIip5TBgwcDtQXNnTp1Amo7bj/33HMArFy5Ek9PT1xdXdHr9bRr1465c+cCtXVQycnJrFq1Cg8PD2JiakMiv86ONfPv5u82YqeAxpIkNaDWeA0F/nyVo5n7wuazmSzanUJWcRVj4pfQr4kMjUKicwMlS45pKanQUpSXRWxsLPv370er1fLll1/i7e2NtbU1Fy9eJCwsjGvXruHp6cmMGTPIy/ulkWJGRgZ9+vTh/Pnz9/EszfwZXJ97lhv/mYWsRmcaq1bAzlZq1LvUKJVKU8dwAKVSWaeztua2OJGfnx+vv/666e/t27ezcOFC9Ho99evX/81ifTP/bv7WmJgQQg88DewGLgLfCiGS/s59mvlrub24VQA2otz0WlMvOX4OMj6K05FbpsdgMDB37lxWrlxJeXk577zzDhcvXsTV1ZWNGzdiYWGBk5MTW7duxcbGxhwf+wdg16cP3gteR+dqjwDybeHb/o4EdRhNflY+R48e5eOPP+bbb7/FwsKCzMxMxo0bR0hICBMmTGDw4MHEx8ezYsUKrl+/TteuXU1Fz0OGDGHSpElUVFSY5KjMmPk1f3udmBBihxCiiRDCTwjx+v9+h5kHiV8Xtzb2cWJrqp5qvaBcK8iq0NA14nEsNI7oawQeVo1MNSgTJ07EwcEBR0dH5s+fz8iRI6moqEAIwfDhw00xMbMRe7ix69OHsIPHCEq+SLuTF1m44AitvFqh1Wp55ZVXyM3NpUWLFsjlchYsWMCpU6dITEz8XbqK/v7+BAcH369TM/MQ8LfWif1RzHViDx4NXt6OAIqPfEPFhf24WEo41GSTV27Exdqa1MIqk7yNUq7CztqJakMp/gH+pKWlUVpaakqbXr58OdOmTcPPzw93d3dyc3N58803GThwoKnbrHlp8SEn4VuIncfh8xl0XFnB4O6tOZNWiIuLC7a2tlhbW7Nt2zb0ej2SJGFra0tKSgoBAQG4uroik8mIiIhg3bp1VFdX8+6777Jw4UJyc3N/c5dxcXGsWrXqoZOiMvP7uZ91YmYecjztLajJTqUy9SieY95HMXAJN3Q2TG7jSKneDkcbd1549AOaN3kEpULFzMeWo5HbEB8fz/PPP4+joyNKpZJFixYxa9YsevbsSUJCAi1atCAnJ4e1a9diZ2dnXlp8QPD19a2j7vCHSPgWtk5l/7l0zufq0RthstcFLq5/DSsrKzIyMoiNjeXixYtUV1fj6+tLTU1tp+ibN2+aPDGZ7I/9LD2MWopm/jrMRszMb1JxNo9VWgsm3EhjVKN2dFVYIlNbUi23ZtkZGdnF2bTw74aPSxMeiXwcmSTnjY1PUlZZAhaQJbJwcHBArVabUu8/++wzAEaNGoVCoWDNmjU88cQT5qXFe4gQ4m9RNE/dtJ2Vme/yduJjbLxa27VZZqiB2HkmsWmtVsuAAQMIDg7m0qVL6HQ6xo8fj06nY9asWVRWVnLu3Dm0Wi0hISGsX7++jlhwWFgYERERdcSCb+9Sbubfh9mImbkrFWfzKN50CXWlHgmwRuIlLBhsYUm37j149pkp2Ns40S3yl2RTG0sHXh26Clcnb3SlOpZ/sJwbWTcIDAzk0UcfxWAw0L59ex599FFKSkqws7OjadOmTJw4kW+//Za9e/dy8eJFc+r930BGRgb+/v6MHDmSkJAQxo0bR0xMDMHBwbz66qt3zK+qqqJHjx58+umnNG7cmFtCBEajkUaNGpGfn8/WrVtp3rw5kZGRtGraji1pj3C1xMjhC9s4ceUkADNiLQh84yJlZWU0bNgQg8FAXl4e9vb2NGnShMrKSsaNG4e7uzvx8fE0atSIBg0aoNFoSExMRKfTmfrY/VbczMy/G7MRM3NXSndnIHS1d+sx3qHsuXIESa9lAhLJJ/djZWWFi6sz6fmJAJy89BM6vZa560ZwLS8VZKD2UlNtrObixYts2bIFb29vPvzwQ0JDQ5k8eTLdunVDq9Uik8mQy+XMmDGDdu3a3c/Tvid06NCBPxL77dmzJ8XFxf/v/V66dInJkyeTlJTEO++8Q1xcHAkJCRw4cKBOt4Ly8nL69OnDsGHDeOqppxg+fDhff/01AHv27CE8PBwXFxfatGnD8ePHOXv2LAEurdh1bjNONu60CepD68BeAHSKmsrFVwLx9/enffv2WFgoWLHCi5mvZFNWlo6NjSXdu3fHwcGB+fPn06xZM5KTk9FqtYSGhnL16lXc3d0pLi6mrKyMli1bAv+/fmRm/lncd8UOMw8mhuIa0/MIj0C6NmrNI1+OwdnKkdBmodjZ2bFuw9eMHj6OomNZeNar5pOv7Ni6U8vpmw5cuVhNaVwpcls5Li4u5OTksHPnTiZOnEhRURG5ubns3LkTFxcX0xLiCy+8wN69e+/ZOWZkZNC7d28SExP/9DZ+rxitr68vcXFxODs7/+F97Nix408d28VD+zi0bhVlhQXUKNV4urnRokULAL799ls+++wz9Ho92dnZXLhwgbCwMAD69evHiy++yBNPPAHA2LFj6devH88++yxffvklY8aMAeDGjRsMGTKE7Oxs8q4X42TjUWf/bvY+7D5/kI3xRQRFFtH/UR/OxisZPvwUjo5yGjaUc/p0Ndk5W3j88cd57rnnTIK/ISEhnDt3jtGjR5uaPJoxczfMnpiZuyK3V9f5+6lmQzk4YS3fjH+fq1evEh0dTUREBDv3zOGrtU4seMMRW1s5GoWeEI2Bvi95EvJVCJ6tPCkuLkalUhEREcHx48fZsWMHDRs2xMHBgSFDhvDtt9/St29foqKi6Nix4992ThkZGaY4yp/lt+IvFRUV9OrVi/DwcEJCQnBxceGll16iadOmhISEcP36dc6ePVvnPUajkdGjR/Of//zHpEIRHR1NcHAwoaGhJu/rVrLF7bGs/+WdXTy0jx8/+4CygnwQgoqimxiqKrl4aB/p6eksXryY2NhYEhIS6NWrVx3R3datW7Nr1y5TLMrHxwc3Nzf27t3LyZMn6dGjBwDPPPMMTz/9NOfPn2dMrxfRGbSmbViqbZg95Cum9HuJi2mZfPfdd+Rkv8+YMbasXlOP95Z5MeEpJyoqjPywZQ7PPvssI0eO5O2338bS0pI2bdpQXl5OXFwcjz/+OPb29tjY2PxjtBTN/HWYjZiZu2LbzRdJ+cvX46Vdi+n21Vh6rBjHY489RlRUFABzZk9l5IhUpk3L5PUFuWRn6zh0oIIfJl8m8clEbuy+QXBwMPb29gQGBgKwevVqIiMjCQ0Nxc/PD7lczvz58+9LQoder+eJJ54gMDCQgQMHUllZybx580zG5/YuxJcvX+b555/n4MGDREVFceXKFdN2du3ahVKpRCaTsWXLFtRqNYcPH6aystLk/Rw/fhyo9WAGDx6Mg4MD+fn5piakgYGBpqaQBQUF6PV60tLSyMzMZPLkySZjCLXemb29/W+e16F1q9Bra+qMCWHk0LpVlJaWYmVlhZ2dnckjvp158+bh4ODAlClTTGNPPvkkjzzyCL179zbFo0pKSvDyqtVQvFRyBOnnr4tGaUGNthKFSkbLQaF8//33zJw5k+qabOLjq0hK+sVg+vgo2bAhDX9/fzZu3MgXX3yBm5sb27dvp1u3bjRt2tQ094svvmD8+PFERERQUVHx0GspmvlrMC8nmrkrVpGuQG1szFBcw8cjX8e2m69pHGqzxfbty+Gz5V7o9TBp4g169balrMyAhUZG/QB/jOVGhBCsW7eO1q1bExAQQEBAAOHh4TRo0ADAtJR4K83+7+SW0Tpz5gy+vr6kpKTQpk0brK2tiY2NpVOnTmzfvp05c+Zw+fJl2rZty08//YSDgwMGg4GBAwdy4sQJNm7cyMGDB+nbrz86o2DngWMYy2+iVKuJjo6mrKwMmUxGSUkJK1aswGg08uabb7Jz506uXLmCXC7H3d2dXbt2oVarMRqNnDp1Cm9vby5fvozRaCQ0NJRGjRqh1+s5ceIEs2bNon79+sD/Xp4sK7x7mnxZYQHh4eFERkYSEBCAj48PrVu3rjPHaDTy3nvvMXbsWF588UXefvtt+vbti9FoZNiwYaZ5c+fOZdCgQTg4ONCpUyduZF3D2lFNSP2WrNg7n/RdZwga9DFXrlzB1tYWjdqDc+cSsbCQERyswd1dyYqvfH5336zg4GBT7O7NN980aymaqUUI8cA8oqOjhZkHm+IffhCpHTuJCwGB4pVGjcXo4R5iT2xDsSe2oXjsMTsxcpSDcHaWicOH24iKigoREhIi/P39hRBCvP7662LhwoVCCCEiIyNFamrqnzqGefPmiSZNmojWrVuLoUOHikWLFomzZ8+K5s2bi9DQUNG/f39x8+ZNcfHiRdG0aVPT+w4dOiQAcfiZBkK8aid6BloLmUwShYWFQgghYmNjhbe3t3j55ZdFs2bNhIWFhXBwcBALFy4UeXl5wsPDQ+zbt0/06tVLHDlyRLh51RNqVz8hKdRCprEWVmGPCLceTwsfvyYCEHK5XERERIiQkBABiJ49ewq1Wi2sra3FY489JhQKhejRo4d4++23hUKhEDKZTHTu3FlIkiRkMpkYN26cOHv2rJDL5eLatWsiODhYFBQUCCGEqF+/vsjPz//Na/Tp5NFi8eBedzw+nTy6zrz69euL06dPiyZNmogRI0aIoKAgMXr0aBEdHS2CgoLEnDlzhBBCnDp1SqjVapGfny8qKytF9+7dxSeffCIaNWokvv/+e3HkyBFhMBiEn5+fyMvLEz/88INo1qyZiIiIEO7u7iIxMVGcOPmpcHBQCCcnufDzU4mlSz3E3n3BIit78+/63NetWyfCw8NFcHCw6Nmzp8jLy/tD3xszDy9AnPgNu2H2xMz8bkq2biV79hzEz/ETY0kJykQjMmGLUfpF5LWsTDDhqRsY9FH07t3blKwxZMgQBg0axIABA36Xiv38+fNZs2YNLi4u+Pj4EB0djYeHB2+++aYppnbixAl8fHxo1aoVO3fupH379kydOpXGjRtTUFCAVqvlxMnP0OtW88EH57GylGjoXwb5gkfq6dh5UbDvq9d5e/1h8vPzKSgo4P333+fUqVN07dqVJ598kurqajQaDTKZjKpLRSQei2dMv+G80e0FJn03EyGMCG01KvdGGNQ2FBdVAmAwGEg6f4FA/1rZpBMnTpiWEPfv349MJmP//v3ExsZiMBgQQuDm5oanpyeZmbXNHr766isMBgPdu3fnxo0bpKSksGDBgv9Z59V26Eh+/OwD9Nra5UkBqNQa2g4dedf5ly5dYuXKlbRo0YKbN2/i6OiIwWCgc+fOTJs2jc2bN2Nvb095eTlDhw5l5MiRjBw5kpycHD755BO6dOlCeXn5HZmLkiTx+eef8+WXX/LOO+8wZuxeqqtOMOAx+R/uWD5kyBBzDaGZOzAbMTO/m7yl75oMGECkhQVzL+YwdqMvZQOUHD9+nX79vPDzs+OTj1fRtm1b5s6dS/v27QH+UPzr9oabOp0O/+AwjpXYcePgQhz92zD/iy858/2nZGZmYmVlhV6vp169ekCtMrpKpQKgR48QPv98NkOHWnP8eCUaCxnJTWyAMqAGATz+4hJ8fP0QQtCqVSsOHTrEsGHDyM3NZdmyZTzzzDPY2NjgaGnHxBefJr/8JgJBdt4VpjQdwgcn1qIz6Cn68WNA4GrnQ6VMAQJ0Bi2JF2uXwORyObm5uUiSZIqzVVdXY2dnh0wm4+bNm6xdu9akWKFSqUw3AMHBwWRkZHDt2jWcnZ1N7W1+CwufBrx74CTuFmrSc3Jp6OlOqZDx2ZmnGThwIM1GNeO9M++RVZ7F5D2TcfF04bXXXqN///7Mnj0bNzc3JEkiKyuL5ORkzp8/j7u7O4GBgbi5ubFq1Sq6detG165dWbBgAefPn6eyspKXX34ZqJu5qNVqTUvHhQUarK170bnTB7/vS2fGzP/AnNhh5nejz86u83eohQU2MhltP4mjT58z5OYK1q8vw97eky5duqBWq3njjTfYuXMnXbp0MWUfrlmzhlatWhEaGkpGRgahoaGmbVaczSP7zZNsf3kVnZxiMFwsJfZyKdWekRSXlmHU16C1cGLmpvN4xHQzeSyWlpYm1Y/t27ebkh4iI5PZv6+YG9e1SBIUFhpITK4hrYEVP6X9YgjkcjkGg4HExERTTOuWMUxNTaWmpgZtaTXCaEQmk1Hf3oPPTq4nvzQPa4UalXsjkMuRJBkt/btjMOoxCD0gIUSt1ySTyUwZhjU1NaYiXl9fX8rKykzHYvx5H/v27SMtLQ2Ampoaqqursba2rpOZ99+yFK/eyGTpyjXklJSx7cgJEpNTSEhIYNPuTby47kWyK2o/z7ziPAoKCgjsHEjXrl0pLCykb9++JCQkEBkZibe3t0lBfsiQIRQUFNCrVy/efvttWrVqRf369enTpw8ODg7MmDEDgBEjRtC5c2fOnz/Pp59+asp+zMjIICMjg/3795tT5838JZiNmJnfjcKjbh3Q+aoqigxGVkdG0qhRI2QyGYMHDyYuLo6xY8fi7+/PsGHD6N69O61ateL69etMmjSJ8PBwtm7dypAhQ+rITN1SCblVoyaqDRRvusTRH1LRG2oNgSRXUnXlJJVVVXwae5GSkhKsrKxwd3fnyy+/JDU1laKiIrp16waAk3MxMrnEmjXFtGhhiY+Pki1bShk2KYur1SG4W6t5oWOtZ2Nra4tarcbHxwdnZ2c8PDwoKipi//79tGjRgsKyYt7pNZMWPuHsGbeK59qMZf35nZRWlaHNuYzSqR6Waht2nV0DgJXGDmuNHTaWjrXnIwRNmjRBo9Gg1+vRaDRIkkR8fDxCCDQaDc7Ozjg4OCCXy+nfv78peSE2NpYWLVogSRJKpdKUIXh7luLms5m0fnMvDV7ezmMfH8XFw7tOXVhUVBSRkZGkXEih9HopwljrDWZ+kYlRZ2Td1nV07doVIQRLly7F39+fffv20aZNG9NnfunSJaqqqnjxxRfZtWsXjRs3Jjg4mLVr1zJw4ED8/f3Jz8/n6tWrLF26lMjISB5//HG0Wi0ZGRmcOXOGgwcPMnToUNasWWPKXDRj5s9iNmJmfjeuzz2LpNGQaW/N3sB6rPJwpsTWhvGXLpGenk5QUBBubm6oVCqTysKcOXM4ePAgo0aNQqlUsn79egYPHszq1avvMGJ3Uwmpqqqid0k1VVdOISk1yK3sUbr4kvXl0ySumImPj09t4fW6dWRlZdG8eXMcHByYM2cOAEU3HcjP07FnTzm9+9iy4isfZr7iymcfhDOo7YtosWXkU8u5cuUKy5YtIzo6mgMHDrBx40ZSU1OZOXMm3bp1w83NDaVCTst6kXw18C0ABoZ04+sh7xDqFYTKpT6i4iavDH6fRh5hKGRKqmrK8XUNJMArEoDCwkIuXbqEUqlEpVJhNBqxsbHBxsYGpVKJq6srwcHBhIWF4eTkxOLFizl16hQymYyamhpOnjzJ9OnTsbCw4OzZszg7O5tqyH7d9y23tJpinYzNZzNNdWErVqyoFdzVQM66HDK/zERXrENfqgcjKEIVKBQKZDIZ1dXV5OXlYTAY+Pzzz3FycsJoNDJw4ECUSiWSJFFUVERFRQU//vgjZWVlbN68GX9/f8rKyigtLaWgoIALFy5QUlJCXFwcTz75JIMGDUKj0VBYWMiGDRuYMmUKhYWF9/7LbOYfg9mImfnd2PXpQ8XYESTWc6XU2Q29nSMBQUG0bt2amJgYLC0t/2vLDAcHB7799luaNWtGcXExQog6CR6/pRIyY8NLqFx8qbmRhLGmgsqUIxgqS1BpLJHJZCxbtoyRI0fi6+tLcXExy5YtM3li9eqPx9JSgW8DJe7uSgCMehV5CY8CcLM8n9U/1B7z2rVrTV6Hs7Mz5eXlbNy4EQCFQoF3PR92px0GoEavpUpXDXIJd183rsUfRqou4/y+t7G3sKd1UG+6RAwhpH5zUrLOms6/Q4cO2Nvb8+STT+Lu7o6joyM1NTW4uLhQWFiITqdj2rRplJWV4eTkhKenJ40bN6Z58+Y4OTmxZ88ebt68yf79++tc21/3fYNaz2/R7hRTXZiNjQ2pqamISkGbli6srnDBU1LwUQNvJAmqjlVx6dIlrKys8PLyoqSkBKVSiUajwcrKCplMxkcffYRKpcLV1ZWqqiqmTJmCtbU1tra2VFdXo1KpWLlyJTY2NjRs2BA/Pz+8vb2JiIggJiaGCxcu8OKLLzJr1ixeeuklOnXqVKebsxkzfxRzYoeZP8TplPMYJAmtixc+qNm2bRs9evRgx44daLVa7Ozs0Gq15Obm4uDgwIIFC2jfvj2rV6/mkUce4fDhw6xcuRInJ6c7Ejzk9uo6huypZkOZ3mYsxQo90Z+Mwmg04DX+Uwq2vUPl5RMo9RWUlNSwfv162rdvz5w5c+jSpQudO3dGq9WSnp6Oq8sjVFVp0Gj0jBlznXre1ozqMpUtO1I5m/YFcpmCT7bMY2PguwQFBfHII4/wwQcfYG9vj1KppHv37qbj+XrTOob0G8SU7+dS396L94bMZeHpL7h0/QqzZs3CQS7nUHYKVdJlbG29uF50DQmB9c9LfwUFBZw4cQJJkli6dClQaxyNRiMlJSU4ODhw9OhRBg4ciFqtprKyksrKSuzs7Lh8+TKSJNG9e3eEEHUUNgCyiqvu+nmpS6+zd+85JEmiadOmqFQqmjfxo2NCEaf0peQZjMy5XgNCwjrTkUpNFZWVlXh7e5Obm4ulpSXFxcUmOajc3FzKysqQJAlPT0/y8/MpLCxEJpNRWlqKu7s76enp6PV6WrduzfTp0xkzZgzx8fFkZGRga2tb5/jWrFnz//g2mjFj9sTM/EFuFdEKpQovLy/8/f3ZsmULVVVVVFRUcOLECdq0acOuXbuorKxky5Yt/Pjjj8THxzNnzhyGDBnCmjVrGDFiBGvWrGHw4MGmbd9VJWTFWPq99xgtqosYrpbRKfs8EaPn0WfoGKZPewaVSmXKfhw1ahQHDx4EYPDgwaZEj5s3S1mw4AvmP7EPQ2FzdvyYQ7vg/jzVfT6udt409gnl7bff5rvvvuPJJ59kyZIl1NTUUFBQwKpVqxg9ejQA+fn5SBZyUq9cIiU/jR2yOMKbRxIdGoJFdgZ5Wi3PdG3Ny371KC7KQCn02EqCYl0NdmqwUQkqKyuRyWSo1WoUCgX9+/dHqVRSUlLChg0bTMuxTZs2Ra1Wo9PpsLe3x8rKCkdHR7Zt24ZMJsPGxqbO5+Jpb1Hnb4WdG63Hz6O16iolJSX079+fESNGYG9vz38ah9FSZcmnxeXY27jh4fyzJ2wAbU0Ner2empoaDAYDSqUSe3t7unXrhlwux8/PD4DKykpCQkKIj4/nhRdeAMDV1ZX8/HzOnz9PVVUVe/bs4amnnkKtrpUwMxqNXL9+naSkJNNxmmNiZv6/mI2YmT+EjVOtQoSkq82sa9WqFc888wwTJkzAycmJJ554gpKSEjZu3MilS5e4efMmSUlJbN682ZS9JoTgjTfeQAiBr6+vadtWka7YD2hs0m38cNAstkS1Z5uXG60sLbHQVfFS0vfsCK6gkav1fz3OW5qMaWlpKJVKhg4dik9TFZey4knLOU9qVjyf7Z5NfmkmV/LOkZSURFlZGZmZmTz6aO1So0ajwdLSEoCLFy8yYcIEZs6cyeTJkwE4ePAgQT6e5F/LwNdag4VKSb7BwCGlhE6CeW7uRFhYoJEkJMBBI2GlqvVg7Ozs8PDwwMPDA1tbWxQKBTExMTg7OxMTE0O/fv0ICQlBCMGQIUNo164dFhYWZGVloVQq75BceqGbPxbKuu1JYpSZyKlbTyaEIL6eD+VGI5LGnoLSbILr1SaMhNZviYudN3KZwiSpZWNjg9FoZM+ePRiNRtLS0rCxsSEgIIDjx4+TlJSEp6cnkiTh6+tLZmYmzz//PCNHjqSwsJBTp07h5+dn8mzDw8M5ffo0y5cvJzEx0aTmYTZmZv4s5uVEM3+IW0W0qvxMajzqs3XrVvLz8ykqKkKlUrF161YmTJhg0lb8o1hFupqkrS516ow+Kwv4uSYtN4cJlZWkL1rMtpxsJkyYgIODA4cOHaJt27asXr2a9u3bsz1tO++dfY/k4mSGvjwUlaa2ZqxhpAsaayUKlZxvD7/H1IELWXtwMda2lrzzzjscP34co9HIvHnz2Lp1K1VVVbRq1Yphw4bh4eFBcXExL7/8Mnl5eURFRVFTU8P5wwe4WnCTd3YfpFqr45P9x9EZjGgNBl7KycZCklDJBcU1UFIjUMnBzs6OjIwMampqWLlyJdbW1hgMBmxsbDAYDCgUClJTU7l58ybW1tZ88sknCCHIyspi1qxZpozD2+kfWathuGh3ClnFVXjaW2BVrb1jHkClpSUBGg3ujj7kleYQdzmW0PqtyLx5hfzS2uvt6OhIcHAw8fHxVFZWYmVlxZgxY1i/fj02NjasWbOG0aNH4+joyIwZM1AoFLRp04a4uDiOHDlCfn6+yZPbs2cP1dXVODs7Y2lpycmTJ0lNTWXgwIFs2bKFVatWkZKScsdSoxkzvwfpVtHlg0BMTIz4I32WzNwfbrX4KNQa0LvXwyBXYGdnR+fOnU3tPP6S/QQGwW3fz9HXrnKmqgq1TIaDpyctWrRg4MCBjBs3DoPBgL29PQvWLWDpiaWkfZKGXTM7ctbnAODu486xg8eIiIigYcOGJCQk0LlzZ3788Ufs7e1RqVQUFBRgYWHBRx99xMiRI7lw4QKdOnVCkiRTnKigoICGDRuyb98+nnjiCbb9sAW1QkFLv3r8mHQJhUyGn6sTqTn5jHKxo9Sqmt03aqjQg1oOGqVEaU2tAvy6desoKirC1dWV6upq3Nzc0Gq1GAwGbty4gdFoZPPmzfTt25ejR4/e1Xj9N5YuXUpJSckd4zb6Mnp9v43NgVNZtncxswZ/QUFpNh9sf5EXB3yEq4cz+7KW06FDB0aPHo2vry89e/YkIyODsrIyDh2q1Tns0aMH3t7erFixgtjYWKysrOjcuTOrV6+mb9++7N+/n7lz57J//37CwsIYMGAAc+fOveuxDh8+nKVLl5rq0cyYuR1Jkk4LIWLu9pp5OdHMHyawbUcmfLiCmctXMXv+AubOnctzzz33lxowqFuXdqsm7VTjJvzUoiX5+fl4eXnx+uuvs337dqqqqhg/fjzz3p6H5CYh9ALbGFuaLGqC3FpOuayczp07U1JSwjSj4El7ew7t2YNcJqNHjx4MGzaMevXqYWNjw+zZs7G0tCQkJITq6mr69OlDs2bNKCoqYtmyZSQkJDBy5EiOHDmCtY0tZdU1/JR0CblMRkMXR7zsbRHA14UlbLlag0wG9mqI8FBQVlO7pLdlyxaTCLCTkxOFhYXk5OSQm5tLeno61tbW2NjY8Oqrr+Lh4WEyYLeKm28VTf+a24ufO3fujFKprPO6Eh1dFYfwaFqCb/Zu4GflEF0laoUGa2sbGrexvUPZ3snJicOHD9cRHH7yySfZvn07zz//PMeOHeOxxx7D1dXVpGy/cuVK09xRo0Zx+w3zr5cPb8mLmTHzRzEbsX8Jc+fOZfHixb/5elxcHFOnTgW4L2oK8+fPx9/fnzZt2jBs2LDaGqmYCHrmZdM5J5NniwupkcGAa1fpeOokMTExODg4UFhYyNNPPw3U/lDmJeUBYNvMlpITJfhr7XFRWTI5KJqxUUG4qFU0q65mmrMLr7m5oRCC7vXrs3//fjIzM8nLy+P69eu0bdsWd3d3IiMj2b59O5mZmZSUlFBWVoZarSY2Nha1Wo2A2ropwEql4mZFFSczbgDQwMeDVr4WhLjIKNdBQr4MjYUFMpmMMWPG8OSTT1JUVERSUlJtp2wXF4KCgnB1dUWpVFJWVsaFCxdqa7uoVbu4cuUKU6dOrdOW5XZuL34OCwujT58+pviZHaX04SfCSMHOt4oWTfdhLS/EWpaHt5Mfvl7+vL1lPLPfefYOZftp06ZRUlKCvb09L774IgB9+/alvLycMWPG8PLLL3P16lUWL17MoEGDiI6OrmPw+vTpw/fff09ERASHDh0yxcLMmPn/Yo6JmQEgJibGpA6xf/9+rK2tadWq1T3Z9691EqOiovB2sGPJN9/QPyIIP1cnPjtwgmqFnA2vzmHcl6vw9/cHaptRTpo0ybQtpazW87Brbkfu0kyaVGo4IyRcbay5WVGBXi7jtIWS1y6kMMHJiRohmLp4MTsOHKB169Y0bdqUpKQkHB0dqaioIDs7m/HjxzN37lycnZ35+OOPsba2JisrCyEEXbp0IS4ujvLSUirKStHrjMiVKqCGgrIaUm9W1Wo7GitQK5QYjUaMRiNvv/02KpWK8vJy6tWrh729vanNiEajQavVmmSqiouLOX36NAsXLiQ1NRVvb2+cnZ0ZMGAA1dXVTJs2jQkTJgB3tmgJCwv7xUNeGgIlvxg+X3sZqVMAuwXwXCKLfMcSd+bO9i4ZGRmm5ytWrDA9P3fuHOHh4QQEBJjG7OzsTLJit9OkSRPT+QF8+umnpjIDM2b+P5g9sX8wr7/+Ok2aNKFNmzakpKQAtb2XbsUdCwoKTNmBtzoWZ2Rk8Mknn7B06VLTXfPfQsK3tT+qc+058npf+rVohEajwcbGhj59+pAcd4IqrRY/VycAWvrVI7ukjONJZ+nXrx9btmzBysqKmpoaGjZsCNQ22+zUoRMauQa1qxprrYJ9SVeI8PE07baospr9qtqv/X6DCieNLdVCxtDeg9CoNZSXl5tUOyoqKlAoFHz++eeEhIQgk8lITU0lJycHo9FIeXm5qWllRVUVaitrtAYjWl2tor9er0eSJCoqKnB1dSUkJAQPDw8kSaKmpsbk2Tg7O5OYmIhGo0GpVFJdXU1QUBBeXl4EBgZiMBgYMmQIvr6+JhHhDz74gNOnTxMXF8eyZct+n+pF5zkIhQbj7XFwpQV0nvOHP74333yTxx57jIULF9YZ/71evHn50MxfhdmI/UM5ffo069atIz4+nh07dnDq1Knf9T5fX18mTpzIc889R3x8PG3btv3rDy7hW9g69WevQEB1EaTuqh3/GW1lZZ23eNjbolEomLv2e3bs2IFer6e6uhpfX1+WLVtGWFgY8fHxrHhnBXNbzcXDyoNIL0/OXM0k3OeX2JqLjRXHb2RzVafnbFUldpZ2eNu5o9Nr0Wm15GblkpycTGRkJB07diQ1NRWVSsW1a9cwGo0YDAbc3d2xsLDA29uba9eu0b17d2xtbXF1tUYmM6LT1y7/ubvb4eLigru7Ozqh42TcSdIy0hBCIJPL8PPzo6ysjISEBFQqFXq9HpVKhUwmo7i4mOzsbBITE4FageOnn34amUxGw4YN+e677wgPD6dFixZcv36dS5cu/eblzsjIwN/fn5GLtxGyQsW4nXJiPqsg+JMaXr3eBsIG15lfVVVFjx49+PTTT2ncuDH5+flAbZ1Xo0aNyM/PJzg4GHd3d5555hm6dOlCbm7uvbsBMmPmNsxG7B9Gds4WjhxpyxdfdiMmpoSS0p+wtbWlb9++9/vQfiF2Huh+UZho7aNg68UqqnfNpby8nG3btnGjtJyy6hoW7z7ImmNn+S4ukQYujthYWVBVVUV5eTkHDhygV69eGI1GEhIS2Lx5MyUlJbzc72V+HPgjfdu0YPHgXjha1dZ6OVpZ8lKPDgwO9kcnjHzYdy4Hxn9NtFcIjwV1wyCMfDXsbfLy8khJScHHxwcHBwecnJz45JNPcHd354knnmDMmDFYWlqakj5OnTrFsmVjeWqiHAcHGQMfs6NePSUeHiXk5eVh5WxFcXExShclPlN8ADAqjTRu1hij0YhCocBgMFCvXj1cXV1p3bo1OTk5aLVaAgICkCQJnU5nio2NGzeOPXv2cOzYMcLDw/Hz87tDwePXXLp0icmTJ5N05Trv7LpCXJaehMwKDiQX1FnmKy8vp0+fPgwbNoynnnqK4cOH8/XXXwOwZ8+eO/qFnT17lqFDh/L222/fmxsgM2Z+hdmI/QFuX4p7EMnO2UJy8iyqa2prffT6MpKTZ5Gds8U055bMEfA/f/j+Nkpu1PmzqZecvv4Kwt5KoUePHuh0OlKycnGztaGksppzN7K4UVRMwo0cMvIKSUlJwcbGhj179vDxxx9z/vx5HnnkEc6dO0dMTAxFRUVAbU1biVbP4t0HTfsyIsOtRIunjStNvWtbwAwI7kpCbgpu1s6M+mo6Go2GzMxMfH19GTlyJGfPnuXDDz+kadOmeHl58dhjj/HII4+Qn5/PqlWrsLa2RogfMRprKCsz0LqNFa/Ncycjowq1WkaRVIQwCPTlesrOlIEMlG5K0tRpWFtbo9fr0ev1qNVqHnvsMc6fP4+trS1KpZLLly8jk8no2LEjU6ZMAaC0tBQHBwcsLS05depUHQWMW9y6mYnd24i4uEF4e7vcVdE+KSmJCxcumN7Xr18/xowZw8iRtc0zx44dy6pVqwD48ssvGTNmDFDbL6xbt26EhoayaNGiux6DGTP3ArMRu4f8USN4S6/u95J2ZTFGY62HExaq4ciRSqqqKkg8/yZbt24FapcLT58+DWASt/01NjY2dfpb/eXYed8x1C+iI7OGrSfG4TGuX7tO/x6D2LJuLTaWGjzsbKnRG3C0s6OkpAQbGxv8/PzQaDR4eHgwZcoUDAYDU6ZMQaVSYWNjQ3p6OoFtO5Jl6UB5jQ4BlMqt+dG5I/OjJmCUpDr7t1RaYDAa+HHaGqqrq3nppZcwGo3MmTMHBwcHHBwcWLFiBXPnziUsLIzx48fTvHlzwsPDWbp0KU7ORSx6Ox9XVwXXr2uZ91ouCoWETmfEtrstVoFWYARtrharQCtqbtRw5fsraLVaevbsySuvvEJ2djYffvghjRo1wsbGBkmSsLKyIjw8nMGDB9O6dWuUSiWdOnVCr9fTpEkTSktLadmyZZ1zycndedvNjKBGm4tcXkJ2zhaTon1sbCwJCQn06tWrzs1M69at2bVrlykd3sfHBzc3N/bu3cvJkyfp0aMHUFvn9vTTT9/RL8yMmXvNv96IVVRU0KtXL8LDwwkJCWH9+vWcPn2a9u3bEx0dTbdu3ci+rRnk6tWriYiIICQkhJMnT5q2MXbsWJo1a0ZkZCRbttR6Pl999RUDBgyge/fuNG7c2CTlAzBp0iRiYmIIDg7m1VdfNY37+vry0ksvERUVxYYNG/D19eXVV18lKiqK0NBQkpOTMRqNd41V5OTe4NjRCp6eksnid/Kpqjby5LgbPDc9nqZNmwIwY8YMPv74YyIjIykoKLjrNfl1OvRfTuc5tQkFP5Na2ZYp+7KYt2MkXx16Hjc3BcaqPK7o5GjaOJAvr0LIQW5Za6D69+/PtWvXMBgMjB07FkmSyMnJIS4uDgcHB0aMGGHSTTwWn4DeyokPGkxiZb0RXLJpAkBOaS7HMs8DsPnCnlqvTALffmF11OttbGxwcXEhJ6e2aLqmpobKn+N19vb2bN++nZkzZ3IhqfZ8goI0FBQYWP65N+PGOWI0Qv2A+rgPckfpqMT3BV/qT62PykVF9CvRHD16lPT0dF555RVu3LhBw4YN+eSTT9i/fz+NGzemoKCA06dP06FDB9LT09m3bx9eXl7s3LmT1NRUMjMz2b9/Px06dABq418lxZ+ZbmZ+wUjalcUmRXs7Oztyc3PvqAebN28eDg4OJq8PauvBhg8fzqBBg0x9zEpKSu5aD/brGyCznJSZv5t/vRHbtWsXnp6enDt3jsTERLp3784zzzzDxo0bOX36NGPHjmXWrFmm+ZWVlcTHx/PRRx8xduxYKioqCA8PZ8+ePVRWVvLMM88wdepU2rRpw9y5c9m1axfvvfce58+fJz8/n48++oiIiAj27dvHRx99REJCAnv37qV///40a9aMrKwscnJyOHPmDNXV1eTn5/Ptt99SVlaGu7s7ixcvRiaTYW9vT1RUFMHBwYwcOZLw8HDc3bxZtqyAsDA1EhAVaUFJiYFmTV1ITk42/djEx8dTXl7OtGnTyMjIwGg08uSTT7JixQq2bt3KiBEjkMvlODs706RJk7/+oocNhj7LwM4HkIh3DOO1RTo++9yDYY87EB6h5PzVIxSe20d5dTn6Mj1IUK4o5631tb28qqqqUKvVDBkyhNjYWNOS3KhRo0y6iampqUiShF5I5G9dRObyieR//wZGXQ0yC1tGbHwZv8WdOXL1NKPbD2bM0FFEDWmLt7c3WVlZfPrpp1y5coVXXnmFjIwMwsLCTC1FMjMz0Wq1jBgxgurqav4zOx2tTjB4iB3nE6oZNPAqb7yRh0Ihp15SPRz8HHBo40Dqi6lcnHIRQ7mBsNIwoqKi6Nu3Lw4ODnh5eVFQUHBHivstEhIS8PT0vOtrt1Ndk/2b4+Hh4URGRhIQEMDjjz9+Rz0YwHvvvWdqfAl168FuMXfuXHM9mJkHAyHEA/OIjo4W94Lvz9wQrRbGCt+XtonI6V8JFw9v8eKLL4qDBw+K8+fPCxsbGxEeHi7Cw8NFSEiI6Nq1qxBCiPbt24vY2FjTdnx8fMTKlSuFs7OzCA4ONs1XqVTi8OHDYsWKFaJTp05izJgxQgghHBwcRO/evYUQQkyfPl1oNBoRGhoqLCwsxOTJk03b9PX1FeXl5WLFihVCoVCICxcuiKqqKuHm5iZat24thBAiISFBREZGCr1eL1xcXMSyZctEVvZm4eioEF5eCuHbQCm8vZVCpZLE/AVPCiGE+PDDD8W4ceOEEELMnTtXLF26VAghxO7du8WAAQOEEELcvHlTGI1GIYQQy5cvF9OnT/87PwohhBA7f4gRe2Ibij2xDcUHH3qKRo1U4vEn7IS7u0pISkkoXZQCOcJjlIew9bMVMplMODo6iqCgIHHw4EERExMjPDw8BCDS09OFEELExMSI4cOHi5deekkAwu2Jt0X9l7YJq9Auwr7DWOE99RvRamHtZzl8+HDxww8/CCGEaNasmdi0aZMQQoiqqipRUVEh9u3bJ3r16iWOHDkioqKixNWrV4UQQjzzzDPitddeE0IIsW3bNgGIbduai48+9hYNG1qKy1fWibKyMhEUFCTe++E90fzN5kLtrRadVncSGxI2iKCgIHHmzBmRnp4uJEkSx44d+81rVFJSIgYOHPi7rufhw21M1/P2x+HDbf7U53Pq1CnRps2fe+8TTzwh8vLy/tR7zZi5BRAnfsNu/OuKnW91wL3VQPCm0hn7x5dQY5PNf/7zHzp16kRwcDDHjh276/uln2MpFw/to7zoJmfXLKfo5k3aN2/GtJdexsHBgVatWjFlyhSSk5NRq9WmJRhJkujYsSPp6els2bIFJycnDh48SKNGjdi8eTNHjhwhJycHFxcXrl27BtQWvrq4uKDRaGjYsCEVFRUAHDlyhMuXL9O4cWMKCwtxcnLCw70f5eUS48fXp+sjggtJFsyYkcKY0XMBiI6OZtOmTUBtwL5fv348++yzdwTshwwZQnZ2NlqtlgYNGvw9H8RtKCxvmp4HBGho2cqS2D3lODsrqPCwpupaFdYR1uSszQEJrCytkMvlqFQqHn30UbRaLWVlZYSEhNC3b1+cnZ1xcnJizZo1uLq64ujsikODUMoKsqi+moCxuhxLJw/y924n9Gs9N2/eJDg4mA4dOtyhYn+LWyr2P/74o8kbOnjwoOl69urVCwcHB5o338bly1/zxBOF+DWs7ZU2YMAASIdhmmE8MuYR5g2fB8C5Aec4dOgQffv2pX79+v9VG9HW1pYNGzb8ruvZ0G8Gycmz6iwpymQWNPSb8bvefztvvvkmH3/8sSlD8Y9i7hdm5u/mX7ec+OsOuPqyQmpQcEoRwgsvvMCJEyfIz883GTGdTlcn82r9+vVcPLSPTxbMRS2X4e1gRyu/epw/fYrnnp7Cd999h6+vL/Hx8fj6+tKtWzd+/PFH0/s/+ugjjh07hpWVFZIkkZ+fT0lJCU8//TTx8fGmpc3AwEDgF6MJmBQcbgXn33//fVNh7C3hWCGga9dP6NzpMmfOxCBJMlM/J7lcjl6vBx6sgL1C5lbn78GD7fEPUFNabqT8fDm2Ebb4jPcheHkw0bOiMRgMbNmyhYMHD+Lg4MCcOXMIDQ3lgw8+YM2aNSaVi1dffRWj0YiupooXmlogzziGpn4YFko5Ffs+JXbHD5w/f57x48f/z/P08PBAo9Fw9uzZv+UaWFlZ/WXb8nDvR0DA62jUnoCERu1JQMDreLj3+8PbuiUndavjtRkzDxr/OiP26w64uvwMri8bxsm3R/Daa68xb948Nm7cyEsvvUR4eDgREREmBYL4+HgKCwuJ6tyNj/YcRqs3cPTyVS7lFXDjZjHxiUnMnz+fy5cvm4xgcnIyAQEBhISEoNfrKS4uxsfHh8LCQjIzM4mKisLV1ZU9e/awYsUK8vLy6NmzJ40bN+bbb38p/p00aRLx8fEkJyfz5ptvYmVlxdChQ8nJyeHy5cu8+uqrbNiwAYPBwGOPPYalpSXbtm0zyRw1btzYlHp+KxFk0KBBDBw4kOrqamJiYujSpQsFBQV3Ddj/nfgHvoSE2vT3kiX5XL2qp0Qr4drfFc8RnhQvLyb35VwuvHUBLy8vrKysTOoeFRUVFBcX0759e2JjYwkNDeXq1asANG7cmLKyMs7FbsIp9zQ9/O15ZeyjqBUynJ2d70ji8Pb2ZvPmzcBvJ3Hs378fgHbt2rF27VoAdu7c+Utqf9u2bN68mcrKSioqKvj+++9p27btb47fjZEjR5oSh/4MHu79aN36EJ07XaZ160N/yoD9FTzomp1mHn7+dcuJnvYWZN5myCwaRqNyb4x/38mc+uAXDT6ZTMYXX3xh0hMEiIiIoGnTplw/c4JBMbV6dOdvZJNbUo5cJkMmgYOTEz4+PkyePJlLly4hSRL+/v7Mnz+f4cOHY2NjQ7du3dDr9TRo0IDPPvuMiRMn4unpyZQpU6iuriY5OZkvvviCiRMnEhoayoQJE0zK5X379iU5OZn69evj5ORkUjOPiIhg6NChDB8+HEmSeOaZZ1i7di1lZWXExMTQsGFDnnjiCSorK1mxYgVhYWEsXLiQmpoajhw5gr+/P+7u7kRHR9OrVy/KysqwtrZGq9WSm5uLm1tdb+mv5NYPbNqVxVTXZDN/XiQN/WZwplLBe2feQ5GjoH+f/siFnOPHj1NVVWUqGfg1v2490qRJE06ePMmuXbu4evUqDRo0YNKkSRQVFRESEoK7u7spcxNqs0+feuop5syZg1KprLOE5+bmxrZt2+jRowdffvklr776KsOGDSM4OJhWrVpRr149AKKiohg9ejTNmjUDarP7IiMjAe46frs24S1+bxLHw8791Ow088/gH9tPrKKigsGDB3Pjxg0MBgOzZ8+mUaNGjJwwhStZhUgaG5x6PYdMZUnmx2Ow0ijQVVfh7OzMggULGDduHEqlEr1ej0KhwM3NDblcTnl5OXl5tUrpztaW6A1GiquqkbjV1AJUKhVyuRy1Wo1KpaKsrIyqqioUCgVqtRqDwVBn+UqSJAYMGICDgwOff/45Tk5O3Lx5E2dnZwoKCrC1taW0tNRUuyOXyzEajVhZWVFdXU2DBg24du0aO3fupGvXrshktV6Gp6cnSUlJuLi4mLwAR0dHCgoK+Oyzzzh8+DDbt28nKCiIS5cuUVBQQOvWrfnmm2+wt7dHkiQ+//xzLl68yDvvvPOXfC5/htv7YmVmZrJt2zbGjRuHtbU1X375JRMmTGD16tV88MEHxMXFsWXLFqqrq+nevTsAy5cvx+P/2Dvv6CiqPgw/szW99xAIPaRAgNBr5KMovVlAiihIU0AFRYqIWFAUqYKIICKgIL2IGHontJAQQg2k955stt3vjzUjoQkC1jzn5MBOZmZnZrNz5/7K+3p78/TTT8sVd39n8vPzefHFF+87B/Z34/333+ebb77Bw8MDPz8/GjZsyNatW5k1axZhYWFkZmZSo0YNJk+eTFhYGLNmzWL+/Pk0bdoUpVKJu7s7o0aN4ty5c8ydO5e9e/ei0WgqBrf/MPfyE/vXzsTKSue3bdsGWJ7QO3TogI+LA8UF+SSnJ5K5+WO8gpohGXVolA40a92aAQMGMGzYMDQajVyUYTDqGD7cmrffvkDr1q54uFYiOjaRzMJiOtVswU+XDqGQFPh5+qC01XLlyhXUajUlJSUoFAocHR3R6XQYjUbc3NwQQqDX63F0dMRoNFJQUMD27dvR6y1OvDk5OahUKgoLC5EkSd5Phw4diIqKkgdRd3d32Z7DbDZz8eJFOdd25MgRJk6cSExMDAqFgrfffpvs7Gy6dOnCuHHjWLx4MdeuXSMnJ4fRo0ezatUq6tWrx86dO/+S4o57cfPsytfXl9q1a8tq8mFhYTg6OvLNN98wfPhwsrOzUalUdO3aVd6mbt26/PTTT3dt7v678SBFHLcSHx9Ply5dZM3FP8qZM2dITk7mqaeeeqDtbtbsNBqNNGjQgIYNG/7udmWSVXZ2drzxRvkClIoZWgX34rENYpIkTQOGAhm/LnpbCLH9cb0fYLGlPzWH1KJU7PPsubzjMi5vutClSxecnZ2JiorCZDKhVIBkMlJdnYnp5EVMRoNlEIiK4stXx6ArsYQbS0p+CztOnnwBIeDYsSyMRlAoJMxmwe6rlryFAon41N/klAy/KpkrFJa0Y9ksKjMzkzp16pCRkUFOTo5cuGEwWI5BkiS5dLSkpAS12mLfIYQgKSlJtgBRKpXcuHEDtVotD5hr1qzB2tqatLQ0OnXqxI0bN1CpLB9xVlYWycnJjBkzBkmSsLOzIykpCQcHB5KTkzl+/LhcjffKK6/w2muvlXPn/Stx/FWpo4zmzZvTtm1brK2tWbt2LQ0bNiQ0NJSjR48CllBcREQEeXl5ODo6MnPmzNuaeu9E2XUv+8z+y5w5c4bIyMj7HsQuHkvlyKYrbNm7imqOYSSey6dWE69ymp1ff/01/fr1w9nZWf5+jB07Fnt7e8DSgzlnzhzeeOMN9u7dK8/QFi1ahFKpZOXKlcybN69Ck7GCcjzumdhsIcTds7qPkG1XtzHt8DR0JkuYLt8xH5/JPuTG5zJ48GD8/PywtramtKSY0Y2UdKmuRqWAVsuKsVaBylhKUm4u6VguilZSYEYw8CUfft6dwbWreqytASyDV1nw0PirNJRKqcJoNKOQFEhKSbbhcHNzQ6FQkJOTg7W1Rbw2Oztb1i9UKBRy6LFM6aBVq1ZYWVmxa9culEol1atXJyYmhqSkJACsra3R6/VoNBpMJhMlJSWYzWbOnTtHYWEharWaDz74gKFDh8rViBMnTuSdd95h4MCBVKpUSQ5njhgxgtGjR+Pm5ibP7O6mxvBX0a5dO7Zs2SLf+LZs2UJmZiZWVlYMGzaMBg0alFu/nIfW7xAfH0/Hjh1p0qQJJ0+epHHjxpw7d46SkhL69OnDu+++y+7du5k7d65c8LFr1y4WLlzIhg0bGDFiBCdOnCi3/l+N0Wikf//+nDp1iqCgIFasWMGsWbPYsmULJSUlNG/enMWLFyNJEm3btqVJkybs2bOH3Nxcli5dSpMmTZg6dSolJSUcPHiQiRMnEhsbW26WFBwczNatW/H39+d/bTpxKfYqeqMeb2d/PJ382PPdBeqHVyc4xGLemZ+fz7Vr14iNjeX69ev3/fncPEPr06fPXzrLrODvyb/mkXPOqTnyAAZgyDHgXuKOh7UHTZo0ISkpCZPJhKRUsTfZmdE7dMw9pkcACgkMxTrclUoUv86MjAgCtVYkpebz1FN2CAF6PZSWChwcFFhZgVqhwsveDZVCic6oR6VQolQo8PT0xNHRESEEmZmZpKWllesVK7sZA3z++ecIIXBwcJBnZVeuXGH37t2AJf8VHBwMQP/+/RFCEBAQgEqlQqFQoNfrkSRJDjvWqVMHOzs7RowYgbe3N6GhoYAlz1JUVETNmjVl+3qAGTNmYGNjw969e/nkk0/kmded1Bj+Km51KB4yZAg7d+7k2rVrj0TSSFZ4j4nh008/JTIykqioKPbt20dUVBTh4eFcuHBBlvlatmwZQ4YMASz5n1vX/6uJi4tj5MiRxMbG4uDgwMKFCxk9ejQnTpwgOjqakpIStm7dKq9vNBo5fvw4n3/+Oe+++y4ajYbp06fzzDPPcObMGZ555pl7vl/PBmOY0GsRE3p9QULmRc5c3U9xcQnFJcWcPxNJA7sMHBSlaNHz8ccf89Zbb2FjY3EWKC4uluXYMjIy0Ol0crViVlbWXasV4+Pj5e/Fzf+/H8rsie71/wr+OTzumdhoSZIGApHA60KInFtXkCRpGDAMkKu7/gipRanlXusSdZxeeZooolAqlXTu3JmSkhJ2797NhYxCSgoFqQUmlBKYzKBBIstoxEmpZLKXNxNTUzipK0Ha9ts+NRrQ6SAvzzKLGtnkGboEhPPsmnEYTUaKDSVIkoLMzEzMZjM+Pj6UlJSQm5uLQqHA2toak8lE7dq1yc7Oxmg0snLlSuzt7VEoFCgUCjQaDVZWVpjNZpRKJVZWVsyfP58ff/yRefPmARAeHs758+cRQqDValm5ciXDhg3DbDYzbdo0xo4dy40bN3jqqae4evUq8fHxDBw4kBdeeIHq1aszYMAA2V/sTu683bt3p3v3v6Yk+248yOzq99h4OolPdsaRnFuCi8jD3btSOYX3L7/8EqPRSEpKCufPn6du3boMGDCAlStX8sILL3DkyBFZ2f1u6/+ZpKRukis7c7Kd8fFxk+Wknn/+eebOnUvVqlX5+OOPKS4ulpu7y/KGvXr1AizN8HeqlPw9th1YQ9S1gwAU6vJpWD2cD9cNBSSqOFqiFo29YefVVBbN/5zg0N/y8/b29nh4eMivDQaDLABwp0GsbJZZ1s9ZXFzM3LlzuXLlCsHBwQ88y2zWrBknT57k/PnzuLm5ERkZyYQJE+RZZo8ePUhISLjNQdvOzo4xY8awdetWrK2t2bRp02Ot4K3g7jzUICZJ0i+A1x1+NQn4AngPS9ztPeBTYMitKwohvgS+BEt14h89Fi9bL1KKftOMsw+xZ9DwQUj81ix87tw5jEYj1tZWWEvWVLfP4EaeCVcbBbp8FQaTkift7enm6Mj76Wl4qVQoHFTkqQ1Ikplvv63C4kVZHD+hQ6twZ9flQ5xIPEf0mG28sWMmZ1JiObnzCLb1PfD09GT8+PGEh4cTHh6Om5sbnp6enDhxglq1arF//35UKhXffPMNtWvXBiw2KadPn6agoIB+/fqRnp4ui/SuXLmS5557jhdffJFPP/2UTz/9VNba+/DDD/Hx8SExMZHu3bvj6+vLuHHj+Prrr+VzX7FihXzjLeNh1Rj+idyq2JKWryPXoGDj6STqOemZNWsWJ06cwNnZmcGDB8th1xdeeIGuXbtiZWVF3759UalUctP5ndb/syiz3ylT5yjVp2E0WRTry1oXJEli5MiRREZG4ufnx7Rp08od552a4W/lZgsf+M3GZ+/evVxOO83rPeahUVvx+ebXMJlNv74vhHgqyS0R7LxmwiygdSVBjcaNiYiIICEhAS8vL65evUqlSpXIy8tDCMHVq1fJysrixo0bzJ49G4D27dvTp08f4uLiWLp0Ke+//z716tWjZcuWciFUWeh01apVVK5cmYSEBGJjY3F3dyczM5Pnn38eT09PUlNT8ff3l127lUoliYmJ6PV6iouLMRgMbNmyRR5Ejx8/jqenJ3Z2dsyePRuFQkFRURFr1qzBysqKyMhInn76aXJzc4mOjmb06NHs3bu3YnD7k3iocKIQ4n9CiOA7/GwSQqQJIUxCCDOwBGj8aA75zoxpMAYrpVW5ZSWq8o3N1pakFpJQoNU4MOdJK7b0s8XBSonKxRFJUuChUgMQoNXS1MEBjb0njg7evDstGJAYPjyQVi3boLFxQuGgxcHWkpRuWqsBbcLbYlvf8lTZqFEjQkNDqVevnpzc1mg0PPXUU7L6QaVKlXB1dZWPr1KlSqxcuZLevXszYcIEORQIUKVKFRQKhZyT+Oijj9Dr9fL2Tz75JKGhoXe1jb8T/0U1hlsVW8BS0PHJzri7KrwLIfDy8sLHx4cZM2bIEl33UoSfOHEiGzZseOznc7P9ThnpaQY2b5oKwKpVq+TP99bm7ntxqxq9v78/p06dAuDUqVNcu3YNsFSOVqrihY2tDak5N7iWFsPF5NNM6T0XK5WCg9eNnEkzU8NZorqzxMZz+axcuVIunnnmmWdISEggPT2d+vXrY2NjQ+vWrQkPD6dy5co83/slXu0yi11b9tM8rC2uzr/NMvPz87Gzs2PChAkUFhbyyy+/4OzsTEBAAEOGDMHa2prXXnuN6OholEolWVlZHD9+HA8PD1JSUoiMjGT06NFoNBpee+01Zs2aRXBwMK6urvIstbi4mCeffFKeqX/wwQdERkYiSRLHjh0jMjLSkqaQJM6ePYvZbCYrK4uzZ8/SunVrlixZ8oc/2wruj8dZnegthCibGvUEHi4b+zt0rtYZQK5O9LL1IqhpEPEHriMkyxNkjRo1qFGtFvb5NXEzqglzewkcK7F/5FSo+zR5W7aQPvtzjCkpfNe0GR7jxuJ4U6l2GU88cfv7j6V8573juwAAo+JJREFUxdTNOYfly5ff8ZhvDd2UvX7vvfcASzMsWAasOXPmEBgYKIf93nrrLd5666077vduyx+ER1WqfTNTp06ldevW/O9//3tk+3xQblVsuXn5zQrvbm5uFBUVsWTJEj755BMaN25MTEwMOTk5rFmzhnfffZesrCxycnIICAjAz8+PmjVrMnfuXAYPHsyqVavYsGEDkydPfqwFH3dSrPfzU7N27VU+/7wOgYGB92zuvhvh4eF89NFHhIaGMnHiRHr37s2KFSsICgqiSZMmsrtBp06deO+z93jzh1742tTA2c6DKn7VKPE7iREFWhdvfBV5FBQUUd1FQZcQR05RjeTkZAYMGEBhYSHNmzfn4sWLsm5pdHQ069evp2PrbhzZc4ZG1d14u+9XZBWkMmfzOC5O6IpGG4e7swOurq5MmTIFFxcXWrVqhUajYffu3UybNo2ioiI+++wzVq5cSV5enjzjtLKykvPC/v7+6PX6crPMstlofHw8paWljBkzhuHDh6PRaBgzZgzOzs4IIWjatKksF1b2b1mRFljCs7t27fqDn2wF98vjzIl9LElSKJZwYjzw8mN8L8AykJUNZmXszD/AsdMHMStKUZi02Bb6Y2f2ptmAAGiSW25dx65d7zho/R3QaDQsXLjwrz6Mh2L69Ol/9SHg42RNYk4xIJAkBSpHT3xeXIiPkzXx8fEcOXKEpk2bcvToUYqKisjOzkahUODu7s5TTz2FRqNhwYIF9O7dm/DwcIQQ1KhRg+3bt1OzZk0AgoKCUCgUXLhwAZPJRLt27YiKinosuTIrrbfs5A3g5aVm2XI/rLQ+tGjxmxfcjBkzmDFjxm3bl0logWWmVvYg5eLiIudNy7hZA7SMbVe3IQ2V8DP5AiW4bXbAOhOKUfP222+zfPlyMnHFzV5iajsVawqbYr6ejyRJ8sBxq/mro6Mjc+fOpUerIaTnJHE1LYZqXkEAZBem8+3lRlStfZTs4gKOao+Sr8/HWm3JN69fv54ePXqQn5/Prl27aNeuHatWrcLPz08euG4OjSoUCiRJ4tSpU3Tq1EmuHgZLfk4IweHDh7GxscFgMPDee+9RpUoVwsPDmT9/Pp06dUKhUMjnoFKp5P/fKzxbwaPjsQ1iQogBj2vfD0LHXq2o6luTI5uuUJhdip2Llmbdq1OryZ1SeX9P7jXrupnPP/+cYcOGyZVfj4qrV6/Su3dvFi1axNixY9HpdFhbW7Ns2TJq167N8uXL2bhxI0VFRVy6dIk33ngDvV7Pt99+i1arZfv27bi4uDB48GC6dOlCnz59Hunx3Q9lpfQ+NUNIPXYCtVctDJnxCIMeh8CWjJ/zMfM3zObixYuYnjbh3dSb9CnpBAQEsG7dOjQaDUIINBoNer2eo0ePUrduXVxcXPDx8SE3N5fS0lKuXr2KJEnUqVOHOnXqoFarH2vBx6NUrP8j3FoV3MijETsP7qR1y9aYzWZycnIwm83o7Gzo/H0JhSURuLi4oFAoeP7553n99ddJTU3FbDYzd+5cunbtSlZWFl26dMFQ5Mi1tBhSc+I5cekXOtbvh6eTHzvP7SfqUB6mUoHS2aJek5+dz6Fjh1Cr1Vy+fJn27duza9cujh07Ru3atUlOTsba2prp06eTkJBASUkJrq6u9O3bF4VCwfbt29m0aZM86Bw/fpxq1aphMpn47rvv5ObzLVu2yIOUlZUVNWvWlHtAe/TogV6vZ+PGjXz55Ze4uLgAFUUgj527ebT8FT9/lp/Yv5UqVaqIjIyMR7Kva9euiaCgIHHhwgURGhoqzpw5I/Ly8oTBYBBCCLFr1y7Zg2zZsmWievXqIj8/X6SnpwsHBwfxxRdfCCGEGDt2rJg9e7a4du2acHJyEmvXrn0kxyeEEFOmTBG7du267/Mp8+zacCpRNJqyQfi/uVV0GbFCqFVaEVy5qXB19BKAsPKzEhpPjUCB8KrsJSpVqiQAsWTJEmE0GoWVlZXw8PAQOTk5wtbWVlhbWwtvb29RrVo1Ub9+fVG7dm2hUCjEkiVLhBBCDBo0SCxbtuyRnfetJKds/NVDrLo4eLClSE7Z+Nje61ZCloeI4OXB8s/Ud6aK8PBw4eLiIvz8/ERwcLCoVKmSUCgUwsbGRjg6Ogpra2thb28vbG1thZ+fn6hRo4bw8/MTvr6+omrVqsLGxkY0a9ZMfDRslbC3dhIONq5ibLfZYv7LEeLdft8JTxc/UeuTWkLjpRH29e2FpJSEpJWEZ3NP4erqKmxtbUVoaKgAhJ2dnWjevLmwsrISTZo0EVlZWaJ58+bCzc1NhIeHi0WLFgmNRiO0Wq3w8/MTDRo0EFZWVkKr1YoBAwYIjUYjHB0dxcCBA4Wtra1wdnYWgYGBwtnZWfj5+Ql7e3shSZLYs2ePyMrKElqtVvTv318EBQWJr7/+WgwaNEgAsl/d+PHjxXvvvfenfT7/FqjwE/tzuZNuY1xc3H03m7Zq1Yri4mIGDx5MdHS0/CS5YMECwsLC7thgO3fuXBITE2nTpg0eHh7s2bPnDx17Wfn59evxpF9Pol2nzvy8bTOBgYEkJCQwaNAgWdj45n638PBw7O3tsbe3l6u4ygSMH0Xv1J3yab8Xniw6nU7+znhMuaWkiywq31RKn3psC/NmL+BEeh4GYymVPQLo2/IVPv5xBFVDa5FXLY2EOQn4vODD0h5LadCgAdOmTeOzzz5Dp9OhVqtZvXo1L7zwAvHx8Rw7doykpCS++eYb7O3tCQ8PR6FQyAUfbdu2BSwFH40bN5Y9yx4F3l7d/zKV+purgtM2pLGvZB9tW7eldevW8jpHjx6lduXKTO7dm/Xx8ahUKkpKSjh69ChLliyR9RTDwsK4evWqrNbRc0g4B850R62wooa3ZRarpJRidQHu7o7U+qgW+gw98SnxWPtbk3stF1vvxrh2GEPasdWoVDFUqVKFoKAgpk+fLqt95OTkkJeXx969e2WXgzJ9UhcXFz7//HMmTpxI3bp1eeKJJ4iMjGTv3r1ER0ezfPly1q1bR9WqVdHpdJhMJqKiovD390ej0eDs7MyPP/6IwWDA09OT5cuXs3r1arp06QJU5MkeB/+aZue/E2W6jWfPniU6OppOnTo9ULMpWHzHnJ2dOX/+PO+99x4nT56U179Tg21Zg+j69esfagCbuP6crPIv1DbkSA7MX2VRjJ8yZQrh4eFER0ezZcuWO5ZpA3LzNVhyDrfmBa5evUr9+vU5duwYzZo1o379+jRv3py4uDjAUgjTo0cP2rdvj7+/P/Pnz8fJyYnx48fTtGlTOWcxePDgu1baFZ1OJ3f9JUy5pQCY8vVYGVQUnU7n2PEvmTFjHDNmFjDzEye0Gg0udu6W45UUxO2IIvkrS54pPTUdvV4ve5S5ubmhVCpxdHRkxowZuLi4EB0dTX5+PkajkRs3buDl5UVJSQlvv/02/fr1w8nJSf78zp07h5fXPyeU/XvcWhWcapOKUSr/eUtmMx4pqXit+Jb32rcHLA39hYWFcm7qTq0JtZp4UTXEDa2NEhDYKTNoZvctJqmYhEUJXJp4ieQVyehT9UgaCUmjJufSKVJXv425RmuwcQYrew4dOsSgQYP45ZdfuHjxIk8++SReXl6o1WqWLVvGhg0bsLOzo0qVKpw5c4bevXuTl5cn/51t376dK1eu0KFDB06fPk2fPn04efIk169fJzU1lQ4dOhAYGIjBYGDBggWUlJTg4+PDjz/+CIBarZa/DxV5skdPxSD2CCk6nU7KR8ex25rHyuXf4u7iRtWqVfnpp58YOXIkdnZ2WFlZsX79ernq78yZM6SmptK4cWNGjx7NhQsXANi3bx9xcXEEBgYyZcoUtFot58+fByyxdxsbG2xtbTlx4gTnz59n7ty5mEwmevToQXh4+B86/lvLzyWlCtcek1i5ciVz5sxh/fr1shzV3SoubyU9PZ3vv/9e7n8bM2YMISEhTJs2jfXr19O8eXOqVKmCTqejQYMGzJ8/n59++okdO3aQm5vLzz//zKRJk1i7di2TJk2iWbNmt/W73Yn8nfEIg7ncMiHgxomVnIv6GK3WjJ2dgiJ9FgajHiuXq+QUplOsL6SGV10ktYTKRUX6mnRGjx6NnZ0drq6udO7cGa1WS5cuXXB3d2fu3LnUqFGDBQsW4O/vz2uvvUa/fv148skn+eijj4iIiKBmzZr069cPsBQLNGvW7D4/kb8/Z1afIXlKMokfJqJP1WO2N7Nu/Tqyr10DIRAZGRzft4+TSUkcyMyky+DBrF+/niNHjpCcnEzv3r05cODAXR9G/Ot4U7ulG6MWtWPQyxLVPBIpTtXzQa4jJx2qEpYIGq0Crz5++A57E/eeb2MqzKbw7E8Y89M5f+YU169fx9vbm2effRaTyUT//v15++23sbGxwc7O7p6KJPb29jRu3Jjq1auzefNmTp48yffff0+9evUoKirCbDYzc+ZMudezd+/eXLhwgdTUVFJTU++63woeHRWD2CPi5if/67lJPFWrDVNaj8TXxZuzZ8+ye/duYmNj0el0VK9e/bYQ2/Hjx3n//ffJysoC4Nq1a9jb28szsTJzxmvXrpGYmEhSUhJFRUXY29tz6dIlXn31VZRKJRs3bvzDM7E7lZ8rNFY49ZjM119/jZWVFRMnTqR+/fr39TS5efNmPvzwQzw9PTly5IgsvTV37lxWrFghN4Vu376d0tJSfHx8mDRpElqtloEDB9KyZUu2b9+Oo6MjlSpVAiAkJOS+VCXKZmC3ku67hqrVBNVraHhhcAILFmRiZSXh4HcSK40tXk6VCanTHJWDCvRgo7YhICAApVJJcnKyLHO1c+dOevToQV5eHseOHaN9+/YUFBQwYMAAHBwcqF27NmvWrCEsLIy9e/fy008/ydv9WyhTrL8Uc4nrx6/jlunGsLrD8HL2otW+/Tzz/Q903PkzVmYznRzsmZKaQlRuLs2bN6dZs2aMGjUKZ2dnXn31Vbmp/1a6du3K6tWr8fX15YXZO1ii6467lRXhwgYF8JSVPZJe4LPRnYT535K1Yy6m4jyKoiNQOXkhaa1RqVQUFRUxadIkFAoFAwcO5NNPP8Xd3b1caf2tmM1mwsPDOXDgAOfPn6ddu3bk5OSQmJiIwWCQVXhKS0vp3bs3kiQRGBjIW2+9Rc2aNW+ruqzg8VCRE3tEZG2+gMJgCRm42bhwLOEsnvautHGvT8z58+j1enr16iVX8N0stfPEr41n9erVkwcHIUS5kFwZ+fn56PV6nnjiCXQ6Henp6SQnJ8vrFRYW/uFzuNkwtKz0HMDPy53vNm2S+8bKqhW///57mjVrhk6nQyfpaDGnBQVOBdiE23BlzRX69OmDlZUVlStXZt++fRgMBlJSUpg8eTKOjo6Eh4fj7e3NoEGDmDRpEm3btsXR0ZHQ0FCuXLki59PKRJLLzvF+BlClk7bcQObn6M3yPh/x9PDnCaij5dLFUoKDrbh4sRSDQXD1RhaHd09HZyjmh11zUdoo0Sg05Gbmsm/fPnJzc6lXrx4xMTGYzWaSk5Nxd3dHq9Wi0+lo0qQJer2e+fPn06NHD86dO8e4cePo2LHjYy+z/7Mpy5vG7lqDrUcoP8fl0KO+bznFeqWbG9z0tzjc1Y0G1tYsLy5m1apVTJs2DTs7u3IPc2UtAG3btpVziDqdjkGDBsn519zcXPQaDdcrV6bKjRsAlArBiX37cB/2FXmHVmPMT0eXGIv3kAXk/vgOrspinn76aezt7VGpVGzcuJHo6GgOHDhAREQEYFEEuXjxImBpNXBzcyMmJoaoqCjc3d1JSUnh0KFDdOnSBXt7e44cOYKtrS0mkwkvLy8yMzNRKpXExsYCsG7dOjllcPN3sk+fPn9Jde6/mYqZ2CMg9sAepOLfXhcbdNhorNkcu5s5u5fh6emJwWAgPT0dZ2dngoODy92I1WqLSohSqZTLdatWrUpeXh6BgYFMnjwZa2tr7Ozs5AR0QUEBPj4++Pn5yV/wstDIHw0nju9YG2u1stwya7WS8R0toRKDwUDNmjXp3bs3y5cvp06dOhw4cIAZP87A3MFM9LfRCAT5+nzMZjO+vr40bdqU3bt3c/36dWxtbWnevDlqtZqsrCz0ej0pKSmkp6fL4UmFQiHbx9zvgHUnHDr6I6nL/3lLKgVJSQa6dXNg6dd+vDzclSVf+bF1W1ViY0xk5CXh6uaMtVbDmZH+nH9RhY0aHDSCqKgo7OzskCQJf39/TCYTS5Ys4YcffuCdd95h1KhRXLt2DU9PT5KTk8nOzubatWs0aNCA+vXrExMTI4eDy3jppZfkZf7+/vJspMw3a+/evXJBwN+FW/OmBTojE9efY+PpJHkdlUqF03PPIllZUXqT6a6k0aB9QG+6iIiIcgVEAHn5+WxxtwhT7y4sxEWppNRsJv37yZiK8zAWZKHxqkHq16MwZsYTGBh4237t7e3LDS5t2rTh2rVrsupNZmYmubm5DBw4kOzsbGrXrs2VK1e4fPkyJpOJOnXqUFpaKhcKVfDXUTGIPQIOrFlBsTFffh3gXo3tg5dwdMRaFj43g7S0NOzs7Lhw4QI7d+7EYDDIT5qhoaHyl8zV1VUOm7Vq1YpatWpx/vx5hg0bRklJCd7e3uTn51O1alUuXrzIqlWr0Ol0sqyQr68vu3bt+sPhxB71ffmwVwi+TtZIQHOVNWNKbEhaHMe6jyNJT8vkxo0bfPfdd9SrV4+8vDz69u1L35Z9ufTpJXRJvyXnNRoNzz33HBcuXEChUKDValEoFLi6uuLh4SFXgmVmZnL48OHbBqtjx44RExPzh84DwLa+B069aiJpTQghMBdnob+8HV8PJ4KDnQHYt6+I4S8nMvzlJBISjTg5O3F1ywdMaqXm3W3XefK7IooNkJqcSOcOT3D8+HFcXFzIy8tDoVBw9epVnnzySV544QW+/PJL6tatS1FREefPn+fZZ59l1qxZREREEBUVRefOnW8rXvjqq6/ueIO9m3L734Gb86ZavyAKzuwgZf9qPtp8mi1bLAVA/v7+xNnY4P3edL7V68kzmVH5+JDUti35vz6w3SprdTdu9pErw9XVlQPnztHl2lVMQrCrWnVerFwZFxsVKkMhVt618ajdkB92R9I4rCHvvPOO7Inn5eVFWFgY4eHhXL16FScnJ77//nsGDRpEWFgYBoOBixcvUqtWLb799lsuXbpEUJCl0XrevHk0bdqUjz/+mNjYWKysrNi7d6/8Xe7fv798jH369LnvvHEFD0dFOPERUJCVSZTNPhq5PYlKoeZCxlXe37sQCQlrd3sWf7OEjRs3PpDsz6BBg6hTp47cMFy1alXc3NyoWbOmLI3k5+cn68gBDBs2jE6dOuHj4/NQA1mP+r5cPJbKnu8uYNRbBpfiPD1KoUGlVHHw4EGsrKwICwtj8ODBmE9Z8gqGTAOlKZYQntFkZP369SgUCkpKSiwO2QYD27ZtQ6vV4uHhQWhoKDqdDn9/f1lRwt/fn+eee45XXnmF5cuXExkZSXx8PIMHDwYsFYll//+9m4Qx8RgFm6Yifh089AY9Vnk6qhT35mjeJtb+cIPPPqtFRkYDFiyIJDk5magdy2hXRfDeHiOvN9Ow+KSezjVVODsKFh5VEBERQf/+/cnIyJBFZEtLS8nKysLBwYH9+/czc+ZMcnNzZV3FEydOsHLlSuLi4pg5cyZCCKZNm8aiRYuYNWsWYWHlXdft7OzkWUJhYSF9+vQhOjqahg0bsnLlSiRJYvr06Q/UsvGouDlvqvWqgdqtCgUnNnLmSiS9Wlv+rt944w2efvppvlQq6TxoIG4rV1JzdwRxgweTk2MxsujatSt9+vRh06ZN9zS6vNUQ1cnJidGjR2NTVERXvWWGJllZ8eGns1h4B7WdHg+pSALc1VC1Ikz496BiEHsE2Lu6cSPTEguv69yGNlUb0chvDheNp+k6ZwoAYWFhDyT7UxaHt7Ky4sqVK/zvf/9j06ZNjBw58q4371deeYVXXnnlkZzTkU1XMOrLJ72VkgpHG3cWL17MBx98QGhoKPXr16dh1YYcm3QMJEj5LoXS5FJsvGy4fPkyrq6uskr4k08+iVKpJCIiAhsbG9atW4e3tzeenp7Ur18frVbLqVOnyql7zJ8/H7j/asibSZ/9uTyAyZjNmGYdZE/1RqSlRfPBByqSkiyVkEIIOnx6nO5+ejRKWHhCj94EOy4bGdWoCI1GQ9OmTdFoNBiNRhISEggKCkKj0WBjY0PlypVxcXFhx44dzJw5k/r168tmoyaTiSeeeIL3338fPz8/Dh06dF/ncPr0aWJiYvDx8aFFixYcOnSIli1bMnr0aKZOtYj8DhgwgK1bt8qitWUtG9u3b+fdd9/ll19+eeBrdzd8nKw5v305hdERKG2cUDq4YV2tD+brJ3nttdfknq/8/Hz54cPGxob4+Hh++uknlEoloaGhzJs3j6ioKCIjI1mxYgWtWrVi7969aDQaOZwKtxuiAqgkidAbCSBJqLy976pxWsaj0gGtMNP8e1IxiD0CWj07kJ+/nM+NolhuFFkGM5VGS4dho//wPouLiwkPD5f12xYuXMiIESPkm8LjpjD7ztV9hcV55OersbOzo02bNrzzzjukpKdgKrGEmIouFOHexZ3i2GKa/685PTv2ZMqUKZSWljJ69Gi+//57DAYD7dq1Izs7m3Xr1hEeHs7p06cZN24cK1asYOzYsY/kHIwpt4vjli23rRuCyWQiPj4eV1dX7O3tycrKYtjwkexf8SE1XRXEpJtRShDmo6Cylyvu7pYiDhcXF4QQzJ49mwEDBuDh4YFWq+X69ev069ePFi1akJCQQL169Th8+DDNmzcnPz+f999/HwBPT08SEhLu6xwaN24sh5hDQ0OJj4+nZcuW7Nmz57H5g92LXn6lnLxwAO8X5oLZTMryMdj61MTDxfqe293s0FzmxACWh7uymejevXuxs7MrN4iVFcJERESQl5eHm5ubRej36acBMALJQPKvBRV3wmg0MmfOHLno4o9iMBhwdHR86P1UcHesrKyoVKmSXCdwP1QMYn+AOylyXNY4sGXnFnSlempW8mbxF19Qp1X4I1XkSE5Olr3J/mi48H6xc9GWG8hc7b0Y1XkmC7e9TlU3G/q8MIqu3Tw5dmwhLVtasWWrEbNGSbHORL3+9ah8qjK71+xmaeJSqlatyoULF6hfvz5JSUlcunQJhULBmjVrsLOzo2HDhgCPTN2jDJW3N8bk38RxcXdhbN0abFerUJiMODk5MWbMGAAiIyM5efIkXy1ZgiHfTAMvBQuesiK1UGBjpabHqBm8e3Q6N27cYOzYseTl5TF37lw+++wzcnJyqFWrFmfPnkWr1RIVFcWWLVvIy8vDbDbj6OhISUmJXJ348ccfywand+Oll16iZcuWaLVa/P39iYyMRKlUMm3aNPr06cPQoUPlhvE/4g/2oJSpnxTs2s7QwHCsHVxYW1KMe1BzOoR4E3/6srzuZ599RlJSEi1btkSj0QCWisNatWpRq1YtWZ0jPj5eVueYP38+ixYtQqlUsnLlynIhxpsNUctaT1xdXeXq3btRWlrK5cuXqVWrFpIkodVqKS4uxtraGn9/f9LS0uQZuK2tLVWqVEGSJOLi4rC1taWgoACj0Yi3tzcqlYobN26gVqsxm814e3uj0+lQKBRy83pMTAw1atRAq9Vy+fJlWR3f09MTd3dLM/2pU6fw9PSUjXJr1KjxQDfsfzNCCLKyskhMTKTqAxQAVRR2/AHupMjx7iefEp+eSWpePv4NGnMlr0he/1EpcpTluh73AAbQrHt1VJK+3DIlpTioC9nQ3cBXX33OgvmjyM8vxNlFia2NhK3CjJ1kxc99fubC/gv4+vrK6h7ipiq1surDMsq+xA9TjXgnPMaNRbKyqEkkOdlxzs8dnUZtCUNlJMvvm5OTw+HDh3nxxRf5aefPmNW2xBcoeb6uhhdaV2FlvDurz+no27cvTZo04fjx4xw+fFgu2hg4cCBvvfWWPHjcXFH3/PPPA5YChaVLlwLlPb7uxldffYW/v/9tyydPniwPWBqN5r79wR6GW9VPlAYzY4waop9pxLONKxPs6ygrw588eZL169fj4+PD9u3bZRWW36NspjZu3DjOnDlz1xyZTqe7rwHsTtt5eHgQHByMUqkkIyMDDw8PAgMDCQoKwmw2l8u9CSGoU6cOlStXJjU1lfz8fHx8fHBxcSEoKEgW973X+QQGBhIYGEh6err8d202m7G1tSUoKAh7e3syMjIe6Dz+zUiShKur6wMby1YMYvdJSuomDh1qRcTuGuj1H/LTT5t48803OXDgAI6OjuzZs4cmTZoQEhLC7t27y1XW3Sm8c/DgQZ599lkAgoODy/UP/fDDD/cszf4zqNXEi3D7Bdgp0gEzdop0mtl9i0YqJiM1kfT0eNauy6R6DS3fr8kjPcNIq9Y2FBUVsWrVKkpKSlAoFAQHB99XPmvq1Kn3VY04derU+87xOHbtivd701H5+BDn7YL5pn47dWEuklGP0mSktLQUKysr+vTpg6enJyqtDRM/Xox2Rh4+0+LwqR4km2GePHmSJUuWYGdnd1czzKNHj952LK6uruzZs4c6deqQk5PDiBEjAEhOTiYgIIDMzEyaNWsmFwe0bdv2jgPAyy+/jJOTE507d2bv3r34+/uTkJDAjz/+KD8oLFmyhEaNGtGyZUuysrLk5W3btuXNN9+kcePG1KpViwMHDty2/ztxs/pJE7967Lx0gOLiEpI3x8gViU5WWuaMH8P0lwaiLC7AqC/FwcGhXGjQxsbmvioSf4/7GcBMxQaMGcWYDSauX7qGhERaWhomk6VatcxZ4MyZM8TExFBQUEBCQgKJiYkUFRWRk5NDQUGB3MycnZ1NcnIyOp2O5OTkcmocMTExlJZaBvjr169z/vx5oqOjiYqKIjY2Fr1ez9mzZ0lKsrQhpKSkYDAYsLGxQa/X3/H4/6s86MMJVIQT74tbLeA9PHOYv8CFhBt6Jk+eTLt27ViwYMFD2b+X8XewvC+jlvc1auWVt4GLHmlHvMkTSYpn4UI/ADp0sGPypFSGDXOlXz9nZrw3m27durF8+XJSU1MxGo1UqlQJNzdLb0+dOnXkgg03Nze6deuGm5ubXI14Lx7Uk6zMI27zs10tulM3IZlM2Fw6y6I1Fh3Ip59++raKT0CuRqxTpw5AOfPMW9c/d+4c1atXv+04FAoFL7zwAuPGjQMs4cKFCxdiY2NDXFwcW7dupXPnznh4eDBt2jRWrlzJt99+y8GDB+UZ2fz58+UHghdffJHNmzdz5syZckUfe/fuJTs7W54pPIqij5ubxkO8atM14Ak6LnsBNxtnGrVuRNrVy9RSGFh+NoYivQEXW2tKSkqJPfBbxEClUtGyZUumT5/OunXrHuvftKnYgCmnFGESlBr1eNm5UWIoRSEkMjIycHJywmQyUVhYiJ2dHW5ubhQXF5OZmSmHFp2cnEhOTqZatWoolUpcXFywsbGRlXPuhNlsplKlSphMJjnVUNZfVlBQgK2tLQqFQp6BWVlZlYtQVPDHqJiJ3Qe3WsBnZhrRaEqpXz+S8ePHy7btD2L/3qJFC3744QcAzp8/z7lz54B7W97f3Fvz+eef3/ML9UhoNxXUtyTs1dbQ/FUkyRKKS0428M47abz+hjuvjE5i/BsZqNVqnn32Wfbs2YOrqyvR0dH4+vqi0WiIjo5m69at2NjYYG9vz/fff4+bmxstW7Zk4cKFxMXF4e/vT1hYGDdu3KB+/fr3Lfp7L+xd3cq9drG1YXynNvLy5cuXc/HiRSIiIli/fr1cxg+WWfPQoUPLbX+39Q0GAy+++OJteQ5JkmjXrh1gaY5vrDWxY/qbrJryBkqFgiZNmgCwZs0aDh48eF/nVFb0oVAo5KIP4IGjAr+H0klb7vWrzQeyf9gqNo3+ilWrVlFTKsXVWsPrHVszpGUYpQYjEzq2ZteKpZw5c4YhQ4bg7+9PRkYGUVFRDB06FCsrq9ve5357x+5FSUE+huxi+YFFrVBhrbbCYDJgJdQUFhaSnp5OcXExRqORoqIiiouL5dJ/Z2dLD6GVldVtsySlUsn169dlhR2tVktRkSVtUFJSQmlpKZmZmVy7do2SkhIMBgP5+fkUFhYiSRKOjo4Ad5yBRUZGyiLej4v7iWK0bdv2dx8kbyY+Pp5Vq1Y97KE9FBWD2H1wqwX8tWt6Ro1MZtCg47z77rtMnjyZoUOHEhwcTMeOHe+rD2zkyJFkZGTIihxBQUE4OjqWe8ovq3Qro6wPLDw8/M8ZxOo+TXyDyQQvLgUkcPSDrnOhThc0GjcSExW8+24aEya4U7myhrnzqrNv/3dMnz6dt99+W97NqVOnKCwsZOfOnXz55ZfUqVOHzZs306FDB4YNGyavd+PGDdavX8+JEyeYNGkSNjY2nD59+r5Ff+9Fq2cHotKUvxmrNFpaPTvwnts1bNiQqKgoObf1e+zcuZO6devStWtXHB0dyc3NZfXq1Wg0Gnr06EHHtm1o0akzH3y/kYSsHBZu/RkhBGu+XESXLl3o3LmzHFIpK+xJSkri5Zdflp/ap0+fzssvv8zRo0cZNmwYQgiUSiUzZszg9ddf59lnnyU9PZ2FCxcydOjQhy76uKP6iVqBQ0d/wNInWUYlZ0fq+Xnz6c8HmL1xu/xdeOONN/jiiy+oX7/+PXUSN2zYQGho6H2HOm+mpCCf/Ix0FDfd1vbdKKb9tyfpuy6N7qtj2HoulcLCQmrVqoW3tzdms5nMzEy54rfs2kuSdNssyd7eHp1Oh9FopLi4GGdnZ0wmEzqdjqysLDlHGRgYiJ2dHUII0tPTsbOzA8o7WN+677CwMObOnfvA5/wgTJ8+vZyV0aOgYhD7h2Cl9S73ulEjG5Z8VYlvvmnM3r17eeedd9iyZQvW1ta8+uqrVK1alW3bthEcHEytWrXk6rs+ffrwzDPP0LhxY0JDQxk+fDjnz59n+vTpXLp0iZYtW9KtWzdiY2NZtWoVEREReHp6Mn/+fIKCgsjMzCQuLo6ePXvKlYp/VGLqvqnTBVxrkNfwWy5t8ST2mWlce64f2al5vDe9hGnvBFO9uhVGgzufzrKnXcs3eeX54ZzdG0n6F2fIz80jPz+f1atXEx4eLn+5x44dy7Fjx8qVmgcHB2Nvb4+7uzuOjo5yCOx+RX/veRqtwukwbDT2bu4gSdi7udNh2GjqtLr39Tt58iT79+8vZzVzP9StW5dx48YxduxY0tPTWbNmDVevXqUwLRndTT1Pg1qEYRaC3Zs3AJZwX1nRR1xcHJ9//jm+vr5cv35d7i0bPXo0ixcvpk2bNrfZ+pS1AHz22WdMmTLlkRR9lKmflM3IlE5anHrVxLa+Rf/z1lnu/wJr8tZTbZn4dHdWrVrFG2+8QUBAAFFRUZw+fZoZM2bIn2fbtm3l469VqxZRUVH3LOy4F4XZlvyfWVgGi60XM5h/IoeUQj0CSC82sehkPgcTdGg0Gry8vNBoNLi5Wdwmyj7j2rVrY21tiUCo1WqqVKmCyWRCpVJRs2ZN1Go1Qgji4+Pp27cv0dHRlJaW8tprr/H0008THh5OcXExQgjee+89Zs+ezaBBg5gwYQJGo5GOHTvSpUsXunTpIs/IbpYYO378+B0tiuLj42nVqhUNGjSgQYMGsrJLmWpInz59CAgIoH///ncMU94cxZg+fTqNGjUiODhYfhAqY+3atbflTe/23m+99RYHDhwgNDSU2bNnP/Bn9iioyIndB/eygC+rVNy2bRtgqUJr37797zairlu3jl69elG1alXS09Np3bo1sbGxjB8/nvDwcFatWkVgYCDvv/8+Li4u5URkX331VT777DP27Nkj55keJ+aCAlKmTOVGfj5jkpMY6epKXn4+JXo977yjZfv27Xy74kMcpAIqKzVkq3I5lR7D94c2YzaY0JXq6NKlC5GRkeTn55Oens60adOIj48vl+O6uWqxTKqq7P+PomqxTqvw3x20Hpb4+Hg6depEw4YNOXXqFP7+/vj6+jJp0iRmzZpFiJcbJy5dldf/5Kd9KCSJs1fjST13AaPRyOLFi+VZeqVKlcjLy6OwsJBnn32W0tJSdu/ezZQpU7h48SI2Njbs3btXNnbs168fzs7OjBs3jrS0NLl46GGxre8hD1q3UtYnadT/lju7n1nuo8b0699IsakAW5Ujc47dQH+LkLzOKFgVU8QT1WNQq9X31XNpb29PSkoKMTExGAwGlEolly5dYty4cXz66adUrlyZUaNGsWjRIiRJIjIykhEjRsj5y8TERJYtW0ajRo3Iz89n27ZtFBUVceXKFd5++23Zd6yMgIAADhw4gEql4pdffpHX8fDwYNeuXVhZWXHp0iWee+45OfR3t6b4u/GgzfJ3e++PPvqIWbNmlXuQ+rOpGMTuwJ36wCJP1mXrli3odKWEBDoyraoruRETeTcpkXiTia1bt6JSqVi1ahXp6el89NFHXLx4keLiYk6dOsWMGTMwGo306tWLESNGcPjwYXQ6HT169ODs2bP4+fkRERHByJEjsba2ZvXq1bz99tusW7eOL7/8EqPRKFdU/dlK6MasLK4qlbyenMwHXt4oJPBTqfixUWP6Z2UyYMAAfH19ccpRcSz9Kl1qtyUxL43Fx9egVCh5un5H9iQdZ8yYMZjNZvkp935zP/804uLicHjeAXUHNacWnaKo4Ld2CxsHxztu07FhPb47fBJra2tatWrFvn37UKlU+Pv7c/z4cebMmcNPP/1E3759GTx4ME2aNMHT05PS0lI6duxI48aN6d+/P1qtlhkzZjB27FjCwsJYtmyZ/B53U4d5WMoeDA6sWUFBVib2rm60enbgY39guBWlSoXJaERvsoRPUwvvXPmXXmAgJCTktuVlnmBgmYGVfc9UKpWscRkfH09GRgZjx45l/fr1BAYGUlhYyNGjR8v5kpWWllK7dm3UajV9+/aVw6p5eXm8+uqrd3RHLyMvL++ODuoGg4HRo0dz5swZlEqlrLoPd2+KvxsP2ix/r/f+q6kYxO7AzbOroqIievbsSVJSEkplFab2fYqZ8+bR03QGjSSRZzLR2dGJ8OeeY9zixXTt2lX2+WrTpg0NGjQgKSmJpUuXotFomDRpEgUFBXz66ae89NJLLFmyhIKCAlxdXbGxsWHPnj3Ur1+flJQUWrRowY0bN0hOTv5TKxVjD+yRb0ilai0Z+lJGJyUxx8eXGlotp0uKSTUa6XP0KOrq1Th37hydO3dm/sdzMJpNFusUhQJbjQ0CgcokMXHiRN577z00Gg1RUVGMHz+e4ODgx34ufwpRP0DEdMhLZLfCA42LmhI/y6xdU09D+tF0EjIsYdMbJiUqZXmnACEEzrWDaWy0SHLNnDmTF198kV9++QV3d3eWLFnCkiVLcHd3Z+/evZhMJjn/kpaWBlhuOH9VFSv8ObPc38POxZX8jHSEEOhNOjztVKQW3j6D93G6t7rI7+Ho6EjlypU5ePAggYGBmM1mnJycOHPmzB3Xt7W1lf9f5o6+YcMG4uPjZfHgm7nbOrNnz8bT05OzZ89iNpvLFcfcHO7+vXynTqdj5MiRD1RNfa/3/qupyIndxMVjqXzz9iHOri1kw9qtDB0wms8++4wqVaowffp0bG1teX/BArIMBno7OrG5ajWMQuAhQfiZs1SuXJmioiJcXFyoWbMm169fZ926dVy/fl1ukhwxYgSnT59m+PDhJCcno9fr6datG506dUKn07Ft2zZSU1OxsbFhzpw51KpV63crFR8lsQf28POX8ynIzAAhKMrJRq3V4GxlxakSy415bW4ur7q7s71FC7Zv346npyfvvPMOk54aTffAdkxoPZQjI35Ao1Tzy5DlfNj3Lezs7Ojbty+5ubn07t2bTz75hC1btpCfb1H/P3jwoGwcCZYn3rJQ6eDBg8tpKP6thFajfoAtr0JeAiBYpjEhbml1kbQSCUkJ9OnTB6G1wsnZGZVKTb6uFKPJckNY8eMGoqOjEULQvXt38vLyqFGjBkIIFi9ejLu7O5mZmQwbNoyaNWuyd+9ejh49Snp6uiy0LISQe8RatWr1SHrE/klY2zvg4O6B8tew9MgmHlipyt/ibrYW+qNoNBo2bNjAihUrWLVqFQ4ODlStWpW1a9cCloeSs2fP3nHbvLy833VHv9s6eXl5eHt7o1Ao+Pbbb/+w6WbZgPUg1dR3e+/HdR96ECoGsV8pU20vzC7F08mPCT2+wJRkxdpFc1j7/SoGDhzI+PHj2VK5Cg2trfkuJ5vu166RbzbzTU4OXQ8fIiEhAUdHR4YOHcq5c+eIjY2lUaNG3Lhxo1zprlqt5pNPPsHHx4eaNWuyf/9+Vq9eTWlpKZ9//jlBQUEUFxdTtWpVuVKxVatW5byLbq5UfJQcWLOiXG4DLDmp59s0YnN+Hlvz8ygwm/GytsFj3NhyXzLrYDdQlL+D31zF9q8kYjoYfsuVZqiUGLIMFF+2VI7mn85HoVWg8dIwd+5cYmJiyCsoROHgRIqdGwKoHRDA5MmTLT5cTk4AeHh4EBcXJw9C1apVw2g08uGHH1JYWIhGoyEsLIx+/fpx9epVwsLCUKvVfPDBB5w4cQJPT08qV65M7dq1qVWrFnl5eXdUjvm3YW3vgHuVqnhVr8nA9vX5qHdd2VrI18maD3uF0KO+70O/j62tLVu3bmX27Nls3ryZ7777jqVLl1KvXj2CgoLYtGnTHbebMGHCXd3Ryyoj77bOyJEj+eabb6hXrx4XLlwoN8O7XyRJwsnJ6Q9VU9/pvevWrYtSqaRevXoVhR1/NTertucWZWKrdaBhraewURn54fBcCnQWe5Tm1jZE/GrEJxBoJYn/2dnxYcMwwk5GotPp+Pnnn7G2tqakpIRly5bRpk0bDhw4gL+/P3PmzJH7R3744QeaNm1KpUqV8PX1lfMbS5YswWQy0bNnTzw9Pbl48SL+/v4sXbpUnp08SsX6m7m5XPpmzNZWfNkwjMGRJwj38mauvpRlU6fSuXNneR1tFQe01Zx+c1VWSDh0rmYpCLjzg+k/n7zEci/djSaueWnIisgiaWkSalc1SlslGqVFQzAuLg4XFzUfzTQwb67FCqSsf8jX1/c27UiFQkGHDh348ccfkSSJ5557jm3btlGrVi20Wi3PPPMMSqWSunXrMn78eGbNmlWuN2zy5MnUrVuX/v37P1Zh4L8rZdZCjwp/f39ZDd/JyamcncvNkYQybp1tNWvWrFw+qczZIisrS25Qv9s6NWvWLPf3MXPmTKC8EzYgRy1u5eb3KLM+upW75U3v9t5qtZrdu3ff8f3+LCoGsV+5Wew2OfsaG48uRpIUlJQW0MJXQd3KziyNVnEiPx8Bv5byCkqE4HhJCc8k3KCwsBCFQsGxY8f45ptveOGFF3BzcyMvLw8hBDNmzMDW1pY9e/bw8ssvI0kSCoUChUKBwWCgtLSUmTNnyvHmFi1asGTJkj9V/Nfe1c0SSvwVuSnYzZ2G32/l3K/Lb+5oKfsy3Oz1BZDw0W/iuw/iA/aorDP+FBwr/RpKtPBCXj6RSgm/ly1qJjUyGtI0oSt2pU7ciPwFDw8Vq1ZbftesuYIjRyT0+kwaNmxIzZo15TLr0tJSFL/KZO3bt49WrVpx8uRJLl68SLNmzYiNjZWFZNevX8+WLVtwcHBg/vz5nD9/ngEDBlCtWjV0Op2cL3scwsAVPDybN29m0qRJfP3114/tPYYMGUJxcfHvanb+E6kIJ/6KnctvidFqNSvz8oihvDTqeVq1ac6Oy2a+PJyJo6MjE6dOpbKHB4VCkGw0olUo8KtVi1OXLmFvb09YWBhbt26V7SUGDRpEq1atUKvVvPTSS3zzzTeoVCpat25NSUkJQUFBXL9+naSkJDQaDYMHD6akpARnZ2dWrlxJo0aN/lTx3z/aFPyf5RZVkydKdHiYzHirHamR0ZC2V5/DrtQZkLCvvO22zUtLBZmZ1wCLMHDZA8yUKVPkQQx+S7YfPnwYpVJpeYgymxk5ciRPPvkkM2bMYODAgQghcHNzw2QysW/fPgB5/Qr+nnTr1o0LFy6U05l81Hz99ddERET8KxXzKwaxX2nWvToqjQKdVRoFDpcwq0pBgtCwerz+xliq96pLIoks2LeAhMwMhrz8Mj/v28ebU6bgGxhIkyZNMBgMXL58mZiYGOzs7JAkiU2bNnHy5EnGjRuHj48Pn3zyCSaTiW7dunHt2jWKiooICQlh//79PPfcc5jNZho0aEBhYSE2Nja8/vrrf+p1+KNNwY8ao9FI//79qVOnDn369KG4uJiTJ0/Spk0bGjZsSMeOHUn51S/sbkULxcXFPP300wQGBtKzZ0+aNGki99WsXr2akJAQgoODefPNN+X3tbOzY9KkSdSrV4+mTZvK1X9XrlyhadOmhISEMHnyZFmFgbpPW1RMHP0ACf8qVbi+cyU/9ztIj+yhfL9nDik58QC8MfE86elGzsfoePWVJJZ9nYOVlcSlS8U0btyYNWvWoNPpiIqK4uWXX0av15Oenl7uusjvi0WiLC8vj40bN7Jw4UK2bNlCgwYN8PX15fDhw6Snp5dzEHj55Zd58803ad++PUlJSf/K4o4K/ntUDGK/UquJF+H9Ayh2iAfFb47GBQUFqNQ2tKzVAdcnXclNzbXkOPw1TJw4kS+++IJt27axbt06GjVqRM+ePdHpdNjZ2aHRaFizZg3jx49n6dKl3LhxgyeffBKw9J6U6SQqlUqysrLYunUrO3bsICIigtq1a9O2bdu/pGy6Tqtwhi1YxutrtjBswbLHOoDFx8ffsdQ+Li6OkSNHEhsbi4ODAwsWLOCVV15h3bp1nDx5kiFDhjBp0iR5/Qexu0lOTubNN99k9+7dnDlzhhMnTrBx40bA0iPYtGlTzp49S+vWrVmyZAkAY8aMYcyYMZw7d07ux5Gp+zSMi4ZpuZZ/61oMGwuzS+nf5g28nf0BEEKBj4+KTZvyKSw0U8lPhZ2d5Svo4uJCfHw8CoWCN954A3t7ewCeeOIJeUZWWFgIWHJnffr04dy5cwwbNgwXFxdiY2OpUqUKTzzxBJcvX6ZVq1bs2bOHwMBAfv75Z7y8vLC3t8doNHL69Gk2bdr0ry3uqOC/RcUgdhO1mnhhUpSvzEtLS2PJkiV8u+BbMjZl4NLWBadWTiyZvYTs7Gw5zFMWwomIiLhtv+PGjaNZs2aARVrJy8uLffv2Ua9ePapWrcqZM2eYPHky9erVQ6vV4ujoyLPPPsumTZvkBGppaeltT+X/NopOp5Py0XFSZh7Hx9GTUJuagMWTa+fOnURHR9O+fXtCQ0OZMWMGiYm/FVU8iN3NiRMnaNu2Le7u7qhUKvr378/+/fsBS/l0WV7q5n0dOXKEvn37AtCvX7/bjj0+Pl6W/CmbPartBZ9vfo2o+MO89/1gCvIgLc2IXi9ITzdSWGjGaBRo1EpKSkpo1qwZTZs2lVVMOnXqxP79+0lPT+fYsWP069cPIQTR0dFs27aNwsJCFi9eTFBQEP3796dfv37s3buXIUOGEBkZSYsWLdixYwcJCQnyefwXizsq+HdTMYjdQlmlWBk1atRgxIgRDBg1gOrvVAcFFEQVIDQWy4b169czbtw4goODMZvNtG7dGrAMamUl8VZWVnzxxRdyOFGj0VBaWkpgYCAKhYJ69erJWont2rUjICCA3bt3061bNznUVdZP9ti1Ev8CjEYjzzzVm9D2jXlp2QR0hlIMRgPtenSkfp16TJgwAbVaTVBQEE5OTnTs2BGFQsG+ffvkkJgQgqeffppWrVqRnJxMkyZNyM3NBX4LHcbGxpYTWS0LHU6bNo3Vq1eTlpaGWq3m6tWrNG3alPHjx3PixIlyIbx7cevs8UrpfqRfv2FpuQloVDa4O7tho1ZhNAq0koLnKjthMJhwtLbm0KFDJN/kRN22bVskScLHx4cOHTpw6tQpzGZLlKB69eo8+eSTbNiwgcjISBISEjAajZjNZn7++WdiY2PZsGEDDg4Oj931uYIK/koqBrFbaNeu3W3JT6NkJNrZUilnW8sWtYsalUJFSUkJV65cQaPR4OLiQl5eHmq1mnfeeUfetkxMs0aNGqhUKnr27MmsWbPknpDCwkJiY2NlZW9ra2scHBxITU0lJCSEwYMHM3fuXNmg799IXFwc/Sp3Ys9L32KnteXH6J1kFGXTN6gTJTmFODg4kJGRQUZGhtzvtG7dOlxdXeWQ2Nq1a3F2dubw4cM4Oztz8uRJ6tWrx7Jly3jzzTf54osvMBqNnD9/nqysLPbt20dRURGNGjWidu3aNG3a9LbQ4WeffSZr6zVt2lTWuFuzZg3w28yxhmsV0r84QyUvX9l1wL9JJ76K+IVMzOgU4GzngVZrzaKG3XnmogPCCJprEupkBQIwFhTIDctlfxtlwq/p6ekcPHgQJycnOb/l6urKoUOHeP3118nIyJCtasxmM1qtFjc3NwICAmQjxgr+HixfvpzRo0f/1Yfxr6JiELuFm200ADS2Gs65nyPR3hK6KogqwMrFijW71xAdHU2nTp0YPXo0J06cIDo6+jZVcaPRKFu2+/v7ExUVRVxcHPv370etVhMfH4/RaJQHzvfff5/IyEiioqLYt2+fLPj7Z1Un/hlsu7qNDus6UPebugzYPgB3b3caOgUA0CuoPccSzyIhMS1iHhfTrnLy5EkcHR1Zt24d165dY9OmTXTu3Bmj0cixY8c4fvw43333naxDWWaLceTIEc6fP09+fj49e/bExcWF1NRUXn/9dQYPHowkSbz99tskJiZy4sQJvvrqK0pKSjhw4AB9+/bl4MGDxMbGUlxcjI+PD5999hl169Zl9OjRSAIadWpOx88HYDQbMeXrMRSW0jikIVVqBvDpvEVkxx6lWCHYYa2nQAEKGxXuV/bfdj2sJYlCg4FmzZqh1+tRKpWYzWZWr15NWFgYPj4+DBgwALPZLFclrl27lgYNGnDgwAHs7OxkBYWyB6Xg4GCeffbZf2U12gMR9QPMDoZpTpZ/o374q4+ogkdMxSB2B8psNKZNm8bb49+mQUAfJKMzQoC1tweGWNi/eD8HDhzA0dHxgU0IT5w4wQ8//MDZs2e5ePEi9erVk9f/4YcfaNCgAfXr1ycmJobz58//qef+uNl2dRvTDk8jpSgFgSC9OJ08fR56u9+KaVxtnKnvE8jPQ76mmnsVunTpQnJyMjNmzCAkJIR33nkHZ2dnMjIy0Ov1XLlyBVtbW1588UUGDhxInTp1MJvN1K1bF3t7e5566inZN8rNzQ1JktixYwc2NjbUr1+f9PR0FAoFarVangU5ODiwbNkyvL0tNjzR0dG89dZbbNiwAYPBgFFv5H/VmxOTfpnqLpUBSCvMpLVnAyq9uJDS3FTkWCJgzE8nt6AYpacn2wryy33xJMBFq+XIkSNy4c/mzZtRqVSykej27dsBWLx4MWazGZVKxdatWyksLKS4uJinnnpK7sMbOXIkV65cYfv27bi5uTFt2jTA0sha1vrxKAWA/7bcIglGXoLl9UMOZGX5z8GDB1OrVi369+/PL7/8QosWLahZsybHjx+/q53KzWzbto1mzZqRmZkpy4XVq1eP3r17y16BgwcP5tVXX6V58+ZUq1ZNlogqLCykXbt2NGjQgJCQEFkhJD4+njp16jB06FCCgoLo0KEDJSUlt733v4mKQex32Hg6iTV73Mm/9CaFFz6itOBdPJ5fSKm9L5MnT2b69OmMHDmSdevWce7cuYcyIbx27RqzZs0iIiKCqKgoOnfu/JeKuj5q4uPjeabtM+hM5c9Jn6XnvYKFSGoFG8//QgOfQLKKc4nOuMSVjOsMGzaMdevWkWNQcSLmCi++PhVdoxdw8fDCwcGBSZMm0alTJ7KyssjMzJRddr/77jsiIyP5/vvvyc/P58MPP8TKyorc3FxSUlIwm81s2LCBpUuX8sknn1BYWEiVKlUIDg6mpKSEhg0bMm7cOFQqFSqVStYtBPj5ha/xtvdApVDiYecKgEJScDk5nhOfDEJhbQ+K34R+VS6VMJYW0f/qFfIBlXSTPJckoXBwKHdNnJyc6Ny5M/v3779NGkilUtGjRw+Cg4N5+umn0Wg0j/iT+pdwiyQYYHkdMf3O6z8Aly9f5vXXX+fChQtcuHCBVatWcfDgQWbNmsUHH3wg26mcPn36NpNYgA0bNvDRRx/JDxq9evXixIkTnD17ljp16rB06VJ53ZSUFA4ePMjWrVt56623AEuefcOGDZw6dYo9e/bw+uuvy6HmS5cuMWrUKGJiYnBycrrN6uXfRoVix+/wyc44Sgy/CW0aC7IQ1vacUAUzcXw1vvrqK6C8mObvCdS2aNGCH374gfDwcM6fP8+5cxYdjLKS+5sFf8vkZMqENv8M/7DHidF8+0Cu8dKw8cDPHE45TTVrXwY37E2TKo0Yt/0DwDKzaNNrEJfsQ9EXbwWzkfNr3seYk4FSIZGYmMiYMWNkJZDAwECUSiWFhYVcv36dsLAwoqOjWbx4sdxzlpmZidlsxt3dHScnJ7KysqhWrZp8czp8+DCnT58mLCxMLuzo06cPX375JTY2NtSoVgM3WxemRczDaP7t72Ne/3fJUOpISMsiccFAvPp9hDEvDUmhoNn07Rx66wnytmwhffbnGFNSUHl7M33cWBx/tcGYP3++LBv07bff8u233952vW5WRbmVx2W38o/kFkmw313+AFStWlW2cwkKCqJdu3ZIkiQbuN7NTgVg9+7dREZG8vPPP+Pw68NLdHQ0kydPJjc3l8LCQjp27Civ36NHDxQKBYGBgXLfohCCt99+m/3796NQKEhKSpJ/V7VqVUJDQ4H/RhVqxSD2OyTnln+SM2TEk753GSmSxLuVXfniiy/YuHGjXDp/v2KagwYNIjAwkICAAIKCgnB0dKRmzZqy4K+fn59cJAC/Cf6W5cb+qSiFkoRFCeiu69D6aqk0tBJ+w/1I/zad68k3UHk5si49F0fbquiFhISEEg3LvpiLVb2nsKocgrEgC6WNI8asRGwqh5CXl8fixYvx8fEhPz+fyMhIUlJSsLa2JiYmhkqVKnHq1CnatGnDTz/9xOXLl7G3t+f7779n0qRJPPvss3h4eODi4kJ6ejoeHh5ygU2NGjWoVq0aV65coU2bNrKihkNHf/JW55Y7N2u1lt3KaMZ37MyIyTPL/c6Ykyyrpzt27SoPWhU8Rm6RBCu3/CG52frkTgau97JcqV69OlevXuXixYtyeHfw4MFs3LiRevXqsXz58nIPIze/V9ls67vvviMjI4OTJ0+iVqvx9/eXoza32rL828OJFYPY7+DjZE3STQOZdbWGWFdriK+TNYfeegKAsLCwBxLTtLKyYuXKlVhZWXHlyhX+97//UaVKFeDuuoKPS/D3z6Y4uZjaL9ZGXV1N4tJEsiKyKDxVyPufvs8nL82hSc1ObDnxNc+3HQ+AQGAuVmMo1aHbtxzHZk+juxGFMTsZlGoKU+M5Hl9A69at5eubk5Mjz8QWLFhAo0aN2LRpE9u2bSM4OJioqCgkSaJx48bExcWhUqnIzc2VKwNv3LhBSUkJOTk5DB8+HJVKRbVq1ZgxYwaHDh1i3759NB3YDh9HT/Qmi/GifxV/3NzdWbhhKSWr5tO0fgu2W9siAT5uzohKlR+pEG0F90G7qZYc2M0hRbW1Zflj5l6WK1WqVOGTTz6hV69erF27lqCgIAoKCvD29sZgMPDdd9/J295r/x4eHqjVavbs2cP169cf16n87XmonJgkSX0lSYqRJMksSVLYLb+bKEnSZUmS4iRJ6ni3ffzdGd+xNtbq8iaGD+tJVCbEWa9ePXr27MnChQv/1XmNqKgoZs+ezeeff46TkxMTW0/E29Yb52bOGGINmFPNLJ64mNTsBJZHvM+pK3v56udpmIWl2ONqcizmknwwGTAbdCAEwmxEobEGfQmSJPH8889jY2ND5cqV6dOnD56enoAl91BWZq5SqTh27BgA7u7uCCEwGo2Ulpbi4OCAEAIhBK1atQJg1qxZNG7cWM6jlYUVq1WrhiRJCBsFkiSx9twO1C9VQ22t4ejRo0RFRfHMk23o1LYF1z7qzNQmat58bcyffdkruEUSDEc/y+tfFVUeJ/eyXAEICAjgu+++o2/fvly5coX33nuPJk2a0KJFCwICAn53//379ycyMpKQkBBWrFhxX9v8ayn74v6RH6AOUBvYC4TdtDwQi/mGFqgKXAGUv7e/hg0bir8jG04liuYfRgj/N7eK5h9GiA2nEv/qQ/rHcPbsWTFjxgzxzjvviDFjxghHR0cxY8YMcfbsWRERESF69OghmjZtKq5duyYA4WrvJea/HCGa1u4knG09hEKhFHPHbBHfREQJSWsjlLbOAkkhUKqF0sZRVK5RW1hbW4tGjRqJoKAgUatWLTFs2DCRmpoqnJ2dhSRJonXr1kKr1Qo/Pz9x7NgxoVQqhaurq3jrrbeEWq0WgPD39xc+Pj5CkiTx3nvvCY1GIzp37ixWrVol1qxZI5o3by5sbW2Fs7Oz0Ov1Qgghli5dKpRKpRBCiC1btohXX31V1K1bV4SEhIhWrVqJS5cu/ZWX/l/J+fPn/+pDqOAxc6fPGIgUdxk3JPEI1K0lSdoLvCGEiPz19cRfB8gPf329E5gmhDhyr/2EhYWJMoHWCv4dzJ49m7y8PAByc3OZM2cOQ4YMITg4mBOR23F0uMzmzYmMHFmDOZ8mIpkdcXeoRHL2NfKKszALE/7ePqTn5VJcXIykUKGwcUAhSVTy8SK8RRO+/vprVCoVtWrV4sqVKyxfvpxVq1axdetWrK2tkSSJoqIinJycsLe3Jzk5GZVKRaVKldDpdKSlpaHRaKhWrRrR0dG4u7tTVFSETqfj/PnzdO3alfT0dIqLizEYDHJC397entOnT8vl0BU8fmJjY6lTp85ffRgVPEbu9BlLknRSCBF2p/UfV4m9L3BzRjXx12W3IUnSMEmSIiVJiszIyLjTKhX8gykbwMooU5oYP/4NEhOO0aUrTH3Hk2+/vUp2fiHpeUlUcfXg+fA3UPxahm4nsjh58iR2dnY8+0xfrNFjozCSHH+Zffv2Ub16dRwdHUlOTqZatWqMHDmSAwcOYG9vT5UqVSguLkahUNCuXTsMBgNCCGrVqkVJSQnFxcWYzWaKi4tJS0vDzs4ODw8P9HpLrmvcuHE89dRTPPXUUygUCpRKJZMmTSIqKopBgwaVs0v5T4d0KqjgL+J3BzFJkn6RJCn6Dj/dH8UBCCG+FEKECSHC3N3dH8UuK/gbcbMWpZOTE6NHj6ZTp05UqqThnWnuWFkpqFFDy5SpnphM4OyoYFv3fRQnTaJPHUsjcGxaKaGhobLhaFBQkFyEodPpyM7OxtraGp1OR/v27XF0dGTZsmXY2tpy6dIlatSoQbNmzdi4cSMKhYLg4GDs7Ozo2bOnXE0qSRJNmjQhICAAV1dXzGYzAQEB7N+/n2bNmhEfH49KpUKr1fLZZ59hZWXF5cuX5fPLzMys8OyqoIK/gN8dxIQQ/xNCBN/hZ9M9NksC/G56XenXZRX8x7iTFqVKpcJsNvHBB+kMeSGBd6elUVpqxs1NSW6+Gav38tly0UDnmira+FuKahwcHNDr9RQWFvLuu+/KEkx5eXlYW1uTlZWF2WymT58+lJaWMnToUIqKilCpVGRlZTFz5kzGjBlDcnIy0dHRXLp0iR07dpQrR963bx+JiYmyJcrGjRvR6XTMmDGDli1b4uDggCRJbN26lSpVqtCgQQO5RPro0aOMGjXqz7uwFVRQAfD4wombgWclSdJKklQVqAkcf0zvVcHfmFu1KB0dHWnXrh0JCQa6dXPg62V+2NhKHD1agrOzEiFgRQ1f5nayZvr+UjrVskatUdOuXTtKSkpYsmQJ06ZNo6CgAF9fX9ngcefOnej1ejQaDQMHDkSn07Fr1y5cXFzIzs7mmWeeYceOHQC89tprpKamYjQasbe3x2QyoVAoWLx4MSkpKXKjqa+vLx06dGDy5Mk0aNAADw8PzGYzjRs3pqCggCFDhvDpp58C0KVLF1599dW/5iJXUMF/mIctse8pSVIi0AzY9msBB0KIGOAH4DzwEzBKCGG6+54q+DdTt25dOoWF4pNxHfOx3ZxetRRPD0dCaloah//3P3siI0tISDCgVsDQyykM2lhCnl7BK9MXEhRcl/Pnz1OvXj3Gjx/PlStXAEu+rUOHDgDUqVMHjUaDTqdj7dq1lJaW0rp1a6pUqUKlSpVo374958+fR6VSsWPHDurVq0e9evVkZwCTyST7hbVs2RKAkydPcvXqVV544QVefPFFGjVqhLOzM9euXSMqKgo/Pz9q1KjxZ1/OCiqo4CYeahATQmwQQlQSQmiFEJ5CiI43/e59IUR1IURtIcSOhz/UCv6pxB7Yw89fzqcgMwOEoCgnG32RHvGzB8osQICtUqKmnZazNQOIbNMWnVGQVmDEpskA7OzsWLp0KWfPnuXQoUOyPUrLli2pXdvSr+fm5kZQUBC2trbMmjWLvn37otPp+PbbbykpKZHlfbRaLdHR0URHRzNw4EC5h8fFxUU+3h49eqDVamnVqhWnTp2iqKiIoqIi2rVrVxE+rOChqLBiefRUKHZU8Ng5sGYFRn15x+ycohJ+vmTLC3scOBqfRWO1LWvzcjnjZOKpcWMxGAxcvHiRoKCgu+73bhqUZTYkZ86cQaFQ4OfnxxNPPHHPYyzzC3vmmWdkvzCwzMbat29PrVq1cHJy4uuvvwaQ3Z8r+Huz7eo25pyaQ2pRKl62XoxpMIbO1Tr/1YdVwSOkQsW+gsdOQVbmbcvc7W35JfoC3bMyKdBo6O/iwrx6oczTqGk9eTKhoaEcPnz4nvsdOXIkGRkZBAYGMnny5HIalMuXL5dNJjt06ED37vcupv38889lv7Cbqw7r1KlDlSpViIqKYv/+/RXhw38Qt9r+pBSlMO3wNLZd3fZQ+/07WLFUcBN364L+K37+rood/2auXbsmateuLQYNGiRq1qwp+vXrJ3bt2iWaN28uatSoIY4dOyaOHTsmmjZtKkJDQ0WzZs3EhQsXhBBCGI1G8frrr4ugoCAREhIi5s6dK4QQ4t133xVhYWEiKChIDB06VCwaMUjMerqzqObuIlrVrCoqOTsKD3tbMbF3F9GzZ09Ro0YNMWnSpAc+dqPRKEpKSoQQQly+fFn4+/uL0tLSP3QdioqKhNlsFkIIsXr1atGtWzchhEWJY86cOX9onxU8eh5EsaP92vYieHnwbT/t17Z/qGO4du2aUCqVIioqSphMJtGgQQPxwgsvCLPZLDZu3Ci6d+8u8vLyhMFgEEIIsWvXLtGrVy8hhBDLli0To0aNEuvXrxctW7YU2dnZQgghMjMz5f1PmjRJ/i4NGjRI9OnTR5hMJhETEyOqV6/+UMf+T+BBFTsqwokVcPnyZdauXcvXX39No0aNZG+kzZs388EHH7BixQoOHDiASqXil19+4e233+bHH3/kyy+/JD4+njNnzqBSqcjOzgZg9OjRTJ1qEVkdMGAApqAAVAUFACgVEmPbt+TQ1QS+2nuEs+fO4eLiQvXq1Rk3bhyurq73fdzFxcWEh4fLDcwPo0F58uRJRo8ejRCiImz4LyG1KPWBlj8If7UVSwW/UTGI/UeJiooiIiKC69ev4+zsjBAChULxQF/IX375RVZ5h9+KI/bs2cPHH39McXEx2dnZBAUF0X3YaBbuPkSQrxf2bu50bRZOwbafZOfkatWqkZCQ8ECDmL29PY9KpqxVq1acPXv2keyrgr8HXrZepBSl3HH5w/JXW7FU8BsVObH/IFFRUWzZskWWhJIkiS1bthAVFXXPL2R0dDRbtmy5p9u0Tqe7o9N1nVbheNcMYPDMOQxbsIyqoQ1vuxH8nvN1BRU8CGMajMFKaVVumZXSijENHr+jwO9Zsfz4448MHDiQmJgYgNusWCq4fyoGsf8gERER5cIbAAaDgYiIiDuuf7cvZPv27Vm8eLE8+GRnZ8sD3M1O1/8Erl+/zrPPPnvPdQ4ePCibYlbw96dztc5Maz4Nb1tvJCS8bb2Z1nzan1Kd+LitWCr4jYpw4n+QW0V5f2/5hAkTGDRoEDNmzKBz599uAC+99BIXL16kbt26qNVqhg4dyujRoxk6dOgDOV3/mfj7+xMZGYmbm1u55QkJCWzcuPG+9rF37140Gg3Nmzd/DEdYwaOkc7XOj3zQ8vf3Jzo6Wn5984Pdzb+7ePGivLzMNHfw4MEMHjwYgPr163P+/HkARowYwYgRI257r1tncYWFhY/iFP5VVMzE/iPcXBa8cOFC1q9fz9WrV1m/fj0mk4mkpCRyc3OJi4vj/fffp3nz5pSWlhIdHU3jxo3p1q0bBoOBzZs38/rrrwPwwQcfcODAARQKBU2aNJGbfw8ePEi3bt0oLS3l6NGjdO7cmV69epGUlCQPFG3btmXr1q3y8e3du1fODzwsQgjMZvMDbdOvXz9KS0uxtramUaNGNGrUCCsrK6ysrBgzpnz4ae/evfzyyy/UrFkTKysrbGxs+OyzzwDLwO7j44ObmxsajYbGjRvL2w0ePBiNRoOdnR0BAQHUrVv34U+2ggr+69ytbPGv+Kkosf9j3E+Z/IYNGwQgateuLerWrSs8PDxEaGiomDx5sqhdu7bQaDSiWrVqYvbs2UIIIQYOHCicnJxEUFCQaNGihejdu7cwGAyiTZs2Yvjw4aJhw4aiZs2a4vjx46Jnz57C3t5e9O3bVwghRJs2bcSECROEEEJ8/vnnwtvbWyQnJwudTid8fX3LlRM/ymtQq1YtMWDAABEYGCgGDx4sGjZsKAIDA8XUqVPl9apUqSIyMjJEcXGx6NSpk1i0aJGoUaOG2Lx5s9BqtcJkMglHR0fh5OQk1q1bJ4KDgwUgGjduLDZu3Cg0Go3w9PQUWq1WaDQasX//frFt2zahVCpFTk6OePHFF4VKpRLXr18XOTk5QqlUiqNHj4qTJ08KpVIpLl++LIqKioSDg4MICQl55Nfh306FKea/nwctsa+Yif1LuHz5Mq+//joXLlzgwoULrFq1irXrxvPii2rGjQsnP/9d/P29uHDhAp9++qlciXjq1CnAEgbZu3cvu3btshhWnjiBp6cn0dHR3Lhxg7p168pViA4ODkRGRtKyZUtatGhBbGws9vb27Ny5k6ysLAC6desGQEhICEFBQXh7e6PVauUqxMfBpUuXGDlyJDExMXz66adERkYSFRXFvn37iIqKktcrLCyka9euPPfcc7zczJnnq+exa0ZvMOn55cspgKWs+YknniAqKoqqVavi5ubGqlWrkCSJ4cOHo9Fo+OCDD2jVqhVPPfUUtra2/PLLL4Cl+qxy5co4OTnh7OzMyZMnWbt2LVWqVKF69erY2Njwv//977Fcgwoq+K9RkRP7h1J0Op38nfGYcktJF1n4+1Qu17fSsKETcXGT8a1USFqakStXk7h+PQNHRzt0OktfVbVq1Zg3bx75+fnY2NjICWVvb28SEhLQ6/XUqFGDtLQ0PvroIz788EP0ej1VqlRBp9OxevVqAK5du4YkSVhbW5OQkMDRo0fp1q0bxcXFKJVKQkND6dWrFzt27ECtVvPcc88BsGLFCj788EPOnTvHM888I+cN/ug1qOxdiaZNmwLwww8/8OWXX2I0GklJSeH8+fNy+K579+5MmDCB/iFqojbOxVSnOyu/+RGTyOezBV/iam+pmkxMTOSZZ54hMTGRvLy8+1bruLlXTaFQyAabFVRQwaOnYib2D6TodDq56y9hyrXoEZry9ah0luVguXHm5OzAbC5BoQCTSbBjewFCwLx5tYiNjcVkMnHw4EGaNGkie3OFh4fj6elJ8+bNkSQJjUZDcHAw1atXR6VSkZOTQ3BwMHFxceh0OiRJwsXFhdTUVLRaLUqlEqPRiFarpXPnzpSUlODr68vx48dZsGABjRo1oqioiHnz5jF8+HC6d+/OggULiI6OZvny5fIs7o9eAyuDiqLT6Vy7do1Zs2YRERFBVFQUnTt3LtcW0KJFC3766SfObv+aLeY2KBx9LZYsZjPHE420rGrDli1bGDVqFN27d8dsNvPxxx+XG4z8/PxYvHgxADt37qSoqEhW1L8TvXv35vr161y7dg2dTnfXStAKKqjgwagYxP6B5O+MRxjKFy4IYVlehsGQU+73JTqBUgm+lfL59L13QQgMCdfQ5qSj0WgwGo289NJL3Lhxg9WrV6PX6zGZTFy5coUbN27IShbnzp3j4sWLODk5YWNjQ3p6Or6+vuj1ermU2Gg0cuDAAUJCQkhNTUWtVuPt7Y1CocDa2pq0tLSHDjPe6xrk5+dja2uLo6MjaWlpso9YGdOnT8fZ2ZmRP6ZgwGLY2aRJEwBy8/LYc7mEqlWrcvDgQV577TVGjRrFwYMH5e3t7e3p0qULQgisrKzo1q0bbdq0kdUX7kRYWBjPPvsstWvXxs3NDVdXV+zt7e/7fCuooII7UzGI/QMpm33ca7laYVPud717OyJJMHxYEpcij6FSKKjq7kKgqwPKX6v5hgwZQqVKlVi9ejUjR44EkL27AgIC0Ol0NG7cGK1Wi06nIz8/n3r16lFUVMSECRNwdHSktLQUg8HAvHnzOHfuHD179pR7zABq165N7dq1yzVVw4M3O9/rGtSrV4/69esTEBBAv379aNGixW3rzZkzh2KjxK5du+TjUqvVjBw5kiHDXubEiRNs2LABX19fDh48iJubG46Ojuh0Orp27cqOHTuwtbVl165dvP/++3KO66uvviqXf0tLS2Ps2LEAzJo1C71eT2ZmJoWFhbJvWQX/LOLj4wkODr7v9adNm8asWbMe4xH9t6nIif0DUTppy93E/Ry9iXjxG5ROlkFh2rRp/K9ZbQ4e0hB9vpTaAVpKSwVKpURBDtTxdyMhK4cd5+JQSBIGkwkXe1uat2lDTk4OK1as4OjRo0iShJOTEx4eHqSmphIWFsaNGzcoLi6mpKQEKysrEhISGDlyJMuWLcPGxobS0lKMRiPDhw+nX79+7Nu3D7Va/adfg1v7a8qIj4+X/z/w2T7klVgGzrS0NLy8vCyDlbXla9G9e/c7qt/XqlWr3EC1ePFiZs+e/bvH3KVLF2JiYjCbzQQEBPDhhx/+7jYVPBx5W7aQPvtzjCkpqLy98Rg3FseuXf/qw6rgEVIxE/sH4tDRH0ld/qOT1AocOvrLr+PT9EwIU7F6kTcJN/QcO1CEjwOUFAtWHD7J1fQsbLVq2gXWQAhBel4BXbp04ciRI2zbtg1JkuTw37x58ygoKODMmTNkZ2dTXFzMm2++yejRo8nOzuaLL75Ar9djNpt5/vnn8fHxISMjg08//ZSsrCzi4+M5ceIEMTExREdHs3Dhwj/lGvwe7Z7shlopcfDgQX744QfatWuHWinR7sluD3QsK1euxN3d/XfXi4yMpKSkhNLSUs6ePYtCUfH1e5zkbdlCypSpGJOTQQiMycmkTJlK3pYtD71vo9FI//79qVOnDn369KG4uBh/f38yMy22Q5GRkeX0EstYsmQJTz75JCUlJdjZ2cnL161bJzdBV/BgVHyL/oHY1vfAqVdNedahdNKiq+fOurWXWDB8N+s+jqSSsz1PaMy0isylqbXgRXfB7n62FOhKUauUvNGpNWn5hWw8FY27vS1OtpZiho4dO7Jt2zbS0tLYtm0boaGhtGrViqeffhoHBwcCAgLw9vbGy8uLDz74gObNm3Pw4EFMJhNhYWF4eHiQmJhIs2bN8PHxYf/+/XzyySd0796dqKgo8vPz+fnnnwkJCXmoZuc7XQOnXjWxre9x3/uoW7cuXbv3pHPnzowdO5aQkBC6du/52JuQ4+PjqVWr1gOFpH6PyMhInnnmGV599dVHts9/OumzP0fcovMpdDrSZ3/+0PuOi4tj5MiRxMbG4uDgcF8PZvPnz2fr1q1s3LgRa2vrhz6GCixUhBP/odjW95Bv2BePpbLnuwsY9ZZCh+I8PQZcuaivzyXnI/xiC/u8HajsrcXVVUJplPg55hLudra0qlWV6t6efHnwJPHx8fj4+PDee+/xxhtvyIr1Op2OFStWsHTpUvr378+0adPk48jJyWHYsGEoFApSUlLKyTm5ublhNBrLFXHAH1Os/71r8EepW7fuP0Y5oyxnWNavdzNhYWF8//33f/Yh/a0xptyuYH+v5Q+Cn5+fnGt9/vnnmTt37j3XX7FiBX5+fmzcuPGxhNf/y1TMxP4FHNl0RR7AyhACduteZJq7KyUKBUKSSFcqyVeVonKwwSgp0KiU2Dk5keXkiUqrxWg0kpOTQ1FREWfOnGHLli0UFRXJ5emOjo7lRH11Oh2XLl1i5syZnDt3ji5dupSTeyoLlz1sEcc/nXsVAly9epXAwEBq165N06ZNcXR0xM7OjrCwMC5cuMDgwYOpXbs2Hh4e+Pn5MXDgQEJCQrCyskKtVlO/fn3i4uLYu3cvzZo1o0uXLhw/fpyQkBBsbW2xtbWV14mPj6dRo0bY2dlhY2ND7dq1OXz4MHv37iU0NBR3d3fs7e1xdHSkf//+/3jbD9WvD033u/xBkCTpttcqlUr++7/V6aHM1igxMfGO+7iXM0QF96ZiEPsHU6aH+MWP03l3zUCWR3zAhcSTfPXzNNLzErkcH0/2tRIKogpI/T6VG/NvIIyCHAqYt+I7EvIKWbb7EOu2/yQLi+bk5BAfH4+dnR2BgYEUFhbK5fQ9e/bEy8uLxMRErl69St++fTEYDKxdu5bCwkL27dv3F1+RfxZ6vZ7OnTtjNptxdHREkiQiIyOJiIhACMHw4cMBi/lno0aNSExMZNGiRezbt4/CwkJ27NiBnZ0db7/9drn9BgQEsG/fPvLy8ti0aZO8joeHBzt37iQzM5OzZ8+iVqvl8OOlS5coLS0lJiaGOnXqcPbsWQ4dOvSnX5NHice4sUi3uA5IVlZ4jBv70Pu+ceMGR44cAWDVqlW0bNkSf39/Tp48CcCPP/5Ybv369euzePFiunXrRnJyMgCenp7ExsZiNpvZsGHDQx/Tf5WKcOI/nMuXLzNt8Ns4Kn34ZP1IIi/vZkKvLzh3/TA7znyDtoOWgM8DkJQSOYdySPk2hWqfVuPcuXN4e3szc+ZMnnnmGc6cOcPzzz/PnDlzeOWVV7C1tcVoNMpOsiEhIRQXF1NUVCTfHC9cuMCcOXOYNWsWsbGxtG3blipVqsjHtnjxYsLCwsoZ/P1XKS0uYs7QgaQmJ/LVwZPk6vQY9To0Gg2BgYGcOXMGo9FIQEAAkiRhNptxcnLC1taWvLw8oqOjadWqFR988AEffvghBw4coLS0FJVKhZeXF6NGjSI+Ph6NRsOpU6f48MMPOXToEMXFxYBFReTgwYMsXbqUbdu2UVxcjBACSZKYMWMGtWvXxtXVlcqVK1O/fn0uXrxIfHz8I20DKCtk+LOU2MuqEB9HdWLt2rVZsGABQ4YMITAwkBEjRtC4cWNefPFFpkyZcseijpYtWzJr1iw6d+7Mrl27+Oijj+jSpQvu7u6EhYVVKNT/Ue4mqvhX/FQIAP8+G04liuYfRgj/N7eKBhO+E95+/iLuaIpY9Moe0bhmezHoiYli/ssR4r2B3wlnXzdR69Nawr6BvdD6aoW2klZovDSi/dr2olevXuLnn3++bf/r1q0TjRs3FsHBwcLHx0d8+OGHQgiLqO/BgweFEEJERESI//3vf/I2rVq1EqdPn/5Tzv+fyK7vVwkvR3sxoVMb4ePkIF5s1UioFAohKVWiUtWaYuzYsUKtVgsvLy8xaNAg0blzZ2E0GoUQQjz33HOidevWYu3atWLXrl2icuXKonHjxmL06NHiiSeeEL6+vsLPz0/88ssvomnTpqJJkybyOp988om4cOGC8PX1FT4+PsLb21s0a9ZM9O3bVzg4OAhJkoRCoRCBgYEiMDBQdO7cWQghxKhRo8QTTzwhli1b9kivg62trbC1tX2ofdxLANhsNguz2fxQ+6/gr6dCAPhfzMbTSUxcf46k3BIEkJavI0snOK8xEd4/ALWVCpVSjZ2LluY9amJvY0PWhizs6thR8/2aVBlTBYzc1dn2bq7MZdzs+PxPyXFNnTpVFua9uQT6fjhz5gzbt2+XX2/evJmPPvrovrYtOp1OykfHSXzrAIqIUkpKTSw7FEm/JqE421hhNJtRCDPJiTf44osvAIuwcnx8PH369OHcuXOcPXsWg8FATEwMr732GuPGjSM7OxuwFM0kJCRQWFiI2Wxm2rRpnD17lvPnz8vreHh48PTTT5OSkkJGRgYZGRmoVCpcXV2pVKmSbFlTo0YNSkvv3Dz+OCgsLKRNmzbY29tjbW1NlSpV2LRpE/Hx8VSvXh1HR0ccHR1xcHCgZ8+eFBcXM336dHx9fblx4wanT58mPj4eIQRxcXGcPXuWM2fOcO7cOTlyUMF/h4pw4j+IT3bGUWIwlVsmhOCTnXEceusJajTwoFOXEPr0aUF8fDxWkhXpp9IxZZu4+MtF1Bo1NtjwwYAPuHTpEjk5Odja2vL6669TWFiItbU1JpMJNzc38vLymD9/PgqFgvXr11NSUgJYJJu+++47kpOTGTZsmKwf+PLLL9O6dWsOHDhAUVHRQ4v7PgpMJhPTp0//w9ufOXOGyMhInnrqKcCizF+mzn8vynQdy2SxrBU2OFs742CtIj4zhyquTkgSuNnZoHnhK1z3fURWVhYKhYJDhw5x6NAhnJ2dGT16NKdPn8bJyYmPPvqIsLAwmjVrRnJyMu+//z6SJGFlZUVubi6hoaEYjUZsbGw4f/48ycnJDBkyBK1Wi62tLc7OziQlJZGWliY3sqvVajQaDQqF4k8t4rCysmLDhg2oVCr0ej2NGjXi9ddfZ9euXVy9ehVADn2W9RWOHj2a0aNHk5aWhlarpbS0tJyJq7Ozc7lQdgX/HSpmYv8gknNLHmg5gF6nx7HIkUDvQOwN9hh1Rg4ePMiiRYu4ceMGQ4YMoaioiJdffpkPPvgANzc3goODadCgAXZ2dgwfPpyoqCg8PT0BGD16NIsXL6ZNmzaUlJSU6/XSaDRERkY+tLjv/VBW1HKnhtM333yTBg0asHbtWgYPHixXUwLMmzePBg0aEBISwoULFwA4fvw4zZo1o379+jRv3py4uDj0ej1Tp07l+++/JzQ0lO+//57ly5czevRowGJw+eqrr9K8eXOqVatW7j3upOuoUWpY3nsWkdcTSckroLKLE61C6uLn5c6pU6fYsmULsbGx9O/fn9WrV5Oens7UqVMJCQlh5syZ9OnTh+XLl6PVaklISODnn3+mffv25OXl0b59e6ysrNBqtWRmZsrrvPLKK0yfPp38/HymTZuGyWQiLi5O3lav11NYWIivry9TpkyRP8v58+cTERHxSJpvN55OYsK7e4h8ax9Cb0YYzBSeSmPSpElUrVoVPz8/rl+/TlJSEpmZmXh7e8vl688//zxqtZqDBw+yZ88emjRpwo0bN8jJyZFVY8DScuDi4vLQx1rBP5OKQewfhI9T+QZJlaMnPi8ulJcvX76cPn36AJbQ2c6dO6lWrRrx8fGcPn2a9u3bs2jRIiRJIjQ0FDs7O3bt2kX16tVZtGgR48aNAyx6iaGhoSxZskSeyRw8eJCwsDD27NnDm2++yfXr19m9ezcxMTHs3bsXe3v7P91D7G4Np66urpw6dYpnn332tm3c3Nw4deoUI0aMkPXsAgICOHDgAKdPn2b69Om8/fbbaDQapk+fLhe9PPPMM7ftKyUlhYMHD7J161beeustefnddB3drD14sWUj9l+8RkjlSuxNyCb569EEBQWxadOmO24zYcIEJk6cSP369W8L2ZaVaDdr1ozZs2dz+PBh6tevL/9+5MiRfPPNN9SrV48LFy5ga2t7r8v5yNl4Oomf155naImSNp89aVko4Kv3FnDiwFHy8vLIz88HLJ9ZaWkpkiSVKz0vK3IZMWIEubm5+Pj44O3tjY2NjTx7LFOXqeC/SUU48R/E+I61mbj+XLmQorVayfiOteXXN2vFpTo5ob7JPuTmXFZZHmvKlCmEh4ezYcMG4uPj71hVVUZZziwyMhI/Pz+mTZv2p+bMbvUPq+TlKzecGgwGuUx57ty5DBkypFzjdRm9evUCoGHDhqxfvx6Aw4cPyzMWSZLIycnho48+wsvL657H06NHDxQKBYGBgeVyMXfTdSwyFWClUTOkUycuVGrFkud60KO+b7l93qr52KxZMy5evCi/LgvLZmVlybOPCRMmMGHChNuOr2bNmuU0HmfOnAlA27Zty33O8+fPv+d5/lE+2RnHHLMGa8r3VBWVFKIssgw+Bw8eRAhBUlISgFx+fuTIEVatWoXBYKBx48YcOXIEhUKBJElkZ2djNBrLuQDcKRwqfq2+LPu3gn8nFYPYP4iyG94nO+NIzi3Bx8ma8R1ry8vLtOLKpHZM6WkY0tPJ27LlrmXFeXl5ssr8zTfQ9u3bs3jxYsLDw1GpVGRnZ8tPu25ubnLTc9nM73Fza57JlK9HFBspOp2OVV1XXnjhBebNm0dSUtI9b1hlg2uZ9xlYBgalUkl0dLQ8kL/11lt3FRG+dV9Q/ibq0NG/3LGCRdex0jNhvFF/Kw/L5s2bmTRpEl9//fVD7+txkpxbgge32s0ItsftIyo1DqPRSNdf/y61Wi1PPPGEvFbz5s3RarWoVCpyc3PlgpzExETc3d1RqVSkpKTI1/3y5cs4OTlRWlqKtbU1mZmZ2NjYYDabcXFx+d0Hkgr+uVQMYv8wetT3ve3pvYw7acVhNpM++/O7DmITJkxg0KBBzJgxg86dO8vLX3rpJS5evEjdunVRq9UMHTqU0aNHM3ToUIKDg/Hy8qJRo0aP7LzuRnx8PJ06dSLQyp9zCbHUcqvK510m0WvlKFILM6nboRHvz5vJ1KlTadasmbzdvHnz2LJlC9euXZNDbHq9nk6dOmEymTCZTFhZWaHX6zl16hQqlYrQ0FBq1KhBYWEho0ePJjw8nJ9++olXX32VyMhILl26RMOGDX/3mMuksMpmjUonLQ4d/R9aIquM+y0w+avxcbImPVfg9etMLO61nRjNRkoMpTh5ulDjvXC8vLxITk7m2LFj1K1bF29vbzIzM+nTpw87duwgNzeXt956iyFDhvDUU0+h0Wiws7PD3t6egoICtFqtPFCVzdCEENja2mJlZfWPLPYYPHgwXbp0+dMeEP/pVAxi/yJu1YTzVWvYXLWavPzmmYW/vz/R0dEAdwxXqVQqPvvsMz777LNy+5wxY8YdKw1vbmi+NVz1sM3OcXFxfNj/VRp1msTr2z/im1OWsKGLtRP1PGrz7rvvUlpaSocOHWTVkLLcV9OmTdm8eTNjxoxBrVazdetWvLy8WLBgAdOmTUOj0fDaa68xf/58JEkiICBAPt7w8HByc3NZsWIFX3zxBVevXuXjjz++r2N+FLqO/3SGV8rl/JXTuDmHy8uEgJl7F3Ho6hFKiku4EZ8ASEx5czoqlYq0tDTMZjOHDx8mLy8PHx8fdDpLU3hsbCwGgwGtVkteXh4lJSWUlJRgMpkoKCgALKFrFxcXkpOTcXFx4eKxVI5sukJhdil2Llqada9OrSYVs7J/ExXZ0H8Rj1Mr7s8k9sAevhz1Ap8+25VVU97Ay8ODpsEWhfteQe05kXQOACcrexYMmEFsbCzh4eFotVri4+NRKpVy7mvOnDlyePHgwYOMGDGC4OBgFi1aJAsQ16hRg+eff57Tp08zY8YMueDDxcWFLl26sGDBAp577jkmTZoka+PdXEQDf54KxT+F2AN7yN75HRn5pzmRuQOwNKauPbuRpIzLjOnzFWqVFntrZ0xmI1fOJ2BrY0+fPn2oUqUKH3zwAVqtFiEEU6ZMwc3Nja+++krWJywuLsba2hohBB4eHri7u+Pu7m5pfv017H3tdDZ7vrtAYbYlP1mYXcqe7y5w8VjqQ51bfHw8derUYejQoQQFBdGhQwdKSko4c+YMTZs2pW7duvTs2ZOcnBwuXLhA48aNy20bEhICwMmTJ2nTpg0NGzakY8eOpDwCYeL/IhWD2L+Ix6kV92cRe2APP385n4LMDBCCopxsSosKKPQvlv3DpF/DU5JCuqt/2J1yX2VFLNHR0WzZsuW+RVfvlvv6N3Kzx9XDcGDNCiassVRc3iiKxSgM/BD/McdTd2Ky90NS22I2m8gutBTE2GgcyMvPITU1levXr7NkyRJ0Oh0pKSkYjUaSk5OZNWsWJpOJkpKScnnPvLw8zGYzOTk55Y7h1I6E24SxjXozRzZdeejzu3TpEqNGjSImJgYnJyd+/PFHBg4cyMyZM4mKiiIkJIR3332XgIAA9Ho9165dA+D777/n/+3deVhV1frA8e/mHEZBBhGBwFlmkcEJEcXMtJsDmopdx7zmlFrdn9c0b2lO95amXiyzuk45lGPXtCwnzDEVBFERNQVNxQkUQaZzOOv3x4EdKCiOgK7P8/DI2fvsddZZgOvsvd/1vpGRkeh0OkaPHs3atWuJjY1l8ODBTJw48ZH79TySk9gzxLZLF1ymTkHr6gqKgtbVFZepU6pUJdvd332DPr9kiPqN2zn8d81M7Ho0YsPv0TRza4xWa8r21T8/0CW7soJYiu6vPE/unKwe1+RVJDOt9MwovnVeIDn9HNPXDMEgDNSyq42JouFi+lk0JloOHDiAoiicOnWKunXrUr9+fXWNYlFi5KKzr+zsbBRFIT8/n/T0dKysrEq81u0b+aV1QT0zexT16tUjICAAMEa6njlzhps3b9K2bVsABg4cyK5duwDo3bu3WianaBI7efIkx44do0OHDgQEBDBt2rQSGe6l8pOT2DPGtksXGu3YjveJRBrt2F6lJjAo/T8/KzNT1u85QNO/tmVD4jbeXPIeGlszqvnfv5ryyZMnSUxMZNu2bYwbN47evXvj7OxcIuS/Xbt2JCYmqouai0tPT+fw4cPqY71eX+7UU1VZVlYW7du3VxeGF61jK2uRORizuTRr1gw/Pz++TzgJhSetn+3Yh6WpKRPWbWbu1j286tuZib3+y6TXvwEEAfXDyNPlYKo1Iz09HYPBwLBhw6hevToXL14kKCgIHx8f8vLyMDMzo3r16ty6dYsXXniBgIAAgoKC8Pf3p169evj7+1OtWjU8PT2xdjAv9b2Vtf1BFD8712g03Lx5s8znRkZGsnr1ak6dOoWiKDRq1AghBL6+vsTHx6sps7Zs2fLI/XoeycAOqVKxqeFovJRYyCAENhbm/KNnF4Z+vpi6detiZWVFSkpKieOKn1kV35eXl0fv3r156aWXAGMxw5iYGBwdHdUAFQcHBw4dOlSivaJsFT169CAmJkbd/qzVfcrKyqJbt27k5OTQuHFjdDod7du3JzU1lTNnztCyZUtSU1N5/fXXcXNzw8fHh5MnT3Lp0iXc3Nw4evQoERER3Lhxg+PHj9OgQQMsLCy4kZmPrqCADXHHuXgjA+fqNkyO6EjtsxcZvf8LfOqHA3Dl5h/8te1YIkKH8Mm6YcyfP5+xY8cyatQoBgwYQOfOncnNzS1xz1EIgY+Pz33fW0i3BiWKxQJozUwI6dbgsY+jra0t9vb27N69m7CwMJYtW6aelTVo0ACNRsPUqVPVRfOenp5cu3aN/fv3ExISgk6n49SpU/j6+j72vj3r5CQmVZii8Png4GAOHz6Mr68v7w/7Gy9170kTN2dOX7lO03puZOTkke38Z6h0Ufh8dnY2eXl5tGrVir1795KZmYmbmxsnT56kX79+xMTEcOHCBbKysti0aRMff/wxWVlZ/OUvf0Gv15OcnEznzp05c+YM586dU4tIWlpasnjxYurVq8eHH35ITk4Oe/bsYcKECeTk5BATE8Nnn33GoEGDqF69OjExMVy+fJlPPvmkyoVFW1hYMHz2cH5t8SsFwwrIH53PsE+HEVgtEA8PD/bt26eeBZ0+fZoZM2Zw6NAhOnbsiI+PD7/88gtpaWnExsbi6+vL6dMnsbNT0OsLQAFzBwNuN2yxtbHhiwNHMROCtPxsah1bTGbt9thb16SJTzAh3RoQMuB7oqKiGDt2LNHR0XzyyScoisLBgwcZPXq02ufyppgqikJ8WtGJS5cuZfjw4WRnZ1O/fn0WL16s7ouMjOQf//iHem/MzMyMtWvXMmbMGDIyMtDr9bzzzjtyEnsYZaW3r4gvWYrl+ZKcnCwAtcTLG2+8IWbOnClcnWuJHiHBYlZkZ/HlyEEiolNHsWbNGiGEEHXq1BFRUVFCCCGmTJmiHp+RkSEGDhwoZs6cKZycnISfn58QQojFixcLT09P9fgaNWqIsLAwIYQQLVq0EHXr1hUFBQXiwIEDokGDBkIIIbZu3Sp69OihHv/WW2+pfS7+eODAgaJnz56ioKBAHD9+XD2+sjpy5IiYPXu2mDRpkjAzMxNHNn4p/vcfH1HzRXuBgrBwtxCAqBFWQ9g72gtFUYSFhYVITU0Vjo6OwsLCQiQnJwtHR0fh5eUlbGxsRI0aNYSbm5to2rSpAIS1tYn42xAH0X+AnVAUxJw5dUWDBq7ipZdeEvn5+UII488wOTlZJCcni9q1a6v92759u4iIiBA5OTnCyclJnD9/XgghxKRJk8SkSZOEEELs2LFDZGVlPfWxk56eBy3F8khnYoqi9AImA95AcyFETOH2usAJ4GThU38TQgx/lNeSng2plzdw9swscvNSuZFuj6uro5o6ql+/fkRFRWFqbsHsb9epC1X3FUtEm51fwNJLTswe/yPmFwUmJhpycnLIyMjg22+/xczMjLy8vHsGKtSvXx8AU1NTDAYDJiYmuLi4kJycjLOzMzVr1kSn05Xr/ZSVeqqySUhIYOPGjer7EkKwMeYcG08p5GUVoJgqNJzakGODjpF3K4+abWqi36rHwcGB6Ohorl+/joWFBXl5eVy/fp3Ro0cTFxdHcnIyiYmJbN68mcBAf0JDLcm+XcD+/YXFOM115OYaU2SZmpoSHR3NuXPn1H4VVUgOCQlRKyQXXbKtiMwwUtXzqJcTjwE9gC9L2XdGCBHwiO1Lz5DUyxtISpqIwWDMPp6XfwV9QQaplzfg4twN+DOpbWnJav8Xd5Eb2fmY3y5AYwVpt/MwKCZk1fDmgw8mYmZmRps2bdQ6XGXRaDQAZGZmqlGJH3zwAVqtlsuXL983h2RxVSX8fvv27XeNiQ5TrG46obVJRVEUsk4Y7ztpqmkw8Tchc71xfNatW6dGcB48eBCNRsOSJUvIzMwkLy8PU1NTGjRogL29Cdu23cbBQUNQkCXnzhlfz9YOTpw4QePGjWnatCleXl5qH0qrkGxlZfXUM8NIVdcjTWJCiBOATK4plcvZM7PUCazI1Ss6ftjwIcOGdVM/icfFxXH+/Hlat25NcHAwmzZtMpZN+cuH6G9d5+beb8m/eALz2v5QoGPMex8QXM14X0Gn05GamoqJiQlJSUnY2NiQlZXF+++/z/Tp08nIyCAjI4P8/Hy15EpAQABmZmYYDAZGjRqFo6Mj169fZ8yYMfzyyy9cunSJ8PDwKn02ULz2VnHBfsH8tv43DPkGbu69iVktM3Iv5JL6ZSrVq1cHjHXVQkNDOXPmDOvWrUMIwdWrV2ncuDFeXl6cOXMGPz8/srOhWTNLGnmYM3CgA6mXjRGgX33ZjNDQ3Xe9dkpKClqtluXLl9+1r6zMMM7Ozk89G79UuT3JwI56iqLEAbeAfwoh7v4tBhRFGQoMBahdu/YT7I5U0XLz7s5I4O5uypo1Z5k711v9JD5v3jzAGB6/cOFCTE1NOX/+PPHR6wAwsbDGZdB/yD59gMyYDVw9cZA4TRZ5eXkcP34ca2trNBoNs2bN4pNPPiEnJwetVsvUqVMZOnQoMTExmJmZ4enpaWw3Pp79+/cTGhrKqlWrGDZsGGAstbJ3717Cw8Pp27cvBQUFd/W/qrC1tS0xkb3//vsA2Fcz4DuxHrnFSplYaCwYUXsEs0fOVlOTpaSk0LlzZ+bOncvvv/+ubi/uzjPt2bNdMTGxpH6DsU/yrUnPufuuE1MUZZuiKMdK+ep2j8NSgdpCiEDg78BKRVGql/ZEIcRXQoimQoimNWvef92PVHVZmN+d/kqjgY8+CuDEiRMsGjSIi527sNnSioy3RuHmaLxftmTJEv75z3+iXE1CY+OIjf/LAJg51QUTDTU9AklJScHExISff/6ZjIwMNm/eTEpKCg4ODsTHxxMYGMjUqVOpXbs2FoVZTXr06KHejwsJCWH06NE0bNiQadOm0bNnTyIiInB0dOTYsWOYmZkRGRnJoEGD1NIlVSn1VPv27TE1NS2xzRQdncVuJl9Px0WnRxECF1NbJreazIu1Xyy1neI5N+/k4twNL6/pWJi7AgoW5q54eU1XLxU/SFuSVF73PRMTQrz0oI0KIfKAvMLvYxVFOQN4ADH3PFB6ptVvMLbEJ3UjE+o3GFtqGZmCjIwSZWQa1bLh4Jks8lJPo7V1InX5OEy0przeqTUffvghQgjMzMyoW7cu//3vf+9KNzVp0iTi4uL46KOPANi9ezdHjx4lPj6eS5cuERgYSEFBAT/88ANHjx6lc+fOai8r8/2u8vD39weM98YyMjKwtbWlfaNq+J/Oxj8jh1e1NaD9h+DfWz2m+ART3gnHxblbmZOWJD0JyuP441QUZScwtlh0Yk0gXQhRoChKfWA30FgIkX6vdpo2bSqKLyyVnj3FoxMtzF2o32AsLs7dOP1ie/SFBREBYrNv0/+PPwi1tyetVi1yc3MZMmQIH06aRM3mXbhx9iiGzOs42NrQt08vZs+ejVar5d133yUqKgpXV1dq1qzJwYMHCQ8P59KlS+Tm5pKeno6trS3Jycm4urqSlZWFk5MT3t7euLu7c/DgQRISEmjYsCEeHh7cvHmTy5cvq8dLFevEiRN4e3tXaB+KLq0+zrPImJgYvvnmG6Kioh5bm1VVaT9jRVFihRBNS3v+o4bYdwfmATWBHxVFiRdCdATaAFMURdEBBmD4/SYw6flQ1if1c+fOMfTCH/iYW3AiL5cahRGEh27eRJ+RgRCCjz/+GENBASG1IO6Snhs5gutXLrFw4UIWL16s1pVycHDg+vXrpKSkYGVlxT//+U8WL15MamoqBQUF3L59Gz8/P1q1asXWrVu5fPkyqampaDQaNS9fVlYWNWvWxMPDg/r167NixQrq169fJRc0P89O7I5m93ffkJl2HZsajoT1GYB3WLv7H/gE6PV6tNrS/8tt2rQpTZuW+n+0dB+PlDtRCPG9EMJNCGEuhKhVOIEhhFgnhPAVQgQIIYKEEBsfT3elZ5XGyYnk/Hxet7NjU736WJqYYALYmxqjBufOncuLL76IVqtVo2E1Gg1NmzZl4cKFeHp6Ym5uzoABAzA1NWXIkCG0b9+eIUOG8MUXX+Dn50fNmjWxt7fH09OT+fPnk5CQQKNGjfjqq69o2LAhjRs3ZsiQIbz99tvo9XpGjRqFoijqurNNmzYxfvz4ih0oqdzurIiQef0aW776jBO7ox/ba5w9e5bAwEAOHDigZp8JCwszRtNiTF82fPhwWrRowbhx4zh48CAhISEEBgbSqlUrTp40LqXduXOnevm6rOekpKQQFhZGUFAQQUFB7Nu3Tz22KHq2KK9lVb/8/SBkAmCpUqjxSiAuWi1BhZnIW1lZIYDbZgYcajrg5+eHoihotVo1dY+VlZUa0VqjRg0KCgrUSr8HDhxg//79/PTTT6SmptKuXTvGjh1LtWrVOHv2LO++++5dpTuCgoJQFIUXXnhBbQuMC5pFYb6+yrygWSqptIoI+vw8dn/3zWNp/+TJk7z22mssWbKE999/n3nz5hEbG8usWbMYOXKk+rwLFy6wb98+Zs+ejZeXF7t37yYuLo4pU6aoUaLFlfUcJycntm7dyuHDh1m1ahVjxoxRj4mLi2Pu3LkkJiZy9uxZ9u7d+1jeY1UgcydKFSYhIUENNCi4qUNYaNBVK0B7W4PWRou4BjmKgTx9Jh9/+TEvt36Zn3/+WT3exMREXbhcfK3izZs36datm1q1uXv37rzwwgvMmTMHRVFwdXVl48aNNG7cmPz8P8t1FL/UU/yTbFVZ0CyVVFY5mLK2P4hr167RrVs31q9fT+3atdm3bx+9evVS9+fl/Tl59urVS/09zcjIYODAgZw+fRpFUUpdlF/Wc3Q6HaNGjSI+Ph6NRlOiInvz5s1xc3MDICAggJSUFFq3bv3I77MqkGdiUoUoSoNUtHYpEysuZ+UxNsyeyAlaohxvo7HRYMgzoLuh49CxQ4wYMQK9Xo+Hh4faztGjR0lISACMf+RpaWkYDAYcHR0B2LRpExqNhgkTJhAXF6cet2TJEhRF4fLly3z00Uekp6eTnJxMbGzsUxwF6UmyqeH4QNsfhK2tLbVr12bPnj0YDAbs7OzUsirx8fGcOHFCfW7xxdnlKcxa1nPmzJlDrVq1OHLkCDExMSU+gN1ZGqZ4qaFnnZzEpApRWhqkGjVqcCE6k9MTTmPINqCx1qCppsHUyRRhKwgODsbV1ZWuXbsCxrOiX3/9lSlTpgDGXIg1atTA1taW5cuXG8/wCgowMTHh1KlTrF+/nitXrnDlyhUSEhIwGAzUrl2b2NhYCgoKSExM5J133gGM/xFUlTVgUunC+gxAa1aydpjWzJywPgMeuW0zMzO+//57vvnmGzZt2kS9evVYs2YNYPy9PHLkSKnHlVWYtTzPycjIwMXFBRMTE5YtW1alF98/TnISkypEaWmQTExM6NW9F43+1YiaXWqiu6bDsq4l+nQ9ZgYzYmNjURSFhIQEHBwcmDlzJmPHjmXt2rVs2rQJV1dX5s2bh4uLC1ZWVnz77bcsXLiQXbt2ERISwvjx4wkKCiI+Pp7vvvsOBwcHUlNTefHFF/n888+ZPHkyu3fvZtCgQfTt25cxY8Zw6tQpxo0bx9q1awE5mVUl3mHteHnoKGwca4KiYONYk5eHjnps0YnVqlVj06ZNzJkzh8jISBYuXEiTJk3w9fVVi4jeady4cUyYMIHAwMC7zpaKLomX9ZyRI0eydOlSmjRpQlJSkky/VUjeE5MqxJ1pkIqYm2RjYTCQ72CK0AvsQuzQXdXh7+7P/PnzAeMZW1G15eL3yMCY+fzw4cPMnz+fWbNm8d///le9Ua7Vatm2bRvvv/8+69atY8qUKWptMLj7k3Fqaip79uwhKSmJrl27ytD6Ksg7rN1jD6kvvvDbzs5OLaj69ttv3/XcO3+nQkJCStzLKsoPmZaWptZJK+s5jRo1Ui+dA3z88ccAhIeHl0hYXfT7/LyQk5hUIdq3b1+iNIidnR2jR42her4Xgw504kL+7/zH7v/waufFF//3BZYplupC0KLquKXp0aMHAMHBwaxfvx4o38300lSVMitS1fbDDz8wceJEFi1aVNFdqZLk5USpQvj7+9OlSxdsbW0BsLKwxvpWI5TrDoCCZZ4NlgXWfOb+Da/WfxW4d5mWIkU3uIvf3C7PzfR7tQUyKlF6crp27UpSUhKtWrWq6K5USXISkyqMv78/7777LpMnT6bWrRDMskomgE7PvMKyz433ForKtDyMsm6UF9XIKq/Jkycza9ash+qDJElPhpzEpEohKz3vrm217Nz5Zf8avL29uXHjBiNGjHiotsu6Ud6uXTsSExMJCAhg1apVD913SZIqzmNJAPy4yATAz6+l7+8tMZGlZV5mweaJ/Gv4cgbOCK2wfk2fPp2lS5fi5OSEu7s7wcHBdO/enbfeeotr165hZWXF119/jZeXF4MGDaJ69erExMRw+fJlNc9iUbHNHTt24O7ujqmpKYMHD6Znz55MmTKFjRs3kpOTQ6tWrfjyyy9RFIXw8HBatGhBdHQ0N2/eZOHChYSFhVXYOFQWlSEBsPRkPWgCYHkmJlUKId0aoDUr+euoKMbtFSU2NpbvvvuO+Ph4fvrpJzUKbejQoWWmGCqKaCyeZ3H9+vWkpKSQmJjIsmXL2L9/v/r8UaNGcejQIY4dO0ZOTg6bNm1S9+n1eg4ePMjcuXPV8jGSJJUkJzGpUvBo4Uy7vl5YOxiDKerUqUP0T/vxaOFcYX3avXs33bt3x8rKiurVq9O1a1dyc3PVFEMBAQEMGzaM1NQ/K1aXFtG4Z88eevXqhYmJCc7OzrRr92fId3R0NC1atKBx48bs2LGD48ePq/uKR1qmpKQ8nTct3VNKSgp+fn6Pvd2YmJgSuRArUvG+PMx94BkzZpR4/KQDVmSIvVRpeLRwrtBJC+B23FVu/ZJCwc08bh07R75ryc95xVMMleZBIhpzc3MZOXIkMTExuLu7M3ny5BKRk6VFWkoPpvjPU2NnTvWOdakW6FQhfakKpVj0ev0j92XGjBklEhsXZdt/UuSZmCQVuh13lZvrT1Nw03hvrpmjMfPC9f3nyMzMZOPGjVhZWaHVaunfvz9w7xRDRUJDQ1m3bh0Gg4ErV66wc+dOAHXCcnR0JCsrS80KIj0ed/48C27mcXP9aW7HXX0s7ReVYTl06BBnzpyp9KVYwsPDefvttwkICMDPz4+DBw8CxrOt/v37ExoaSv/+/Uv0BSAxMZHw8HDq169fomhnREQEwcHB+Pr68tVXXwEwfvx4cnJyCAgIoG/fvgBYW1s/UD8flDwTk6RCt35JQegM6uPGzp508WxHiy5huHjVplmzZoDxMt/PP/9MkyZN0Ol09OnThyZNmpTZ7muvvcb27dvx8fHB3d2doKAgbG1tsbOz480338TPzw9nZ2e1fenxuPPnCSB0Bm79kvLIZ2MnT56kT58+LFmyhCZNmtC+fXsWLFhAo0aNOHDgACNHjmTHjh3An6VYNBoNt27dKjV7THFlZZgpKsViYWHB6dOnef311ykKhIuLi+P48eO4uroSGhrK3r17S12Skp2dTXx8PLt27WLw4MFq5pHExET27NmDpaWl+iGrSFJSEtHR0WRmZuLp6cmIESMwNTVl0aJFODg4kJOTQ7NmzXjttdf497//zWeffVbmlYry9vNByElMkgoVfWIvbkyrAYxpNYClNrtYunQp58+fx93dnTfffFONUtywYQNbt269K0rR399frQY9a9YsrKysGDJkCMuXLyc/P585c+YwePBgpk2bpkYpHjp0iEuXLhEdHV0iStHJyQkPDw8ZpfgASvt53mt7eRUvw+Lj40NWVlaVKcXy+uuvA9CmTRtu3brFzZs3AeOCa0tLy1Lf76uvvoq5uTnm5uY4OTlx5coV3NzciIqK4vvvvwfgjz/+4PTp09SoUeOeY/ckSsbIy4mSVEhjZ17q9uO3zz5ylGLnzp2pX78+q1evZsGCBaxevVpGKT5hZf08y9peXsXLsABVqhRL8bp7xR+XJwtO8bZ37tzJtm3b2L9/P0eOHCEwMLBcmXCeRMkYOYlJUqHqHeuimJb8k9ALHd8eWkodCw3nYg88dJTizp07iYiIYN68eQwePFhGKT4Fpf08FVMTqnes+0jtFi/DsnLlSqpXr15lSrEULerfs2cPtra2atq3B5WRkYG9vT1WVlYkJSXx22+/qftMTU3LnZ/0cZCTmCQVqhbohF2PRmjszBEIbutvcejaZm7kXSEv+zZbvvqME7ujgft/+n6YKMW1a9dy9OhR3nzzTRml+BgU/3mC8QzMrkejxxKdWLwMyw8//MCKFSuqRCkWCwsLAgMDGT58OAsXLnzg44t06tQJvV6Pt7c348ePp2XLluq+oUOH4u/vrwZ2PHFCiErzFRwcLCSpMvhy5CAxq/erYlbvV8U7HVoLF1sb8a8encR/hvQVDRs2FDNnzhQhISFi9erVQgghDAaDiI+PF0IIMXDgQLFmzRq1rWrVqolJkyaJfv36iVdffVUUFBSIy5cvC3t7e7FmzRpx48YN4eTkJLKzs0VmZqbw9fUVkyZNEkII0bZtW3Ho0CEhhBDXrl0TderUearjUNkkJiZWdBeeiLVr14oBAwY80dco/rtUmZX2MwZiRBnzhgzskKRSZKZdV793s7elibsLn27ZjbWFGSEdOgGwYsUKRowYwbRp08oVpejv78+ZM2dklKJUgizF8mhk7kRJKsVXb71B5vVrd223cazJ0M8Xl6uN0vIuvvzyy4wbN47U1FROnTrFli1bCAsLk3kXy0nmTnz2ydyJkvQYhPUZgNasZBSb1sycsD4DynV8WXkX27Zty7lz59DpdIwdO5ZJkyapx8i8i5L04OQkJkml8A5rx8tDR2HjWBMUBRvHmrw8dBTeYe3KlU+urLyLubm5mJubY2ZmxsaNG2XeRUl6RPKemCSVwTusHd5h7e7/xEKplzdw9swscvNSSU42YKIpeW9L5l2UpMdPnolJUjlMnz4dDw8PWrdureayK54vr0ULH7ZsGUtu3iU++fgKCUev8sX8Nbi712LZsmVs3LgRS0tLhBC88MILdOjQgVdeeUU9ozty5Ajvvfcefn5+DB06VH3drVu3Mm3aNJo1a0b9+vXZtm0bIPMuSlIROYlJ0n2Up67Y3/5mxtw5F9RjDAXw+ut26PUZDB06lGbNmnH06FG8vLzw8/Pj0qVLbNmyRW3Ly8uLjz/+WL2/VbSQ1dHREWtrazIzM7G1tUWr1d4V0dixY0cZ0fgURUVF4e3t/VjWQdWtW5fr16/f8zlLlixh1KhR5X7OggUL+Oabbx65b1WFvJwoSfdR/P4WcFfGDoCsrBPodH9eAgwNtaL9Szb06+9ARLcrrFy5knfeeYd27drxxhtvAMb7VkXH9+zZk08++YSPPvqI9PR0NeBDURRmzZpFhw4dOHHiBP7+/jRu3BiAadOmMW3atLv6WzyBq6Oj43N9TywhIYHt27eTkZGBra0t7du3x9/f/5HanD9/Ptu2bVNzAFY2w4cPfyzt3Kt0TGUiz8QkqRT/i7tI6L93UG/8j/xn22mSLt8qsf/OjB1LlzZn0WJ3df+KFTeYODEVC3OXR87YMX78eAICAoiIiMDW1hZn54qtuVZVJCQksHHjRjIyMgBjqqSNGzeSkJDw0G0OHz6cs2fP8sorrzBnzhzS09OJiIjA39+fli1bqm2XtT0tLY2XX34ZX19fhgwZUubvxuLFi/Hw8KB58+bs3btX3X7t2jVee+01mjVrRrNmzUrsK1IUeJSUlETz5s3V7SkpKeoHoNjYWNq2bUtwcDAdO3ZUA4zCw8N55513aNq0KdOnT6devXpqCqlbt26VeFxZyElMku7wv7iLTFh/lIs3cxBArqMHP2zYwKr9v5eoK1Y8X169+v9H8tk7kquioX6Dserjh60r9uWXXxIfH8/evXvV2kzS/W3fvv2u/3B1Oh3bt29/6DYXLFiAq6sr0dHRvPvuu0yaNInAwEASEhKYMWMGAwYYl2CUtf2jjz6idevWHD9+nO7du3P+/Pm7XiM1NZVJkyaxd+9e9uzZQ2Jiorrv7bff5t133+XQoUOsW7eOIUOGlNlXLy8v8vPzSU5OBox5EyMjI9HpdIwePZq1a9cSGxvL4MGDmThxonpcfn4+MTExTJo0ifDwcH788UcAvvvuO3r06IGpqelDj9+TUPnPFSXpKZv5y0lydH8mVzV3boilZxiDuoQT7FVXvf9UPGPHpUuX0Otzycq6jRACa2stHh5enEi04fbt22oxzdatW+Pj48P58+fRaDT0798fvV5PcHAwfn5+1KpVi4sXLzJ16lTmzZtHZmYmc+bMYcWKFURERHD+/HksLCyoV68ex48fx8TEBDs7Ozw9PTl+/Dj5+fnMnj37vvdQngdFZ2Dl3f4w9uzZo9YDe/HFF0lLS+PWrVtlbt+1axfr168HjCVO7O3t72rzwIEDhIeHU7NmTQAiIyPVsivbtm0rMandunWLrKysMvvXu3dvVq1axfjx41m1ahWrVq3i5MmTHDt2jA4dOgBQUFCAi4uLekxkZKT6/ZAhQ/jkk0+IiIhg8eLFfP311w81Tk+SPBOTpDtcuplz1zbbVpE4D/mSPXv2sHLlSsaOHUu9evX4+eefOXLkCK1btyYrK5fU1KukpPzB7dtgbl6Lnj17smXLFrKzs5kzZw779u0jKSmJWrVqkZOTw+nTp1m3bh2xsbGcOXOGkJAQTE1NycvLIzo6Gr1er14+XLp0KQaDgdzcXPLy8koslC4oKCArK4uJEyfy4YcfPrWxqszKytD+sJnbKwODwcBvv/2mXsa+ePHiPc/OIyMjWb16NadOnUJRFBo1aoQQAl9fX7WNo0ePsmXLFvWY4omFQ0NDSUlJYefOnRQUFODn5/dE39/DkJOYJN3B1a704oB3bp8+ug8N7cwJtrZk/8aNOGtMuLB+PUOGDMHU1JQ9e/Zw48YNunbtilarZdiwYWRkZGBubs4ff/xBmzZtcHV15eeff0av19OhQwe+/vpr/Pz80Gq1/O9//0Or1bJkyRKGDh3KF198gbW1NSYmJqSkpBAVFYWHhwd6vZ6//e1vgLGKdGZm5hMfo6qgffv2d136MjU1pX379o/tNcLCwlixYgVgDKhxdHSkevXqZW5v06YNK1euBGDz5s3cuHHjrjZbtGjBr7/+SlpaGjqdTr1kDfDyyy8zb9489XFZaw6LNGjQAI1Gw9SpU9UzLE9PT65du6Zmf9HpdCUWyt9pwIAB/PWvf1UDkiobOYlJ0h3+0dETS1NNiW2Wphr+0dFTfRz73b9ZuWIda1xqs8DFnQyDAVFgYMTf/86M7t3x9vamYcOGAOTk5NCvXz+6d+9OQUEBcXFxKIpCYGCgmlYKYNmyZdy+fVt9jVGjRuHo6MigQYPIyMhg9uzZbN++XZ38AgMDmTt3Lrm5ueqnZ1NT0/sGkjwv/P396dKli3rmZWtrS5cuXR45OrG4yZMnExsbi7+/P+PHj2fp0qX33D5p0iR27dqFr68v69evp3bt2ne16eLiwuTJkwkJCSE0NLREHsGoqChiYmLw9/fHx8eHBQsW3LePkZGRLF++nN69ewPGemhr167lvffeo0mTJgQEBLBv374yj+/bty83btxQq0JXOmWlt6+IL1mKRaosvj98QbT613ZR971NotW/tovvD18QQgix6cwm0WFNB+HyurMY5FpDJHp6iURPL1FbqxUKCFMQDaysBCBsbGwEIMLCwsTAgQPFN998I1atWiWEEEJRFPH3v/9dvP3222LRokXC+KcoRMOGDUX16tWFTqcTn376qQBEjRo1hJOTk1AURaSlpYmQkBBhamoq2rZtKy5fviwURRFLly4VQgiRlJQkNBpNxQzaU/CslmKpzNasWSP69ev31F5PlmKRpMcgIvAFIgJfKLHtx7M/MnnfZHILchGARR5gY9xnr9WSYTCQYTBwOTeXOnXq4Ofnh6OjI2vWrGHfvn2sWLGCsLAw9RNxaXx9fcnJycHKygq9Xo+trS1jx44lNzeXRYsWUatWLYQQ6o14jUZTZluS9KhGjx7N5s2b+emnnyq6K2V6pFIsiqLMBLoA+cAZ4A0hxM3CfROAvwEFwBghxC/3a0+WYpEqs5fXvkzqbeN6mpyUHNKnpWCjmGCv0XA6L49mVlZc1em5juCWomBubo6/vz/dunVj3rx5ZGdnk5OTg16vRwhBXl4ebdu2JS4uTg0F1+l0DBkyhBkzZuDk5ERBQQGWlpbodDoiIiJYs2YNFhYWWFtbk52djVarxdLSkitXrjzTJViKyFIsz76nXYplK+AnhPAHTgETCl/QB+gD+AKdgPmKosiPjFKVdvn25RKPc83AQMkPgQUmCllAo0aNsLW15bfffgPg6tWruLm5cevWLZYsWUJ+fj5gzMih1+txd3fHx8eHgoIClixZgrOzMwUFBVhbW+Pr64vBYFAXpJqbm9OxY0eys7Np0qRJieAAWYJFet480uVEIcSWYg9/A3oWft8N+E4IkQckK4ryO9Ac2I8kVVHO1ZzVM7HbJ29j/ZIDr3o60n+nYO4l0DsFsDt+DdZW1dHnCRwcHLh9+za5ubnk5+eTnp5OUFAQACYmxs+PV69eJTIyksWLF5OWloajoyOfffYZx44dQ6fTERMTQ3Z2NmZmZtSpUwcANzc34uLiaNy4MdeuXcPGxkbtoyzBIj1vHuc9scHAqsLvX8A4qRW5ULjtLoqiDAWGAqVG6khSZdHXri/Hk45jqbdk943dnOc8sf7VaNdmKjc+WIiFxhwzrTmtPF8lIvRvtOvrxYJVn2AwGDAzM+PTTz+lZ0/j57zia3u2bNlCQEAA+fn56uJlvV7P8uXLOXbsGO7u7nh7e6PT6cjNzeX06dOsX7+ezp07M27cOL766iu1LVmCRXre3PdyoqIo2xRFOVbKV7diz5kI6IEVD9oBIcRXQoimQoimRSvUJamySUhI4MKBC1jprVBQaFS7EZmHMxlqP5Qb0RoSzu7HVGuBg40zB05tITs7h+2rEli7di1WVlbY2Nio63KEEBgMBgCcnJxwdnbm8OHDREdHq9ubNm1Kbm4uDg4OnD17Vs3YUJSiys7OjqysLH744YcKGA1JqjzuO4kJIV4SQviV8rUBQFGUQUBnoK/4M0rkIuBerBm3wm2SVCXdmYfPxcUFX19fJr4xkZkr/05tJ+MasqEdp6A1MeUfi7vydlRX7OzsAGjTpg3bt2+nSZMm+Pr6qqVW6tSpQ40aNfDx8aFfv36YmJhga2vLgAED8PX1xd7enqCgIGrVqoW5uTl2dna4uLjQp08fOnbsSGBg4FMfi+dZSkrKY8taUVTap7LYuXMnnTt3fuR25s6dS3Z29mPoUfk80uVERVE6AeOAtkKI4r3+AVipKMpswBVoBBx8lNeSpIpUWr69Nm3a0KZNG+rldyArPU/dPul1Yy0nawdzBs4IvWe7S5cuJSsrC2tra9LS0mjevDmNGzfGxMSEXbt2ldg+c+ZMAM6dO1dqW7IEy92KV9u2MHehfoOxuDh3u/+BT0F8fDwxMTH85S9/eWqv+TTKq8ydO5d+/fqppYuetEeNTvwM40qZrYqixCuKsgBACHEcWA0kAj8DbwkhCspuRpIqt3vl4Qvp1gCtWck/Ja2ZCSHdGpSr7c6dOxMQEEBYWBgffPCBmiuxrO1S+aRe3kBS0kRy8y4Bgty8SyQlTST18oZHalev19O3b1+8vb3p2bOnetYxZcoUmjVrplbnLrowFR4eznvvvUfz5s3x8PBg9+7d5Ofn8+GHH7Jq1SoCAgJYtWqVWkKliJ+fn/pBJCIiguDgYHx9fUvcA7W2tmbixIk0adKEli1bcuXKlbv6O3nyZPr3709oaCj9+/cnJSWFsLAwgoKCCAoKKjVbx6FDhwgMDGThwoVERESo27du3Ur37t0BGDFiBE2bNsXX11fN4xkVFcWlS5do164d7dq1e6RxLq9HmsSEEA2FEO5CiIDCr+HF9k0XQjQQQngKITY/elclqeLcKw+fRwtn2vX1wtrBGFRh7WBOu75eeLQo36Szc+dO4uPjSUxMZNCgQffdLpXP2TOzMBhKJnM2GHI4e2ZWGUeUz8mTJxk5ciQnTpygevXqzJ8/HzCmCTt06JBanXvTpk3qMXcufTAzM2PKlClERkYSHx9fInN8aRYtWkRsbCwxMTFERUWRlpYGwO3bt2nZsiVHjhyhTZs2ZWaZT0xMZNu2bXz77bc4OTmxdetWDh8+zKpVqxgzZkyJ5+7bt4/hw4ezYcMGBg8eTFJSEteuXQOMdc4GDx4MwPTp04mJiSEhIYFff/2VhIQExowZo5aqiY6OfrgBfkAyd6IklcP98vB5tHBm4IxQ3lrwIgNnhJZ7ApOenNy81AfaXl7u7u6EhhovE/fr1489e/YAEB0dTYsWLWjcuDE7duwokVT3UZc+REVFqWdbf/zxB6dPnwaMeRCL7mPdq+2uXbtiaWlMYK3T6XjzzTdp3LgxvXr1KlHa5cSJEwwdOpSNGzdSu3ZtFEWhf//+LF++nJs3b7J//35eeeUVAFavXk1QUBCBgYEcP368RDtPk0w7JUnl5O/v/1iTx0pPloW5S+GlxLu3PwpFUe56XFSdOyYmBnd3dyZPnlyiOnd5lj5otVo1OhX+jETduXMn27ZtY//+/VhZWREeHq7uMzU1Vftzr7aLl1eZM2cOtWrV4siRIxgMBiwsLNR9Li4u5ObmEhcXh6urKwBvvPEGXbp0wcLCgl69eqHVaklOTmbWrFkcOnQIe3t7Bg0aVOL9Pk3yTEySpGdS/QZjMTEpWT7HxMSyRLXth3H+/Hl1ucTKlStp3br1fatzl8bGxqZE2Zy6dety+PBhAA4fPqxWZM7IyMDe3h4rKyuSkpLULDAPKyMjAxcXF0xMTFi2bJkaKQvGpRs//vgjEyZMUAOFXF1dcXV1Zdq0aWo5llu3blGtWjVsbW25cuUKmzf/ecfozvf1pMlJTJKkZ5KLcze8vKZjYe4KKFiYu+LlNf2RoxM9PT35/PPP8fb25saNG4wYMQI7OzvefPNN/Pz86Nixo1r9+17atWtHYmKiGtjx2muvkZ6ejq+vL5999hkeHh4AdOrUCb1ej7e3N+PHj6dly5aP1P+RI0eydOlSmjRpQlJSUomzNIBatWqxadMm3nrrLQ4cOAAYy7EULboHaNKkCYGBgXh5efHXv/5VvbwKMHToUDp16vTUAjseKQHw4yYTAEuSdC8yAXDFGDVqFIGBgWrx1SfpQRMAy3tikiRJUpmCg4OpVq0an376aUV3pVRyEpMkSZLKFBsbW9FduCd5T0ySpCqlMt0CkR6vh/nZyklMkqQqw8LCgrS0NDmRPYOEEKSlpZUI+S8PeTlRkqQqw83NjQsXLqgZJKRni4WFBW5ubg90jJzEJEmqMkxNTalXr15Fd0OqROTlREmSJKnKkpOYJEmSVGXJSUySJEmqsipVxg5FUa4BpVf8q5ocgesV3YlKTo7RvcnxuTc5Pvf3LIxRHSFEzdJ2VKpJ7FmjKEpMWalSJCM5Rvcmx+fe5Pjc37M+RvJyoiRJklRlyUlMkiRJqrLkJPZkfVXRHagC5Bjdmxyfe5Pjc3/P9BjJe2KSJElSlSXPxCRJkqQqS05ikiRJUpUlJ7EnQFGUmYqiJCmKkqAoyveKotgV2zdBUZTfFUU5qShKxwrsZoVRFKWXoijHFUUxKIrS9I59z/34ACiK0qlwDH5XFGV8RfenMlAUZZGiKFcVRTlWbJuDoihbFUU5XfivfUX2sSIpiuKuKEq0oiiJhX9fbxduf6bHSE5iT8ZWwE8I4Q+cAiYAKIriA/QBfIFOwHxFUTQV1suKcwzoAewqvlGOj1Hhe/4ceAXwAV4vHJvn3RKMvxfFjQe2CyEaAdsLHz+v9MD/CSF8gJbAW4W/N8/0GMlJ7AkQQmwRQugLH/4GFNUW6AZ8J4TIE0IkA78DzSuijxVJCHFCCHGylF1yfIyaA78LIc4KIfKB7zCOzXNNCLELSL9jczdgaeH3S4GIp9mnykQIkSqEOFz4fSZwAniBZ3yM5CT25A0GNhd+/wLwR7F9Fwq3SUZyfIzkOJRfLSFEauH3l4FaFdmZykJRlLpAIHCAZ3yMZD2xh6QoyjbAuZRdE4UQGwqfMxHjKf6Kp9m3yqA84yNJj5MQQiiK8tyvGVIUxRpYB7wjhLilKIq671kcIzmJPSQhxEv32q8oyiCgM9Be/LkY7yLgXuxpboXbnjn3G58yPDfjcx9yHMrviqIoLkKIVEVRXICrFd2hiqQoiinGCWyFEGJ94eZneozk5cQnQFGUTsA4oKsQIrvYrh+APoqimCuKUg9oBBysiD5WUnJ8jA4BjRRFqacoihnGYJcfKrhPldUPwMDC7wcCz+1ZvmI85VoInBBCzC6265keI5mx4wlQFOV3wBxIK9z0mxBieOG+iRjvk+kxnu5vLr2VZ5eiKN2BeUBN4CYQL4ToWLjvuR8fAEVR/gLMBTTAIiHE9IrtUcVTFOVbIBxjaZErwCTgf8BqoDbGMk69hRB3Bn88FxRFaQ3sBo4ChsLN72O8L/bMjpGcxCRJkqQqS15OlCRJkqosOYlJkiRJVZacxCRJkqQqS05ikiRJUpUlJzFJkiSpypKTmCRJklRlyUlMkiRJqrL+Hxaij1XPX0yTAAAAAElFTkSuQmCC\n",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {
+ "needs_background": "light"
+ },
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "plt.figure(figsize = (7, 7))\n",
+ "\n",
+ "for label in unique_labels:\n",
+ " indices = [i for i in range(len(labels)) if labels[i] == label]\n",
+ " plt.scatter(tsne[indices, 0], tsne[indices, 1], cmap = palette(unique_labels.index(label)),\n",
+ " label = label)\n",
+ " \n",
+ "labels = x\n",
+ "for label, x, y in zip(\n",
+ " labels, tsne[:, 0], tsne[:, 1]\n",
+ "):\n",
+ " label = (\n",
+ " '%s, %.3f' % (label[0], label[1])\n",
+ " if isinstance(label, list)\n",
+ " else label\n",
+ " )\n",
+ " plt.annotate(\n",
+ " label,\n",
+ " xy = (x, y),\n",
+ " xytext = (0, 0),\n",
+ " textcoords = 'offset points',\n",
+ " )\n",
+ "plt.legend()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Stacking models\n",
+ "\n",
+ "More information, you can read at https://malaya.readthedocs.io/en/latest/Stack.html\n",
+ "\n",
+ "If you want to stack zero-shot classification models, you need to pass labels using keyword parameter,\n",
+ "\n",
+ "```python\n",
+ "malaya.stack.predict_stack([model1, model2], List[str], labels = List[str])\n",
+ "```\n",
+ "\n",
+ "We will passed `labels` as `**kwargs`."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "WARNING:tensorflow:From /usr/local/lib/python3.7/site-packages/albert/tokenization.py:240: The name tf.logging.info is deprecated. Please use tf.compat.v1.logging.info instead.\n",
+ "\n",
+ "INFO:tensorflow:loading sentence piece model\n"
+ ]
+ }
+ ],
+ "source": [
+ "alxlnet = malaya.zero_shot.classification.transformer(model = 'alxlnet')\n",
+ "albert = malaya.zero_shot.classification.transformer(model = 'albert')\n",
+ "tiny_bert = malaya.zero_shot.classification.transformer(model = 'tiny-bert')"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[{'makan': 0.0044827852,\n",
+ " 'makanan': 0.0027062024,\n",
+ " 'novel': 0.0020867025,\n",
+ " 'buku': 0.013082165,\n",
+ " 'kerajaan': 0.8859287,\n",
+ " 'food delivery': 0.0028363755,\n",
+ " 'kerajaan jahat': 0.018133936,\n",
+ " 'kerajaan prihatin': 0.9922408,\n",
+ " 'bantuan rakyat': 0.909674}]"
+ ]
+ },
+ "execution_count": 11,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "string = 'kerajaan sebenarnya sangat prihatin dengan rakyat, bagi duit bantuan'\n",
+ "labels = ['makan', 'makanan', 'novel', 'buku', 'kerajaan', 'food delivery', \n",
+ " 'kerajaan jahat', 'kerajaan prihatin', 'bantuan rakyat']\n",
+ "malaya.stack.predict_stack([alxlnet, albert, tiny_bert], [string], \n",
+ " labels = labels)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.7.7"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/malaya/__init__.py b/malaya/__init__.py
index 87cc37dc..2a8eaee3 100644
--- a/malaya/__init__.py
+++ b/malaya/__init__.py
@@ -13,8 +13,8 @@
home = os.path.join(str(Path.home()), 'Malaya')
-version = '3.9'
-bump_version = '3.9.2'
+version = '4.0'
+bump_version = '4.0'
version_path = os.path.join(home, 'version')
__version__ = bump_version
path = os.path.dirname(__file__)
diff --git a/malaya/model/bert.py b/malaya/model/bert.py
index f325a209..5dd7740f 100644
--- a/malaya/model/bert.py
+++ b/malaya/model/bert.py
@@ -729,16 +729,16 @@ def __init__(
segment_ids = segment_ids,
input_masks = input_masks,
logits = logits,
+ vectorizer = vectorizer,
sess = sess,
tokenizer = tokenizer,
label = label,
)
- self._vectorizer = vectorizer
self._softmax = tf.nn.softmax(self._logits)
self._batch_size = 20
def _base(self, strings_left, strings_right):
- input_ids, input_masks, segment_ids = bert_tokenization_siamese(
+ input_ids, input_masks, segment_ids, _ = bert_tokenization_siamese(
self._tokenizer, strings_left, strings_right
)
@@ -1066,6 +1066,7 @@ def __init__(
segment_ids,
input_masks,
logits,
+ vectorizer,
sess,
tokenizer,
label = ['not similar', 'similar'],
@@ -1076,6 +1077,7 @@ def __init__(
segment_ids = segment_ids,
input_masks = input_masks,
logits = logits,
+ vectorizer = vectorizer,
sess = sess,
tokenizer = tokenizer,
label = label,
@@ -1092,7 +1094,7 @@ def _base(self, strings, labels):
mapping[no].append(index)
index += 1
- input_ids, input_masks, segment_ids = bert_tokenization_siamese(
+ input_ids, input_masks, segment_ids, _ = bert_tokenization_siamese(
self._tokenizer, strings_left, strings_right
)
@@ -1113,6 +1115,69 @@ def _base(self, strings, labels):
results.append(result)
return results
+ @check_type
+ def vectorize(
+ self, strings: List[str], labels: List[str], method: str = 'first'
+ ):
+ """
+ vectorize a string.
+
+ Parameters
+ ----------
+ strings: List[str]
+ labels : List[str]
+ method : str, optional (default='first')
+ Vectorization layer supported. Allowed values:
+
+ * ``'last'`` - vector from last sequence.
+ * ``'first'`` - vector from first sequence.
+ * ``'mean'`` - average vectors from all sequences.
+ * ``'word'`` - average vectors based on tokens.
+
+
+ Returns
+ -------
+ result: np.array
+ """
+ strings_left, strings_right, combined = [], [], []
+ for no, string in enumerate(strings):
+ for label in labels:
+ strings_left.append(string)
+ strings_right.append(f'teks ini adalah mengenai {label}')
+ combined.append((string, label))
+
+ input_ids, input_masks, segment_ids, s_tokens = bert_tokenization_siamese(
+ self._tokenizer, strings_left, strings_right
+ )
+
+ v = self._sess.run(
+ self._vectorizer,
+ feed_dict = {
+ self._X: input_ids,
+ self._segment_ids: segment_ids,
+ self._input_masks: input_masks,
+ },
+ )
+ if len(v.shape) == 2:
+ v = v.reshape((*np.array(input_ids).shape, -1))
+
+ if method == 'first':
+ v = v[:, 0]
+ elif method == 'last':
+ v = v[:, -1]
+ elif method == 'mean':
+ v = np.mean(v, axis = 1)
+ else:
+ v = [
+ merge_sentencepiece_tokens(
+ list(zip(s_tokens[i], v[i][: len(s_tokens[i])])),
+ weighted = False,
+ vectorize = True,
+ )
+ for i in range(len(v))
+ ]
+ return combined, v
+
@check_type
def predict_proba(self, strings: List[str], labels: List[str]):
"""
diff --git a/malaya/model/xlnet.py b/malaya/model/xlnet.py
index d7762d08..1a443787 100644
--- a/malaya/model/xlnet.py
+++ b/malaya/model/xlnet.py
@@ -752,24 +752,22 @@ def __init__(
X = X,
segment_ids = segment_ids,
input_masks = input_masks,
+ vectorizer = vectorizer,
logits = logits,
sess = sess,
tokenizer = tokenizer,
label = label,
)
- self._vectorizer = vectorizer
self._softmax = tf.nn.softmax(self._logits)
self._batch_size = 20
def _base(self, strings_left, strings_right):
- input_ids, input_masks, segment_ids = xlnet_tokenization_siamese(
+ input_ids, input_masks, segment_ids, _ = xlnet_tokenization_siamese(
self._tokenizer, strings_left, strings_right
)
- segment_ids = np.array(segment_ids)
- batch_segment[batch_segment == 0] = 1
return self._sess.run(
- self._vectorizer,
+ self._softmax,
feed_dict = {
self._X: input_ids,
self._segment_ids: segment_ids,
@@ -793,6 +791,16 @@ def vectorize(self, strings: List[str]):
input_ids, input_masks, segment_ids, _ = xlnet_tokenization(
self._tokenizer, strings
)
+ segment_ids = np.array(segment_ids)
+ segment_ids[segment_ids == 0] = 1
+ return self._sess.run(
+ self._vectorizer,
+ feed_dict = {
+ self._X: input_ids,
+ self._segment_ids: segment_ids,
+ self._input_masks: input_masks,
+ },
+ )
@check_type
def predict_proba(self, strings_left: List[str], strings_right: List[str]):
@@ -1111,6 +1119,7 @@ def __init__(
segment_ids,
input_masks,
logits,
+ vectorizer,
sess,
tokenizer,
label = ['not similar', 'similar'],
@@ -1121,6 +1130,7 @@ def __init__(
segment_ids = segment_ids,
input_masks = input_masks,
logits = logits,
+ vectorizer = vectorizer,
sess = sess,
tokenizer = tokenizer,
label = label,
@@ -1138,7 +1148,7 @@ def _base(self, strings, labels):
mapping[no].append(index)
index += 1
- input_ids, input_masks, segment_ids = xlnet_tokenization_siamese(
+ input_ids, input_masks, segment_ids, _ = xlnet_tokenization_siamese(
self._tokenizer, strings_left, strings_right
)
@@ -1159,6 +1169,70 @@ def _base(self, strings, labels):
results.append(result)
return results
+ @check_type
+ def vectorize(
+ self, strings: List[str], labels: List[str], method: str = 'first'
+ ):
+ """
+ vectorize a string.
+
+ Parameters
+ ----------
+ strings: List[str]
+ labels : List[str]
+ method : str, optional (default='first')
+ Vectorization layer supported. Allowed values:
+
+ * ``'last'`` - vector from last sequence.
+ * ``'first'`` - vector from first sequence.
+ * ``'mean'`` - average vectors from all sequences.
+ * ``'word'`` - average vectors based on tokens.
+
+
+ Returns
+ -------
+ result: np.array
+ """
+
+ strings_left, strings_right, combined = [], [], []
+ for no, string in enumerate(strings):
+ for label in labels:
+ strings_left.append(string)
+ strings_right.append(f'teks ini adalah mengenai {label}')
+ combined.append((string, label))
+
+ input_ids, input_masks, segment_ids, s_tokens = xlnet_tokenization_siamese(
+ self._tokenizer, strings_left, strings_right
+ )
+
+ v = self._sess.run(
+ self._vectorizer,
+ feed_dict = {
+ self._X: input_ids,
+ self._segment_ids: segment_ids,
+ self._input_masks: input_masks,
+ },
+ )
+ v = np.transpose(v, [1, 0, 2])
+
+ if method == 'first':
+ v = v[:, 0]
+ elif method == 'last':
+ v = v[:, -1]
+ elif method == 'mean':
+ v = np.mean(v, axis = 1)
+ else:
+ v = [
+ merge_sentencepiece_tokens(
+ list(zip(s_tokens[i], v[i][: len(s_tokens[i])])),
+ weighted = False,
+ vectorize = True,
+ model = 'xlnet',
+ )
+ for i in range(len(v))
+ ]
+ return combined, v
+
@check_type
def predict_proba(self, strings: List[str], labels: List[str]):
"""
diff --git a/malaya/path/__init__.py b/malaya/path/__init__.py
index a3fc7016..7de4c010 100644
--- a/malaya/path/__init__.py
+++ b/malaya/path/__init__.py
@@ -274,7 +274,7 @@
},
'alxlnet': {
'model': 'v34/emotion/alxlnet-base-emotion.pb',
- 'quantized': 'v34/emotion/alxlnet-base-emotion.pb.quantized',
+ 'quantized': 'v40/emotion/alxlnet-base-emotion.pb.quantized',
'vocab': 'tokenizer/sp10m.cased.v9.vocab',
'tokenizer': 'tokenizer/sp10m.cased.v9.model',
},
@@ -857,36 +857,42 @@
PATH_SIMILARITY = {
'bert': {
'model': home + '/similarity/bert/base/model.pb',
+ 'quantized': home + '/similarity/bert/base/quantized/model.pb',
'vocab': home + '/bert/sp10m.cased.bert.vocab',
'tokenizer': home + '/bert/sp10m.cased.bert.model',
'version': 'v36',
},
'tiny-bert': {
'model': home + '/similarity/bert/tiny/model.pb',
+ 'quantized': home + '/similarity/bert/tiny/quantized/model.pb',
'vocab': home + '/bert/sp10m.cased.bert.vocab',
'tokenizer': home + '/bert/sp10m.cased.bert.model',
'version': 'v36',
},
'albert': {
'model': home + '/similarity/albert/base/model.pb',
+ 'quantized': home + '/similarity/albert/base/quantized/model.pb',
'vocab': home + '/albert/sp10m.cased.v10.vocab',
'tokenizer': home + '/albert/sp10m.cased.v10.model',
'version': 'v36',
},
'tiny-albert': {
'model': home + '/similarity/albert/tiny/model.pb',
+ 'quantized': home + '/similarity/albert/tiny/quantized/model.pb',
'vocab': home + '/bert/sp10m.cased.bert.vocab',
'tokenizer': home + '/bert/sp10m.cased.bert.model',
'version': 'v36',
},
'xlnet': {
'model': home + '/similarity/xlnet/base/model.pb',
+ 'quantized': home + '/similarity/xlnet/base/quantized/model.pb',
'vocab': home + '/xlnet/sp10m.cased.v9.vocab',
'tokenizer': home + '/xlnet/sp10m.cased.v9.model',
'version': 'v36',
},
'alxlnet': {
'model': home + '/similarity/alxlnet/base/model.pb',
+ 'quantized': home + '/similarity/alxlnet/base/quantized/model.pb',
'vocab': home + '/xlnet/sp10m.cased.v9.vocab',
'tokenizer': home + '/xlnet/sp10m.cased.v9.model',
'version': 'v36',
@@ -896,31 +902,37 @@
S3_PATH_SIMILARITY = {
'bert': {
'model': 'v36/similarity/bert-base-similarity.pb',
+ 'quantized': 'v40/similarity/bert-base-similarity.pb.quantized',
'vocab': 'tokenizer/sp10m.cased.bert.vocab',
'tokenizer': 'tokenizer/sp10m.cased.bert.model',
},
'tiny-bert': {
'model': 'v36/similarity/tiny-bert-similarity.pb',
+ 'quantized': 'v40/similarity/tiny-bert-similarity.pb.quantized',
'vocab': 'tokenizer/sp10m.cased.bert.vocab',
'tokenizer': 'tokenizer/sp10m.cased.bert.model',
},
'albert': {
'model': 'v36/similarity/albert-base-similarity.pb',
+ 'quantized': 'v40/similarity/albert-base-similarity.pb.quantized',
'vocab': 'tokenizer/sp10m.cased.v10.vocab',
'tokenizer': 'tokenizer/sp10m.cased.v10.model',
},
'tiny-albert': {
'model': 'v36/similarity/albert-tiny-similarity.pb',
+ 'quantized': 'v40/similarity/albert-tiny-similarity.pb.quantized',
'vocab': 'tokenizer/sp10m.cased.v10.vocab',
'tokenizer': 'tokenizer/sp10m.cased.v10.model',
},
'xlnet': {
'model': 'v36/similarity/xlnet-base-similarity.pb',
+ 'quantized': 'v40/similarity/xlnet-base-similarity.pb.quantized',
'vocab': 'tokenizer/sp10m.cased.v9.vocab',
'tokenizer': 'tokenizer/sp10m.cased.v9.model',
},
'alxlnet': {
'model': 'v36/similarity/alxlnet-base-similarity.pb',
+ 'quantized': 'v40/similarity/alxlnet-base-similarity.pb.quantized',
'vocab': 'tokenizer/sp10m.cased.v9.vocab',
'tokenizer': 'tokenizer/sp10m.cased.v9.model',
},
diff --git a/malaya/relevancy.py b/malaya/relevancy.py
index 2bb710a9..245aa01e 100644
--- a/malaya/relevancy.py
+++ b/malaya/relevancy.py
@@ -5,12 +5,32 @@
label = ['not relevant', 'relevant']
_transformer_availability = {
- 'bert': {'Size (MB)': 425.6, 'Accuracy': 0.872},
- 'tiny-bert': {'Size (MB)': 57.4, 'Accuracy': 0.656},
- 'albert': {'Size (MB)': 48.6, 'Accuracy': 0.871},
- 'tiny-albert': {'Size (MB)': 22.4, 'Accuracy': 0.843},
- 'xlnet': {'Size (MB)': 446.6, 'Accuracy': 0.885},
- 'alxlnet': {'Size (MB)': 46.8, 'Accuracy': 0.874},
+ 'bert': {'Size (MB)': 425.6, 'Quantized Size (MB)': 111, 'Accuracy': 0.872},
+ 'tiny-bert': {
+ 'Size (MB)': 57.4,
+ 'Quantized Size (MB)': 15.4,
+ 'Accuracy': 0.656,
+ },
+ 'albert': {
+ 'Size (MB)': 48.6,
+ 'Quantized Size (MB)': 12.8,
+ 'Accuracy': 0.85265,
+ },
+ 'tiny-albert': {
+ 'Size (MB)': 22.4,
+ 'Quantized Size (MB)': 5.98,
+ 'Accuracy': 0.843,
+ },
+ 'xlnet': {
+ 'Size (MB)': 446.6,
+ 'Quantized Size (MB)': 118,
+ 'Accuracy': 0.885,
+ },
+ 'alxlnet': {
+ 'Size (MB)': 46.8,
+ 'Quantized Size (MB)': 13.3,
+ 'Accuracy': 0.9123,
+ },
}
@@ -26,7 +46,7 @@ def available_transformer():
@check_type
-def transformer(model: str = 'xlnet', **kwargs):
+def transformer(model: str = 'xlnet', quantized: bool = False, **kwargs):
"""
Load Transformer relevancy model.
@@ -41,6 +61,9 @@ def transformer(model: str = 'xlnet', **kwargs):
* ``'tiny-albert'`` - Google ALBERT TINY parameters.
* ``'xlnet'`` - Google XLNET BASE parameters.
* ``'alxlnet'`` - Malaya ALXLNET BASE parameters.
+ quantized : bool, optional (default=False)
+ if True, will load 8-bit quantized model.
+ Quantized model not necessary faster, totally depends on the machine.
Returns
-------
@@ -58,5 +81,6 @@ def transformer(model: str = 'xlnet', **kwargs):
'relevancy',
label,
model = model,
+ quantized = quantized,
**kwargs
)
diff --git a/malaya/similarity.py b/malaya/similarity.py
index d02e54ea..0b0d2bc5 100644
--- a/malaya/similarity.py
+++ b/malaya/similarity.py
@@ -454,6 +454,15 @@ def encoder(vectorizer):
},
}
+_vectorizer_mapping = {
+ 'bert': 'import/bert/encoder/layer_11/output/LayerNorm/batchnorm/add_1:0',
+ 'tiny-bert': 'import/bert/encoder/layer_3/output/LayerNorm/batchnorm/add_1:0',
+ 'albert': 'import/bert/encoder/transformer/group_0_11/layer_11/inner_group_0/LayerNorm_1/batchnorm/add_1:0',
+ 'tiny-albert': 'import/bert/encoder/transformer/group_0_3/layer_3/inner_group_0/LayerNorm_1/batchnorm/add_1:0',
+ 'xlnet': 'import/model/transformer/layer_11/ff/LayerNorm/batchnorm/add_1:0',
+ 'alxlnet': 'import/model/transformer/layer_shared_11/ff/LayerNorm/batchnorm/add_1:0',
+}
+
def available_transformer():
"""
@@ -466,7 +475,9 @@ def available_transformer():
)
-def _transformer(model, bert_class, xlnet_class, quantized = False, **kwargs):
+def _transformer(
+ model, bert_class, xlnet_class, quantized = False, siamese = False, **kwargs
+):
model = model.lower()
if model not in _transformer_availability:
raise ValueError(
@@ -501,13 +512,18 @@ def _transformer(model, bert_class, xlnet_class, quantized = False, **kwargs):
)
selected_class = bert_class
- selected_node = 'import/bert/pooler/dense/BiasAdd:0'
+ if siamese:
+ selected_node = 'import/bert/pooler/dense/BiasAdd:0'
if model in ['xlnet', 'alxlnet']:
tokenizer = sentencepiece_tokenizer_xlnet(path[model]['tokenizer'])
selected_class = xlnet_class
- selected_node = 'import/model_1/sequnece_summary/summary/BiasAdd:0'
+ if siamese:
+ selected_node = 'import/model_1/sequnece_summary/summary/BiasAdd:0'
+
+ if not siamese:
+ selected_node = _vectorizer_mapping[model]
return selected_class(
X = g.get_tensor_by_name('import/Placeholder:0'),
@@ -552,5 +568,6 @@ def transformer(model: str = 'bert', quantized: bool = False, **kwargs):
bert_class = SIAMESE_BERT,
xlnet_class = SIAMESE_XLNET,
quantized = quantized,
+ siamese = True,
**kwargs
)
diff --git a/malaya/text/bpe.py b/malaya/text/bpe.py
index 68bb9580..7e732d9b 100644
--- a/malaya/text/bpe.py
+++ b/malaya/text/bpe.py
@@ -143,7 +143,7 @@ def _truncate_seq_pair(tokens_a, tokens_b, max_length):
def bert_tokenization_siamese(tokenizer, left, right):
- input_ids, input_masks, segment_ids = [], [], []
+ input_ids, input_masks, segment_ids, s_tokens = [], [], [], []
a, b = [], []
for i in range(len(left)):
tokens_a = tokenizer.tokenize(transformer_textcleaning(left[i]))
@@ -164,6 +164,7 @@ def bert_tokenization_siamese(tokenizer, left, right):
segment_id.append(0)
tokens.append('[SEP]')
+ s_tokens.append(tokens[:])
segment_id.append(0)
for token in tokens_b:
tokens.append(token)
@@ -182,7 +183,7 @@ def bert_tokenization_siamese(tokenizer, left, right):
input_masks = padding_sequence(input_masks, maxlen)
segment_ids = padding_sequence(segment_ids, maxlen)
- return input_ids, input_masks, segment_ids
+ return input_ids, input_masks, segment_ids, s_tokens
SEG_ID_A = 0
@@ -324,7 +325,7 @@ def tokenize_fn(text, sp_model):
def xlnet_tokenization_siamese(tokenizer, left, right):
- input_ids, input_mask, all_seg_ids = [], [], []
+ input_ids, input_mask, all_seg_ids, s_tokens = [], [], [], []
for i in range(len(left)):
tokens = tokenize_fn(transformer_textcleaning(left[i]), tokenizer)
tokens_right = tokenize_fn(
@@ -332,6 +333,7 @@ def xlnet_tokenization_siamese(tokenizer, left, right):
)
segment_ids = [SEG_ID_A] * len(tokens)
tokens.append(SEP_ID)
+ s_tokens.append([tokenizer.IdToPiece(i) for i in tokens])
segment_ids.append(SEG_ID_A)
tokens.extend(tokens_right)
@@ -354,7 +356,7 @@ def xlnet_tokenization_siamese(tokenizer, left, right):
input_ids = padding_sequence(input_ids, maxlen)
input_mask = padding_sequence(input_mask, maxlen, pad_int = 1)
all_seg_ids = padding_sequence(all_seg_ids, maxlen, pad_int = 4)
- return input_ids, input_mask, all_seg_ids
+ return input_ids, input_mask, all_seg_ids, s_tokens
def xlnet_tokenization(tokenizer, texts):
diff --git a/malaya/train/model/transformer/layer.py b/malaya/train/model/transformer/layer.py
new file mode 100644
index 00000000..e69de29b
diff --git a/malaya/zero_shot/classification.py b/malaya/zero_shot/classification.py
index a30171ea..b1b3926c 100644
--- a/malaya/zero_shot/classification.py
+++ b/malaya/zero_shot/classification.py
@@ -44,5 +44,6 @@ def transformer(model: str = 'bert', quantized: bool = False, **kwargs):
bert_class = ZEROSHOT_BERT,
xlnet_class = ZEROSHOT_XLNET,
quantized = quantized,
+ siamese = False,
**kwargs
)
diff --git a/session/emotion/quantize-emotion-model.ipynb b/session/emotion/quantize.ipynb
similarity index 67%
rename from session/emotion/quantize-emotion-model.ipynb
rename to session/emotion/quantize.ipynb
index 1484535b..14686d0b 100644
--- a/session/emotion/quantize-emotion-model.ipynb
+++ b/session/emotion/quantize.ipynb
@@ -120,13 +120,16 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
+ "WARNING:tensorflow:From :11: FastGFile.__init__ (from tensorflow.python.platform.gfile) is deprecated and will be removed in a future version.\n",
+ "Instructions for updating:\n",
+ "Use tf.gfile.GFile.\n",
"bert-base-emotion.pb ['Placeholder', 'Placeholder_1']\n",
"xlnet-base-emotion.pb ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
"alxlnet-base-emotion.pb ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
@@ -139,11 +142,9 @@
"source": [
"transforms = ['add_default_attributes',\n",
" 'remove_nodes(op=Identity, op=CheckNumerics, op=Dropout)',\n",
- " 'fold_constants(ignore_errors=true)',\n",
" 'fold_batch_norms',\n",
" 'fold_old_batch_norms',\n",
" 'quantize_weights(fallback_min=-10, fallback_max=10)',\n",
- " 'quantize_nodes(fallback_min=-10, fallback_max=10)',\n",
" 'strip_unused_nodes',\n",
" 'sort_by_execution_order']\n",
"\n",
@@ -154,14 +155,17 @@
" \n",
" if 'bert' in pb:\n",
" inputs = ['Placeholder', 'Placeholder_1']\n",
+ " outputs = ['dense/BiasAdd']\n",
+ " \n",
" if 'xlnet'in pb:\n",
" inputs = ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
+ " outputs = ['transpose_3']\n",
" \n",
" print(pb, inputs)\n",
" \n",
" transformed_graph_def = TransformGraph(input_graph_def, \n",
" inputs,\n",
- " ['logits', 'logits_seq'], transforms)\n",
+ " ['logits', 'logits_seq'] + outputs, transforms)\n",
" \n",
" with tf.gfile.GFile(f'{pb}.quantized', 'wb') as f:\n",
" f.write(transformed_graph_def.SerializeToString())"
@@ -173,58 +177,58 @@
"metadata": {},
"outputs": [],
"source": [
- "def load_graph(frozen_graph_filename, **kwargs):\n",
- " with tf.gfile.GFile(frozen_graph_filename, 'rb') as f:\n",
- " graph_def = tf.GraphDef()\n",
- " graph_def.ParseFromString(f.read())\n",
+ "# def load_graph(frozen_graph_filename, **kwargs):\n",
+ "# with tf.gfile.GFile(frozen_graph_filename, 'rb') as f:\n",
+ "# graph_def = tf.GraphDef()\n",
+ "# graph_def.ParseFromString(f.read())\n",
"\n",
- " # https://github.com/onnx/tensorflow-onnx/issues/77#issuecomment-445066091\n",
- " # to fix import T5\n",
- " for node in graph_def.node:\n",
- " if node.op == 'RefSwitch':\n",
- " node.op = 'Switch'\n",
- " for index in xrange(len(node.input)):\n",
- " if 'moving_' in node.input[index]:\n",
- " node.input[index] = node.input[index] + '/read'\n",
- " elif node.op == 'AssignSub':\n",
- " node.op = 'Sub'\n",
- " if 'use_locking' in node.attr:\n",
- " del node.attr['use_locking']\n",
- " elif node.op == 'AssignAdd':\n",
- " node.op = 'Add'\n",
- " if 'use_locking' in node.attr:\n",
- " del node.attr['use_locking']\n",
- " elif node.op == 'Assign':\n",
- " node.op = 'Identity'\n",
- " if 'use_locking' in node.attr:\n",
- " del node.attr['use_locking']\n",
- " if 'validate_shape' in node.attr:\n",
- " del node.attr['validate_shape']\n",
- " if len(node.input) == 2:\n",
- " node.input[0] = node.input[1]\n",
- " del node.input[1]\n",
+ "# # https://github.com/onnx/tensorflow-onnx/issues/77#issuecomment-445066091\n",
+ "# # to fix import T5\n",
+ "# for node in graph_def.node:\n",
+ "# if node.op == 'RefSwitch':\n",
+ "# node.op = 'Switch'\n",
+ "# for index in xrange(len(node.input)):\n",
+ "# if 'moving_' in node.input[index]:\n",
+ "# node.input[index] = node.input[index] + '/read'\n",
+ "# elif node.op == 'AssignSub':\n",
+ "# node.op = 'Sub'\n",
+ "# if 'use_locking' in node.attr:\n",
+ "# del node.attr['use_locking']\n",
+ "# elif node.op == 'AssignAdd':\n",
+ "# node.op = 'Add'\n",
+ "# if 'use_locking' in node.attr:\n",
+ "# del node.attr['use_locking']\n",
+ "# elif node.op == 'Assign':\n",
+ "# node.op = 'Identity'\n",
+ "# if 'use_locking' in node.attr:\n",
+ "# del node.attr['use_locking']\n",
+ "# if 'validate_shape' in node.attr:\n",
+ "# del node.attr['validate_shape']\n",
+ "# if len(node.input) == 2:\n",
+ "# node.input[0] = node.input[1]\n",
+ "# del node.input[1]\n",
"\n",
- " with tf.Graph().as_default() as graph:\n",
- " tf.import_graph_def(graph_def)\n",
- " return graph"
+ "# with tf.Graph().as_default() as graph:\n",
+ "# tf.import_graph_def(graph_def)\n",
+ "# return graph"
]
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
- "g = load_graph('xlnet-base-emotion.pb.quantized')\n",
- "x = g.get_tensor_by_name('import/Placeholder:0')\n",
- "x_len = g.get_tensor_by_name('import/Placeholder_1:0')\n",
- "x_len2 = g.get_tensor_by_name('import/Placeholder_2:0')\n",
- "logits = g.get_tensor_by_name('import/logits:0')"
+ "# g = load_graph('xlnet-base-emotion.pb.quantized')\n",
+ "# x = g.get_tensor_by_name('import/Placeholder:0')\n",
+ "# x_len = g.get_tensor_by_name('import/Placeholder_1:0')\n",
+ "# x_len2 = g.get_tensor_by_name('import/Placeholder_2:0')\n",
+ "# logits = g.get_tensor_by_name('import/logits:0')"
]
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
@@ -233,47 +237,27 @@
},
{
"cell_type": "code",
- "execution_count": 21,
+ "execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
- "test_sess = tf.InteractiveSession(graph = g)"
+ "# test_sess = tf.InteractiveSession(graph = g)"
]
},
{
"cell_type": "code",
- "execution_count": 22,
+ "execution_count": 11,
"metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "CPU times: user 2.54 s, sys: 187 ms, total: 2.72 s\n",
- "Wall time: 1.8 s\n"
- ]
- },
- {
- "data": {
- "text/plain": [
- "array([[-1.9607849, -3.294118 , -2.1960788, 10.039215 , 5.1764708,\n",
- " 3.7647057]], dtype=float32)"
- ]
- },
- "execution_count": 22,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
+ "outputs": [],
"source": [
- "%%time\n",
- "test_sess.run(logits, feed_dict = {x: [[1,2,3,3,4]], x_len: [[1,1,1,1,1]],\n",
- " x_len2: [[1,1,1,1,1]]})"
+ "# %%time\n",
+ "# test_sess.run(logits, feed_dict = {x: [[1,2,3,3,4]], x_len: [[1,1,1,1,1]],\n",
+ "# x_len2: [[1,1,1,1,1]]})"
]
},
{
"cell_type": "code",
- "execution_count": 23,
+ "execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
@@ -283,7 +267,7 @@
},
{
"cell_type": "code",
- "execution_count": 24,
+ "execution_count": 13,
"metadata": {},
"outputs": [
{
@@ -297,7 +281,7 @@
" 'albert-tiny-emotion.pb.quantized']"
]
},
- "execution_count": 24,
+ "execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
@@ -307,61 +291,6 @@
"quantized"
]
},
- {
- "cell_type": "code",
- "execution_count": 25,
- "metadata": {},
- "outputs": [],
- "source": [
- "from b2sdk.v1 import *\n",
- "info = InMemoryAccountInfo()\n",
- "b2_api = B2Api(info)\n",
- "application_key_id = 'd3c416cf4cb1'\n",
- "application_key = '0007c73b0ef09cbff76ebdd5b14f2e0044d6d44b74'\n",
- "b2_api.authorize_account(\"production\", application_key_id, application_key)\n",
- "file_info = {'how': 'good-file'}\n",
- "b2_bucket = b2_api.get_bucket_by_name('malaya-model')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 26,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "bert-base-emotion.pb.quantized\n",
- "albert-base-emotion.pb.quantized\n",
- "xlnet-base-emotion.pb.quantized\n",
- "tiny-bert-emotion.pb.quantized\n",
- "alxlnet-base-emotion.pb.quantized\n",
- "albert-tiny-emotion.pb.quantized\n"
- ]
- }
- ],
- "source": [
- "for file in quantized:\n",
- " print(file)\n",
- " key = file\n",
- " outPutname = f\"v40/emotion/{file}\"\n",
- " b2_bucket.upload_local_file(\n",
- " local_file=key,\n",
- " file_name=outPutname,\n",
- " file_infos=file_info,\n",
- " )"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 27,
- "metadata": {},
- "outputs": [],
- "source": [
- "!rm *.pb*"
- ]
- },
{
"cell_type": "code",
"execution_count": null,
diff --git a/session/entities/quantize-entity-model.ipynb b/session/entities/quantize.ipynb
similarity index 87%
rename from session/entities/quantize-entity-model.ipynb
rename to session/entities/quantize.ipynb
index aa67b9dc..7b2c81c6 100644
--- a/session/entities/quantize-entity-model.ipynb
+++ b/session/entities/quantize.ipynb
@@ -12,7 +12,7 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
@@ -58,7 +58,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 3,
"metadata": {},
"outputs": [
{
@@ -121,13 +121,16 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
+ "WARNING:tensorflow:From :11: FastGFile.__init__ (from tensorflow.python.platform.gfile) is deprecated and will be removed in a future version.\n",
+ "Instructions for updating:\n",
+ "Use tf.gfile.GFile.\n",
"xlnet-base-entity.pb ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
"alxlnet-base-entity.pb ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
"albert-tiny-entity.pb ['Placeholder', 'Placeholder_1']\n",
@@ -140,11 +143,9 @@
"source": [
"transforms = ['add_default_attributes',\n",
" 'remove_nodes(op=Identity, op=CheckNumerics, op=Dropout)',\n",
- " 'fold_constants(ignore_errors=true)',\n",
" 'fold_batch_norms',\n",
" 'fold_old_batch_norms',\n",
" 'quantize_weights(fallback_min=-10, fallback_max=10)',\n",
- " 'quantize_nodes(fallback_min=-10, fallback_max=10)',\n",
" 'strip_unused_nodes',\n",
" 'sort_by_execution_order']\n",
"\n",
@@ -155,14 +156,16 @@
" \n",
" if 'bert' in pb:\n",
" inputs = ['Placeholder', 'Placeholder_1']\n",
+ " outputs = ['dense/BiasAdd']\n",
" if 'xlnet'in pb:\n",
" inputs = ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
+ " outputs = ['transpose_3']\n",
" \n",
" print(pb, inputs)\n",
" \n",
" transformed_graph_def = TransformGraph(input_graph_def, \n",
" inputs,\n",
- " ['logits'], transforms)\n",
+ " ['logits'] + outputs, transforms)\n",
" \n",
" with tf.gfile.GFile(f'{pb}.quantized', 'wb') as f:\n",
" f.write(transformed_graph_def.SerializeToString())"
@@ -170,7 +173,7 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
@@ -212,7 +215,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
@@ -225,7 +228,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
@@ -234,7 +237,7 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
@@ -243,24 +246,24 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 13,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "CPU times: user 2.78 s, sys: 179 ms, total: 2.96 s\n",
- "Wall time: 1.93 s\n"
+ "CPU times: user 2.58 s, sys: 615 ms, total: 3.19 s\n",
+ "Wall time: 2.68 s\n"
]
},
{
"data": {
"text/plain": [
- "array([[8, 7, 0, 0, 8]], dtype=int32)"
+ "array([[2, 2, 2, 0, 0]], dtype=int32)"
]
},
- "execution_count": 15,
+ "execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
@@ -273,7 +276,7 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
@@ -283,7 +286,7 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": 15,
"metadata": {},
"outputs": [
{
@@ -297,7 +300,7 @@
" 'albert-base-entity.pb.quantized']"
]
},
- "execution_count": 17,
+ "execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
@@ -307,52 +310,6 @@
"quantized"
]
},
- {
- "cell_type": "code",
- "execution_count": 18,
- "metadata": {},
- "outputs": [],
- "source": [
- "from b2sdk.v1 import *\n",
- "info = InMemoryAccountInfo()\n",
- "b2_api = B2Api(info)\n",
- "application_key_id = 'd3c416cf4cb1'\n",
- "application_key = '0007c73b0ef09cbff76ebdd5b14f2e0044d6d44b74'\n",
- "b2_api.authorize_account(\"production\", application_key_id, application_key)\n",
- "file_info = {'how': 'good-file'}\n",
- "b2_bucket = b2_api.get_bucket_by_name('malaya-model')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 19,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "bert-base-entity.pb.quantized\n",
- "tiny-bert-entity.pb.quantized\n",
- "alxlnet-base-entity.pb.quantized\n",
- "xlnet-base-entity.pb.quantized\n",
- "albert-tiny-entity.pb.quantized\n",
- "albert-base-entity.pb.quantized\n"
- ]
- }
- ],
- "source": [
- "for file in quantized:\n",
- " print(file)\n",
- " key = file\n",
- " outPutname = f\"v40/entity/{file}\"\n",
- " b2_bucket.upload_local_file(\n",
- " local_file=key,\n",
- " file_name=outPutname,\n",
- " file_infos=file_info,\n",
- " )"
- ]
- },
{
"cell_type": "code",
"execution_count": null,
diff --git a/session/sentiment/quantize.ipynb b/session/sentiment/quantize.ipynb
index c95c99a9..418ef0d5 100644
--- a/session/sentiment/quantize.ipynb
+++ b/session/sentiment/quantize.ipynb
@@ -58,7 +58,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 3,
"metadata": {},
"outputs": [
{
@@ -95,7 +95,7 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 12,
"metadata": {},
"outputs": [
{
@@ -109,7 +109,7 @@
" 'alxlnet-base-sentiment.pb']"
]
},
- "execution_count": 7,
+ "execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
@@ -121,45 +121,77 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# with tf.gfile.GFile('alxlnet-base-sentiment.pb', \"rb\") as f:\n",
+ "# graph_def = tf.GraphDef()\n",
+ "# graph_def.ParseFromString(f.read())\n",
+ "\n",
+ "# with tf.Graph().as_default() as graph:\n",
+ "# tf.import_graph_def(graph_def)\n",
+ "\n",
+ "# op = graph.get_operations()\n",
+ "# x = []\n",
+ "# for i in op:\n",
+ "# try:\n",
+ "# #if 'pooler' in i.values()[0].name:\n",
+ "# x.append(i.values())\n",
+ "# except:\n",
+ "# pass\n",
+ " \n",
+ "# x[-100:]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
- "albert-tiny-sentiment.pb\n",
- "WARNING:tensorflow:From :14: FastGFile.__init__ (from tensorflow.python.platform.gfile) is deprecated and will be removed in a future version.\n",
+ "WARNING:tensorflow:From :11: FastGFile.__init__ (from tensorflow.python.platform.gfile) is deprecated and will be removed in a future version.\n",
"Instructions for updating:\n",
"Use tf.gfile.GFile.\n",
- "xlnet-base-sentiment.pb\n",
- "albert-base-sentiment.pb\n",
- "tiny-bert-sentiment.pb\n",
- "bert-base-sentiment.pb\n",
- "alxlnet-base-sentiment.pb\n"
+ "albert-tiny-sentiment.pb ['Placeholder', 'Placeholder_1']\n",
+ "xlnet-base-sentiment.pb ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
+ "albert-base-sentiment.pb ['Placeholder', 'Placeholder_1']\n",
+ "tiny-bert-sentiment.pb ['Placeholder', 'Placeholder_1']\n",
+ "bert-base-sentiment.pb ['Placeholder', 'Placeholder_1']\n",
+ "alxlnet-base-sentiment.pb ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n"
]
}
],
"source": [
"transforms = ['add_default_attributes',\n",
" 'remove_nodes(op=Identity, op=CheckNumerics, op=Dropout)',\n",
- " 'fold_constants(ignore_errors=true)',\n",
" 'fold_batch_norms',\n",
" 'fold_old_batch_norms',\n",
" 'quantize_weights(fallback_min=-10, fallback_max=10)',\n",
- " 'quantize_nodes(fallback_min=-10, fallback_max=10)',\n",
" 'strip_unused_nodes',\n",
" 'sort_by_execution_order']\n",
"\n",
"for pb in pbs:\n",
- " print(pb)\n",
" input_graph_def = tf.GraphDef()\n",
" with tf.gfile.FastGFile(pb, 'rb') as f:\n",
" input_graph_def.ParseFromString(f.read())\n",
" \n",
+ " if 'bert' in pb:\n",
+ " inputs = ['Placeholder', 'Placeholder_1']\n",
+ " outputs = ['dense/BiasAdd']\n",
+ " \n",
+ " if 'xlnet'in pb:\n",
+ " inputs = ['Placeholder', 'Placeholder_1', 'Placeholder_2']\n",
+ " outputs = ['transpose_3']\n",
+ " \n",
+ " print(pb, inputs)\n",
+ " \n",
" transformed_graph_def = TransformGraph(input_graph_def, \n",
- " ['Placeholder', 'Placeholder_1'],\n",
- " ['logits', 'logits_seq'], transforms)\n",
+ " inputs,\n",
+ " ['logits', 'logits_seq'] + outputs, transforms)\n",
" \n",
" with tf.gfile.GFile(f'{pb}.quantized', 'wb') as f:\n",
" f.write(transformed_graph_def.SerializeToString())"
@@ -167,7 +199,7 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
@@ -209,7 +241,7 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
@@ -221,7 +253,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
@@ -230,7 +262,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
@@ -239,7 +271,7 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
@@ -249,7 +281,7 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
@@ -259,7 +291,7 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": 14,
"metadata": {},
"outputs": [
{
@@ -273,7 +305,7 @@
" 'tiny-bert-sentiment.pb.quantized']"
]
},
- "execution_count": 17,
+ "execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
@@ -283,52 +315,6 @@
"quantized"
]
},
- {
- "cell_type": "code",
- "execution_count": 18,
- "metadata": {},
- "outputs": [],
- "source": [
- "from b2sdk.v1 import *\n",
- "info = InMemoryAccountInfo()\n",
- "b2_api = B2Api(info)\n",
- "application_key_id = 'd3c416cf4cb1'\n",
- "application_key = '0007c73b0ef09cbff76ebdd5b14f2e0044d6d44b74'\n",
- "b2_api.authorize_account(\"production\", application_key_id, application_key)\n",
- "file_info = {'how': 'good-file'}\n",
- "b2_bucket = b2_api.get_bucket_by_name('malaya-model')"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 19,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "albert-base-sentiment.pb.quantized\n",
- "xlnet-base-sentiment.pb.quantized\n",
- "albert-tiny-sentiment.pb.quantized\n",
- "bert-base-sentiment.pb.quantized\n",
- "alxlnet-base-sentiment.pb.quantized\n",
- "tiny-bert-sentiment.pb.quantized\n"
- ]
- }
- ],
- "source": [
- "for file in quantized:\n",
- " print(file)\n",
- " key = file\n",
- " outPutname = f\"v40/sentiment/{file}\"\n",
- " b2_bucket.upload_local_file(\n",
- " local_file=key,\n",
- " file_name=outPutname,\n",
- " file_infos=file_info,\n",
- " )"
- ]
- },
{
"cell_type": "code",
"execution_count": null,