Skip to content

Instantly share code, notes, and snippets.

@natzir

natzir/SERP-analysis.ipynb

Created October 7, 2019 14:22
Show Gist options
  • Save natzir/64f03d3008185024ee621967085a3efe to your computer and use it in GitHub Desktop.
Save natzir/64f03d3008185024ee621967085a3efe to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
SERP-analysis.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<h1>SERP Analysis using Spacy</h1>\n",
"<br>Based on the idea of Rory Truesdale: <a href=\"https://www.slideshare.net/RoryTruesdale/brightonseo-2019-mining-the-serp-for-seo-content-customer-insights\">https://www.slideshare.net/RoryTruesdale/brightonseo-2019-mining-the-serp-for-seo-content-customer-insights</a>\n",
"Scraping the SERP thanks to: <a href=\"https://stackoverflow.com/questions/56392962/scrape-google-search-results-titles-and-urls-using-python\">https://stackoverflow.com/questions/56392962/scrape-google-search-results-titles-and-urls-using-python </a>"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Requirement already satisfied: es_core_news_sm==2.1.0 from https://github.com/explosion/spacy-models/releases/download/es_core_news_sm-2.1.0/es_core_news_sm-2.1.0.tar.gz#egg=es_core_news_sm==2.1.0 in c:\\users\\nturr\\anaconda3\\lib\\site-packages (2.1.0)\n",
"[+] Download and installation successful\n",
"You can now load the model via spacy.load('es_core_news_sm')\n"
]
}
],
"source": [
"import pandas as pd\n",
"from collections import defaultdict\n",
"import string\n",
"from selenium import webdriver\n",
"from bs4 import BeautifulSoup\n",
"import time\n",
"from bs4.element import Tag\n",
"import spacy\n",
"!python -m spacy download es_core_news_sm\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"from IPython.display import display"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"query = 'vestidos de fiesta'\n",
"number_of_results = 20\n",
"language = 'es'\n",
"country = 'ES'\n",
"google = 'google.es'"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Download <a href=\"http://chromedriver.chromium.org/downloads\">Chrome Driver</a> and replace 'C:/Users/nturr/chromedriver' with your driver path."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"driver = webdriver.Chrome('C:/Users/nturr/chromedriver')\n",
"driver.get('https://www.'+ google + '/search?q='+query.replace(' ', '+') \n",
" + '&num=' + str(number_of_results) \n",
" + '&hl=' + str(language)\n",
" + '&gl=' + str(country)\n",
" + '&pws=0')\n",
"time.sleep(3)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['Vestidos de Fiesta de Mujer · Moda · El Corte Inglés',\n",
" 'colección - Vestidos de fiesta | Marfil Barcelona',\n",
" 'Vestidos de fiesta y de gala en Zalando',\n",
" 'Vestidos de fiesta - Nueva colección 2019 | Aire Barcelona',\n",
" 'Veneno en la piel: Vestidos para invitadas y novias hechos ...',\n",
" 'OUTLETS de vestidos de fiesta en Barcelona - ¡HASTA 70 ...',\n",
" 'Vestidos de Fiesta Cortos y Largos | Rosa Clará',\n",
" 'Tiendas Vestidos de Fiesta Baratos Barcelona - DreamyDress',\n",
" 'Vestidos de fiesta | Atrevidos vestidos largos y cortos de fiesta ...',\n",
" 'Las 11 mejores tiendas de vestidos de fiesta en Barcelona',\n",
" 'Vestidos de fiesta para Mujer en C&A Online',\n",
" 'Vestidos de Fiesta Barcelona y Hospitalet de Llobregat',\n",
" 'Fiesta - Vestidos de Mujer 2019 | Mango España',\n",
" 'Vestidos de Fiesta Online | Vestidos Invitada Boda | Atelier ...',\n",
" 'Vestidos de fiesta mujer | Nueva Colección Online | ZARA ...',\n",
" 'Tienda online de vestidos de fiesta - Invitada perfecta',\n",
" 'Vestidos de fiesta. Vestidos de fiesta cortos y largos. Matilde ...',\n",
" 'Vestidos de fiesta 2019. Colección Primavera Verano 2019 ...',\n",
" 'Vestidos para Fiesta de Mujer | Vestidos Coctel - Pronovias']"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"['Encuentra tu vestido de fiesta para tus bodas de día o noche, de cóctel, para bautizos o comuniones. Encontrarás vestidos cortos, largos, de encaje, lentejuelas.',\n",
" 'Vestidos de fiesta para la mujer cosmopolita, sensual y femenina que busca lucir el look de invitada perfecto. Descubre las nuevas colecciones.',\n",
" 'ENVÍO Y DEVOLUCIÓN GRATIS | Vestidos de fiesta para todo tipo de celebraciones en Zalando.',\n",
" 'TU VESTIDO IDEAL | Encuentra el vestido de fiesta perfecto, Aire Barcelona dispone de una variedad de vestidos de fiesta largos, cortos, de encaje y mucho\\xa0...',\n",
" 'Vestidos para invitadas y novias hechos con amor. Made in Barcelona. Outlet de vestidos de fiesta y vestidos de graduación.',\n",
" '12 feb. 2019 - Descubre estas increíbles TIENDAS OUTLET de vestidos de fiesta en Barcelona. ¡Con descuentos de hasta el 70%!',\n",
" 'Elegantes y sofisticados vestidos de fiesta cortos y largos, para mujeres que buscan ser las invitadas ideales en eventos o celebraciones tipo cóctel.',\n",
" 'Navegar por una amplia selección de vestidos baratos en Barcelona. Una variedad de vestidos Barcelona están disponibles con la mejor calidad y bajo precio.',\n",
" 'Llamando a todas las chicas elegantes: lúcete y disfruta de la atención que te mereces con la selección de fabulosos vestidos de fiesta de boohoo. Desde\\xa0...',\n",
" \"26 may. 2019 - Luce un 'modelito' de alfombra roja con los fantásticos conjuntos que te ofrecen las mejores tiendas de vestidos de Barcelona.\",\n",
" 'Los vestidos de fiesta y de cóctel que te ofrecemos desbordan siempre elegancia a precios asequibles♥ Encuentra el tuyo en C&A Online.',\n",
" 'Tienda especializada en Vestidos de Fiestas para bodas, eventos, fiestas, en Barcelona y Hospitalet, vestidos de fiesta, sastrería, Fuentecapala, Outlet, vestidos\\xa0...',\n",
" 'Últimas tendencias en vestidos de mujer. Nuevos modelos cada semana: vestidos cortos, largos, de fiesta o de noche. Envío y devoluciones gratis.',\n",
" 'Vestidos de fiesta online y vestidos de invitada de boda con diseños increíbles para que te conviertas en la invitada perfecta.',\n",
" 'ENVÍO GRATUITO. El vestido perfecto para esa fiesta o celebración. Descubre las siluetas que están de moda esta temporada.',\n",
" 'Mariquita Trasquilá, su tienda de moda online. Fiesta y Lowcost. Vestidos, Blusas, Camisetas, Zapatos, Bolsos, Complementos ¡Haz tu pedido online!',\n",
" 'Vestidos de fiesta de Matilde Cano. Vestidos para fiestas, bodas, vestidos de celebración, vestidos de madrina para que luzcas espléndida.',\n",
" 'Vestidos de Fiesta 2019. Colección Primavera Verano 2019 Balcón del Mar de Sonia Peña, diseñadora de modas de los mejores vestidos de fiesta, vestidos de\\xa0...',\n",
" 'Descubre los vestidos para fiestas y celebraciones de Pronovias. Vestidos largos.']"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"['https://www.elcorteingles.es/moda/fiesta-mujer/vestidos/',\n",
" 'https://www.marfilbarcelona.com/c/vestidos/de-fiesta/',\n",
" 'https://www.zalando.es/vestidos-fiesta/',\n",
" 'https://www.airebarcelona.com/coleccion/fiesta/coleccion-2019/',\n",
" 'http://www.venenoenlapiel.com/es/',\n",
" 'https://www.salir.com/los-mejores-outlet-de-vestidos-de-fiesta-en-barcelona-art-1325.html',\n",
" 'https://www.rosaclara.es/c/vestidos-de-fiesta/',\n",
" 'https://www.dreamydress.es/vestidos-de-fiesta-barcelona.html',\n",
" 'https://es.boohoo.com/mujer/vestidos/vestidos-de-fiesta',\n",
" 'https://www.zankyou.es/p/las-mejores-tiendas-de-vestidos-de-fiesta-en-barcelona-177419',\n",
" 'https://www.c-and-a.com/es/es/shop/mujer-categorias-vestidos-y-monos-vestidos-de-fiesta',\n",
" 'http://www.vestidosfiestabarcelona.es/',\n",
" 'https://shop.mango.com/es/mujer/vestidos-fiesta_c97893681',\n",
" 'https://www.atelierbadajoz.com/categoria-producto/fiesta/',\n",
" 'https://www.zara.com/es/es/mujer-vestidos-fiesta-l1581.html',\n",
" 'https://mariquitatrasquila.com/',\n",
" 'https://www.matildecano.es/vestidos-de-fiesta/',\n",
" 'https://www.soniapena.com/es/vestidos-de-fiesta',\n",
" 'https://www.pronovias.com/es/vestidos-fiesta/']"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"soup = BeautifulSoup(driver.page_source,'lxml')\n",
"result_div = soup.find_all('div', attrs={'class': 'g'})\n",
"\n",
"links = []\n",
"titles = []\n",
"descriptions = []\n",
"\n",
"for r in result_div:\n",
" # Checks if each element is present, else, raise exception\n",
" try:\n",
" link = r.find('a', href=True)\n",
" title = None\n",
" title = r.find('h3')\n",
"\n",
" if isinstance(title,Tag):\n",
" title = title.get_text()\n",
"\n",
" description = None\n",
" description = r.find('span', attrs={'class': 'st'})\n",
"\n",
" if isinstance(description, Tag):\n",
" description = description.get_text()\n",
"\n",
" # Check to make sure everything is present before appending\n",
" if link != '' and title != '' and description != '':\n",
" links.append(link['href'])\n",
" titles.append(title)\n",
" descriptions.append(description)\n",
" # Next loop if one element is not present\n",
" except Exception as e:\n",
" print(e)\n",
" continue\n",
"\n",
"display(titles)\n",
"display(descriptions)\n",
"display(links)"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"\"Vestidos de Fiesta de Mujer · Moda · El Corte Inglés colección - Vestidos de fiesta | Marfil Barcelona Vestidos de fiesta y de gala en Zalando Vestidos de fiesta - Nueva colección 2019 | Aire Barcelona Veneno en la piel: Vestidos para invitadas y novias hechos ... OUTLETS de vestidos de fiesta en Barcelona - ¡HASTA 70 ... Vestidos de Fiesta Cortos y Largos | Rosa Clará Tiendas Vestidos de Fiesta Baratos Barcelona - DreamyDress Vestidos de fiesta | Atrevidos vestidos largos y cortos de fiesta ... Las 11 mejores tiendas de vestidos de fiesta en Barcelona Vestidos de fiesta para Mujer en C&A Online Vestidos de Fiesta Barcelona y Hospitalet de Llobregat Fiesta - Vestidos de Mujer 2019 | Mango España Vestidos de Fiesta Online | Vestidos Invitada Boda | Atelier ... Vestidos de fiesta mujer | Nueva Colección Online | ZARA ... Tienda online de vestidos de fiesta - Invitada perfecta Vestidos de fiesta. Vestidos de fiesta cortos y largos. Matilde ... Vestidos de fiesta 2019. Colección Primavera Verano 2019 ... Vestidos para Fiesta de Mujer | Vestidos Coctel - Pronovias Encuentra tu vestido de fiesta para tus bodas de día o noche, de cóctel, para bautizos o comuniones. Encontrarás vestidos cortos, largos, de encaje, lentejuelas. Vestidos de fiesta para la mujer cosmopolita, sensual y femenina que busca lucir el look de invitada perfecto. Descubre las nuevas colecciones. ENVÍO Y DEVOLUCIÓN GRATIS | Vestidos de fiesta para todo tipo de celebraciones en Zalando. TU VESTIDO IDEAL | Encuentra el vestido de fiesta perfecto, Aire Barcelona dispone de una variedad de vestidos de fiesta largos, cortos, de encaje y mucho\\xa0... Vestidos para invitadas y novias hechos con amor. Made in Barcelona. Outlet de vestidos de fiesta y vestidos de graduación. 12 feb. 2019 - Descubre estas increíbles TIENDAS OUTLET de vestidos de fiesta en Barcelona. ¡Con descuentos de hasta el 70%! Elegantes y sofisticados vestidos de fiesta cortos y largos, para mujeres que buscan ser las invitadas ideales en eventos o celebraciones tipo cóctel. Navegar por una amplia selección de vestidos baratos en Barcelona. Una variedad de vestidos Barcelona están disponibles con la mejor calidad y bajo precio. Llamando a todas las chicas elegantes: lúcete y disfruta de la atención que te mereces con la selección de fabulosos vestidos de fiesta de boohoo. Desde\\xa0... 26 may. 2019 - Luce un 'modelito' de alfombra roja con los fantásticos conjuntos que te ofrecen las mejores tiendas de vestidos de Barcelona. Los vestidos de fiesta y de cóctel que te ofrecemos desbordan siempre elegancia a precios asequibles♥ Encuentra el tuyo en C&A Online. Tienda especializada en Vestidos de Fiestas para bodas, eventos, fiestas, en Barcelona y Hospitalet, vestidos de fiesta, sastrería, Fuentecapala, Outlet, vestidos\\xa0... Últimas tendencias en vestidos de mujer. Nuevos modelos cada semana: vestidos cortos, largos, de fiesta o de noche. Envío y devoluciones gratis. Vestidos de fiesta online y vestidos de invitada de boda con diseños increíbles para que te conviertas en la invitada perfecta. ENVÍO GRATUITO. El vestido perfecto para esa fiesta o celebración. Descubre las siluetas que están de moda esta temporada. Mariquita Trasquilá, su tienda de moda online. Fiesta y Lowcost. Vestidos, Blusas, Camisetas, Zapatos, Bolsos, Complementos ¡Haz tu pedido online! Vestidos de fiesta de Matilde Cano. Vestidos para fiestas, bodas, vestidos de celebración, vestidos de madrina para que luzcas espléndida. Vestidos de Fiesta 2019. Colección Primavera Verano 2019 Balcón del Mar de Sonia Peña, diseñadora de modas de los mejores vestidos de fiesta, vestidos de\\xa0... Descubre los vestidos para fiestas y celebraciones de Pronovias. Vestidos largos.\""
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"text = titles + descriptions\n",
"textp = \" \".join(text)\n",
"textp"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Word</th>\n",
" <th>POS</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <td>0</td>\n",
" <td>Vestidos</td>\n",
" <td>ADJ</td>\n",
" </tr>\n",
" <tr>\n",
" <td>1</td>\n",
" <td>de</td>\n",
" <td>ADP</td>\n",
" </tr>\n",
" <tr>\n",
" <td>2</td>\n",
" <td>Fiesta</td>\n",
" <td>PROPN</td>\n",
" </tr>\n",
" <tr>\n",
" <td>3</td>\n",
" <td>de</td>\n",
" <td>ADP</td>\n",
" </tr>\n",
" <tr>\n",
" <td>4</td>\n",
" <td>Mujer</td>\n",
" <td>PROPN</td>\n",
" </tr>\n",
" <tr>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <td>671</td>\n",
" <td>Pronovias</td>\n",
" <td>PROPN</td>\n",
" </tr>\n",
" <tr>\n",
" <td>672</td>\n",
" <td>.</td>\n",
" <td>PUNCT</td>\n",
" </tr>\n",
" <tr>\n",
" <td>673</td>\n",
" <td>Vestidos</td>\n",
" <td>ADJ</td>\n",
" </tr>\n",
" <tr>\n",
" <td>674</td>\n",
" <td>largos</td>\n",
" <td>ADJ</td>\n",
" </tr>\n",
" <tr>\n",
" <td>675</td>\n",
" <td>.</td>\n",
" <td>PUNCT</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>676 rows × 2 columns</p>\n",
"</div>"
],
"text/plain": [
" Word POS\n",
"0 Vestidos ADJ\n",
"1 de ADP\n",
"2 Fiesta PROPN\n",
"3 de ADP\n",
"4 Mujer PROPN\n",
".. ... ...\n",
"671 Pronovias PROPN\n",
"672 . PUNCT\n",
"673 Vestidos ADJ\n",
"674 largos ADJ\n",
"675 . PUNCT\n",
"\n",
"[676 rows x 2 columns]"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"nlp = spacy.load('es_core_news_sm')\n",
"doc = nlp(textp)\n",
"\n",
"data = []\n",
"for token in doc:\n",
" data.append([token.text, token.pos_])\n",
" \n",
"df = pd.DataFrame(data, columns = ['Word', 'POS'])\n",
"\n",
"df"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Removing stopwords and normalizing"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\nturr\\Anaconda3\\lib\\site-packages\\pandas\\core\\indexing.py:576: SettingWithCopyWarning: \n",
"A value is trying to be set on a copy of a slice from a DataFrame.\n",
"Try using .loc[row_indexer,col_indexer] = value instead\n",
"\n",
"See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
" self.obj[item_labels[indexer[info_axis]]] = value\n",
"C:\\Users\\nturr\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:4: SettingWithCopyWarning: \n",
"A value is trying to be set on a copy of a slice from a DataFrame.\n",
"Try using .loc[row_indexer,col_indexer] = value instead\n",
"\n",
"See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
" after removing the cwd from sys.path.\n"
]
},
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Word</th>\n",
" <th>POS</th>\n",
" <th>Count</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <td>0</td>\n",
" <td>vestidos</td>\n",
" <td>ADJ</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>2</td>\n",
" <td>fiesta</td>\n",
" <td>PROPN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>4</td>\n",
" <td>mujer</td>\n",
" <td>PROPN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>5</td>\n",
" <td>·</td>\n",
" <td>PROPN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>6</td>\n",
" <td>moda</td>\n",
" <td>PROPN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <td>667</td>\n",
" <td>fiestas</td>\n",
" <td>NOUN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>669</td>\n",
" <td>celebraciones</td>\n",
" <td>NOUN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>671</td>\n",
" <td>pronovias</td>\n",
" <td>PROPN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>673</td>\n",
" <td>vestidos</td>\n",
" <td>ADJ</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>674</td>\n",
" <td>largos</td>\n",
" <td>ADJ</td>\n",
" <td>1</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>364 rows × 3 columns</p>\n",
"</div>"
],
"text/plain": [
" Word POS Count\n",
"0 vestidos ADJ 1\n",
"2 fiesta PROPN 1\n",
"4 mujer PROPN 1\n",
"5 · PROPN 1\n",
"6 moda PROPN 1\n",
".. ... ... ...\n",
"667 fiestas NOUN 1\n",
"669 celebraciones NOUN 1\n",
"671 pronovias PROPN 1\n",
"673 vestidos ADJ 1\n",
"674 largos ADJ 1\n",
"\n",
"[364 rows x 3 columns]"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"remove = ['ADP','PUNCT','SYM','DET','CONJ','SPACE','PRON', 'AUX']\n",
"df = df[~df.POS.isin(remove)]\n",
"df.loc[:,'Word'] = df['Word'].str.lower()\n",
"df['Count'] = 1\n",
"df"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Selecting most frequent POS and removing Words with non-alphanumeric characters "
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Word</th>\n",
" <th>POS</th>\n",
" <th>Count</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <td>5</td>\n",
" <td>aire</td>\n",
" <td>PROPN</td>\n",
" <td>2</td>\n",
" </tr>\n",
" <tr>\n",
" <td>6</td>\n",
" <td>alfombra</td>\n",
" <td>NOUN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>7</td>\n",
" <td>amor</td>\n",
" <td>NOUN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>8</td>\n",
" <td>amplia</td>\n",
" <td>ADJ</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>9</td>\n",
" <td>asequibles</td>\n",
" <td>ADJ</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <td>153</td>\n",
" <td>vestido</td>\n",
" <td>ADJ</td>\n",
" <td>4</td>\n",
" </tr>\n",
" <tr>\n",
" <td>154</td>\n",
" <td>vestidos</td>\n",
" <td>ADJ</td>\n",
" <td>54</td>\n",
" </tr>\n",
" <tr>\n",
" <td>155</td>\n",
" <td>zalando</td>\n",
" <td>PROPN</td>\n",
" <td>2</td>\n",
" </tr>\n",
" <tr>\n",
" <td>156</td>\n",
" <td>zapatos</td>\n",
" <td>PROPN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>157</td>\n",
" <td>zara</td>\n",
" <td>PROPN</td>\n",
" <td>1</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>152 rows × 3 columns</p>\n",
"</div>"
],
"text/plain": [
" Word POS Count\n",
"5 aire PROPN 2\n",
"6 alfombra NOUN 1\n",
"7 amor NOUN 1\n",
"8 amplia ADJ 1\n",
"9 asequibles ADJ 1\n",
".. ... ... ...\n",
"153 vestido ADJ 4\n",
"154 vestidos ADJ 54\n",
"155 zalando PROPN 2\n",
"156 zapatos PROPN 1\n",
"157 zara PROPN 1\n",
"\n",
"[152 rows x 3 columns]"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_group = df.groupby('Word').agg({'POS' : lambda x:x.value_counts().index[0], 'Count':'sum'}).reset_index()\n",
"df_group = df_group[df_group['Word'].str.isalpha()]\n",
"df_group"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Ploting most frequent words"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x1800 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"df_group.sort_values('Count').tail(50).plot(kind='barh',x='Word',y='Count',figsize=(15,25), title=\"50 most frequent words\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Ploting most frequent words by POS"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th></th>\n",
" <th>Count</th>\n",
" </tr>\n",
" <tr>\n",
" <th>POS</th>\n",
" <th>Word</th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <td rowspan=\"5\" valign=\"top\">ADJ</td>\n",
" <td>amplia</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>asequibles</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>atrevidos</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>baratos</td>\n",
" <td>2</td>\n",
" </tr>\n",
" <tr>\n",
" <td>cortos</td>\n",
" <td>7</td>\n",
" </tr>\n",
" <tr>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <td rowspan=\"5\" valign=\"top\">VERB</td>\n",
" <td>made</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>mereces</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>navegar</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>ofrecemos</td>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <td>ofrecen</td>\n",
" <td>1</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>152 rows × 1 columns</p>\n",
"</div>"
],
"text/plain": [
" Count\n",
"POS Word \n",
"ADJ amplia 1\n",
" asequibles 1\n",
" atrevidos 1\n",
" baratos 2\n",
" cortos 7\n",
"... ...\n",
"VERB made 1\n",
" mereces 1\n",
" navegar 1\n",
" ofrecemos 1\n",
" ofrecen 1\n",
"\n",
"[152 rows x 1 columns]"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_POS = df_group.set_index(['POS', 'Word'])\n",
"df_POS.sort_index(inplace=True)\n",
"df_POS"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA5kAAAJOCAYAAAA9GsMFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOzde7SddX3v+8/XRIMIBEWkEa1xa6wtd4meqoBgrfYYL1VRdGuJl1NKL1p2j1Z2a5XaocbLrhzptootR6l2S/WotaWtWgREFCVcI2jtscZaxBa8xCDKUfyeP9aMXYaVhMAvmWslr9cYa2TO5/p75j+MN79nPrO6OwAAADDCXaY9AAAAAHYdIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEgO1QVTfN+vtRVX1v1vvnTrb5uar6cFVtqKqNVXV+VT1q1jGWV1XP2m99VZ26lXOeWVX/NDnf8+dY/9+q6uuT851VVUu2cQ33mJz37+ZYt35yTRur6ttV9amqOrmq7jJrm9Oq6t238yMDYDcjMgFgO3T3Xpv+kvxrkifPWvaeqnpQkouTrEvywCT3TfLBJB+tqkdudrh9J8d5TpJXVtUvbeG0VyX5jSSXb76iqp6Q5NQkv5BkeZL/kuQPt3EZxye5Jcnjq2rZHOuf3N17J3lAkjVJXp7kz7dxTABIIjIBYLTTkny6u3+/u7/Z3Ru7+y1J/iLJ6+faobs/neSaJAdvYf3/7O7zknx/jtWrk/x5d1/T3d9K8kdJnr+NMa5O8rYkVyd57pY26u4N3f3hJCckWV1Vc44PAGYTmQAw1i8med8cy/8qyaOras/ZC2vGo5MclOSKO3C+gzIz07nJVUkOqKr95tq4qn46ybFJ3jP5O3FbJ+juzyb5tyRH34HxAbCbWTztAQDALubeSa6fY/n1mfmfu/ectezGJJ3k60lOncxWbq+9kmyY9X7T672TfGOO7U9McnV3X1tV307yhqo6oru3FbhfS3KvOzA+AHYzIhMAxroxyVzfc1yW5EdJvpXkPpNl9+7uH97J892UZJ9Z7ze93riF7U9M8o4k6e6vVdWFmbl9dluReWCSb96JcQKwm3C7LACM9Y9JnjnH8mdl5ruaNw8+3zVJDpv1/rAk/97dt5nFnDzhdkWS/z55Gu3Xk/xvSZ5TVVv8H89V9fDMROYnh44cgF2SyASAsf4wyaOq6jVVda+q2ruqXpyZGcSX35EDVtXdqmqPJJXkrlW1x6yfFDk7yYsmP5tyzySvSPLOLRxqdZKPJfm5JIdP/g5OsmeS/32O8+5TVU9K8t4k7+7udXdk/ADsXkQmAAzU3f+c5KjMzCiuz8x3MZ+R5AndffEdPOxHk3wvyaOSnDl5fczkfP+Q5A1Jzk/ylcnfqzY/wCRSn5XkjO7++qy/L2fmyberZ23+N1W1MclXk/x+kj9O8oLNL/UOXgsAu7jq9t8IAOD2q6o/TnKX7j5l2mMBYP4xkwkA3G5VtW+SJyRZO+2xADA/iUwA4HaZfD/zS0k+k5nf/QSA23C7LAAAAMOYyQQAAGCYLf4mFlt273vfu5cvXz7tYQAAAEzFZZdddmN37z/XOpF5Byxfvjxr13reAQAAsHuqqq9saZ3bZQEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMIunPYCFaN11G7L81HOnPQwAAGAXtn7NqmkP4Q4xkwkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMAs2MqvqlKrac9b7v6uqfefY7rSqeunOHR0AAMDuacFGZpJTkvw4Mrv7id397SmOBwAAYLe3eBonrarXJ/lKd7918v60JBszE73PSrIkyQe7+1VVdY8kf5XkfkkWJfmjJAckuW+S86vqxu4+rqrWJ1nZ3TdW1e8nOTHJV5PckOSyyXkOT/K2zMTpl5K8sLu/VVUvSXJykh8muba7n70TPgYAAIBdzrRmMt+b5IRZ75+VmRhckeQRSQ5PcmRVHZPkl5J8rbsP6+6Dk/xDd78lydeSHNfdx80+cFUdmeTZSY5I8vQkD5+1+uwkL+/uQ5OsS/KqyfJTkxwxWX7yXAOuqpOqam1Vrb315g134tIBAAB2XVOJzO6+Isl9quq+VXVYkm8lOTTJ45NckeTyJA/NTHSuS/K4qnp9VR3d3dsqvKMzMwt6c3d/J8mHk6SqlibZt7svnGz3riTHTF5fneQ9VfW8zMxmzjXmM7t7ZXevXLTn0jt45QAAALu2qdwuO/H+JMcn+anMzGwuT/K67n775htOZiefmOR1VfXR7n71No7d2zmWVZkJzqck+YOqOqi754xNAAAAtmyaD/55b2Zuaz0+M8H5kSQvrKq9kqSqDqyq+1TVfZPc3N3vTvKmJA+b7L8xyd5zHPcTSZ5WVXevqr2TPDlJJjOg36qqoyfb/UqSC6vqLknu393nJ/ndJPsm2Wv85QIAAOz6pjaT2d3XTCLwuu6+Psn1VfWzST5dVUlyU5LnJXlwkjdW1Y+S/CDJr08OcWaSv6+q62d/L7O7L6+qc5JcmeQrSS6addrVSd42+emTf0nygsw8TOjdk9tpK8mbPaUWAADgjqnu7b2zlCXLVvSy1adPexgAAMAubP2aVdMewhZV1WXdvXKudQv5dzIBAACYZ0QmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDCLpz2AheiQA5dm7Tz+YVQAAIBpMZMJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYRZPewAL0brrNmT5qedOexjsAOvXrJr2EAAAYEEzkwkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYRZEZFbVTdMeAwAAANu2ICJze9SMXe66AAAAFoIFFWNVtVdVnVdVl1fVuqp66mT58qr6fFW9NcnlSe5fVS+qqi9W1QVV9Y6q+pPJtg+YHOPqyb8/PVn+zKr6XFVdVVWfmN5VAgAALFwLKjKTfD/J07r7YUmOS/I/qqom634mydndfUSSHyT5gyQ/n+QXkzx01jH+ZLLdoUnek+Qtk+WvTPKE7j4syVM2P3FVnVRVa6tq7a03b9gBlwYAALDwLbTIrCSvraqrk/xjkgOTHDBZ95XuvmTy+hFJLuzub3b3D5K8b9YxHpnkLyev/yLJUZPXFyd5Z1X9apJFm5+4u8/s7pXdvXLRnkuHXhQAAMCuYqFF5nOT7J/kyO4+PMm/J9ljsu67s7arzXfcik6S7j45ySuS3D/JlVW1350fLgAAwO5loUXm0iT/0d0/qKrjkjxgC9t9NsljquqeVbU4yTNmrftUkmdPXj83ySeTpKoe1N2f6e5XJrkxM7EJAADAdlg87QFsp/ck+ZuqWpvkyiRfmGuj7r6uql6b5DNJvpbk2iSbvkj5kiRnVdXLktyQ5AWT5W+sqhWZmQU9L8lVO+wqAAAAdlELIjK7e6/Jvzdm5juVczl4s/d/2d1nTmYyP5jko5NjrE/y2DnO8fRhAwYAANhNLbTbZbfHaVV1ZZLPJflykg9NeTwAAAC7vAUxk3lHdPdLpz0GAACA3c2uPJMJAADATiYyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDC77NNld6RDDlyatWtWTXsYAAAA846ZTAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhm8bQHsBCtu25Dlp967rSHwRasX7Nq2kMAAIDdlplMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIbZpSOzqo6tqkdNexwAAAC7i102MqtqcZJjk4hMAACAnWTxtAdwe1TViUlemqSTXJ3kFUnOSrJ/khuSvKC7/7Wq3pnkm0mOmPz76CS3VtXzkrw4yb9uYb9nJnlVkluTbOjuY3bi5QEAAOwy5n1kVtVBSX4/yaO7+8aquleSdyU5u7vfVVUvTPKWJL882eUhSR7X3bdW1WlJburuN02O9Tdb2O+VSZ7Q3ddV1b5bGMdJSU5KkkX77L+jLhcAAGBBWwi3yz42yfu7+8Yk6e5vJnlkkr+crP+LJEfN2v593X3rFo61pf0uTvLOqvrVJIvm2rG7z+zuld29ctGeS+/wxQAAAOzKFkJkVmZuk92a2eu/ux3H7iTp7pMzcwvu/ZNcWVX7bdcIAQAASLIwIvO8JM/aFH6T22U/leTZk/XPTfLJLey7Mcnes97PuV9VPai7P9Pdr0xyY2ZiEwAAgO0077+T2d3XVNVrklxYVbcmuSLJS5KcVVUvy+QBPlvY/W+SvL+qnpqZB/9sab83VtWKzMyanpfkqh12QQAAALuweR+ZSdLd78rMw35me+wc2z1/s/dfTHLo7djv6XdyiAAAAGRh3C4LAADAAiEyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIZZPO0BLESHHLg0a9esmvYwAAAA5h0zmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhm8bQHsBCtu25Dlp967rSHsVOsX7Nq2kMAAAAWEDOZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAyzS0VmVZ1SVXvOev93VbXvHNudVlUv3bmjAwAA2PXtUpGZ5JQkP47M7n5id397iuMBAADYrSye9gA2qarXJ/lKd7918v60JBszE8LPSrIkyQe7+1VVdY8kf5XkfkkWJfmjJAckuW+S86vqxu4+rqrWJ1nZ3TdW1e8nOTHJV5PckOSyyXkOT/K2zMTpl5K8sLu/tXOuGgAAYNcyn2Yy35vkhFnvn5WZGFyR5BFJDk9yZFUdk+SXknytuw/r7oOT/EN3vyXJ15Ic193HzT5wVR2Z5NlJjkjy9CQPn7X67CQv7+5Dk6xL8qq5BldVJ1XV2qpae+vNG+781QIAAOyC5k1kdvcVSe5TVfetqsOSfCvJoUken+SKJJcneWhmonNdksdV1eur6uju3lb1HZ2ZWdCbu/s7ST6cJFW1NMm+3X3hZLt3JTlmC+M7s7tXdvfKRXsuvXMXCwAAsIuaN7fLTrw/yfFJfiozM5vLk7yuu9+++YaT2cknJnldVX20u1+9jWP34LECAACwmXkzkznx3szc1np8ZoLzI0leWFV7JUlVHVhV96mq+ya5ubvfneRNSR422X9jkr3nOO4nkjytqu5eVXsneXKSTGZAv1VVR0+2+5UkF86xPwAAALfDvJrJ7O5rJhF4XXdfn+T6qvrZJJ+uqiS5Kcnzkjw4yRur6kdJfpDk1yeHODPJ31fV9bO/l9ndl1fVOUmuTPKVJBfNOu3qJG+b/PTJvyR5wQ69SAAAgF1YdbuLdHstWbail60+fdrD2CnWr1k17SEAAADzTFVd1t0r51o3326XBQAAYAETmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDLJ72ABaiQw5cmrVrVk17GAAAAPOOmUwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMs3jaA1iI1l23IctPPXfaw9gu69esmvYQAACA3YCZTAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMs2Ajs6r2r6rPVNUVVXX0du57eFU9cUeNDQAAYHe1ICOzqhYn+YUkX+juI7r7ou08xOFJRCYAAMBgU4vMqlpeVV+oqndV1dVV9f6q2rOqjqyqC6vqsqr6SFUtm2x/QVW9tqouTPLbSd6Q5IlVdWVV3b2qHl9Vn66qy6vqfVW112S/h1fVp6rqqqr6bFUtTfLqJCdM9j2hqu5VVR+ajOOSqjp0Wp8LAADAQrZ4yuf/mSQv6u6Lq+qsJL+Z5GlJntrdN1TVCUlek+SFk+337e7HJElVfSPJyu7+raq6d5JXJHlcd3+3ql6e5Heqak2Sc5Kc0N2XVtU+SW5O8spN+06OdUaSK7r7l6vqsUnOzsxs549V1UlJTkqSRfvsv+M+EQAAgAVs2pH51e6+ePL63Ul+L8nBST5WVUmyKMn1s7Y/ZwvH+fkkP5fk4sl+d0vy6cxE7PXdfWmSdPd3kmSyzWxHJXnGZJuPV9V+VbW0uzds2qC7z0xyZpIsWbai78jFAgAA7OqmHZmbx9rGJNd09yO3sP13t7C8knysu5/zEwtnbnu9PUF4m+q8nfsBAAAwy7Qf/PPTVbUpKJ+T5JIk+29aVlV3raqDbsdxLkny6Kp68GS/PavqIUm+kOS+VfXwyfK9Jw8N2phk71n7fyLJcyfbHJvkxk2zngAAANx+047MzydZXVVXJ7lXkjOSHJ/k9VV1VZIrkzxqWwfp7huSPD/J/5oc65IkD+3u/y/JCUnOmBzvY0n2SHJ+kp/b9OCfJKclWTnZd02S1UOvEgAAYDdR3dO5K7Sqlif52+4+eCoDuBOWLFvRy1afPu1hbJf1a1ZNewgAAMAuoqou6+6Vc62b9kwmAAAAu5CpPfinu9dn5kmyAAAA7CLMZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDTO3psgvZIQcuzVq/OwkAAHAbZjIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADLN4Syuq6owkvYXVtyT5UpL3dPfGHTEwAAAAFp4tRmaStdvY76AkH0jyi0NHBAAAwIK1xcjs7ndta+eq+ruxwwEAAGAh2+p3MqtqdVVdXlXfnfytraoTN63v7ifu+CECAACwUGztO5knJjklye8kuTxJJXlYkjdWVbr77J0zRAAAABaKrc1k/kaSp3X3+d29obu/3d0fT/KMyToAAAD4CVuLzH26e/3mCyfL9tlRAwIAAGDh2lpkfu8OrgMAAGA3tbWfMPnZqrp6juWV5L/soPEAAACwgG01MudYVknul+T3dsxwAAAAWMi29juZX9n0uqoOT/JfkzwryZeT/D87fmgAAAAsNFv7CZOHJHl2kuck+UaSc5JUdx+3k8YGAADAArO122W/kOSiJE/u7v83Sarqv+2UUQEAALAgbe3pss9I8vUk51fVO6rqFzLznUwAAACY0xYjs7s/2N0nJHlokguS/LckB1TVn1bV43fS+AAAAFhAtjaTmSTp7u9293u6+0mZebLslUlO3eEjAwAAYMHZZmTO1t3f7O63d/djd9SAAAAAWLi2KzIBAABga0QmAAAAw2ztJ0zYgnXXbcjyU8+d9jC2y/o1q6Y9BAAAYDdgJhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYRZsZFbVyqp6y7THAQAAwH9aPO0B3FHdvTbJ2tu7fVUt7u4f7sAhAQAA7PamOpNZVcur6gtV9WdV9bmqek9VPa6qLq6qf66qR1TVParqrKq6tKquqKqnTvY9tqr+dvL6XlX1oaq6uqouqapDJ8tPq6ozq+qjSc6uqkVV9cbJsa6uql+bbLesqj5RVVdOxnH01D4UAACABWw+zGQ+OMkzk5yU5NIk/zXJUUmekuT3klyb5OPd/cKq2jfJZ6vqHzc7xh8muaK7f7mqHpvk7CSHT9YdmeSo7v5eVZ2UZEN3P7yqliS5eBKgT0/yke5+TVUtSrLn5oOc7HtSkizaZ/+R1w8AALDLmA+R+eXuXpckVXVNkvO6u6tqXZLlSe6X5ClV9dLJ9nsk+enNjnFUkmckSXd/vKr2q6qlk3Uf7u7vTV4/PsmhVXX85P3SJCsyE7dnVdVdk3you6/cfJDdfWaSM5NkybIVfWcvGgAAYFc0HyLzllmvfzTr/Y8yM75bkzyju/9p9k5VdcDst3Mcd1MIfnez7V7c3R/ZfOOqOibJqiR/UVVv7O6zt+sqAAAAWBBPl/1IkhdXVSVJVR0xxzafSPLcyfpjk9zY3d/ZwrF+fTJjmap6yOQ7nw9I8h/d/Y4kf57kYeMvAwAAYNc3H2Yyt+WPkpye5OpJaK5P8qTJuk2zlacl+b+r6uokNydZvYVj/VlmbsG9fHKsG5L8cpJjk7ysqn6Q5KYkJ46+CAAAgN1BdS/MrxdW1TOSPKW7txSUO8ySZSt62erTd/Zp75T1a1ZNewgAAMAuoqou6+6Vc61bCDOZt1FVT0nymiQvnPZYAAAA+E8LMjK7+8NJPjztcQAAAPCTFsKDfwAAAFggRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwzIL8CZNpO+TApVm7ZtW0hwEAADDvmMkEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMIunPYCFaN11G7L81HN3+nnXr1m1088JAACwPcxkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGCYnR6ZVbW+qu69k861b1X9xs44FwAAALv+TOa+SUQmAADATrJDI7OqnldVn62qK6vq7VW16Pasr6oXVdUXq+qCqnpHVf3JZPmTq+ozVXVFVf1jVR0wWX5aVZ012f5fquolk1OsSfKgyfHfONn2ZVV1aVVdXVV/OFl2j6o6t6quqqrPVdUJO/JzAQAA2FUt3lEHrqqfTXJCkkd39w+q6q1Jnrut9VX1j0n+IMnDkmxM8vEkV012+2SSn+/urqr/I8nvJvk/J+semuS4JHsn+aeq+tMkpyY5uLsPn5zz8UlWJHlEkkry4ao6Jsn+Sb7W3asm2y2d43pOSnJSkizaZ/8RHxEAAMAuZ4dFZpJfSHJkkkurKknunuQ/bsf6RyS5sLu/mSRV9b4kD5nsc78k51TVsiR3S/LlWcc7t7tvSXJLVf1HkgPmGNPjJ39XTN7vlZnovCjJm6rq9Un+trsv2nzH7j4zyZlJsmTZir79HwMAAMDuY0dGZiV5V3f/959YWPX8bax/2laOeUaSP+7uD1fVsUlOm7Xullmvb83c11ZJXtfdb7/Niqojkzwxyeuq6qPd/eqtjAMAAIA57MjvZJ6X5Piquk+SVNW9quoBt2P9Z5M8pqruWVWLkzxj1j5Lk1w3eb36doxhY2Zun93kI0leWFV7Tc55YFXdp6rum+Tm7n53kjdl5lZdAAAAttMOm8ns7mur6hVJPlpVd0nygyS/ua313X1JVb02yWeSfC3JtUk2THY7Lcn7quq6JJckeeA2xvCNqrq4qj6X5O+7+2WT74J+enKL7k1JnpfkwUneWFU/mozj18d8CgAAALuX6p5/Xy+sqr26+6bJTOYHk5zV3R+c9rg2WbJsRS9bffpOP+/6Nat2+jkBAAA2V1WXdffKudbN19/JPK2qrkzyucw83OdDUx4PAAAAt8OOfPDPHdbdL532GAAAANh+83UmEwAAgAVIZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBh5uXTZee7Qw5cmrV+sxIAAOA2zGQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgmMXTHsBCtO66DVl+6rk7/bzr16za6ecEAADYHmYyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADDMvIvMqlpeVZ+b9jEAAADYfvMuMgEAAFi45mtkLqqqd1TVNVX10aq6e1U9qKr+oaouq6qLquqhSVJVB1TVB6vqqsnfo7Z0jMn2h1fVJVV19WS/e06Wv6Sqrp0sf++UrhsAAGBBm6+RuSLJ/+zug5J8O8kzkpyZ5MXdfWSSlyZ562TbtyS5sLsPS/KwJNds5RhJcnaSl3f3oUnWJXnVZPmpSY6YLD958wFV1UlVtbaq1t568wSM39gAABBESURBVIaxVwsAALCLWDztAWzBl7v7ysnry5IsT/KoJO+rqk3bLJn8+9gkJyZJd9+aZMNkdvI2x6iqpUn27e4LJ8vfleR9k9dXJ3lPVX0oyYc2H1B3n5mZ0M2SZSt6xEUCAADsauZrZN4y6/WtSQ5I8u3uPvxOHOPu29h+VZJjkjwlyR9U1UHd/cPtOB8AAMBub77eLru57yT5clU9M0lqxmGTdecl+fXJ8kVVtc+WDtLdG5J8q6qOniz6lSQXVtVdkty/u89P8rtJ9k2y1465FAAAgF3XQonMJHlukhdV1VWZ+d7lUyfLfzvJcVW1LjO3xR60jeOsTvLGqro6yeFJXp1kUZJ3T45xRZI3d/e3d8A1AAAA7NKq29cLt9eSZSt62erTd/p5169ZtdPPCQAAsLmquqy7V861biHNZAIAADDPiUwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMIunPYCF6JADl2at36wEAAC4DTOZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMMziaQ9gIVp33YYsP/XcnX7e9WtW7fRzAgAAbA8zmQAAAAwjMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMM28is6qWV9XndsBxj62qR40+LgAAALc1byLzzqiqxVtZfWwSkQkAALATbC3OpmFxVb0ryRFJvpjkxCQvTfLkJHdP8qkkv9bdXVUXTN4/OsmHq+qLSV6R5G5JvpHkuZN9Tk5ya1U9L8mLk/xrkrOS7J/khiQv6O5/rapnJnlVkluTbOjuY3bOJQMAAOw65ttM5s8kObO7D03ynSS/keRPuvvh3X1wZqLxSbO237e7H9Pd/yPJJ5P8fHcfkeS9SX63u9cneVuSN3f34d19UZI/SXL25BzvSfKWybFemeQJ3X1YkqdsPrCqOqmq1lbV2ltv3rADLh0AAGDhm2+R+dXuvnjy+t1JjkpyXFV9pqrWJXlskoNmbX/OrNf3S/KRyXYv22y72R6Z5C8nr/9ico4kuTjJO6vqV5Ms2nyn7j6zu1d298pFey69A5cGAACw65tvkdlzvH9rkuO7+5Ak70iyx6z13531+ozMzHoekuTXNttum+fs7pMzc7vt/ZNcWVX7bf/wAQAAdm/zLTJ/uqoeOXn9nMzcApskN1bVXkmO38q+S5NcN3m9etbyjUn2nvX+U0mePXn93E3nqKoHdfdnuvuVSW7MTGwCAACwHebbg38+n2R1Vb09yT8n+dMk90yyLsn6JJduZd/Tkryvqq5LckmSB06W/02S91fVUzPz4J+XJDmrql6WyYN/Jtu9sapWJKkk5yW5atxlAQAA7B6qe/M7VNmWJctW9LLVp+/0865fs2qnnxMAAGBzVXVZd6+ca918u10WAACABUxkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAyzeNoDWIgOOXBp1q5ZNe1hAAAAzDtmMgEAABhGZAIAADCMyAQAAGAYkQkAAMAwIhMAAIBhRCYAAADDiEwAAACGEZkAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGEZkAgAAMIzIBAAAYBiRCQAAwDAiEwAAgGFEJgAAAMOITAAAAIYRmQAAAAwjMgEAABhGZAIAADDM4mkPYCFad92GLD/13J1+3vVrVu30cwIAAGwPM5kAAAAMIzIBAAAYRmQCAAAwjMgEAABgGJEJAADAMCITAACAYUQmAAAAw4hMAAAAhhGZAAAADCMyAQAAGGbBRmZV7V9Vn6mqK6rq6O3c9/CqeuKOGhsAAMDuakFGZlUtTvILSb7Q3Ud090XbeYjDk4hMAACAwRZP68RVtTzJPyT5TJIjknwxyYlJfjbJHyfZK8mNSZ7f3ddX1QVJPpXk0Uk+nOS3k9y9qq5M8sgkRyf5wyRLknwpyQu6+6aqeniS/yvJPZLckuQXk7x6su9RSV6X5MtJTk9y9yTfm+z7Tzv2EwAAANj1TC0yJ34myYu6++KqOivJbyZ5WpKndvcNVXVCktckeeFk+327+zFJUlXfSLKyu3+rqu6d5BVJHtfd362qlyf5napak+ScJCd096VVtU+Sm5O8ctO+k2Ptk+SY7v5hVT0uyWuTPGP2QKvqpCQnJcmiffbfcZ8IAADAAjbtyPxqd188ef3uJL+X5OAkH6uqJFmU5PpZ25+zheP8fJKfS3LxZL+7Jfl0ZiL2+u6+NEm6+ztJMtlmtqVJ3lVVK5J0krtuvkF3n5nkzCRZsmxFb89FAgAA7C6mHZmbx9rGJNd09yO3sP13t7C8knysu5/zEwurDp3jHHP5oyTnd/fTJrfxXnA79gEAAGAz037wz09X1aagfE6SS5Lsv2lZVd21qg66Hce5JMmjq+rBk/32rKqHJPlCkvtOvpeZqtp78tCgjUn2nrX/0iTXTV4//05eEwAAwG5r2jOZn0+yuqrenuSfk5yR5CNJ3lJVSzMzvtOTXLO1g0y+v/n8JP+rqpZMFr+iu784+V7nGVW16aE+j0tyfpJTJw8Nel2SN2TmdtnfSfLx0RcJAAAsTD/4wQ/yb//2b/n+978/7aFMxR577JH73e9+uetdb/ONwi2q7ul8vXByW+rfdvfBUxnAnbBk2Ypetvr0nX7e9WtW7fRzAgDA7uzLX/5y9t577+y3335zPdtll9bd+cY3vpGNGzfmgQ984E+sq6rLunvlXPtN+3ZZAACAeev73//+bhmYycwDU/fbb7/tnsWd2u2y3b0+M0+SBQAAmLd2x8Dc5I5cu5lMAAAAhpn2g38AAAAWjOWnnjv0eLfnuStf//rXc8opp+TSSy/NkiVLsnz58px++ul5yEMeMmQMF1xwQe52t7vlUY961JDjmckEAACYp7o7T3va03LsscfmS1/6Uq699tq89rWvzb//+78PO8cFF1yQT33qU8OOJzIBAADmqfPPPz93vetdc/LJJ/942eGHH56jjjoqL3vZy3LwwQfnkEMOyTnnnJNkJhif9KQn/Xjb3/qt38o73/nOJMny5cvzqle9Kg972MNyyCGH5Atf+ELWr1+ft73tbXnzm9+cww8/PBdddNGdHrPbZQEAAOapz33ucznyyCNvs/wDH/hArrzyylx11VW58cYb8/CHPzzHHHPMNo9373vfO5dffnne+ta35k1velP+7M/+LCeffHL22muvvPSlLx0yZpF5Bxxy4NKs9ZuVAADAlHzyk5/Mc57znCxatCgHHHBAHvOYx+TSSy/NPvvss9X9nv70pydJjjzyyHzgAx/YIWNzuywAAMA8ddBBB+Wyyy67zfLunnP7xYsX50c/+tGP32/+G5dLlixJkixatCg//OEPB470P4lMAACAeeqxj31sbrnllrzjHe/48bJLL70097znPXPOOefk1ltvzQ033JBPfOITecQjHpEHPOABufbaa3PLLbdkw4YNOe+887Z5jr333jsbN24cNma3ywIAANxOt+cnR0aqqnzwgx/MKaeckjVr1mSPPfb48U+Y3HTTTTnssMNSVXnDG96Qn/qpn0qSPOtZz8qhhx6aFStW5IgjjtjmOZ785Cfn+OOPz1//9V/njDPOyNFHH33nxrylaVa2bOXKlb127dppDwMAANjBPv/5z///7d07iFxlGMbx/0NcGVEhXqJIxksEi7XQCCKBWMQgEi8YCwVFwUKwsVBQRG1EIYWN2tiIihZeUaPByuAFrbzECyoxeEF0ScjG1aA2EfW1mBNcw8ZkJgfPzOb/g2Xm++YMvLs87DfvuQ3T09Ndl9Gphf4GSbZU1fkLbe/pspIkSZKk1thkSpIkSZJaY5MpSZIkSf/hcL7EcJTf3SZTkiRJkvaj1+sxNzd3WDaaVcXc3By9Xm+o93l3WUmSJEnaj36/z8zMDLt27eq6lE70ej36/f5Q77HJlCRJkqT9mJqaYsWKFV2XMVE8XVaSJEmS1BqbTEmSJElSa2wyJUmSJEmtyeF4l6RDleRXYFvXdWjROBH4sesitGiYJ7XFLKlN5kltMk/j4fSqWrbQC974ZzTbqur8rovQ4pDkQ/OktpgntcUsqU3mSW0yT+PP02UlSZIkSa2xyZQkSZIktcYmczSPdl2AFhXzpDaZJ7XFLKlN5kltMk9jzhv/SJIkSZJa45FMSZIkSVJrbDIlSZIkSa2xyRxSknVJtiX5OsldXdejyZLkiSSzST6fN3d8ks1Jvmoej+uyRk2GJKcmeSvJ1iRfJLm1mTdPGlqSXpL3k3za5Om+Zn5FkveaPD2f5Miua9VkSLIkycdJXmvGZkkjSfJdks+SfJLkw2bOtW7M2WQOIckS4BHgUuBs4LokZ3dblSbMk8C6febuAt6oqrOAN5qxdCB/ALdX1TSwCril+X9knjSKPcDaqjoXWAmsS7IKeAB4qMnTz8BNHdaoyXIrsHXe2CzpUFxUVSvnfTema92Ys8kczgXA11X1bVX9DjwHrO+4Jk2QqnoH+Gmf6fXAU83zp4Cr/teiNJGqakdVfdQ8/5XBh7nlmCeNoAZ+a4ZTzU8Ba4EXm3nzpIOSpA9cDjzWjINZUrtc68acTeZwlgM/zBvPNHPSoTi5qnbAoHEATuq4Hk2YJGcA5wHvYZ40oub0xk+AWWAz8A2wu6r+aDZxzdPBehi4E/irGZ+AWdLoCng9yZYkNzdzrnVj7oiuC5gwWWDO74CR1JkkxwAvAbdV1S+DAwbS8KrqT2BlkqXARmB6oc3+36o0aZJcAcxW1ZYka/ZOL7CpWdLBWl1V25OcBGxO8mXXBenAPJI5nBng1HnjPrC9o1q0eOxMcgpA8zjbcT2aEEmmGDSYT1fVy820edIhqardwNsMrvVdmmTvDmnXPB2M1cCVSb5jcFnRWgZHNs2SRlJV25vHWQY7wC7AtW7s2WQO5wPgrOYOaUcC1wKbOq5Jk28TcGPz/Ebg1Q5r0YRornF6HNhaVQ/Oe8k8aWhJljVHMElyFHAxg+t83wKubjYzTzqgqrq7qvpVdQaDz0lvVtX1mCWNIMnRSY7d+xy4BPgc17qxlyrPVhhGkssY7JFbAjxRVRs6LkkTJMmzwBrgRGAncC/wCvACcBrwPXBNVe17cyDpX5JcCLwLfMY/1z3dw+C6TPOkoSQ5h8HNM5Yw2AH9QlXdn+RMBkejjgc+Bm6oqj3dVapJ0pwue0dVXWGWNIomNxub4RHAM1W1IckJuNaNNZtMSZIkSVJrPF1WkiRJktQam0xJkiRJUmtsMiVJkiRJrbHJlCRJkiS1xiZTkiRJktQam0xJkiRJUmtsMiVJkiRJrfkbNDrF4Yu8ynYAAAAASUVORK5CYII=\n",
"text/plain": [
"<Figure size 1080x720 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x720 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x720 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAA5wAAAJOCAYAAADbMwhBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOzde9hddX3n/c9XgokIBBVqI7ZGbRQPnAQ8n2vVDrVKPdWqI9UpQ1vtY72speqonT61qeOMWpVaZKw6j23xgCOVVlELSlGEcAwg1qcltjKe0IoiSAW+88e9ojfpnZBAftl37rxe15XrXnvttfb6rv0Pvl1r713dHQAAANjebjfrAQAAAFiaBCcAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAFuhqq6Z9++mqrpu3uPnTtvcv6pOqaqrq+p7VXV6VT183musrqqet9+GqjpuC8c8oaq+OB3v6AWe/+2q+tp0vHdV1fIFtllRVd+pqscv8NybquqD0/KGTc7pmqp62/Tc0VV147Tuu1V1UVX9whbO6+tVdXxV7b6NbzMAS4zgBICt0N17bvyX5J+TPGXeuvdV1b2TnJVkfZJ7Jrlbkg8nOa2qHrbJy+0zvc5zkrymqp68mcNelOQ3kpy/6RNV9aQkxyX52SSrk9wrye8vMPcPkpyU5D9usv9u0/HfM2/1/HPas7tfPO+5z00z75Pk+CR/VVX7bOa8DkzysCS/uZnzAmAXITgBYPt4Xeai7FXd/e3u/l53/0mS/5Xkjxfaobs/l+TSJA/czPNv7+5PJfnBAk+/IMn/7O5Lu/tfk/xBkqM3M9t7kjy9qvaYt+5JmfvfAX97i2d285luytw53THJms1s840kn0hy/215bQCWHsEJANvHzyX5wALr35/kEZvEXmrOI5I8IMkFt+J4D8jcFdCNLkpy16q6y6Ybdvdnk3w1yS/NW/38JH/R3Tdsy0GnK6O/muSHSb68mW3ulrmgPXtbXhuApWfZrAcAgCVi38xF3aa+mrn/g/dO89ZdlaSTfC3JcdNVzG21Z5Kr5z3euLxXkm8tsP17M3db7f9XVXsneWqSR2yyzf+uqvkB+jvd/c5p+aFV9Z3MXdm8IcnzpiuZ811VVUmyMsnnknxw204JgKXGFU4A2D6uSrJqgfWrktyU5F/nrdu3u+/U3febbru9Na5Jsve8xxuXv7eZ7d+b5HFVtX+SZyT5/7t70yurT+vufeb9e+e8587u7n0yF86nJHnUAsfYd9pmj8x9nvVj23ZKACw1ghMAto9PJnnmAuuflbnPdl67nY93aZKD5z0+OMnXu3uhq5vp7n9OcmaS52budtr33pqDdvc1mfsio+dX1aGb2ea6JO9O8rCq2vfWHAeApUFwAsD28ftJHl5Vf1hVd66qvarqJZm7jfV3b80LVtXtq2pFkkqy+/QTJxv/2/3eJC+aforlTklenbnI25L3JHlx5m6lfd+tmSlJpqg9MclrNjP38sxF7dey8O29AOwiBCcAbAfd/aUkj8zclcYNmfvs5tOTPKm7z7qVL3takuuSPDzJCdPyo6fjfSzJG5Kcnrkv7/lyktfewut9MHO3xH6quxf6vOlfb/I7nB/ewmu9Ocl/qKqD5q37TlVdk+TrmftZlF/s7r6lkwRg6Sr/HQAAAGAEVzgBAAAYQnACAAAwhOAEAABgCMEJAADAEMtmPcBSsO+++/bq1atnPQYAAMBMnHfeeVd1936brhec28Hq1auzbt26WY8BAAAwE1X15YXWu6UWAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAyxbNYDLAXrr7w6q487ddZjAAAAS9SGtUfOeoRbxRVOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIaYeXBW1euq6uU7+zEAAAC4uZkH52JRVbvNegYAAIClZCbBWVWvqqovVtUnk9x3WnfvqvpYVZ1XVWdW1QHT+mdW1SVVdVFVfWZat1tVvbGq1lfVxVX1kmn9hqrad1o+vKrOmHfYg6vq76rqS1X1a9M2j62q06vqL5Ksn9Y9r6rOqaoLq+rPhCgAAMCts2xHH7CqDkvyy0kOnY5/fpLzkpyQ5Nju/lJVPSTJ8Uken+Q1SZ7U3VdW1T7TyxyT5J5JDu3uG6rqzltx6IOSPDTJHZNcUFWnTusfnOSB3X1FVd0vybOTPKK7f1hVxyd5bpL3LnAex0xzZLe999vm9wEAAGCp2+HBmeRRST7c3dcmSVWdkmRFkocn+UBVbdxu+fT3rCTvrqr3Jzl5WveEJO/o7huSpLu/vRXH/Uh3X5fkuqo6PXOh+Z0k53T3FdM2P5vksCTnTnPcIck3Fnqx7j4hc5Gc5avW9FYcHwAAYJcyi+BMkk0D7XZJvtPdh/y7DbuPna54Hpnkwqo6JEkt8BpJckN+fJvwils45sbH35+3rpK8p7t/75ZPAQAAgC2ZxWc4P5PkqKq6Q1XtleQpSa5NckVVPTNJas7B0/K9u/vz3f2aJFcl+akkpyU5tqqWTdtsvKV2Q+auUCbJ0zc57lOrakVV3SXJY5Ocu8Bsn0ryjKr6iY2vW1X32B4nDQAAsKvZ4cHZ3ecnOSnJhUk+lOTM6annJnlRVV2U5NIkT53W/7fpy4EuyVysXpTkxCT/nOTiaftfmbb9/SRvqaozk9y4yaHPSXJqkrOT/EF3/58FZrssyauTnFZVFyf5RJJVt/2sAQAAdj3V7eOHt9XyVWt61QvePOsxAACAJWrD2iNnPcIWVdV53X34puv9DicAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIZbNeoCl4MD9V2bdIv8hVgAAgB3NFU4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGEJwAAAAMsWzWAywF66+8OquPO3XWY8Cis2HtkbMeAQCAGXKFEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABD7DTBWVW/VVVfqKr3zXoWAAAAbtmyWQ+wDX4jyc939xXzV1bVsu6+YUYzAQAAsBmLMjir6mVJXjg9PDHJAUnuleSUqnpXkpVJ7pZkdZKrqur5SdYmeWyS5Une3t1/Nr3WK5I8P8lNSf62u4+rqnsneXuS/ZJcm+TXuvvyqnp3ku8mOTzJTyZ5RXd/cPgJAwAALEGLLjir6rAkv5rkIUkqyeeTPC/Jk5M8rruvqqrXJTksySO7+7qqOibJ1d19RFUtT3JWVZ2WuVB9WpKHdPe1VXXn6TAnJDm2u79UVQ9JcnySx0/PrUryyGnfU5IsGJzTMY9Jkt323m+7vgcAAABLwaILzszF3oe7+/tJUlUnJ3nUAtud0t3XTctPTHJQVT1jerwyyZokT0jy5919bZJ097eras8kD0/ygara+FrL573u/+7um5JcVlV33dyQ3X1C5sI1y1et6W0/TQAAgKVtMQZn3fImSZLvb7LPS7r74zd7oaonJ9k0Bm+X5DvdfchmXvf6WzELAAAAm1iM31L7mSRPq6o9quqOSY5KcuYt7PPxJL9eVbsnSVXdZ9r3tCQvrKo9pvV37u7vJrmiqp45rauqOnjUyQAAAOyqFl1wdvf5Sd6d5JzMfX7zxO6+4BZ2OzHJZUnOr6pLkvxZkmXd/bHMfQ5zXVVdmOTl0/bPTfKiqrooyaVJnrrdTwQAAGAXV90+fnhbLV+1ple94M2zHgMWnQ1rj5z1CAAA7ABVdV53H77p+kV3hRMAAIClQXACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhlg26wGWggP3X5l1fm8QAADgZlzhBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCGWzXqApWD9lVdn9XGnznoMWHQ2rD1y1iMAADBDrnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMMQOD86q+q2q+kJVvW9HHxsAAIAdZ9kMjvkbSX6+u6+Yv7KqlnX3DTOYBwAAgAGGBmdVvSzJC6eHJyY5IMm9kpxSVe9KsjLJ3ZKsTnJVVT0/ydokj02yPMnbu/vPptd6RZLnJ7kpyd9293FVde8kb0+yX5Jrk/xad19eVe9Oct10vHsk+dUkL0jysCSf7+6jp9d8TpJXJqkkp3b371bVbkn+Z5LDk3SSd3X3m0a8PwAAAEvZsOCsqsMyF3oPyVzQfT7J85I8OcnjuvuqqnpdksOSPLK7r6uqY5Jc3d1HVNXyJGdV1WmZC8enJXlId19bVXeeDnNCkmO7+0tV9ZAkxyd5/PTcnablX0zy10kekeQ/JTm3qg5J8o0kfzwd/1+TnFZVT0vyL0n27+4HTuexz2bO75gkxyTJbnvvd9vfMAAAgCVm5BXORyb5cHd/P0mq6uQkj1pgu1O6+7pp+YlJDqqqZ0yPVyZZk+QJSf68u69Nku7+dlXtmeThST5QVRtfa/m81/3r7u6qWp/k6929fprj0sxdUb1HkjO6+5vT+vcleXSSP0hyr6p6a5JTk5y20Ml19wmZC94sX7Wmt/pdAQAA2EWMDM665U2SJN/fZJ+XdPfHb/ZCVU/O3O2t890uyXe6+5DNvO7109+b5i1vfLwsyYKfF+3uf62qg5M8KclvJnlWfnxbMAAAAFtp5LfUfibJ06pqj6q6Y5Kjkpx5C/t8PMmvV9XuSVJV95n2PS3JC6tqj2n9nbv7u0muqKpnTutqCsWt9fkkj6mqfafPbT4nyaerat8kt+vuDyX5L0ketA2vCQAAwGTYFc7uPn/68p5zplUndvcF825/XciJmbvd9fya2/CbSZ7W3R+bPne5rqr+LcnfZO7Lfp6b5E+r6tVJdk/yV0ku2sr5vlpVv5fk9MxdWf2b7v7IFK1/XlUbY/z3tvqkAQAA+JHq9vHD22r5qjW96gVvnvUYsOhsWHvkrEcAAGAHqKrzuvvwTdePvKUWAACAXZjgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhi2awHWAoO3H9l1vmBewAAgJtxhRMAAIAhBCcAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDLJv1AEvB+iuvzurjTp31GLDobFh75KxHAABghlzhBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQgvMW1BzvEwAAwDZaFCFVVaur6gtV9c6qurSqTquqO1TVr1XVuVV1UVV9qKr2qKqVVbVhYwRO6/6lqnavqntX1ceq6ryqOrOqDpi2uXdVnT291n+tqmum9XtW1aeq6vyqWl9VT91knuOTnJ/kp2b13gAAAOysFkVwTtYkeXt3PyDJd5I8PcnJ3X1Edx+c5AtJXtTdVye5KMljpv2ekuTj3f3DJCckeUl3H5bk5UmOn7Z5S5K3dPcRSf7PvGP+IMlR3f2gJI9L8t+rqqbn7pvkvd19aHd/edNhq+qYqlpXVetuvPbq7fYmAAAALBXLZj3APFd094XT8nlJVid5YFX9v0n2SbJnko9Pz5+U5NlJTk/yy0mOr6o9kzw8yQd+3IxZPv19WJKnTct/keSN03IleX1VPTrJTUn2T3LX6bkvd/fZmxu2u0/IXOBm+ao1fSvOFwAAYElbTMF5/bzlG5PcIcm7kzytuy+qqqOTPHZ6/pQkf1RVd05yWJK/S3LHJN/p7kO24ZjPTbJfksO6+4dVtSHJium579+60wAAACBZXLfULmSvJF+tqt0zF4dJku6+Jsk5mbtV9qPdfWN3fzfJFVX1zORHX/Zz8LTL2Zm7RTeZuyK60cok35hi83FJ7jH2dAAAAHYdiz04/0uSzyf5RJLLN3nupCTPm/5u9NwkL6qqi5JcmuSp0/qXJnlZVZ2TZFWSjR+6fF+Sw6tq3bTvpscAAADgVloUt9R294YkD5z3+I3znv7Tzezzwcx9BnP+uiuSPHmBza9M8tDu7qr65STrpu2vytznOxfywM2sBwAAYCssiuDcAQ5L8rbpG2i/k+SFM54HAABgydslgrO7z0xy8C1uCAAAwHaz2D/DCQAAwE5KcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABD7BI/izLagfuvzLq1R856DAAAgEXFFU4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGEJwAAAAMsWzWAywF66+8OquPO3XWY8Cis2HtkbMeAQCAGXKFEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGWDLBWVXLZj0DAAAAPzbz4Kyq1VV1eVWdWFWXVNX7quoJVXVWVX2pqh5cVXesqndV1blVdUFVPXXa9+iq+kBV/XWS06Z1vzNtd3FV/f684/zHad1FVfW/pnX7VdWHpu3PrapHTOsfU1UXTv8uqKq9ZvDWAAAA7NQWy1XBn0nyzCTHJDk3ya8keWSSX0zyyiSXJfm77n5hVe2T5Jyq+uS078OSHNTd366qJyZZk+TBSSrJKVX16CTfSvKqJI/o7quq6s7Tvm9J8qbu/vuq+ukkH09yvyQvT/Kb3X1WVe2Z5AebDlxVx0zzZre999vObwcAAMDOb7EE5xXdvT5JqurSJJ/q7q6q9UlWJ7l7kl+sqpdP269I8tPT8ie6+9vT8hOnfxdMj/fMXIAenOSD3X1Vkszb/glJ7l9VG+fYe7qaeVaS/1FV70tycnd/ZdOBu/uEJCckyfJVa/o2nj8AAMCSs1iC8/p5yzfNe3xT5ma8McnTu/uL83eqqock+f78VUn+qLv/bJPtfivJQlF4uyQP6+7rNlm/tqpOTfIfkpxdVU/o7su38ZwAAAB2aTP/DOdW+niSl9R0KbKqDt3Cdi+cboNNVe1fVT+R5FNJnlVVd5nWb7yl9rQkL964c1UdMv29d3ev7+4/TrIuyQEDzgkAAGBJ21mC8w+S7J7k4qq6ZHr873T3aUn+IsnnpttxP5hkr+6+NMkfJvl0VV2U5H9Mu/xWksOnLxO6LMmx0/qXTl9gdFGS65L87agTAwAAWKqq28cPb6vlq9b0qhe8edZjwKKzYe2Rsx4BAIAdoKrO6+7DN12/s1zhBAAAYCcjOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDLJv1AEvBgfuvzDq/NwgAAHAzrnACAAAwhOAEAABgiG0Kzqrat6pq1DAAAAAsHZsNzqp6aFWdUVUnV9WhVXVJkkuSfL2qnrzjRgQAAGBntKUvDXpbklcmWZnk75L8fHefXVUHJPnLJB/bAfMBAACwk9rSLbXLuvu07v5Akq9199lJ0t2X75jRAAAA2JltKThvmrd83SbP9YBZAAAAWEK2dEvtwVX13SSV5A7TcqbHK4ZPBgAAwE5ts8HZ3bvtyEEAAABYWrb5dzirap+qetWIYQAAAFg6tvSzKD9VVSdU1Uer6j9V1R5V9d+TfCnJT+y4EQEAANgZbekznO9N8ukkH0ry5CRnJ7k0yYHd/bUdMBsAAAA7sS0F5527+3XT8ser6utJjuju68ePBQAAwM5uS8GZqrpT5r6VNkm+lmSPqrpjknT3twfPBgAAwE5sS8G5Msl5+XFwJsn5099Ocq9RQwEAALDz29LPoqzegXMAAACwxGzpW2qfN2/5EZs89+KRQwEAALDz29LvcL5s3vJbN3nuhQNmAQAAYAnZUnDWZpYXegwAAAA3s6Xg7M0sL/QYAAAAbmZL31J7QFVdnLmrmfeeljM99g21AAAAbNGWgvP0JK9PcmVc0QQAAGAbbSk4T0vyxiSrkpyU5C+7+8IdMhUAAAA7vc1+hrO739LdD0vymCTfTvLnVfWFqnpNVd1nh00IAADATmlLXxqUJOnuL3f3H3f3oUl+JclRSb4wfDIAAAB2arcYnFW1e1U9parel+Rvk/xDkqcPnwwAAICd2mY/w1lVP5fkOUmOTHJOkr9Kckx3f38HzQYAAMBObEtfGvTKJH+R5OXd/e0dNM9Oaf2VV2f1cafOegxYdDasPXLWIwAAMEObDc7uftyOHAQAAICl5RY/wwkAAAC3huAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwJqmqM6rq8FnPAQAAsJQITgAAAIbYaYOzqlZX1eVVdWJVXVJV76uqJ1TVWVX1pap68PTvs1V1wfT3vtO+d6iqv6qqi6vqpCR3mPe6T6yqz1XV+VX1garac2YnCQAAsBPbaYNz8jNJ3pLkoCQHJPmVJI9M8vIkr0xyeZJHd/ehSV6T5PXTfr+e5NruPijJHyY5LEmqat8kr07yhO5+UJJ1SV620IGr6piqWldV62689upBpwcAALDzWjbrAW6jK7p7fZJU1aVJPtXdXVXrk6xOsjLJe6pqTZJOsvu036OT/EmSdPfFVXXxtP6hSe6f5KyqSpLbJ/ncQgfu7hOSnJAky1et6e1/agAAADu3nT04r5+3fNO8xzdl7tz+IMnp3X1UVa1Ocsa87ReKxEryie5+znafFAAAYBezs99Se0tWJrlyWj563vrPJHluklTVAzN3S26SnJ3kEVX1M9Nze1TVfXbMqAAAAEvLUg/ONyT5o6o6K8lu89b/aZI9p1tpX5HknCTp7m9mLkz/cnru7Mx9NhQAAIBtVN0+fnhbLV+1ple94M2zHgMWnQ1rj5z1CAAA7ABVdV53H77p+qV+hRMAAIAZEZwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQyyb9QBLwYH7r8w6P3APAABwM65wAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYIhlsx5gKVh/5dVZfdypsx4DFp0Na4+c9QgAAMyQK5wAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGGJJBmdVXXMr9zu8qv5ke88DAACwK1o26wEWk+5el2Tdpuurall33zCDkQAAAHZaS/IK50ZV9diq+ui8x2+rqqOn5SOq6rNVdVFVnVNVe83fvqpeV1UnVNVpSd47mzMAAADYee2SVzir6vZJTkry7O4+t6r2TnLdApseluSR3f3vnquqY5IckyS77b3fyHEBAAB2Skv6CucW3DfJV7v73CTp7u9u5pbZUxaKzWmfE7r78O4+fLc9Vo6cFQAAYKe01IPzhtz8HFdMfytJb8X+39/uEwEAAOwilnpwfjnJ/atqeVWtTPKz0/rLk9ytqo5Ikunzm7vk7cUAAACjLOnI6u5/qar3J7k4yZeSXDCt/7eqenaSt1bVHTL3+c0nzG5SAACApWdJBmd37zlv+RVJXrHANucmeegmq8+Y/qW7XzdsQAAAgF3AUr+lFgAAgBkRnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhluTvcO5oB+6/MuvWHjnrMQAAABYVVzgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwxLJZD7AUrL/y6qw+7tRZjwGLzoa1R856BAAAZsgVTgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGEJwAAAAMsWSDs6qumfUMAAAAu7IlG5wAAADM1pIPzqp6bFV9dN7jt1XV0dPyEVX12aq6qKrOqaq9qmq3qvpvVXVuVV1cVf95ZsMDAADsxJbNeoBZqarbJzkpybO7+9yq2jvJdUlelOTq7j6iqpYnOauqTuvuKzbZ/5gkxyTJbnvvt4OnBwAAWPx22eBMct8kX+3uc5Oku82Dgm4AABDiSURBVL+bJFX1xCQHVdUzpu1WJlmT5GbB2d0nJDkhSZavWtM7amgAAICdxa4QnDfk5rcOr5j+VpKFQrGSvKS7Pz56MAAAgKVsyX+GM8mXk9y/qpZX1cokPzutvzzJ3arqiCSZPr+5LMnHk/x6Ve0+rb9PVd1xFoMDAADszJb8Fc7u/peqen+Si5N8KckF0/p/q6pnJ3lrVd0hc5/ffEKSE5OsTnJ+VVWSbyZ52ixmBwAA2Jkt2eDs7j3nLb8iySsW2ObcJA9dYPdXTv8AAAC4lXaFW2oBAACYAcEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGGLJ/g7njnTg/iuzbu2Rsx4DAABgUXGFEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIZYNusBloL1V16d1cedOusxYNHZsPbIWY8AAMAMucIJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBA7PDir6prp7+qqumRHH38hVfW6qnr5rOcAAABYSlzhBAAAYIhFEZzT1c4zq+r86d/Dp/WPrapPV9X7q+ofqmptVT23qs6pqvVVde9pu6dU1eer6oKq+mRV3XVa/7qqeldVnVFV/1RVvzXvmK+qqi9W1SeT3Hfe+kOq6uyquriqPlxVd9rBbwcAAMCSsCiCM8k3kvxcdz8oybOT/Mm85w5O8v8kOTDJ85Pcp7sfnOTEJC+Ztvn7JA/t7kOT/FWSV8zb/4AkT0ry4CSvrardq+qwJL+c5NAkv5TkiHnbvzfJ73b3QUnWJ3ntQgNX1TFVta6q1t147dW3/swBAACWqGWzHmCye5K3VdUhSW5Mcp95z53b3V9Nkqr6xySnTevXJ3nctHz3JCdV1aokt09yxbz9T+3u65NcX1XfSHLXJI9K8uHuvnZ63VOmvyuT7NPdn572fU+SDyw0cHefkOSEJFm+ak3f2hMHAABYqhbLFc7fTvL1zF3NPDxz0bjR9fOWb5r3+Kb8OJjfmuRt3X1gkv+cZMVm9r9x3j4iEQAAYKDFEpwrk3y1u2/K3G2zu92K/a+cll+wFdt/JslRVXWHqtoryVOSpLuvTvKvVfWoabvnJ/n0Zl4DAACALVgst9Qen+RDVfXMJKcn+f427v+6JB+oqiuTnJ3knlvauLvPr6qTklyY5MtJzpz39AuSvKOq9kjyT0l+dRtnAQAAIEl1u7P0tlq+ak2vesGbZz0GLDob1h456xEAANgBquq87j580/WL5ZZaAAAAlhjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADDEslkPsBQcuP/KrPMD9wAAADfjCicAAABDCE4AAACGEJwAAAAMITgBAAAYQnACAAAwhOAEAABgCMEJAADAEIITAACAIQQnAAAAQwhOAAAAhhCcAAAADCE4AQAAGEJwAgAAMITgBAAAYAjBCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhBCcAAABDCE4AAACGWDbrAZaC9VdendXHnTrrMWDR2bD2yFmPAADADLnCCQAAwBCCEwAAgCEEJwAAAEMITgAAAIYQnAAAAAwhOAEAABhCcAIAADCE4AQAAGAIwQkAAMAQghMAAIAhls16gC2pqtcluSbJ3kk+092fnO1EAAAAbK1FHZwbdfdrZj0DAAAA22bR3VJbVa+qqi9W1SeT3Hda9+6qesa0vLaqLquqi6vqjfOef0dVnVlV/1BVvzCtX1FVf15V66vqgqp63LT+6Ko6uao+VlVfqqo3zDv+E6vqc1V1flV9oKr23OFvAgAAwBKwqK5wVtVhSX45yaGZm+38JOfNe/7OSY5KckB3d1XtM2/31Ukek+TeSU6vqp9J8ptJ0t0HVtUBSU6rqvtM2x8yHef6JF+sqrcmuS7Jq5M8obu/X1W/m+RlSf7rArMek+SYJNlt7/22zxsAAACwhCyq4EzyqCQf7u5rk6SqTtnk+e8m+UGSE6vq1CQfnffc+7v7piRfqqp/SnJAkkcmeWuSdPflVfXlJBuD81PdffV0nMuS3CPJPknun+SsqkqS2yf53EKDdvcJSU5IkuWr1vRtOWkAAIClaLEFZ5JsNt66+4aqenCSn83cldAXJ3n8ZvbrJLWF41w/b/nGzL0XleQT3f2cbR0aAACAm1tsn+H8TJKjquoOVbVXkqfMf3L6POXK7v6bJC/N3G2xGz2zqm5XVfdOcq8kX5xe77nTvvdJ8tPT+s05O8kjpttxU1V7zLsFFwAAgG2wqK5wdvf5VXVSkguTfDnJmZtssleSj1TVisxdjfztec99Mcmnk9w1ybHd/YOqOj7JO6pqfZIbkhzd3ddPt8sudPxvVtXRSf6yqpZPq1+d5B+2ywkCAAA7rR/+8If5yle+kh/84AezHmVmVqxYkbvf/e7Zfffdt2r76t75P35YVe9O8tHu/uAsjr981Zpe9YI3z+LQsKhtWHvkrEcAANhurrjiiuy11165y13uks1dxFrKujvf+ta38r3vfS/3vOc9b/ZcVZ3X3Ydvus9iu6UWAABgUfrBD36wy8ZmklRV7nKXu2zTFd5FdUvtrdXdR896BgAAYOnbVWNzo209f1c4AQAAGGJJXOEEAADY0VYfd+p2fb2t/f6Lr33ta3npS1+ac889N8uXL8/q1avz5je/Ofe5z/b5gY0zzjgjt7/97fPwhz/8Nr+WK5wAAAA7ie7OUUcdlcc+9rH5x3/8x1x22WV5/etfn69//evb7RhnnHFGPvvZz26X1xKcAAAAO4nTTz89u+++e4499tgfrTvkkEPyyEc+Mr/zO7+TBz7wgTnwwANz0kknJZmLx1/4hV/40bYvfvGL8+53vztJsnr16rz2ta/Ngx70oBx44IG5/PLLs2HDhrzjHe/Im970phxyyCE588xNf6ly27ilFgAAYCdxySWX5LDDDvt3608++eRceOGFueiii3LVVVfliCOOyKMf/ehbfL199903559/fo4//vi88Y1vzIknnphjjz02e+65Z17+8pff5nkF53Zw4P4rs87vDQIAADPy93//93nOc56T3XbbLXe9613zmMc8Jueee2723nvvLe73S7/0S0mSww47LCeffPJ2n8sttQAAADuJBzzgATnvvPP+3fruXnD7ZcuW5aabbvrR401/Q3P58uVJkt122y033HDDdpx0juAEAADYSTz+8Y/P9ddfn3e+850/WnfuuefmTne6U0466aTceOON+eY3v5nPfOYzefCDH5x73OMeueyyy3L99dfn6quvzqc+9albPMZee+2V733ve9tlXrfUAgAA3Apb+zMm21NV5cMf/nBe+tKXZu3atVmxYsWPfhblmmuuycEHH5yqyhve8Ib85E/+ZJLkWc96Vg466KCsWbMmhx566C0e4ylPeUqe8Yxn5CMf+Uje+ta35lGPetStn3dzl17ZeocffnivW7du1mMAAAADfeELX8j97ne/WY8xcwu9D1V1Xncfvum2bqkFAABgCMEJAADAEIITAABgK+3qH0nc1vMXnAAAAFthxYoV+da3vrXLRmd351vf+lZWrFix1fv4lloAAICtcPe73z1f+cpX8s1vfnPWo8zMihUrcve7332rtxecAAAAW2H33XfPPe95z1mPsVNxSy0AAABDCE4AAACGEJwAAPzf9u4u1LKyjuP499c0vUAvQiMkjjmBXhWmFVPhjWiBlsxcJDQXZobeBJJBENVFUXfdRPSC4htqhRkacRIlFAvrwtfJl3QaGCRoSJjSGpXCmPp7sZfTYbvPzLqY9ay1z/5+YHP2Ps9z4M/5nf/Zz7POWutI0iCyqndYOpGSvATsH7sOAbAN+PvYRego85gW85gOs5gW85gOs5gW85iWqedxelWdPP9Jbxp0Yuyvqg+PXYQgyaNmMR3mMS3mMR1mMS3mMR1mMS3mMS3Lmoen1EqSJEmSBuGGU5IkSZI0CDecJ8Z1Yxego8xiWsxjWsxjOsxiWsxjOsxiWsxjWpYyD28aJEmSJEkahH/hlCRJkiQNwg2nJEmSJGkQbjh7SnJhkv1JDiT56oLxNye5vRt/KMmO9lWujh55XJ7kb0ke7x5XjlHnKkhyU5JDSf64wXiSfL/L6skkH2xd4yrpkcd5SQ6v641vtK5xVSQ5LclvkuxL8nSSqxfMsT8a6ZmH/dFAkrckeTjJE10W31owx3VVIz3zcF3VUJItSf6Q5K4FY0vXG/4fzh6SbAF+BHwCOAg8kmStqp5ZN+0K4B9VdUaSPcB3gM+0r3bz65kHwO1VdVXzAlfPzcAPgVs3GL8IOLN7fAS4pvuoYdzMsfMA+F1VXdymnJV2BPhyVe1N8nbgsST3zv2usj/a6ZMH2B8tvAKcX1UvJ9kK/D7JPVX14Lo5rqva6ZMHuK5q6WpgH/COBWNL1xv+hbOfncCBqnq2qv4D/AzYPTdnN3BL9/wO4IIkaVjjKumThxqpqgeAF44xZTdwa808CJyU5JQ21a2eHnmokap6rqr2ds9fYrZ4OHVumv3RSM881ED38/5y93Jr95i/i6XrqkZ65qFGkmwHPgXcsMGUpesNN5z9nAr8Zd3rg7z+TeronKo6AhwG3tWkutXTJw+AT3enqN2R5LQ2pWmBvnmpnY91p07dk+R9YxezCrpTns4BHpobsj9GcIw8wP5oojtl8HHgEHBvVW3YG66rhtcjD3Bd1cr3gK8A/9tgfOl6ww1nP4uOGswf+ekzRydGn+/1r4AdVXUWcB//PxKk9uyNadkLnF5VHwB+APxy5Ho2vSRvA+4EvlRVL84PL/gS+2NAx8nD/mikqv5bVWcD24GdSd4/N8XeaKhHHq6rGkhyMXCoqh471rQFn5t0b7jh7OcgsP5IznbgrxvNSfJG4J14WttQjptHVT1fVa90L68HPtSoNr1en/5RI1X14munTlXV3cDWJNtGLmvT6q6HuhP4aVX9YsEU+6Oh4+Vhf7RXVf8EfgtcODfkumoEG+XhuqqZc4FdSf7M7JKx85P8ZG7O0vWGG85+HgHOTPLeJG8C9gBrc3PWgM91zy8B7q+qSR9tWGLHzWPuGqhdzK7V0TjWgMu6u3F+FDhcVc+NXdSqSvLu1671SLKT2fvA8+NWtTl13+cbgX1V9d0NptkfjfTJw/5oI8nJSU7qnr8V+Djwp7lprqsa6ZOH66o2quprVbW9qnYwW9/eX1WXzk1but7wLrU9VNWRJFcBvwa2ADdV1dNJvg08WlVrzN7EfpzkALOjDHvGq3hz65nHF5PsYnZXwheAy0creJNLchtwHrAtyUHgm8xuOEBVXQvcDXwSOAD8C/j8OJWuhh55XAJ8IckR4N/Anqm/US2xc4HPAk9110YBfB14D9gfI+iTh/3RxinALd1d598A/Lyq7nJdNZo+ebiuGtGy90b8PSpJkiRJGoKn1EqSJEmSBuGGU5IkSZI0CDeckiRJkqRBuOGUJEmSJA3CDackSZIkaRBuOCVJkiRJg3DDKUmSJEkaxKsgu3czLEU1BQAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 1080x720 with 1 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"size = 10\n",
"df_POS.loc['ADJ'].sort_values('Count').tail(size).plot(kind='barh', figsize=(15,10))\n",
"plt.title('TOP 10 ADJ')\n",
"plt.ylabel('ADJ')\n",
"plt.show()\n",
"df_POS.loc['NOUN'].sort_values('Count').tail(size).plot(kind='barh', figsize=(15,10))\n",
"plt.title('TOP 10 NOUN')\n",
"plt.ylabel('NOUN')\n",
"plt.show()\n",
"df_POS.loc['PROPN'].sort_values('Count').tail(size).plot(kind='barh', figsize=(15,10))\n",
"plt.title('TOP 10 PROPN')\n",
"plt.ylabel('PROPN')\n",
"df_POS.loc['VERB'].sort_values('Count').tail(size).plot(kind='barh', figsize=(15,10))\n",
"plt.title('TOP 10 VERB')\n",
"plt.ylabel('VERB')\n",
"plt.show()"
]
},
{
"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": "3.7.3"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
@angulo4
Copy link

angulo4 commented Sep 5, 2020

Gracias por el código, funciona muy bien en Jupyter.

Thanks, just one problem (I'm not so good with Python), on block [6]:

doc = nlp(textp) # <-- on this line of code

I'm getting this TypError:
"TypeError: Argument 'string' has incorrect type (expected str, got list)"
Don't know what to do.
Thanks
Cheers
Mario

@natzir
Copy link
Author

natzir commented Sep 7, 2020

@angulo4 mira a ver si lo que tienes en "textp" es un string, no sea que algún carácter del texto a analizar lo haya roto.

@angulo4
Copy link

angulo4 commented Sep 7, 2020

Solucionado!
Buscando en Google, encontré una posible solución;

textp = " ".join(text) # codigo original

Lo he cambiado por:
textp = " ".join(str(elem) for elem in text) # codigo modificado

Ahora funciona el resto del código.

Muchas gracias @natzir

Mario

@natzir
Copy link
Author

natzir commented Sep 7, 2020

@angulo4 grande!

@angulo4
Copy link

angulo4 commented Oct 1, 2020 •
edited
Loading

Another update,
Utilizando tu código hoy (lo uso muy a menudo, antes de comenzar con SEMrush y otros), me daba "None" en las descripciones. Revisando el código de los resultados de Google, he visto que en:
description = r.find('span', attrs={'class': 'st'})
'class' ha cambiado de 'st' a 'aCOpRe'

He actualizado esa línea a:
description = r.find('span', attrs={'class': 'aCOpRe'})
y ahora me esta funcionando.
La pregunta, sabes si Google cambia las 'class' a menudo?, o no te había pasado antes?

Gracias y saludos

@natzir
Copy link
Author

natzir commented Oct 1, 2020

pasa muy a menudo @angulo4

@angulo4
Copy link

angulo4 commented Oct 1, 2020

Pues nada, gracias y lo iremos actualizando y comentando @natzir

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment