commit files to HF hub

.gitattributes

@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.mp3 filter=lfs diff=lfs merge=lfs -text
 *.ogg filter=lfs diff=lfs merge=lfs -text
 *.wav filter=lfs diff=lfs merge=lfs -text
+*.csv filter=lfs diff=lfs merge=lfs -text

code/01-get_products_list.py

@@ -0,0 +1,55 @@
+import sys
+import os
+sys.path.append('python-sdk/lib/')
+from meli import Meli
+import pandas as pd
+from tqdm import tqdm
+import json
+
+countries_ids = {
+    'MLA':'Argentina', 
+    'MCO':'Colombia', 
+    'MPE':'Perú', 
+    'MLU':'Uruguay',
+    'MLC':'Chile', 
+    'MLM':'Mexico', 
+    'MLV':'Venezuela', 
+    'MLB':'Brasil'
+}
+
+n_tries = 5
+
+def generate_products_list():
+    meli = Meli(client_id=1234, client_secret="a secret")
+    products_and_categories = {'prod_id': [], 'cat_id': []}
+    for country_id, country in countries_ids.items():
+        try:
+            print('País:',country)
+            # Obtengo las categorías del país:
+            for i in range(n_tries):
+                try:
+                    categories = meli.get('sites/{}/categories/all'.format(country_id)).json()
+                    break
+                except json.decoder.JSONDecodeError:
+                    print('Error 1')
+            # para cada categoría, obtengo sus productos
+            for category_id in tqdm(categories.keys()):
+                for i in range(n_tries):
+                    try:
+                        products = meli.get('sites/{}/search?category={}'.format(country_id,category_id)).json()['results']
+                        break
+                    except (KeyError, json.decoder.JSONDecodeError) as e:
+                        print('Error 2')
+                products_and_categories['prod_id'].extend([product['id'] for product in products])
+                products_and_categories['cat_id'].extend([category_id] * len(products))
+        
+        except KeyboardInterrupt:
+            pass
+            
+        pd.DataFrame(products_and_categories).to_csv('./products/{}.csv'.format(country),index=False)
+        
+        
+if __name__ == '__main__':
+    generate_products_list()
+
+

code/02-remove_duplicates_and_get_reviews.py

@@ -0,0 +1,76 @@
+import sys
+sys.path.append('python-sdk/lib/')
+from meli import Meli
+import pandas as pd
+import argparse
+
+countries_ids = {
+    'MLA':'Argentina', 
+    'MCO':'Colombia', 
+    'MPE':'Perú', 
+    'MLU':'Uruguay',
+    'MLC':'Chile', 
+    'MLM':'Mexico', 
+    'MLV':'Venezuela', 
+    'MLB':'Brasil'
+}
+
+n_tries = 5
+
+def remove_duplicates_and_get_reviews(parts_dir,countries_ids):
+    products_df = pd.concat([pd.read_csv('products/{}.csv'.format(country))
+                             for country in countries_ids.values()]).drop_duplicates(subset=['prod_id'])
+    products_df = products_df.reset_index(drop=True)
+    # offset = 141000
+    prod_len = len(products_df)
+    # products_df = products_df.iloc[offset:,:]
+    meli = Meli(client_id=1212334,client_secret='a secret')
+    reviews_dict = {'prod_id': [], 'cat_id': [], 'review_id': [], 'country': [],
+                    'prod_title': [], 'reviewer_id': [], 'review_date': [],
+                    'review_status': [], 'review_title': [], 'review_content': [],
+                    'review_rate': [], 'review_likes': [], 'review_dislikes': []}
+    
+    for idx, (prod_id, category_id) in products_df.iterrows():
+        print('{}/{} ({:.1f}%)'.format(idx,prod_len,idx/prod_len*100))
+        prod_title = meli.get('items/{}'.format(prod_id)).json()['title']
+        country = prod_id[:3]
+        for i in range(n_tries):
+            try:
+                reviews = meli.get('/reviews/search?item_id={}&limit=100&order_criteria=valorization'.format(prod_id)).json()['results']
+                break
+            except KeyError:
+                print('Error')
+        for review in reviews:
+            reviews_dict['prod_id'].append(prod_id)
+            reviews_dict['cat_id'].append(category_id)
+            reviews_dict['country'].append(country)
+            reviews_dict['prod_title'].append(prod_title)
+            reviews_dict['review_id'].append(review['id'])
+            reviews_dict['review_date'].append(review['date_created'])
+            reviews_dict['review_status'].append(review['status'])
+            reviews_dict['review_title'].append(review['title'])
+            reviews_dict['review_content'].append(review['content'])
+            reviews_dict['review_rate'].append(review['rate'])
+            reviews_dict['review_likes'].append(review['likes'])
+            reviews_dict['review_dislikes'].append(review['dislikes'])
+            reviews_dict['reviewer_id'].append(review['reviewer_id'])
+        if idx % 200 == 199:
+            df_reviews = pd.DataFrame(reviews_dict)
+            df_reviews.to_csv('./{}/reviews_part{:06}.csv'.format(parts_dir,idx+1),index=False)
+            reviews_dict = {'prod_id': [], 'cat_id': [], 'review_id': [], 'country': [],
+                            'prod_title': [], 'reviewer_id': [], 'review_date': [],
+                            'review_status': [], 'review_title': [], 'review_content': [],
+                            'review_rate': [], 'review_likes': [], 'review_dislikes': []}
+    
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--partsdir', type=str, required=True)
+    parser.add_argument('--countries', type=str, nargs="+", required=False)
+    args = vars(parser.parse_args())
+    parts_dir = args["partsdir"]
+    countries_ids = {country_id: country_name for country_id, country_name in countries_ids.items() \
+                    if country_id in args["countries"]}
+    remove_duplicates_and_get_reviews(parts_dir,countries_ids)
+
+if __name__ == '__main__':
+    main()

code/03-get_parts_in_csv.py

@@ -0,0 +1,90 @@
+import numpy as np
+import pandas as pd
+import os
+
+root_path = './28-melisa/melisa/'
+
+def filenames_generator(parts_directories):
+
+	rs = np.random.RandomState(127361824)
+	for directory in parts_directories:
+		print('Extracting from {}'.format(directory))
+		list_parts = np.array(os.listdir(directory))
+		num_parts = len(list_parts)
+		for filename in rs.choice(list_parts,num_parts,replace=False):
+			yield '{}{}'.format(directory,filename)
+
+def drop_dups(df):
+	# Elimino los duplicados y los que tienen valores faltantes:
+	df = df.drop_duplicates(subset=['review_id'])\
+		.reset_index(drop=True).dropna()
+
+	assert df['prod_id'].apply(type).eq(str).all()
+	assert df['cat_id'].apply(type).eq(str).all()
+	assert df['review_id'].apply(type).eq(int).all()
+	assert df['country'].isin(['MLB','MLA','MLM',
+	'MLU','MCO','MLC','MLV','MPE']).all()
+	assert df['prod_title'].apply(type).eq(str).all()
+	assert df['reviewer_id'].apply(type).eq(int).all()
+	assert df['review_date'].apply(type).eq(str).all()
+	assert df['review_status'].apply(type).eq(str).all()
+	df['review_title'] = df['review_title'].apply(str)
+	assert df['review_title'].apply(type).eq(str).all()
+	assert df['review_content'].apply(type).eq(str).all()
+	assert df['review_rate'].isin([1, 2, 3, 4, 5]).all()
+	assert df['review_likes'].apply(type).eq(int).all()
+	assert df['review_dislikes'].apply(type).eq(int).all()
+
+	print('Cantidad de reviews únicos descargados:',len(df))
+
+	# Cambio todos los espacios por espacios simples 
+	# y vuelvo a eliminar duplicados:
+	df['review_content'] = df['review_content']\
+		.str.replace(r'\s+',' ',regex=True)
+	df['review_title'] = df['review_title'].str.replace(r'\s+',' ',regex=True)
+	df = df.drop_duplicates(subset=['review_content',
+				'review_title','review_rate']).reset_index(drop=True)
+	print('Cantidad de reviews con contenido, título y rate únicos:',len(df))
+
+	return df
+
+
+def get_csv(filenames_gen,csv_filename):
+	df = pd.concat([pd.read_csv(filename,lineterminator='\n',sep=',') \
+		for filename in filenames_gen], ignore_index=True)
+	df = drop_dups(df)
+
+	# Guardo en un csv los campos más importantes:
+	df.to_csv(root_path + csv_filename,index=False)
+	print('Guardado OK.')
+
+
+def merge_dfs(csvs_lists):
+	df = pd.concat([pd.read_csv(csv_filename,lineterminator='\n',sep=',') \
+		for csv_filename in csvs_lists], ignore_index=True)
+	df = drop_dups(df)
+	return df
+
+
+def get_parts_in_csv():
+
+	csvs = {'orig.csv': ['parts/', 'parts-2/'],
+			'ven.csv': ['ven_parts/'],
+			'per.csv': ['peru_parts/']}
+
+	generators = {csvfilename: filenames_generator([root_path + 'all_parts/' + part for part in directories])\
+					for csvfilename,directories in csvs.items()}
+
+	for filename,gen in generators.items():
+		print('Generando archivo {}...'.format(filename))
+		get_csv(gen,filename)
+		print()
+				
+	print('Mergeando todos en un único csv...')
+	df = merge_dfs([root_path + csvfiles for csvfiles in csvs.keys()])
+	filename = root_path + 'reviews_all.csv'
+	df.to_csv(filename,index=False)
+
+
+if __name__ == "__main__":
+	get_parts_in_csv()

code/04-process_csv.py

@@ -0,0 +1,188 @@
+import sys
+import pandas as pd
+pd.options.mode.chained_assignment = None  # default='warn'
+import numpy as np
+
+
+root_path = './28-melisa/melisa/'
+MeLi_path = './29-mercado-libre-api-v3/python-sdk/lib/'
+sys.path.append(MeLi_path)
+from meli import Meli
+from get_parts_in_csv import drop_dups
+
+
+# Se ejecutó --> dict(df['category'].value_counts())
+# Se obtuvo el diccionario de ocurrencias de las categorías.
+# Se usa eso para generar un conversor de categorías menos específico:
+
+generalize_categories = {
+
+	'Hogar, Muebles y Jardín': 'Hogar / Casa',
+	'Casa, Móveis e Decoração': 'Hogar / Casa',
+	'Herramientas y Construcción': 'Hogar / Casa',
+	'Industrias y Oficinas': 'Hogar / Casa',
+	'Ferramentas e Construção': 'Hogar / Casa',
+	'Bebés': 'Hogar / Casa',
+	'Animales y Mascotas': 'Hogar / Casa',
+	'Hogar y Muebles': 'Hogar / Casa',
+	'Bebês': 'Hogar / Casa',
+	'Animais': 'Hogar / Casa',
+	'Indústria e Comércio': 'Hogar / Casa',
+	'Industrias': 'Hogar / Casa',
+
+	'Computación': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Accesorios para Vehículos': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Acessórios para Veículos': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Electrónica, Audio y Video': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Electrodomésticos y Aires Ac.': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Celulares y Telefonía': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Informática': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Eletrônicos, Áudio e Vídeo': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Electrodomésticos': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Eletrodomésticos': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Celulares y Teléfonos': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Cámaras y Accesorios': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Consolas y Videojuegos': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Celulares e Telefones': 'Tecnología y electrónica / Tecnologia e electronica',
+	'Câmeras e Acessórios': 'Tecnología y electrónica / Tecnologia e electronica',
+
+	'Deportes y Fitness': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Belleza y Cuidado Personal': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',  
+	'Calçados, Roupas e Bolsas': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Esportes e Fitness': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Ropa y Accesorios': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Salud y Equipamiento Médico': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Beleza e Cuidado Pessoal': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Ropa, Bolsas y Calzado': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Saúde': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Vestuario y Calzado': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Ropa, Calzados y Accesorios': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Ropa, Zapatos y Accesorios': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+	'Estética y Belleza': 'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal',
+
+	'Juegos y Juguetes': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Brinquedos e Hobbies': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Arte, Librería y Mercería': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Instrumentos Musicales': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Arte, Papelaria e Armarinho': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Arte, Papelería y Mercería': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Joyas y Relojes': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Instrumentos Musicais': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Games': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Joias e Relógios': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Souvenirs, Cotillón y Fiestas': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Festas e Lembrancinhas': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Recuerdos, Cotillón y Fiestas': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Antigüedades y Colecciones': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Libros, Revistas y Comics': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Relojes y Joyas': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Antiguidades e Coleções': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Arte, Librería y Cordonería': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Recuerdos, Piñatería y Fiestas': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Música, Películas y Series': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Música, Filmes e Seriados': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Relojes, Joyas y Bisutería': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Música y Películas': 'Arte y entretenimiento / Arte e Entretenimiento',
+	'Livros, Revistas e Comics': 'Arte y entretenimiento / Arte e Entretenimiento',
+
+	'Alimentos y Bebidas': 'Alimentos y Bebidas / Alimentos e Bebidas',
+	'Alimentos e Bebidas': 'Alimentos y Bebidas / Alimentos e Bebidas',
+
+	'Servicios': np.nan,
+	'Serviços': np.nan,
+	'Agro': np.nan,
+	'Otras categorías': np.nan,
+	'Mais Categorias': np.nan,
+	'Otras Categorías': np.nan,
+	'Ingressos': np.nan,
+	'Entradas para Eventos': np.nan,
+	'Boletas para Espectáculos': np.nan,
+	'Autos, Motos y Otros': np.nan
+
+}
+
+countries = ['MLB','MLA','MLM','MLU','MCO','MLC','MLV','MPE']
+
+rates = [1, 2, 3, 4, 5]
+
+cat2n = {
+
+	'Tecnología y electrónica / Tecnologia e electronica': [14396,11189,9350,1473,3237,3188,656,316],
+	'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal': [16032,11399,9522,1228,3260,2295,173,211],
+	'Hogar / Casa': [19477,16601,8613,2175,2321,3188,182,157],
+	'Arte y entretenimiento / Arte e Entretenimiento': [4572,2710,1789,253,487,380,72,30],
+	'Alimentos y Bebidas / Alimentos e Bebidas': [386,462,467,29,22,86,2,2] 
+
+}
+
+
+def get_orig_cat_name():
+	filename = root_path + 'reviews_all.csv'
+	df = pd.read_csv(filename)
+
+	all_categories = pd.read_csv(root_path + 'catid2catname.csv',index_col='cat_id').to_dict()['category']
+	df['category'] = df['cat_id'].map(all_categories)
+	return df.dropna().reset_index(drop=True)
+
+
+def sort_by_score(df):
+    score_map = {2:1., 1:.75, 3:.5, 4:.25, 5:0.}
+    df['rate_score'] = df['review_rate'].map(score_map)
+    diff = df['review_likes'] - df['review_dislikes']
+    vals = np.log(diff - diff.min() + 1)
+    vals_norm = (vals - vals.min()) / (vals.max()-vals.min())
+    lenghts = np.maximum(0,df['review_content'].str.len()-50)
+    lenghts_norm = (lenghts - lenghts.min()) / (lenghts.max() - lenghts.min())
+    df['val_score'] = vals_norm + .5 * lenghts_norm
+    df = df.sort_values(by=['rate_score','val_score'],ascending=[False,False])
+    return df.drop(['rate_score','val_score'], axis=1)
+
+
+def mask(df,ctry,cat,rate):
+    return ((df['country'] == ctry) & (df['category'] == cat) & (df['review_rate'] == rate))
+
+
+def sample_and_get_indices(df,n):
+    diff = df['review_likes'] - df['review_dislikes']
+    vals = np.log(diff - diff.min() + 1)
+    vals_norm = (vals - vals.min()) / (vals.max()-vals.min())
+    lenghts = np.maximum(0,df['review_content'].str.len()-50)
+    lenghts_norm = (lenghts - lenghts.min()) / (lenghts.max() - lenghts.min())
+    df['score'] = vals_norm + .5 * lenghts_norm
+    df = df.sort_values(by=['score'],ascending=False)
+    return df.iloc[:n,:].index.tolist()
+
+
+def process_csv():
+	print('Generalizando categorías...')
+	df = get_orig_cat_name()
+	df['category'] = df['category'].map(generalize_categories)
+	df = df.dropna().reset_index(drop=True)
+	print('Limitando por cantidad de productos...')
+	df = sort_by_score(df)
+	df = df.groupby(['prod_id']).head(30)
+	df = df.reset_index(drop=True)
+
+	print('Sampleando los reviews por categoría, país y rate...')
+	indices = []
+	for cat, n in cat2n.items():
+		for rate in rates:
+			for i, country in enumerate(countries):
+				df_new = df[mask(df,country,cat,rate)]
+				idx = sample_and_get_indices(df_new,n[i])
+				indices.extend(idx)
+		print('Cantidad de índices hasta ahora:',len(indices))
+	df = df.loc[indices,:].reset_index(drop=True)
+	
+	#df.to_csv(root_path + 'reviews_sampled_full.csv',index=False)
+	df = drop_dups(df)
+
+	df_esp = df[df['country']!='MLB']
+	df_esp.to_csv(root_path + 'reviews_esp_full.csv',index=False)
+
+	df_por = df[df['country']=='MLB']
+	df_por.to_csv(root_path + 'reviews_por_full.csv',index=False)
+
+
+if __name__ == "__main__":
+	process_csv()

code/06-train-dev-test-split.py

@@ -0,0 +1,170 @@
+import pandas as pd
+import numpy as np
+import fasttext
+
+to_be_removed_esp = [
+     483197,483223,483442,483512,484318,484498,484586,484729,485267,485810,
+     485903,486030,486154,486295,486525,486749,486773,486786,486791,486811,486814,
+     486827,486842,486848,486855,486854,486869,486908,486917,486918,486922,486925,
+     486931,486934,486935,486940,486941,486948,486956,486958,486959,486961,486962,
+     486966,486968,486973,486974,486978,486981,486983,486987,486992,486999,487001,
+     487003,487009,487019,487018,487021,487022,487026,487028,487029,487033,487037,
+     487041,487044,487052,487060,487062,487065,487072,487074,487076,487081,487084,
+     487085,487088,487092,487094,487095,487099,487103,487105,487106,487109,487110,
+     487111,487112,487115,487122,487126,487133,487135,487138,487142,487143,487146,
+     487154,487155,487156,487162,487164,487173,487179,487185,487189,487197,487198,
+     487199,487204,487207,487210,487211,487216,487219,487225,487229,487233,487234,
+     487235,487243,487244,487245,487251,487252,487253,487254,487255,487256,487258,
+     487264,487273,487276,487282,487290,487292,487294,487298,487304,487303,487308,
+     487318,487321,487323,487326,487327,487329,487330,487331,487334,487337,487346,
+     487343,487347,487349,487350,487360,487361,487365,487366,487375,487379,487380,
+     487386,487389,487391,487393,487396,487397,487399,487400,487401,487402,487411,
+     487412,487414,487416,487417,487419,487421,487424,487425,487427,487433,487435,
+     487437,487443,487449,487452,487453,487455,487456,487459,487463,487465,487466,
+     487468,487471,487479,487481,487483,487485,487486,487487,487488,487489,487490,
+     487493,487494,487496,487498,487500,487501,487502,487505,487506,487512,487517,
+     487519,487528,487525,487529,487530,487537,487538,487541,487542,487545,487554,
+     487555,487556,487558,487567,487569,487573,487574,487578,487582,487586,487587,
+     487592,487596,487602,487603,487604,487607,487608,487609,487612,487613,487616,
+     487617,487618,487621,487623,487624                
+]
+
+to_be_removed_por = [
+    274310,274300,274299,274294,274287,274281,274265,274259,274256,274255,274232,274225,274226,274219,
+    274213,274206,274200,274199,274194,274172,274171,274170,274167,274166,274165,274163,274153,274146,
+    274143,274142,274136,274134,274130,274125,274123,274122,274109,274108,274079,274075,274073,274071,
+    274068,274057,274054,274044,274043,274042,274030,274029,274019,274018,274017,274015,274014,274011,
+    273998,273975,273969,273967,273951,273934,273924,273922,273914,273910,273909,273901,273899,273895,
+    273889,273881,273876,273871,273875,273869,273820,273812,273799,273791,273786,273783,273781,273780,
+    273779,273772,273768,273754,273750,273741,273739,273736,273732,273731,273727,273715,273703,273674,
+    273596,273595,
+]
+
+
+# countries = ['MLB','MLA','MLM','MLU','MCO','MLC','MLV','MPE']
+# esp_countries = ['MLA','MLM','MLU','MCO','MLC','MLV','MPE']
+# rates = [1, 2, 3, 4, 5]
+
+abbreviations = {
+    'Hogar / Casa': 'HOGAR',
+    'Tecnología y electrónica / Tecnologia e electronica': 'TEC',
+    'Arte y entretenimiento / Arte e Entretenimiento': 'ARTE',
+    'Salud, ropa y cuidado personal / Saúde, roupas e cuidado pessoal': 'SALUD',
+    'Alimentos y Bebidas / Alimentos e Bebidas': 'ALIMENTOS'
+}
+inv_abbreviations = {v:k for k,v in abbreviations.items()}
+
+
+def detect_lang_fasttext(df_es,df_pt):
+    ds_es = (df_es['review_content'] + ' ' + df_es['review_title']).astype(str)
+    ds_pt = (df_pt['review_content'] + ' ' + df_pt['review_title']).astype(str)
+    
+    model_predict = fasttext.load_model('../datav2/lid.176.bin').predict
+
+    def apply_lang_detect(text):
+        return dict(zip(*[('lang','prob'),next(zip(*model_predict(text, k=1)))]))
+
+    lang_score_es = pd.DataFrame(ds_es.apply(apply_lang_detect).tolist())
+    lang_score_pt = pd.DataFrame(ds_pt.apply(apply_lang_detect).tolist())
+    
+    lang_score_es.loc[lang_score_es['lang'] != '__label__es', 'prob'] = 0.
+    df_es['lang_prob'] = lang_score_es['prob']
+    df_es = df_es.sort_values(by=['lang_prob'],ascending=False).reset_index(drop=True)
+    
+    lang_score_pt.loc[lang_score_pt['lang'] != '__label__pt', 'prob'] = 0.
+    df_pt['lang_prob'] = lang_score_pt['prob']
+    df_pt = df_pt.sort_values(by=['lang_prob'],ascending=False).reset_index(drop=True)
+    
+    return df_es, df_pt
+
+
+def train_test_split(
+    df,
+    samples,
+    random_seed
+):
+    rs = np.random.RandomState(random_seed)
+    test_indices = []
+    for country in samples.keys():
+        for cat, n in samples[country].items():
+            if n == 0:
+                continue
+#             print(country, cat, n)
+#             print(df.loc[
+#                 (df['country'] == country) & (df['category'] == inv_abbreviations[cat]), "review_rate"
+#             ])
+            idx = df[
+                (df['country'] == country) & (df['category'] == inv_abbreviations[cat])
+            ].groupby('review_rate').sample(n=n,random_state=rs).index.tolist()
+            test_indices.extend(idx)
+
+    df_test = df.loc[
+        test_indices, ['country','category','review_content','review_title','review_rate']
+    ].reset_index(drop=True)
+    
+    train_indices = sorted(list(set(range(len(df))) - set(test_indices)))
+    df_train = df.loc[
+        train_indices, ['country','category','review_content','review_title','review_rate']
+    ].reset_index(drop=True)
+    
+    return df_train, df_test
+
+def main():
+    # Se leen todos los comentarios descargados
+    df_es = pd.read_csv('./reviews_es_full.csv')
+    df_pt = pd.read_csv('./reviews_pt_full.csv')
+    
+    # Se ordenan por relevancia según idioma
+    df_es, df_pt = detect_lang_fasttext(df_es,df_pt)
+    
+    ## ESPAÑOL
+    # Se eliminan los que están en la lista to_be_removed_esp
+    df_es = df_es.drop(set(to_be_removed_esp)).reset_index(drop=True)
+    
+    # Se extrae el conjunto de test
+    es_country_samples = {
+        'MLA':{'ALIMENTOS': 3,'ARTE':30,'HOGAR': 156,'SALUD':210,'TEC':315},
+        'MLM':{'ALIMENTOS': 4,'ARTE':30,'HOGAR': 156,'SALUD':210,'TEC':315},
+        'MLU':{'ALIMENTOS': 4,'ARTE':30,'HOGAR': 156,'SALUD':210,'TEC':315},
+        'MCO':{'ALIMENTOS': 4,'ARTE':30,'HOGAR': 156,'SALUD':210,'TEC':315},
+        'MLC':{'ALIMENTOS': 4,'ARTE':30,'HOGAR': 156,'SALUD':210,'TEC':315},
+        'MLV':{'ALIMENTOS': 2,'ARTE':30,'HOGAR': 156,'SALUD':172,'TEC':353},
+        'MPE':{'ALIMENTOS': 2,'ARTE':30,'HOGAR': 156,'SALUD':210,'TEC':315}
+    }
+    df_es_train, df_es_test = train_test_split(df_es,es_country_samples,random_seed=776436538)
+    
+    # Se extrae el conjunto de dev
+    es_country_samples = {
+        'MLA':{'ALIMENTOS': 10,'ARTE':30,'HOGAR': 200,'SALUD':200,'TEC':300},
+        'MLM':{'ALIMENTOS': 10,'ARTE':30,'HOGAR': 200,'SALUD':200,'TEC':300},
+        'MLU':{'ALIMENTOS': 10,'ARTE':30,'HOGAR': 200,'SALUD':200,'TEC':300},
+        'MCO':{'ALIMENTOS': 10,'ARTE':40,'HOGAR': 200,'SALUD':200,'TEC':300},
+        'MLC':{'ALIMENTOS': 20,'ARTE':60,'HOGAR': 200,'SALUD':200,'TEC':300},
+        'MLV':{'ALIMENTOS': 0,'ARTE':30,'HOGAR': 20,'SALUD':0,'TEC':250},
+        'MPE':{'ALIMENTOS': 0,'ARTE':0,'HOGAR': 1,'SALUD':0,'TEC':1}
+    }
+    df_es_train, df_es_dev = train_test_split(df_es_train,es_country_samples,random_seed=776436538)
+    
+    df_es_train.to_csv('./es/train.csv',index=False)
+    df_es_dev.to_csv('./es/validation.csv',index=False)
+    df_es_test.to_csv('./es/test.csv',index=False)
+    
+    ## PORTUGUÉS
+    # Se eliminan los que están en la lista to_be_removed_por
+    df_pt = df_pt.drop(set(to_be_removed_por)).reset_index(drop=True)
+    
+    # Se extrae el conjunto de test
+    pt_country_samples = {'MLB':{'ALIMENTOS': 23,'ARTE':210,'HOGAR': 1092,'SALUD':1432,'TEC':2243}}
+    df_pt_train, df_pt_test = train_test_split(df_pt,pt_country_samples,random_seed=776436538)
+    
+    # Se extrae el conjunto de dev
+    pt_country_samples = {'MLB':{'ALIMENTOS': 20,'ARTE':200,'HOGAR': 1032,'SALUD':1400,'TEC':1400}}
+    df_pt_train, df_pt_dev = train_test_split(df_pt,pt_country_samples,random_seed=776436538)
+    
+    df_pt_train.to_csv('./pt/train.csv',index=False)
+    df_pt_dev.to_csv('./pt/validation.csv',index=False)
+    df_pt_test.to_csv('./pt/test.csv',index=False)
+    
+    
+if __name__ == "__main__":
+    main()

code/get_data.sh

@@ -0,0 +1,11 @@
+#! /bin/bash
+
+mkdir products parts parts-2 peru_parts ven_parts
+python 01-get_products_list.py
+python 02-remove_duplicates_and_get_reviews.py --partsdir "parts" --countries "MLA" "MCO" "MPE" "MLU" "MLC" "MLM" "MLV" "MLB"
+python 02-remove_duplicates_and_get_reviews.py --partsdir "parts-2" --countries "MLA" "MCO" "MPE" "MLU" "MLC" "MLM" "MLV" "MLB"
+python 02-remove_duplicates_and_get_reviews.py --partsdir "peru_parts" --countries "MPE"
+python 02-remove_duplicates_and_get_reviews.py --partsdir "ven_parts" --countries "MLV"
+python 03-get_parts_in_csv.py
+python 04-process_csv.py
+python 06-train-dev-test-split.py

code/python-sdk/README.md

@@ -0,0 +1,115 @@
+# MercadoLibre's Python SDK
+
+This is the official Python SDK for MercadoLibre's Platform.
+
+## How do I install it?
+
+       clone repository
+       https://github.com/mercadolibre/python-sdk.git
+
+## How do I use it?
+
+The first thing to do is to instance a ```Meli``` class. You'll need to give a ```clientId``` and a ```clientSecret```. You can obtain both after creating your own application. For more information on this please read: [creating an application](http://developers.mercadolibre.com/application-manager/)
+
+### Including the Lib
+Include the lib meli in your project
+
+### Attention
+Don't forget to set the authentication URL of your country in file lib/config.ini
+
+```python
+import sys
+sys.path.append('../lib')
+from meli import Meli
+```
+Start the development!
+
+### Create an instance of Meli class
+Simple like this
+```python
+meli = Meli(client_id=1234, client_secret="a secret")
+```
+With this instance you can start working on MercadoLibre's APIs.
+
+There are some design considerations worth to mention.
+
+1. This SDK is just a thin layer on top of an http client to handle all the OAuth WebServer flow for you.
+
+2. There is JSON parsing. this SDK will include [json](http://docs.python.org/2/library/json.html) for internal usage.
+
+3. If you already have the access_token and the refresh_token you can pass in the constructor
+
+```python
+meli = Meli(client_id=1234, client_secret="a secret", access_token="Access_Token", refresh_token="Refresh_Token")
+```
+
+## How do I redirect users to authorize my application?
+
+This is a 2 step process.
+
+First get the link to redirect the user. This is very easy! Just:
+
+```python
+redirectUrl = meli.auth_url(redirect_URI="http://somecallbackurl")
+```
+
+This will give you the url to redirect the user. You need to specify a callback url which will be the one that the user will redirected after a successfull authrization process.
+
+Once the user is redirected to your callback url, you'll receive in the query string, a parameter named ```code```. You'll need this for the second part of the process.
+
+```python
+meli.authorize(code="the received code", redirect_URI="http://somecallbackurl")
+```
+
+This will get an ```access_token``` and a ```refresh_token``` (is case your application has the ```offline_access```) for your application and your user.
+
+At this stage your are ready to make call to the API on behalf of the user.
+
+#### Making GET calls
+
+```python
+params = {'access_token' : meli.access_token}
+response = meli.get(path="/users/me", params=params)
+```
+
+#### Making POST calls
+
+```python
+params = {'access_token' : meli.access_token}
+
+  #this body will be converted into json for you
+body = {'foo'  : 'bar', 'bar' : 'foo'}
+
+response = meli.post(path="/items", body=body, params=params)
+```
+
+#### Making PUT calls
+
+```python
+params = {'access_token' : meli.access_token}
+
+  #this body will be converted into json for you
+body = {'foo'  : 'bar', 'bar' : 'foo'}
+
+response = meli.put(path="/items/123", body=body, params=params)
+```
+
+#### Making DELETE calls
+```python
+params = {'access_token' : meli.access_token}
+response = meli.delete(path="/questions/123", params=params)
+```
+
+## Examples
+
+Don't forget to check out our examples codes in the folder [examples](https://github.com/mercadolibre/python-sdk/tree/master/examples)
+
+## Community
+
+You can contact us if you have questions using the standard communication channels described in the [developer's site](http://developers.mercadolibre.com/community/)
+
+## I want to contribute!
+
+That is great! Just fork the project in github. Create a topic branch, write some code, and add some tests for your new code.
+
+Thanks for helping!

code/python-sdk/examples/delete.py

@@ -0,0 +1,11 @@
+import sys
+sys.path.append('../lib')
+from meli import Meli
+
+def main():
+    meli = Meli(client_id=CLIENT_ID,client_secret=CLIENT_SECRET, access_token=ACCESS_TOKEN, refresh_token=REFRESH_TOKEN)
+    response = meli.delete("/questions/QUESTION_ID", {'access_token':meli.access_token})
+    print response.content
+
+if __name__ == "__main__":
+    main()

code/python-sdk/examples/example_login.py

@@ -0,0 +1,25 @@
+ #!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from bottle import Bottle, run, template, route, request
+import json
+
+import sys
+sys.path.append('../lib')
+from meli import Meli
+
+meli = Meli(client_id=CLIENT_ID,client_secret=CLIENT_SECRET)
+
+app = Bottle()
+
+@app.route('/authorize')
+def authorize():
+    if request.query.get('code'):
+        meli.authorize(request.query.get('code'), REDIRECT_URI)
+    return meli.access_token
+
+@app.route('/login')
+def login():
+    return "<a href='"+meli.auth_url(redirect_URI=REDIRECT_URI)+"'>Login</a>"
+
+run(app, host='localhost', port=4567, reloader=True)

code/python-sdk/examples/get.py

@@ -0,0 +1,12 @@
+import sys
+sys.path.append('../lib')
+from meli import Meli
+
+def main():
+    meli = Meli(client_id=CLIENT_ID,client_secret=CLIENT_SECRET, access_token=ACCESS_TOKEN, refresh_token=REFRESH_TOKEN)
+
+    response = meli.get("/items/ITEM_ID")
+    print response.content
+
+if __name__ == "__main__":
+    main()

code/python-sdk/examples/post.py

@@ -0,0 +1,12 @@
+import sys
+sys.path.append('../lib')
+from meli import Meli
+
+def main():
+    meli = Meli(client_id=CLIENT_ID,client_secret=CLIENT_SECRET, access_token=ACCESS_TOKEN,refresh_token=REFRESH_TOKEN)
+    body = {"title":"Item De Teste - Por Favor, Não Ofertar! --kc:off","category_id":"MLB257111","price":10,"currency_id":"BRL","available_quantity":1,"buying_mode":"buy_it_now","listing_type_id":"bronze","condition":"new","description":"Item de Teste. Mercado Livre's PHP SDK.","video_id":"Q6dsRpVyyWs","warranty":"12 month","pictures":[{"source":"https://upload.wikimedia.org/wikipedia/commons/thumb/6/64/IPhone_7_Plus_Jet_Black.svg/440px-IPhone_7_Plus_Jet_Black.svg.png"},{"source":"https://upload.wikimedia.org/wikipedia/commons/thumb/b/bc/IPhone7.jpg/440px-IPhone7.jpg"}],"attributes":[{"id":"EAN","value_name":"190198043566"},{"id":"COLOR","value_id":"52049"},{"id":"WEIGHT","value_name":"188g"},{"id":"SCREEN_SIZE","value_name":"4.7 polegadas"},{"id":"TOUCH_SCREEN","value_id":"242085"},{"id":"DIGITAL_CAMERA","value_id":"242085"},{"id":"GPS","value_id":"242085"},{"id":"MP3","value_id":"242085"},{"id":"OPERATING_SYSTEM","value_id":"296859"},{"id":"OPERATING_SYSTEM_VERSION","value_id":"iOS 10"},{"id":"DISPLAY_RESOLUTION","value_id":"1920 x 1080"},{"id":"BATTERY_CAPACITY","value_name":"3980 mAh"},{"id":"FRONT_CAMERA_RESOLUTION","value_name":"7 mpx"}]}
+    response = meli.post("/items", body, {'access_token':meli.access_token})
+    print response.content
+
+if __name__ == "__main__":
+    main()

code/python-sdk/examples/put.py

@@ -0,0 +1,12 @@
+import sys
+sys.path.append('../lib')
+from meli import Meli
+
+def main():
+    meli = Meli(client_id=CLIENT_ID,client_secret=CLIENT_SECRET, access_token=ACCESS_TOKEN, refresh_token=REFRESH_TOKEN)
+    body = {"title":"oculos edicao especial!", "price":1000 }
+    response = meli.put("/items/ITEM_+ID", body, {'access_token':meli.access_token})
+    print response.content
+
+if __name__ == "__main__":
+    main()

code/python-sdk/lib/config.ini

@@ -0,0 +1,24 @@
+[config]
+sdk_version = MELI-PYTHON-SDK-2.0.0
+api_root_url = https://api.mercadolibre.com
+auth_url = https://auth.mercadolibre.com.ar
+oauth_url = /oauth/token
+ssl_version = PROTOCOL_TLSv1
+
+
+# Set the auth_url according to your country
+
+#MLA https://auth.mercadolibre.com.ar // Argentina 
+#MLB https://auth.mercadolivre.com.br // Brasil
+#MCO https://auth.mercadolibre.com.co // Colombia
+#MCR https://auth.mercadolibre.com.cr // Costa Rica
+#MEC https://auth.mercadolibre.com.ec // Ecuador
+#MLC https://auth.mercadolibre.cl     // Chile
+#MLM https://auth.mercadolibre.com.mx // Mexico
+#MLU https://auth.mercadolibre.com.uy // Uruguay
+#MLV https://auth.mercadolibre.com.ve // Venezuela
+#MPA https://auth.mercadolibre.com.pa // Panama
+#MPE https://auth.mercadolibre.com.pe // Peru
+#MPT https://auth.mercadolibre.com.pt // Prtugal
+#MRD https://auth.mercadolibre.com.do // Dominicana
+#CBT https://global-selling.mercadolibre.com // CBT

code/python-sdk/lib/meli.py

@@ -0,0 +1,145 @@
+ #!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from configparser import ConfigParser # SelfConfigParsed has been renamed to ConfigParser in Python 3.2 (https://github.com/python/cpython/blob/cd90f6a3692e0f7ef0a13aae651e19a08d1f9b31/Lib/configparser.py#L1230)
+from ssl_helper import SSLAdapter
+from urllib.parse import urlencode # urlencode is now inside urlib.parse (https://stackoverflow.com/questions/44031471/importerror-cannot-import-name-urlencode-when-trying-to-install-flask-ext-sto)
+import json
+import os
+import re
+import requests
+import ssl
+
+class Meli(object):
+    def __init__(self, client_id, client_secret, access_token=None, refresh_token=None):
+        self.client_id = client_id
+        self.client_secret = client_secret
+        self.access_token = access_token
+        self.refresh_token = refresh_token
+        self.expires_in = None
+
+
+        parser = ConfigParser()
+        parser.read(os.path.dirname(os.path.abspath(__file__))+'/config.ini')
+
+        self._requests = requests.Session()
+        try:
+            self.SSL_VERSION = parser.get('config', 'ssl_version')
+            self._requests.mount('https://', SSLAdapter(ssl_version=getattr(ssl, self.SSL_VERSION)))
+        except:
+            self._requests = requests
+
+        self.API_ROOT_URL = parser.get('config', 'api_root_url')
+        self.SDK_VERSION = parser.get('config', 'sdk_version')
+        self.AUTH_URL = parser.get('config', 'auth_url')
+        self.OAUTH_URL = parser.get('config', 'oauth_url')
+
+    #AUTH METHODS
+    def auth_url(self,redirect_URI):
+        params = {'client_id':self.client_id,'response_type':'code','redirect_uri':redirect_URI}
+        url = self.AUTH_URL  + '/authorization' + '?' + urlencode(params)
+        return url
+
+    def authorize(self, code, redirect_URI):
+        params = { 'grant_type' : 'authorization_code', 'client_id' : self.client_id, 'client_secret' : self.client_secret, 'code' : code, 'redirect_uri' : redirect_URI}
+        headers = {'Accept': 'application/json', 'User-Agent':self.SDK_VERSION, 'Content-type':'application/json'}
+        uri = self.make_path(self.OAUTH_URL)
+
+        response = self._requests.post(uri, params=urlencode(params), headers=headers)
+
+        if response.ok:
+            response_info = response.json()
+            self.access_token = response_info['access_token']
+            if 'refresh_token' in response_info:
+                self.refresh_token = response_info['refresh_token']
+            else:
+                self.refresh_token = '' # offline_access not set up
+                self.expires_in = response_info['expires_in']
+
+            return self.access_token
+        else:
+            # response code isn't a 200; raise an exception
+            response.raise_for_status()
+
+    def get_refresh_token(self):
+        if self.refresh_token:
+            params = {'grant_type' : 'refresh_token', 'client_id' : self.client_id, 'client_secret' : self.client_secret, 'refresh_token' : self.refresh_token}
+            headers = {'Accept': 'application/json', 'User-Agent':self.SDK_VERSION, 'Content-type':'application/json'}
+            uri = self.make_path(self.OAUTH_URL)
+
+            response = self._requests.post(uri, params=urlencode(params), headers=headers, data=params)
+
+            if response.ok:
+                response_info = response.json()
+                self.access_token = response_info['access_token']
+                self.refresh_token = response_info['refresh_token']
+                self.expires_in = response_info['expires_in']
+                return self.access_token
+            else:
+                # response code isn't a 200; raise an exception
+                response.raise_for_status()
+        else:
+            raise Exception("Offline-Access is not allowed.")
+
+    # REQUEST METHODS
+    def get(self, path, params=None, extra_headers=None):
+        params = params or {}
+        headers = {'Accept': 'application/json', 'User-Agent':self.SDK_VERSION, 'Content-type':'application/json'}
+        if extra_headers:
+            headers.update(extra_headers)
+        uri = self.make_path(path)
+        response = self._requests.get(uri, params=urlencode(params), headers=headers)
+        return response
+
+    def post(self, path, body=None, params=None, extra_headers=None):
+        params = params or {}
+        headers = {'Accept': 'application/json', 'User-Agent':self.SDK_VERSION, 'Content-type':'application/json'}
+        if extra_headers:
+            headers.update(extra_headers)
+        uri = self.make_path(path)
+        if body:
+            body = json.dumps(body)
+
+        response = self._requests.post(uri, data=body, params=urlencode(params), headers=headers)
+        return response
+
+    def put(self, path, body=None, params=None, extra_headers=None):
+        params = params or {}
+        headers = {'Accept': 'application/json', 'User-Agent':self.SDK_VERSION, 'Content-type':'application/json'}
+        if extra_headers:
+            headers.update(extra_headers)
+        uri = self.make_path(path)
+        if body:
+            body = json.dumps(body)
+
+        response = self._requests.put(uri, data=body, params=urlencode(params), headers=headers)
+        return response
+
+    def delete(self, path, params=None, extra_headers=None):
+        params = params or {}
+        headers = {'Accept': 'application/json', 'User-Agent':self.SDK_VERSION, 'Content-type':'application/json'}
+        if extra_headers:
+            headers.update(extra_headers)
+        uri = self.make_path(path)
+        response = self._requests.delete(uri, params=params, headers=headers)
+        return response
+
+    def options(self, path, params=None, extra_headers=None):
+        params = params or {}
+        headers = {'Accept': 'application/json', 'User-Agent':self.SDK_VERSION, 'Content-type':'application/json'}
+        if extra_headers:
+            headers.update(extra_headers)
+        uri = self.make_path(path)
+        response = self._requests.options(uri, params=urlencode(params), headers=headers)
+        return response
+
+    def make_path(self, path, params=None):
+        params = params or {}
+        # Making Path and add a leading / if not exist
+        if not (re.search("^\/", path)):
+            path = "/" + path
+        path = self.API_ROOT_URL + path
+        if params:
+            path = path + "?" + urlencode(params)
+
+        return path

code/python-sdk/lib/ssl_helper.py

@@ -0,0 +1,19 @@
+ #!/usr/bin/env python
+# -*- coding: utf-8 -*-
+
+from requests.adapters import HTTPAdapter
+from requests.packages.urllib3.poolmanager import PoolManager
+
+
+class SSLAdapter(HTTPAdapter):
+    """"Transport adapter" that allows us to use SSLv3."""
+    def __init__(self, ssl_version=None, **kwargs):
+        self.ssl_version = ssl_version
+        super(SSLAdapter, self).__init__(**kwargs)
+
+    def init_poolmanager(self, connections, maxsize, block=False):
+        self.poolmanager = PoolManager(num_pools=connections,
+                                       maxsize=maxsize,
+                                       block=block,
+                                       ssl_version=self.ssl_version)
+

code/python-sdk/test/main.py

@@ -0,0 +1,144 @@
+import unittest
+import re
+import sys
+sys.path.append('../lib')
+from meli import Meli
+import requests
+import json
+
+class MeliTest(unittest.TestCase):
+
+    def setUp(self):
+        self.CLIENT_ID = "123"
+        self.CLIENT_SECRET = "a secret"
+        self.ACCESS_TOKEN = "a access_token"
+        self.REFRESH_TOKEN = "a refresh_token"
+        self.NEW_ACCESS_TOKEN = "a new access_token"
+        self.NEW_REFRESH_TOKEN = "a new refresh_token"
+        self.meli =  Meli(client_id=self.CLIENT_ID, client_secret=self.CLIENT_SECRET, access_token=self.ACCESS_TOKEN, refresh_token=self.REFRESH_TOKEN)
+
+
+    #constructor tests
+    def testClientId(self):
+        self.assertEqual(self.meli.client_id, self.CLIENT_ID)
+
+    def testClientSecret(self):
+        self.assertEqual(self.meli.client_secret, self.CLIENT_SECRET)
+
+    def testAccessToken(self):
+        self.assertEqual(self.meli.access_token, self.ACCESS_TOKEN)
+
+    def testRefreshToken(self):
+        self.assertEqual(self.meli.refresh_token, self.REFRESH_TOKEN)
+
+    #auth_url tests
+    def testAuthUrl(self):
+        callback = "http://test.com/callback"
+        self.assertTrue(re.search("^http", self.meli.auth_url(redirect_URI=callback)))
+        self.assertTrue(re.search("^https\:\/\/auth.mercadolibre.com\/authorization", self.meli.auth_url(redirect_URI=callback)))
+        self.assertTrue(re.search("redirect_uri", self.meli.auth_url(redirect_URI=callback)))
+        self.assertTrue(re.search(self.CLIENT_ID,self.meli.auth_url(redirect_URI=callback)))
+        self.assertTrue(re.search("response_type", self.meli.auth_url(redirect_URI=callback)))
+
+    #Mock requests
+    def mockGet(url, path=None, params={},headers={}, data=None, body=None):
+
+        response = requests.Response()
+
+        if re.search("/users/me", url):
+            if "access_token" in params:
+                response.status_code = 200
+            else:
+                response.status_code = 403
+        elif re.search("/authorization", url):
+            response.status_code = 200
+        else:
+            response.status_code = 200
+        return response
+
+    def mockPost(url, path=None, body=None, params={},headers={}, data=None):
+        response = requests.Response()
+
+        if re.search("/oauth/token", url):
+            if "grant_type" not in params or "client_id" not in params or "client_secret" not in params:
+                response.status_code = 403
+            else:
+                if re.search("grant_type=authorization_code", params):
+                    content = {'access_token' : 'a access_token', 'refresh_token' : 'a refresh_token'}
+                elif re.search("grant_type=refresh_token", params):
+                    content = {'access_token' : 'a new access_token', 'refresh_token' : 'a new refresh_token'}
+                response._content = json.dumps(content)
+                response.status_code = 200
+        else:
+            if "access_token" in params:
+                response.status_code = 200
+            else:
+                response.status_code = 403
+
+        return response
+
+    def mockPut(url, path=None, body=None, params={},headers={}, data=None):
+        response = requests.Response()
+        if "access_token" in params:
+            response.status_code = 200
+        else:
+            response.status_code = 403
+        return response
+
+    def mockDelete(url, path=None, params={},headers={}, data=None, body=None):
+        response = requests.Response()
+        if "access_token" in params:
+            response.status_code = 200
+        else:
+            response.status_code = 403
+        return response
+
+    requests.get    = mockGet
+    requests.post   = mockPost
+    requests.put    = mockPut
+    requests.delete = mockDelete
+
+    #requests tests
+    def testGet(self):
+        response = self.meli.get(path="/items/test1")
+        self.assertEqual(response.status_code, requests.codes.ok)
+
+    def testPost(self):
+        body = {"condition":"new", "warranty":"60 dias", "currency_id":"BRL", "accepts_mercadopago":True, "description":"Lindo Ray_Ban_Original_Wayfarer", "listing_type_id":"bronze", "title":"oculos Ray Ban Aviador  Que Troca As Lentes  Lancamento!", "available_quantity":64, "price":289, "subtitle":"Acompanha 3 Pares De Lentes!! Compra 100% Segura", "buying_mode":"buy_it_now", "category_id":"MLB5125", "pictures":[{"source":"http://upload.wikimedia.org/wikipedia/commons/f/fd/Ray_Ban_Original_Wayfarer.jpg"}, {"source":"http://en.wikipedia.org/wiki/File:Teashades.gif"}] }
+        response = self.meli.post(path="/items",body=body,params={'access_token' : self.meli.access_token})
+        self.assertEqual(response.status_code, requests.codes.ok)
+
+    def testPut(self):
+        body = {"title":"oculos edicao especial!", "price":1000 }
+        response = self.meli.put(path="/items/test1",body=body,params={'access_token' : self.meli.access_token})
+        self.assertEqual(response.status_code, requests.codes.ok)
+
+    def testDelete(self):
+        response = self.meli.delete(path="/questions/123",params={'access_token' : self.meli.access_token})
+        self.assertEqual(response.status_code, requests.codes.ok)
+
+    def testWithoutAccessToken(self):
+        response = self.meli.get(path="/users/me")
+        self.assertEqual(response.status_code, requests.codes.forbidden)
+
+    def testWithAccessToken(self):
+        response = self.meli.get(path="/users/me",params={'access_token' : self.meli.access_token})
+        self.assertEqual(response.status_code, requests.codes.ok)
+
+    #auth tests
+    def testAuthorize(self):
+        self.meli.access_token = None
+        self.meli.refresh_token = None
+        response = self.meli.authorize(code="a code from get param", redirect_URI="A redirect Uri")
+        self.assertEqual(self.meli.access_token, self.ACCESS_TOKEN)
+        self.assertEqual(self.meli.refresh_token, self.REFRESH_TOKEN)
+
+
+    def testDoRefreshToken(self):
+        response = self.meli.get_refresh_token()
+        self.assertEqual(self.meli.access_token, self.NEW_ACCESS_TOKEN)
+        self.assertEqual(self.meli.refresh_token, self.NEW_REFRESH_TOKEN)
+
+
+if __name__ == '__main__':
+    unittest.main()

code/python-sdk/teste.py

@@ -0,0 +1,3 @@
+import sys
+sys.path.append('../lib')
+from meli import Meli

es/test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:562aaf7ce44740cd08d1b835be5e71f47988ca786c0b4ef3e71a91c96d3c77c9
+size 6295693

es/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f0ed9275e72b7fa8198a5df1367c4921297c8b602e9f1e46b00c4a2b139f97f
+size 116501559

es/validation.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:393e77005d0f3b138934f2c8a050fe915aa88dbcddc884050762731a7f55d7ac
+size 5143755

pt/test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:604cb4edefd266a285d64714436b1b7367625c0f835599de36450671ea9e0900
+size 6415294

pt/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fab8152362803376e07df7e2659791c4fd030ae1bbddde02c6ed307456f1cf84
+size 62857885

pt/validation.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8edb7907b18fab9ff7404206b380b50a56388940eac9deff774f541361ad275c
+size 5135145