ftfarias

# !pip install watermark
%load_ext watermark
%watermark -udtz -v -m -iv -rg

ftfarias

import boto3
from io import TextIOWrapper, BytesIO
from gzip import GzipFile
import csv
import logging
import collections
from tqdm import tqdm_notebook as tqdm

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

ftfarias

#! /usr/bin/env python
from argparse import ArgumentParser

import boto3
import pandas as pd

CSV_FILENAME = 'ec2_reserves.csv'

ec2 = boto3.client('ec2')

ftfarias

import os
import json
import urllib3

url_delete = 'http://172.31.x.x:9200/my_index/_delete_by_query'
query_delete = {
  "query": { "match": { "fk_xxx": day }}
}

ftfarias

watch -n 2 'ip address'

# Du para diretórios
https://dev.yorhel.nl/ncdu

Freeing disk space on your Linux server
- Get to the root of your machine by running cd /
- Run sudo du -h --max-depth=1.
- Note which directories are using a lot of disk space.

ftfarias

import boto3
import argparse
import elasticsearch
from io import TextIOWrapper
from gzip import GzipFile
import csv

fact_key = "/2018/05/15/mycsv_files"
BUCKET = 'csv_data'
print(f'Reading files at {fact_key}')

ftfarias

"""
{USAGE}
python thisfile.py param1
"""
import sys
import os
import pprint as pp
import argparse
import logging
import traceback

ftfarias

# Combinações e contrações do português
https://pt.wiktionary.org/wiki/Ap%C3%AAndice:Adv%C3%A9rbios_do_portugu%C3%AAs
https://pt.wiktionary.org/wiki/Ap%C3%AAndice:Gent%C3%ADlicos_e_top%C3%B3nimos_em_portugu%C3%AAs    
    
CONTRACOES = [
    # Com a preposição "com" + Artigos definidos
    (('com','um'),'cum'),
    # A preposição "de" + Artigos definidos
    (('de','o'),'do'),
    (('de','a'),'da'),

ftfarias

import requests
from io import BytesIO
from zipfile import ZipFile

# Download the dataset
dk = requests.get('http://www.ssfpack.com/files/DK-data.zip').content
f = BytesIO(dk)
zipped = ZipFile(f)
df = pd.read_table(
    BytesIO(zipped.read('internet.dat')),

ftfarias

import csv
import re

EMAIL_REGEXP = '^[_A-Za-z0-9-+]+(\.[_A-Za-z0-9-]+)*@[A-Za-z0-]+(\.[A-Za-z0-9]+)*(\.[A-Za-z]{2,3})$'

#from tqdm import tqdm_notebook as tqdm

import gensim
import collections
import nltk