Al Johri AlJohri

## downloader.py
import os
import time
import arrow
import requests

API = "https://projects.fivethirtyeight.com/election-night-api/tests/default/projected_winners.json"

NUM_TIMES = 4
TIME_TO_SLEEP = int(60/NUM_TIMES)

## dailykos.py
import io
import us
import csv
import itertools
import requests
import pandas as pd
import numpy as np

def pairwise(iterable):
    "s -> (s0,s1), (s1,s2), (s2, s3), ..."

## nlp_or_ml_in_journo.md

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                AlJohri
                / nlp_or_ml_in_journo.md
            
            
              Created
              September 28, 2018 14:16
            
          
[topic modeling] Reuters: The Echo Chamber
[?] Reuters: The Child Exchange
[classify and count] AJC: Doctors and Sex Abuse
[search and count] LATimes: LAPD Crime Misclassification

LAPD Crime Data
LA Times: LAPD misclassified more than 25,000 serious crimes as minor, audit finds
LAPD Misclassified Nearly 1,200 Violent Crimes As Minor Offenses


[search and show] KQUED: 10 Emails That Detail PG&E’s Cozy Relationship With Regulators
[phonetics and count] [The GOP Can’t Depend On Minority Candidates To Win Minority Votes](https://fivethir


## qb.json
{
    "id": 3120,
    "name": "Kirk Cousins",
    "source": {
        "player_id": 403308
    },
    "pointer": null,
    "story_type": "PlayerCard",
    "project": "nfl",
    "data": {

## diff_russian_troll_tweets.py
import os
import csv
import tqdm
import hashlib
import itertools

def load(filepaths, line_limit=None):
    data = []
    for i, filepath in enumerate(filepaths):
        print(f'[{i+1} of {len(filepaths)}] loading {filepath}')

## latest.csv
office_id,office_type,state_code,latest
AK-G,G,AK,"Bill Walker, an independent, has been the governor of Alaska since 2014. This seat is up for election this year.

Walker is running for reelection.

Mark Begich is running for the Democratic nomination.

Ten Republicans are competing in the primary: Darin A. Colbry, Mike Dunleavy, Thomas Gordon, Scott Hawkins, Gerald Heikes, Merica Hlatcu, Michael D. Sheldon, Mead Treadwell, Mike Chenault and Jacob Kern.

The Alaska gubernatorial primaries take place Aug. 21.

## input.csv
Year,State,Agency Type,Agency name,q01,q02,q03,q04,Population2
2010,ALABAMA,Cities,Decatur,0,0,1,2,57153
2010,ALABAMA,Cities,Hoover,0,1,0,1,74687
2010,ALABAMA,Cities,Jasper,0,1,0,0,14138
2010,ALABAMA,Cities,Lanett,0,0,0,1,7026
2010,ALABAMA,Cities,Madison,0,0,1,0,41426
2010,ALABAMA,Cities,Oneonta,0,1,0,0,7317
2010,ALABAMA,Cities,Riverside,0,0,0,1,2146
2010,ALABAMA,Cities,Slocomb,1,0,0,0,2065
2010,ALABAMA,Cities,Talladega,0,1,0,0,16826

## .block
license: gpl-3.0
height: 940
border: no

## convert.py
# encoding: utf-8

import io
import torch
import json
import numpy as np

vocab = torch.load('data/vocab.pt')
np.savez_compressed(file='data/embedding.npz', embedding=vocab.embed)
with open('data/word2id.json', 'w') as f:

## word2id.json
{"unsupportable": 128318, "Villalon": 118160, "nunnery": 97674, "Plasticine": 144240, "woods": 5729, "clotted": 66309, "spiders": 9036, "Nampo": 181451, "Nampa": 92771, "woody": 51770, "trawling": 24655, "comically": 48133, "spidery": 91990, "canes": 41343, "beautyheaven": 143561, "Fulde": 115294, "Archuleta": 82341, "Kaboul": 24450, "Journey": 22519, "caned": 69502, "Eure": 104348, "Gravesend": 24420, "rumbustious": 177271, "Retreat": 30939, "Euro": 4009, "Valli": 54480, "naturopathic": 117376, "Valle": 13969, "grenadiers": 181508, "pigment": 21858, "Mizell": 146931, "Alya": 89149, "Morten": 34955, "bringing": 2181, "Valls": 18920, "wooded": 12650, "Al-Sidra": 179543, "grueling": 19942, "pipers": 99625, "wooden": 4192, "wholemeal": 35673, "Saco": 85256, "Miers": 96666, "wednesday": 513, "Sack": 49830, "viable": 11637, "LOVEFiLM": 141680, "chameleons": 48272, "Miera": 100719, "Malenchenko": 96215, "Safarova": 54544, "bbqs": 114055, "Pinkerton": 176544, "pizzle": 165680, "snuggles": 50483, "snuggler": 175379,
	import os
	import time
	import arrow
	import requests

	API = "https://projects.fivethirtyeight.com/election-night-api/tests/default/projected_winners.json"

	NUM_TIMES = 4
	TIME_TO_SLEEP = int(60/NUM_TIMES)
	import io
	import us
	import csv
	import itertools
	import requests
	import pandas as pd
	import numpy as np

	def pairwise(iterable):
	"s -> (s0,s1), (s1,s2), (s2, s3), ..."
	{
	"id": 3120,
	"name": "Kirk Cousins",
	"source": {
	"player_id": 403308
	},
	"pointer": null,
	"story_type": "PlayerCard",
	"project": "nfl",
	"data": {
	import os
	import csv
	import tqdm
	import hashlib
	import itertools

	def load(filepaths, line_limit=None):
	data = []
	for i, filepath in enumerate(filepaths):
	print(f'[{i+1} of {len(filepaths)}] loading {filepath}')
	office_id,office_type,state_code,latest
	AK-G,G,AK,"Bill Walker, an independent, has been the governor of Alaska since 2014. This seat is up for election this year.

	Walker is running for reelection.

	Mark Begich is running for the Democratic nomination.

	Ten Republicans are competing in the primary: Darin A. Colbry, Mike Dunleavy, Thomas Gordon, Scott Hawkins, Gerald Heikes, Merica Hlatcu, Michael D. Sheldon, Mead Treadwell, Mike Chenault and Jacob Kern.

	The Alaska gubernatorial primaries take place Aug. 21.
	Year,State,Agency Type,Agency name,q01,q02,q03,q04,Population2
	2010,ALABAMA,Cities,Decatur,0,0,1,2,57153
	2010,ALABAMA,Cities,Hoover,0,1,0,1,74687
	2010,ALABAMA,Cities,Jasper,0,1,0,0,14138
	2010,ALABAMA,Cities,Lanett,0,0,0,1,7026
	2010,ALABAMA,Cities,Madison,0,0,1,0,41426
	2010,ALABAMA,Cities,Oneonta,0,1,0,0,7317
	2010,ALABAMA,Cities,Riverside,0,0,0,1,2146
	2010,ALABAMA,Cities,Slocomb,1,0,0,0,2065
	2010,ALABAMA,Cities,Talladega,0,1,0,0,16826
	# encoding: utf-8

	import io
	import torch
	import json
	import numpy as np

	vocab = torch.load('data/vocab.pt')
	np.savez_compressed(file='data/embedding.npz', embedding=vocab.embed)
	with open('data/word2id.json', 'w') as f: