Slater-Victoroff

"""
Collection of methods built to assist in data augmentation for extraction datasets
"""
from ast import literal_eval
import json
import random
from collections import defaultdict
from functools import partial
from typing import Iterable, Dict, Callable

Slater-Victoroff

"""
Demo script for Needle-in-a-haystack problems
"""
from functools import partial

import numpy as np
from indicoio.custom import vectorize
from scipy.spatial.distance import cdist
from scipy.stats import gmean

Slater-Victoroff

import os
import re

def fix_moodle_output(assn_dir, outdir):
    for (_, _, filenames) in os.walk(assn_dir):
        for filename in filenames:
            # Remove spaces so we can use this as a package name
            name = filename.split("_")[0].replace(" ", "").replace("-", "")
            if not os.path.exists("%s/%s" % (outdir, name)):
                os.makedirs("%s/%s" % (outdir, name))

Slater-Victoroff

[4.8502882289326867e-14, 8.820403952763587e-14, 1.0250590153123516e-13, 1.6954166246183968e-13, 1.908753789966549e-13, 2.004836996365545e-13, 3.5909784960471013e-13, 5.166693869401628e-13, 7.550572170701337e-13, 9.448695131203987e-13, 1.0156517689399418e-12, 1.1252632977953388e-12, 1.2755835948385307e-12, 1.718926371953829e-12, 5.561410412067812e-12, 5.612637550150243e-12, 8.453230126630751e-12, 8.725636132121303e-12, 1.613513316811227e-11, 1.9752204589154072e-11, 3.992415416880054e-11, 5.064793173473998e-11, 6.907684718699313e-11, 7.573645788325202e-11, 1.9359271945838294e-10, 2.5869603101103545e-10, 3.2604244542604767e-10, 6.107383625965949e-10, 3.054349088416707e-09, 3.948608349010387e-09, 4.447405216487758e-09, 1.0212339772102215e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.3008598108779382e-06, 1.300859810

Slater-Victoroff

def firstNonRepeated(s):
    return ([letter for i, letter in enumerate(s) if letter not in (s[:i] + s[i+1:])] + [""])[0]

def first_non_repeated1(s):
    for i, letter in enumerate(s):
	if letter not in (s[:i] + s[i+1:]):
	    return letter
    return ""

def first_non_repeated1(s):

Slater-Victoroff

import json
from collections import OrderedDict

import numpy as np
from scipy.spatial.distance import cdist
from indicoio.custom import vectorize
from nouns import NOUNS

FEATURES = json.load(open("features.json"), object_pairs_hook=OrderedDict)

Slater-Victoroff

{"http://images.neimanmarcus.com/product_assets/T/8/A/P/X/NMT8APX_mz.jpg": [0.0, 0.0, 0.7503045797348022, 0.0, 4.050353527069092, 0.0, 0.0, 3.970454454421997, 2.56343936920166, 0.0, 0.0, 0.0, 0.0, 0.8763924241065979, 2.9059646129608154, 1.7854936122894287, 0.0, 0.0, 0.0, 3.603835105895996, 0.0, 0.0, 0.0, 0.0, 2.9097509384155273, 0.6788361072540283, 0.0, 0.0, 0.0, 0.0, 0.9168252348899841, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.21072697639465332, 0.0, 2.585055112838745, 0.0, 0.18340826034545898, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 4.772193431854248, 0.0, 0.0, 0.0, 0.0, 1.1300804615020752, 0.0, 0.0, 0.0, 1.8126261234283447, 0.0, 0.0, 2.4351160526275635, 1.0090645551681519, 6.7620415687561035, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.6218217611312866, 0.0, 3.785642385482788, 0.0, 0.0, 0.0, 0.4094317555427551, 0.0, 1.046784520149231, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.5131956338882446, 0.0, 2.341181516647339, 3.7282121181488037, 2.7647757530212402, 0.0, 0.0, 0.2037421613931656, 0.0, 0.0, 2.629900932

Slater-Victoroff

id	title	description	google_product_category	product_type	price	sale_price	sale_price_effective_date	link	mobile_link	image_link	additional_image_link	brand	gtin	mpn	identifier_exists	condition	availability	availability_date	item_group_id	color	material	pattern	size	size_type	size_system	gender	age_group	tax	shipping	shipping_weight	shipping_label	multipack	is_bundle	adult	adwords_redirect	adwords_grouping	adwords_labels	custom_label_0	custom_label_1	custom_label_2	custom_label_3	custom_label_4	excluded_destination	expiration_date	promotion_id	display_ads_id	display_ads_link	display_ads_similar_id	display_ads_title	display_ads_value	shipping_length	shipping_width	shipping_height
sku152300450	Washable-Crepe Straight-Leg Pants, Petite, Size: PS (6/8), BLACK - Eileen Fisher	Eileen Fisher Washable-Crepe Straight-Leg Pants, Petite Details From Eileen Fisher, bi-stretch crepe pants with elegant day-to-night texture and remarkable fit memory, great for travel. 29" approx. inseam. Regular rise; yoked waist contours t

Slater-Victoroff

def get_voter_links(outfile="voter_info.txt"):
    start_urls = ("http://usavoters.directory/complete.php?id=%s" % i for i in xrange(128555, 214545))
    with open(outfile, 'a') as sink:
        for url in start_urls:
            document = etree.HTML(requests.get(url).content)
            link_selector = CSSSelector('tr>td>a')
            person_links = link_selector(document)[14:]
            sink.write('\n'.join(link.get('href') for link in person_links if link.get('href')))
            print url

Slater-Victoroff

import requests
import json
import urllib

URLS = {
	'css': 'http://cssminifier.com/raw',
	'js': 'http://javascript-minifier.com/raw'
}

def new_filepath(filepath):