cjauvin

# Christian Jauvin <[email protected]>
# http://cjauvin.github.com
#
# Solutions for:
#   https://www.readyforzero.com/challenge
#   [email protected]

# Problem 1

s = "en3pG3+nz+A2acXKrsyDouhViP9EDQS4JQK6uJqM3rBjKEBKC3yc=AA1=LUQqRPHvQ4dopgkbb/axClP3smzVcaTkRsCqHSG5aKFQ2TbOae0t5r4nWrCVesGK1Z3yEq+dClrDwXXOiAMyW09WdCS+CaKcfu=6kv9dUFBcS4KsUIgwMiXimoBpJZSWlBzILVf4zVA=7GjRP8RXn6uKjbjAPkpFEs/mYJpeOpEnhQfPhjoscgjfL5/SQsU6+jaAf5pg9MQzZdQAEJt7Jm1541fEmnumpjmJMd/MTJ5vzBttBBA7b5rbjDX0nHdTWn8C5suYKfNyYzc6x8S6FIepoEBsMS2mKhx5BRH5jSBrYRem4iQgYARzGnCFot3jPhp3cHj7qjXBWfZZASz8YJqi2d+r393AmdGm1L9NfU2f=FJprbLwJpuE7uT7xAlQA3Ry8aHRNgNkffP29Iqb2DSoQ0PK+9LX0t37HIAhI5zvoP6b4J7yQZEQDgeNlnPQMvSjw9pLWAxQ1VUY+NMU5BLZ2Bxuma1cIsHxcx4PwHcg0u1HYPJAWM2WK=xhJP5aQSc6oNMQK4s2=6guQRVFll6rvWkXTebrdsws7m/Kpa29spUFl8XzFx0ondEMCF3byAyWj875wAI3Hn8ZY92ddTAKj0s+a4X7qSti2lA0GzePHjBjMCD5g9kZYLtB94kkbVZ6eCle/xtto4LHH8GElc5YoUi=mk3nmQ5iOL1zMWfDyRUMLq+HCXbQL9NTejNa/yTdL3sayJOlMW1T7/Jmaz1FMbfBRFzruHeMT41=Zu3nYZJ3nIP22qKrFzNkt/24RuQ+7IMVCI2

cjauvin

import psycopg2, psycopg2.extras
import little_pger as db
from flask import *

application = Flask('autocomplete-tribute')

@application.route('/autocomplete', methods=['GET'])
def autocomplete():
    conn = psycopg2.connect("dbname=autocomplete-tribute user=christian", 
                            connection_factory=psycopg2.extras.RealDictConnection)

cjauvin

import time
from math import sqrt
from joblib import Parallel, delayed

# Results obtained on my dual code Thinkpad laptop by using this modification:
# https://github.com/cjauvin/joblib/compare/parallel_job_batch

start = time.time()
Parallel(n_jobs=1)(delayed(sqrt)(i**2) for i in xrange(50000))
print time.time() - start # 4.2 secs

cjauvin

# Simulator for the simple Boltzmann machine of Coursera NN Lecture 11e
# Christian Jauvin - [email protected]

from collections import defaultdict
import numpy as np

# weights
w_v1_h1 = 2
w_h1_h2 = -1
w_h2_v2 = 1

cjauvin

import time
from sklearn.feature_extraction.text import CountVectorizer

# brown20.txt is the Brown corpus concatenated 20x to itself (~1M lines)
# Results obtained on a 24-core Linux machine
    
start = time.time()
vect = CountVectorizer()
vect.fit_transform(open('brown20x.txt'), n_jobs=1)
print time.time() - start # 307 seconds

cjauvin

import psycopg2, psycopg2.extras
import little_pger as db
from flask import *
 
application = Flask('autocomplete-tribute')
 
@application.route('/autocomplete', methods=['GET'])
def autocomplete():
    conn = psycopg2.connect("dbname=autocomplete-tribute user=christian", 
                            connection_factory=psycopg2.extras.RealDictConnection)

cjauvin

Ext.onReady(function() {
 
    Ext.define('Ubuntu', {
        extend: 'Ext.data.Model',
        fields: [{
            name: 'release',
            convert: function(v, rec) {
                return Ext.String.format('{0} {1} - {2}', 
                                         rec.raw.adjective, 
                                         rec.raw.animal, 

cjauvin

where = {}
 
where['adjective'] = 'Lucid' 
# where adjective = 'Lucid'
 
where['adjective'] = ('Warty', 'Dapper')
# where adjective in ('Warty', 'Dapper')
 
# For sequence values, the rules are: a tuple translates 
# to the 'in' operator (as above), a list to a PG array 

cjauvin

select * from ubuntu where 
    adjective || animal || version ilike E'%lynx%' and 
    adjective || animal || version ilike E'%04%' and 
    adjective || animal || version ilike E'%lucid%'

cjauvin

create table ubuntu (
    id serial primary key,
    adjective text,
    animal text,
    version text
);
 
insert into ubuntu (adjective, animal, version) 
            values ('Warty', 'Warthog', '4.10');
insert into ubuntu (adjective, animal, version)