kunigami

\documentclass{beamer}


\usecolortheme{wolverine}
\useinnertheme{circles}
\useoutertheme{infolines}
\usefonttheme{default}

\title[Short title]{
  Title

kunigami

\begin{table}
  \caption{Table caption}
  \centering
  \begin{tabular}{l | r r | r r}
          & \multicolumn{2}{>{c|}{A} & \multicolumn{2}{c}{B} \\
      Foo & 1 & 2                    & 1 & 2\\
      \hline
      1   & a & b                    & c & d \\
      \hline
  \end{tabular}

kunigami

\begin{table}
  \caption{Table caption}
  \centering
  \rowcolors{1}{loco3blue!17}{loco3blue!7}
  \begin{tabular}{l | r r | r r}
        & \multicolumn{2}{>{\columncolor{loco3blue!17}}c|}{A} & \multicolumn{2}{>{\columncolor{loco3blue!17}}c}{B} \\
    Foo & 1 & 2 & 1 & 2\\
    \hline
      1 & a & b & c & d\\
    \hline

kunigami

################################################################################
# svn_mime_filter.py
# Script to check if any of the file does not match the svn mime type
#
# TODOs:
#       1 - Report only once files that have the same extension
################################################################################

import os
import re

kunigami

#include <iostream>
#include <vector>

using namespace std;

typedef long long i64;

bool is_prime[1123123];
vector<i64> primes;     // lista de primos

kunigami

-- Original source: Real World Haskell, Chapter 27
-- URL: http://book.realworldhaskell.org/read/sockets-and-syslog.html

import Data.Bits
import Network.Socket
import Network.BSD
import Data.List
import SyslogTypes

data SyslogHandle = 

kunigami

{-# LANGUAGE FlexibleContexts, GeneralizedNewtypeDeriving,
             PatternGuards, ScopedTypeVariables #-}

import Control.Concurrent (forkIO)
import Control.Concurrent.STM
import Control.Exception (catch, finally)
import Control.Monad.Error
import Control.Monad.State
import Data.Char (isControl)
import Data.List (nub)

kunigami

import Control.Concurrent

transfer :: MVar Int -> MVar Int -> Int -> IO ()
transfer a b val = do
  valA <- takeMVar a
  valB <- takeMVar b
  putMVar a (valA - val)
  putMVar b (valB + val)

kunigami

Nesse post falarei sobre as desigualdades de cobertura erguidas (em uma horrorosa tradução livre de <em>lifted cover inequalities</em>). A minha ideia inicial era comentar sobre cortes automáticos que resolvedores como o XPRESS inserem no programa para apertar o modelo. Há dois tipos de desigualdades usadas como cortes: as de cobertura erguidas e aquelas resultantes do procedimento de Chvátal-Gomory.

Como esse segundo método faz uso direto do <em>tableau</em> do simplex, antes eu gostaria de escrever um post revisando o primal simplex e o dual simplex. Por isso, ater-me-ei às coberturas erguidas.

Essas desigualdades surgiram a partir do problema da mochila binária. Vamos relembrar o problema: dado um conjunto de n itens -- cada qual com um peso e valor --  e uma mochila de capacidade b, queremos escolher um subconjunto de itens cuja soma dos pesos respeite à capacidade da mochila e cuja soma dos valores seja máxima.

Vamos supor que os valores dos itens sejam dados por <img src="http://www.codecogs.com/eq.l

kunigami

(*
  Implementation of a queue with two lists. The first list represents the front
  of the queue and the second the end, reversed. Insertions are performed at the
  beginning of the second queue (remember inserting at the beginning of the list
  is more efficient). Removals are done at the front list.
  When the removal of an element will leave the front list empty, an operation
  is performed to reverse the second list and assign it to the first.
  The invariant we keep is that if 'left' is empty then 'right' must empty as
  well.
*)