Skip to content

T
Toy-MT-Introduction
Commit ff4244f5 by xiaotong
Options

tex files

parent c484bbe4
正在显示 包含 1132 行增加 和 0 行删除
.gitignore 0 → 100644
*.aux
*.log
*.nav
*.out
*.bak
*.snm
*.gz
*.toc
*.blg
Section01-Introduction/section01.test.tex 0 → 100644
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{beamer}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{shapes,arrows,shadows}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\setbeamertemplate{items}[ball]
\usefonttheme[onlymath]{serif} % fout of math
\definecolor{ugreen}{rgb}{0,0.5,0}
\definecolor{lgreen}{rgb}{0.9,1,0.8}
\definecolor{xtgreen1}{rgb}{0.824,0.898,0.8}
\definecolor{xtgreen}{rgb}{0.914,0.945,0.902}
\definecolor{lightgray}{gray}{0.85}
\setbeamercolor{uppercol}{fg=white,bg=ugreen}
\setbeamercolor{lowercol}{fg=black,bg=xtgreen}
%\definecolor{ublue}{rgb}{0,0.298,0.525}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\setbeamercolor{uppercolblue}{fg=white,bg=ublue}
\setbeamercolor{lowercolblue}{fg=black,bg=blue!10}
%\usetheme{default}
%\usetheme{Darmstadt}
%\usetheme{Madrid}
%\usetheme{Frankfurt}
%\usetheme{Dresden}
%\usetheme{Boadilla}
%\usecolortheme{dolphin}
\usefonttheme[onlylarge]{structurebold}
\begin{CJK}{GBK}{song}
\end{CJK}
\setbeamerfont*{frametitle}{size=\large,series=\bfseries}
\setbeamertemplate{navigation symbols}{\begin{CJK}{GBK}{hei} 第一章 机器翻译简介 \hspace*{2em} 肖桐\&朱靖波 \end{CJK} \hspace*{2em} \today \hspace*{2em} \insertframenumber{}/\inserttotalframenumber}
\setbeamertemplate{itemize items}[circle] % if you want a circle
\setbeamertemplate{itemize subitem}[triangle] % if you wnat a triangle
\setbeamertemplate{itemize subsubitem}[ball] % if you want a ball
\begin{document}
\begin{CJK}{GBK}{you}
\title{\Large{机器翻译简介}}
\subtitle{背景 + 概念 + 方法概述}
\author{\large{\textbf{肖桐\ \ 朱靖波}}}
\institute{
东北大学 自然语言处理实验室 \\
\blue{\underline{\url{xiaotong@mail.neu.edu.cn}}} \black{} \\
\vspace{0.2cm}
\hspace{0.1cm} \includegraphics[scale=0.1]{../Figures/logo.pdf}
}
\date{}
\maketitle
\setlength{\leftmargini}{1em}
\setlength{\leftmarginii}{1em}
\section{机器翻译}
\subsection{基本概念}
%%%------------------------------------------------------------------------------------------------------------
%%% 神经机器翻译
\begin{frame}{如何进行机器翻译？ - 基于神经网络的方法}
\alert{神经机器翻译}：把字符串表示成实数向量，翻译被看做序列到序列转化。这个过程由\alert{编码器-解码器}网络计算，不需要依赖人工先验或者语言学规则 \\
\vspace{0.5em}
\begin{tikzpicture}
\begin{scope}
\small{
\node [anchor=south west,minimum width=15em] (source) at (0,0) {\textbf{source}: 我\ \ \ \ 对\ \ \ \ 你\ \ \ \ 感到\ \ \ \ 满意};
\visible<3->{
\node [anchor=south west,minimum width=15em] (target) at ([yshift=12em]source.north west) {\textbf{target}: I\ \ am\ \ \ satisfied\ \ \ with\ \ \ you};
}
\visible<2->{
\node [anchor=center,minimum width=9.6em,minimum height=1.8em,draw,rounded corners=0.3em] (hidden) at ([yshift=6em]source.north) {};
\node [anchor=west,minimum width=1.5em,minimum size=1.5em,fill=ugreen!20] (cell01) at ([xshift=0.2em]hidden.west) {\footnotesize{.2}};
\node [anchor=west,minimum width=1.5em,minimum size=1.5em,fill=ugreen!10] (cell02) at (cell01.east) {\footnotesize{-1}};
\node [anchor=west,minimum width=1.5em,minimum size=1.5em,fill=ugreen!70] (cell03) at (cell02.east) {\footnotesize{6}};
\node [anchor=west,minimum width=1.5em,minimum size=1.5em,fill=ugreen!50] (cell04) at (cell03.east) {\footnotesize{5}};
\node [anchor=west,minimum width=1.5em,minimum size=1.5em,fill=ugreen!30] (cell05) at (cell04.east) {\footnotesize{.7}};
\node [anchor=west,minimum width=1.5em,minimum size=1.5em,fill=ugreen!10] (cell06) at (cell05.east) {\footnotesize{-2}};
}
\visible<2->{
\filldraw [fill=red!20,draw=white] (source.north west) -- (source.north east) -- ([xshift=-0.2em,yshift=-0.1em]hidden.south east) -- ([xshift=0.2em,yshift=-0.1em]hidden.south west);
}
\visible<3->{
\filldraw [fill=blue!20,draw=white] (target.south west) -- (target.south east) -- ([xshift=-0.2em,yshift=0.1em]hidden.north east) -- ([xshift=0.2em,yshift=0.1em]hidden.north west);
}
\visible<2->{
\draw [->,thick] (source.north west) -- ([xshift=0.2em,yshift=-0.1em]hidden.south west);
\draw [->,thick] (source.north east) -- ([xshift=-0.2em,yshift=-0.1em]hidden.south east);
}
\visible<3->{
\draw [->,thick] ([xshift=0.2em,yshift=0.1em]hidden.north west) -- (target.south west);
\draw [->,thick] ([xshift=-0.2em,yshift=0.1em]hidden.north east) -- (target.south east);
}
}
\visible<4->{
\node [anchor=south] (enclabel) at ([yshift=2em]source.north) {\large{\textbf{Encoder}}};
\node [anchor=north] (declabel) at ([yshift=-2em]target.south) {\large{\textbf{Decoder}}};
\node [anchor=east,align=left] (hiddenlabel) at (hidden.west) {\scriptsize{represe-}\\\scriptsize{ntation}};
}
\end{scope}
\begin{scope}[xshift=16.8em,yshift=6em]
\visible<5->{
\tikzstyle{numbernode} = [fill=blue!30,minimum height=0.8em,minimum width=0.8em,circle,draw,inner sep=1pt]
\tikzstyle{numbernode2} = [fill=ugreen!30,minimum height=0.8em,minimum width=0.8em,circle,draw,inner sep=1pt]
\node[anchor=west,numbernode] (neuron01) at (0,0) {};
\node[anchor=west,numbernode] (neuron02) at ([xshift=0.3em]neuron01.east) {};
\node[anchor=west,numbernode] (neuron03) at ([xshift=0.3em]neuron02.east) {};
\node[anchor=west,numbernode] (neuron04) at ([xshift=0.3em]neuron03.east) {};
\node[anchor=west,numbernode] (neuron11) at ([xshift=2em]neuron04.east) {};
\node[anchor=west,numbernode] (neuron12) at ([xshift=0.3em]neuron11.east) {};
\node[anchor=west,numbernode] (neuron13) at ([xshift=0.3em]neuron12.east) {};
\node[anchor=west,numbernode] (neuron14) at ([xshift=0.3em]neuron13.east) {};
\node[anchor=west,numbernode2] (neuron21) at ([xshift=1.5em,yshift=3em]neuron02.north) {};
\node[anchor=west,numbernode2] (neuron22) at ([xshift=0.3em]neuron21.east) {};
\node[anchor=west,numbernode2] (neuron23) at ([xshift=0.3em]neuron22.east) {};
\node[anchor=west,numbernode2] (neuron24) at ([xshift=0.3em]neuron23.east) {};
\node[anchor=west,numbernode2] (neuron25) at ([xshift=0.3em]neuron24.east) {};
\begin{pgfonlayer}{background}
\visible<5->{
\node[rectangle,draw,thin,inner sep=3pt,rounded corners=1pt] [fit = (neuron01) (neuron02) (neuron03) (neuron04)] (layer1) {};
\node[rectangle,draw,thin,inner sep=3pt,rounded corners=1pt] [fit = (neuron11) (neuron12) (neuron13) (neuron14)] (layer2) {};
\node[rectangle,draw,thin,inner sep=3pt,rounded corners=1pt] [fit = (neuron21) (neuron22) (neuron23) (neuron24) (neuron25)] (layer3) {};
}
\end{pgfonlayer}
% input and output labels
\draw[->,thick,align=center] ([yshift=-0.8em]layer1.south)--([yshift=-0.1em]layer1.south) node [pos=0,anchor=north,inner sep=3pt] (word01) {\footnotesize{input} \footnotesize{(word)}};
\draw[->,thick,align=center] ([yshift=-0.8em]layer2.south)--([yshift=-0.1em]layer2.south) node [pos=0,anchor=north,inner sep=3pt] (word02) {\footnotesize{input} \footnotesize{(word)}};
\draw[<-,thick,align=center] ([yshift=0.8em]layer3.north)--([yshift=0.1em]layer3.north) node [pos=0,anchor=south,inner sep=0] (outputnode) {\footnotesize{output} \footnotesize{(representation)}};
% layer and neuron labels
\node[anchor=west] (layerlabel3) at (layer3.east) {\footnotesize{layer}};
% connections
\draw[-] (neuron01)--(neuron21); \draw[-] (neuron02)--(neuron21); \draw[-] (neuron03)--(neuron21); \draw[-] (neuron04)--(neuron21);
\node[anchor=north] (connections) at ([yshift=-0.7em,xshift=0.5em]layer3.south) {\textbf{...}};
\node[anchor=west] (conntections2) at ([xshift=0.5em]connections.east) {\textbf{...}};
\begin{pgfonlayer}{background}
\visible<5->{
\node[draw,thin,minimum width=11em,align=left,rounded corners=1pt,fill=red!5,drop shadow] [fit = (outputnode) (word01) (word02) ] (netbox) {};
\node[rectangle,draw,thin,inner sep=3pt,rounded corners=1pt] [fit = (neuron01) (neuron02) (neuron03) (neuron04)] (layer1) {};
\node[rectangle,draw,thin,inner sep=3pt,rounded corners=1pt] [fit = (neuron11) (neuron12) (neuron13) (neuron14)] (layer2) {};
\node[rectangle,draw,thin,inner sep=3pt,rounded corners=1pt] [fit = (neuron21) (neuron22) (neuron23) (neuron24) (neuron25)] (layer3) {};
}
\end{pgfonlayer}
\draw [->,very thick,dotted] (enclabel.east) ..controls + (east:2em) and +(west:2em).. ([yshift=-1em]netbox.west);
\node [anchor=north] (netlabel) at (netbox.south) {\footnotesize{sample network of the encoder}};
}
\end{scope}
\end{tikzpicture}
\end{frame}
\end{CJK}
\end{document}
Section02-Words-Trees-Probs/pgfsubpic.sty 0 → 100644
\RequirePackage{pgf}
\input{pgfsubpic.tex}
\endinput
Section02-Words-Trees-Probs/pgfsubpic.tex 0 → 100644
% pgfsubpic.tex
% Version 1.1, 25 Dec 2009
% Copyright 2009 by David Chiang
% This program is free software; you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation; either version 2 of the License, or
% (at your option) any later version.
% This program is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% GNU General Public License for more details.
% You should have received a copy of the GNU General Public License along
% with this program; if not, write to the Free Software Foundation, Inc.,
% 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
% New in version 1.1:
% - the ability to save a subpicture in local variables
% - nodes in subpictures are tracked if the subpicture is placed with arbitrary transforms
% - new \pgffitsubpicture macro to transform a subpicture (preserving aspect) to fit in a desired box
% To do:
% - speed up interpictureshift hook by promoting subsubpictures
\newdimen\pgf@subpicminx
\newdimen\pgf@subpicminy
\newdimen\pgf@subpicmaxx
\newdimen\pgf@subpicmaxy
% Special virtual node for current subpicture's bounding box
\expandafter\def\csname pgf@sh@ns@current subpicture\endcsname{rectangle}
\expandafter\def\csname pgf@sh@np@current subpicture\endcsname{%
\def\southwest{\pgfqpoint{\pgf@subpicminx}{\pgf@subpicminy}}%
\def\northeast{\pgfqpoint{\pgf@subpicmaxx}{\pgf@subpicmaxy}}%
}
\expandafter\def\csname pgf@sh@nt@current subpicture\endcsname{{\pgf@pt@aa}{\pgf@pt@ab}{\pgf@pt@ba}{\pgf@pt@bb}{\the\pgf@pt@x}{\the\pgf@pt@y}} % the transformation at invocation time
\expandafter\def\csname pgf@sh@pi@current subpicture\endcsname{\pgfpictureid}
% Create a pgfpicture inside an hbox for delayed placement
\def\pgfsubpicture{%
\expandafter\global\expandafter\setbox\pgf@hbox=\hbox\bgroup
\pgfinterruptpicture
\pgfpicture
\relax % not sure why. otherwise a curly brace immediately after causes an error
}
\def\endpgfsubpicture{
\global\pgf@subpicminx=\pgf@picminx
\global\pgf@subpicminy=\pgf@picminy
\global\pgf@subpicmaxx=\pgf@picmaxx
\global\pgf@subpicmaxy=\pgf@picmaxy
\global\edef\subpictureid{\pgfpictureid}%
\pgfsetbaseline{\pgf@picminy}%
\endpgfpicture%
\endpgfinterruptpicture%
\egroup
}
% Allocate registers for saving a subpicture. #1 is text, not a control sequence.
\def\pgfnewsubpicture#1{%
\expandafter\newbox\csname pgf@subpic@hbox@#1\endcsname
\expandafter\newdimen\csname pgf@subpic@minx@#1\endcsname
\expandafter\newdimen\csname pgf@subpic@miny@#1\endcsname
\expandafter\newdimen\csname pgf@subpic@maxx@#1\endcsname
\expandafter\newdimen\csname pgf@subpic@maxy@#1\endcsname
}
% saved subpictures are local to the current group
\def\pgfsavesubpicture#1{%
\expandafter\setbox\csname pgf@subpic@hbox@#1\endcsname\copy\pgf@hbox
\csname pgf@subpic@minx@#1\endcsname\pgf@subpicminx
\csname pgf@subpic@miny@#1\endcsname\pgf@subpicminy
\csname pgf@subpic@maxx@#1\endcsname\pgf@subpicmaxx
\csname pgf@subpic@maxy@#1\endcsname\pgf@subpicmaxy
\expandafter\edef\csname pgf@subpic@id@#1\endcsname{\subpictureid}
}
\def\pgfrestoresubpicture#1{%
\edef\act{\global\noexpand\setbox\pgf@hbox\noexpand\copy\csname pgf@subpic@hbox@#1\endcsname}\act
\expandafter\global\expandafter\pgf@subpicminx\csname pgf@subpic@minx@#1\endcsname
\expandafter\global\expandafter\pgf@subpicminy\csname pgf@subpic@miny@#1\endcsname
\expandafter\global\expandafter\pgf@subpicmaxx\csname pgf@subpic@maxx@#1\endcsname
\expandafter\global\expandafter\pgf@subpicmaxy\csname pgf@subpic@maxy@#1\endcsname
\xdef\subpictureid{\csname pgf@subpic@id@#1\endcsname}
}
% Place a previously-created subpicture, lining up its origin with the current origin
\def\pgfplacesubpicture{
\pgfscope
% expand current bounding box to accommodate subpicture
\pgfpathrectanglecorners{\pgfpoint{\the\pgf@subpicminx}{\the\pgf@subpicminy}}{\pgfpoint{\the\pgf@subpicmaxx}{\the\pgf@subpicmaxy}}%
\pgfusepath{use as bounding box}%
%
% make the subpicture a node in the containing picture
\expandafter\gdef\csname pgf@sh@ns@\subpictureid\endcsname{rectangle}
\expandafter\xdef\csname pgf@sh@np@\subpictureid\endcsname{%
\noexpand\def\noexpand\southwest{\noexpand\pgfqpoint{\the\pgf@subpicminx}{\the\pgf@subpicminy}}%
\noexpand\def\noexpand\northeast{\noexpand\pgfqpoint{\the\pgf@subpicmaxx}{\the\pgf@subpicmaxy}}%
}%
\pgfgettransform\pgf@temp
\expandafter\xdef\csname pgf@sh@nt@\subpictureid\endcsname{\pgf@temp}%
\expandafter\xdef\csname pgf@sh@pi@\subpictureid\endcsname{\pgfpictureid}%
%
% align origin of subpicture with origin
\pgftransformshift{\pgfpoint{\the\pgf@subpicminx}{\the\pgf@subpicminy}}%
\pgfqboxsynced{\pgf@hbox}%
\endpgfscope
}
% Hook onto existing macro \pgf@shape@interpictureshift.
% This is called whenever we look up an anchor of a node.
% This hook recursively checks to see if the node's picture
% is a subpicture of another, and if so, adjusts its position accordingly.
% This is slow. It makes drawing trees O(n^2) in the depth of the tree.
% The alternative is to store, for each picture, a list of the nodes
% inside it. But this way doesn't require us to hijack \pgfnode, and
% is robust to re-placement of a subpicture. A compromise would be
% to store, for each picture, a list of the *subpictures* inside it.
\let\orig@pgf@shape@interpictureshift\pgf@shape@interpictureshift
\def\unwind@subpic#1{%
% is #1 the current picture?
\edef\subpicid{#1}%
\ifx\subpicid\pgfpictureid
% yes, we're done
\else
% does #1 have a parent picture?
\expandafter\ifx\csname pgf@sh@pi@#1\endcsname\relax
% no, the original node was not inside the current picture
\fallback
\else
% yes, apply transform and move up to parent picture
{%
\pgfsettransform{\csname pgf@sh@nt@#1\endcsname}%
\pgf@pos@transform{\pgf@x}{\pgf@y}%
\global\pgf@x=\pgf@x
\global\pgf@y=\pgf@y
}%
\unwind@subpic{\csname pgf@sh@pi@#1\endcsname}%
\fi
\fi
}
\def\pgf@shape@interpictureshift#1{%
\edef\fallback{\pgf@x=\the\pgf@x\pgf@y=\the\pgf@y\noexpand\orig@pgf@shape@interpictureshift{#1}}%
\unwind@subpic{\csname pgf@sh@pi@#1\endcsname}%
}
% \pgffitsubpicture{sw}{ne}
% Make the subpicture fit in the rectangle from sw to ne, preserving its aspect ratio.
\def\pgffitsubpicture#1#2{%
% current size
\pgfpointdiff{\pgfpointanchor{current subpicture}{south west}}{\pgfpointanchor{current subpicture}{north east}}%
\pgf@xa=\pgf@x \pgf@ya=\pgf@y
% desired size
\pgf@process{\pgfpointdiff{#1}{#2}}%
\pgf@xb=\pgf@x \pgf@yb=\pgf@y
\pgfmathparse{min(\pgf@xb/\pgf@xa,\pgf@yb/\pgf@ya)}%
\pgfmathparse{min(1,\pgfmathresult)}%
\pgftransformscale{\pgfmathresult}%
%
% current position
\pgfpointanchor{current subpicture}{center}%
\pgf@xa=\pgf@x \pgf@ya=\pgf@y
% desired position
% we scaled transform, so apply reverse scaling to argument
\pgfmathparse{1/\pgfmathresult}%
\pgf@process{\pgfpointscale{\pgfmathresult}{\pgfpointlineattime{0.5}{#1}{#2}}}%
\pgf@xb=\pgf@x \pgf@yb=\pgf@y
\pgfpointdiff{\pgfpoint{\pgf@xa}{\pgf@ya}}{\pgfpoint{\pgf@xb}{\pgf@yb}}%
\pgftransformshift{\pgfpoint{\pgf@x}{\pgf@y}}%
}
% utility functions -- not currently used
\def\pgfnodedelete#1{%
\expandafter\global\expandafter\let\csname pgf@sh@ns@#1\endcsname\relax
\expandafter\global\expandafter\let\csname pgf@sh@np@#1\endcsname\relax
\expandafter\global\expandafter\let\csname pgf@sh@nt@#1\endcsname\relax
\expandafter\global\expandafter\let\csname pgf@sh@pi@#1\endcsname\relax
\expandafter\global\expandafter\let\csname pgf@sh@ma@#1\endcsname\relax
}
\def\pgfnodeifexists#1#2#3{%
\expandafter\ifx\csname pgf@sh@ns@#1\endcsname\relax#3\else#2\fi
}
Section02-Words-Trees-Probs/pgftree.sty 0 → 100644
\RequirePackage{pgf}
\RequirePackage{pgffor}
\RequirePackage{pgfsubpic}
\input{pgftree.tex}
\endinput
Section02-Words-Trees-Probs/pgftree.tex 0 → 100644
% pgftree.tex
% Version 1.1, 25 Dec 2009
% Copyright 2009 by David Chiang
% This program is free software; you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation; either version 2 of the License, or
% (at your option) any later version.
% This program is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% GNU General Public License for more details.
% You should have received a copy of the GNU General Public License along
% with this program; if not, write to the Free Software Foundation, Inc.,
% 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
% New in version 1.1:
% - major restructuring to not do arbitrary nesting of subpicture environments
% - sideways trees
% To do:
% - trees with all leaves at same level
% - if \nodename does not exist as desired, wrap inside a rectangle node
% - don't use pgfsubpic internals
\newdimen\levelsep \levelsep=30pt
\newdimen\subtreesep \subtreesep=2pt
\def\leveldirection{down}
\def\siblingdirection{right}
% definitions of growing directions
\def\pgftree@levelshift{\csname pgftree@levelshift@\leveldirection\endcsname}
\def\pgftree@parentanchor{\csname pgftree@parentanchor@\leveldirection\endcsname}
\def\pgftree@childanchor{\csname pgftree@childanchor@\leveldirection\endcsname}
% these assume that the current subpicture is the child
\def\pgftree@presiblingshift{\csname pgftree@presiblingshift@\siblingdirection\endcsname}
\def\pgftree@postsiblingshift{\csname pgftree@postsiblingshift@\siblingdirection\endcsname}
\def\pgftree@levelshift@down{\pgfpoint{0}{-\levelsep}}
\def\pgftree@parentanchor@down{south}
\def\pgftree@childanchor@down{north}
\def\pgftree@levelshift@up{\pgfpoint{0}{\levelsep}}
\def\pgftree@parentanchor@up{north}
\def\pgftree@childanchor@up{south}
\def\pgftree@levelshift@right{\pgfpoint{\levelsep}{0}}
\def\pgftree@parentanchor@right{east}
\def\pgftree@childanchor@right{west}
\def\pgftree@levelshift@left{\pgfpoint{-\levelsep}{0}}
\def\pgftree@parentanchor@left{west}
\def\pgftree@childanchor@left{east}
\def\pgftree@presiblingshift@right{\pgf@process{\pgf@x-\pgf@subpicminx \advance\pgf@x\subtreesep \pgf@y 0pt}}
\def\pgftree@postsiblingshift@right{\pgf@process{\pgf@x\pgf@subpicmaxx \pgf@y 0pt}}
\def\pgftree@presiblingshift@left{\pgf@process{\pgf@x-\pgf@subpicmaxx \advance\pgf@x-\subtreesep \pgf@y 0pt}}
\def\pgftree@postsiblingshift@left{\pgf@process{\pgf@x\pgf@subpicminx \pgf@y 0pt}}
\def\pgftree@presiblingshift@up{\pgf@process{\pgf@x 0pt \pgf@y-\pgf@subpicminy \advance\pgf@y\subtreesep}}
\def\pgftree@postsiblingshift@up{\pgf@process{\pgf@x 0pt \pgf@y\pgf@subpicmaxy}}
\def\pgftree@presiblingshift@down{\pgf@process{\pgf@x 0pt \pgf@y-\pgf@subpicmaxy \advance\pgf@y-\subtreesep}}
\def\pgftree@postsiblingshift@down{\pgf@process{\pgf@x 0pt \pgf@y\pgf@subpicminy}}
% for convenience if you are using \pgftree directly
\def\drawnode#1{\pgfnode{rectangle}{base}{#1}{\nodename}{\pgfusepath{discard}}}
\def\drawedge{%
\pgfpathmoveto{\pgfpointanchor{\parentnodename}{\pgftree@parentanchor}}%
\pgfpathlineto{\pgfpointanchor{\nodename}{\pgftree@childanchor}}%
\pgfusepath{stroke}}
% local variables that we need to assign to inside a \pgfforeach
\newdimen\pgftree@childx
\newdimen\pgftree@savechildx
\newdimen\pgftree@childy
\newdimen\pgftree@savechildy
\newcount\pgftree@childi
\newcount\pgftree@savechildi
%%% \pgftree{subtree}
\def\pgftree#1{%
\def\nodename{r}%
#1
\pgfplacesubpicture
}
%%% \pgfsubtree{root}{subtrees}
% The first argument draws the root node using PGF/TikZ commands.
% The node must be named \nodename.
% The second argument is an even-length sequence of tokens.
% Token 2n-1 in the sequence draws the nth edge. It should draw an edge from \parentnodename to \nodename.
% Token 2n in the sequence draws the nth subtree. Its root must be named \nodename.
\pgfnewsubpicture{children}
\pgfnewsubpicture{child}
\newdimen\pgftree@lastchildx
\newdimen\pgftree@lastchildy
\def\pgfsubtree#1#2{%
\let\parentnodename\nodename
\pgftree@savechildx=\pgftree@childx
\pgftree@savechildy=\pgftree@childy
\pgftree@savechildi=\pgftree@childi
% Build subpicture with all the children and their subtrees. Save it in "children"
{\pgftree@childx=0pt%
\pgftree@childy=0pt%
\pgftree@childi=0%
\process@children #2}%
\ifnum\pgftree@childi>0%
\begin{pgfsubpicture}%
% Create root node
#1
% Place children
% move down \levelsep
{\pgftransformshift{\pgftree@levelshift}%
% center so that parent is midway between origins of first and last children
\pgftransformshift{\pgfpointscale{-0.5}{\pgfpoint{\the\pgftree@childx}{\the\pgftree@childy}}}%
\pgfplacesubpicture}
% Draw the edges
{\pgftree@childi=0%
\process@edges #2}%
\end{pgfsubpicture}%
\else % leaf node
\begin{pgfsubpicture}%
#1
\end{pgfsubpicture}%
\fi
\global\pgftree@childi=\pgftree@savechildi
\global\pgftree@childx=\pgftree@savechildx
\global\pgftree@childy=\pgftree@savechildy
}
\def\process@children{%
\pgfutil@ifnextchar\egroup
{% No more children, step back to origin of last child
\global\pgftree@childx\pgftree@lastchildx
\global\pgftree@childy\pgftree@lastchildy
\global\pgftree@childi\pgftree@childi
\pgfrestoresubpicture{children}% pass children back to caller
}%
{\@process@children}%
}
\def\@process@children#1#2{% #1 is the edge, #2 is the child
% Build the current child and save it as "child"
{\edef\nodename{\parentnodename-\the\pgftree@childi}%
#2}%
\pgfsavesubpicture{child}%
\begin{pgfsubpicture}
% Place previous children (if any)
\ifnum\pgftree@childi>0%
\pgfrestoresubpicture{children}%
\pgfplacesubpicture
\fi
% Place current child
\pgfrestoresubpicture{child}%
\ifnum\pgftree@childi>0% the first child is always at 0
\pgftree@presiblingshift \global\advance\pgftree@childx\pgf@x \global\advance\pgftree@childy\pgf@y
\fi
\global\pgftree@lastchildx\pgftree@childx
\global\pgftree@lastchildy\pgftree@childy
{\pgftransformshift{\pgfpoint{\the\pgftree@childx}{\the\pgftree@childy}}%
\pgfplacesubpicture}%
\pgftree@postsiblingshift \global\advance\pgftree@childx\pgf@x \global\advance\pgftree@childy\pgf@y
\end{pgfsubpicture}%
% Save the augmented row of children back into "children"
\pgfsavesubpicture{children}%
\advance\pgftree@childi by 1%
\process@children
}
\def\process@edges{%
\pgfutil@ifnextchar\egroup
{}%
{\@process@edges}%
}
\def\@process@edges#1#2{%
\edef\nodename{\parentnodename-\the\pgftree@childi}%
#1%
\advance\pgftree@childi by 1%
\process@edges
}
\def\subtreeof#1{%
% the subpicture which contains a node also contains exactly its subtree
\csname pgf@sh@pi@#1\endcsname
}
Section02-Words-Trees-Probs/section02.test.tex 0 → 100644
% !Mode:: "TeX:GBK"
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{beamer}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usepackage{array}
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\setbeamertemplate{items}[ball]
\usefonttheme[onlymath]{serif} % fout of math
\definecolor{ugreen}{rgb}{0,0.5,0}
\definecolor{lgreen}{rgb}{0.9,1,0.8}
\definecolor{xtgreen1}{rgb}{0.824,0.898,0.8}
\definecolor{xtgreen}{rgb}{0.914,0.945,0.902}
\definecolor{lightgray}{gray}{0.85}
\setbeamercolor{uppercol}{fg=white,bg=ugreen}
\setbeamercolor{lowercol}{fg=black,bg=xtgreen}
%\definecolor{ublue}{rgb}{0,0.298,0.525}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\setbeamercolor{uppercolblue}{fg=white,bg=ublue}
\setbeamercolor{lowercolblue}{fg=black,bg=blue!10}
%\usetheme{default}
%\usetheme{Darmstadt}
%\usetheme{Madrid}
%\usetheme{Frankfurt}
%\usetheme{Dresden}
%\usetheme{Boadilla}
%\usecolortheme{dolphin}
\usefonttheme[onlylarge]{structurebold}
\begin{CJK}{GBK}{song}
\end{CJK}
\setbeamerfont*{frametitle}{size=\large,series=\bfseries}
\setbeamertemplate{navigation symbols}{\begin{CJK}{GBK}{hei} 第二章 词法、语法及概率思想基础 \hspace*{2em} 肖桐\&朱靖波 \end{CJK} \hspace*{2em} \today \hspace*{2em} \insertframenumber{}/\inserttotalframenumber}
\setbeamertemplate{itemize items}[circle] % if you want a circle
\setbeamertemplate{itemize subitem}[triangle] % if you wnat a triangle
\setbeamertemplate{itemize subsubitem}[ball] % if you want a ball
\begin{document}
\begin{CJK}{GBK}{you}
\title{\Large{词法、语法及概率思想基础}}
\author{\large{\textbf{肖桐\ \ 朱靖波}}}
\institute{
\blue{\url{xiaotong@mail.neu.edu.cn}} \black{} \\
\blue{\url{zhujingbo@mail.neu.edu.cn}} \black{} \\
\vspace{1.0em}
东北大学 自然语言处理实验室 \\
\blue{\underline{\url{http://www.nlplab.com}}} \black{} \\
\vspace{0.2cm}
\hspace{0.1cm} \includegraphics[scale=0.1]{../Figures/logo.pdf}
}
\date{}
\maketitle
\setlength{\leftmargini}{1em}
\setlength{\leftmarginii}{1em}
\section{中文分词}
%%% 进一步扩展，基于n-gram LM的方法
\begin{frame}{进一步扩展：基于$n$-gram语言模型的方法}
\begin{itemize}
\item 这种方法也被称作基于1-gram(统计)语言模型的方法\\
所谓统计语言模型就是计算$\textrm{P}(w_1 w_2 ... w_m)$的概率
\end{itemize}
{\scriptsize
\tabcolsep 5pt
\begin{tabular}{l | l | l l l}
链式法则 & 1-gram & 2-gram & ... & $n$-gram \\
$\textrm{P}(w_1 w_2 ... w_n)=$ & $\textrm{P}(w_1 w_2 ... w_n)=$ & $\textrm{P}(w_1 w_2 ... w_n)=$ & ... & $\textrm{P}(w_1 w_2 ... w_n)=$ \\
$\textrm{P}(w_1) \times$ & $\textrm{P}(w_1) \times$ & $\textrm{P}(w_1) \times$ & ... & $\textrm{P}(w_1) \times$ \\
$\textrm{P}(w_2|w_1) \times$ & $\textrm{P}(w_2) \times$ & $\textrm{P}(w_2|w_1) \times$ & ... & $\textrm{P}(w_2|w_1) \times$ \\
$\textrm{P}(w_3|w_1 w_2) \times$ & $\textrm{P}(w_3) \times$ & $\textrm{P}(w_3|w_2) \times$ & ... & $\textrm{P}(w_3|w_1 w_2) \times$ \\
$\textrm{P}(w_4|w_1 w_2 w_3) \times$ & $\textrm{P}(w_4) \times$ & $\textrm{P}(w_4|w_3) \times$ & ... & $\textrm{P}(w_4|w_1 w_2 w_3) \times$ \\
... & ... & ... & ... & ... \\
$\textrm{P}(w_m|w_1...w_{m-1})$ & $\textrm{P}(w_m)$ & $\textrm{P}(w_m|w_{m-1})$ & ... & $\textrm{P}(w_m|w_{m-n+1} ... w_{m-1})$ \\
\end{tabular}
}
\begin{itemize}
\item<2-> \textbf{$n$-gram语言模型}的核心思想就是当前词($w_m$)出现的概率只依赖于前$n-1$个词($w_{m-n+1} ... w_{m-1}$)
\vspace{-2em}
\begin{eqnarray}
& & \textrm{P}_{2\textrm{-gram}}(\textrm{'确实/现在/数据/很/多'}) \nonumber \\
& = & \textrm{P}(\textrm{'确实'}) \times \textrm{P}(\textrm{'现在'$|$'确实'}) \times \textrm{P}(\textrm{'数据'$|$'现在'}) \times \nonumber \\
& & \textrm{P}(\textrm{'很'$|$'数据'}) \times \textrm{P}(\textrm{'多'$|$'很'}) \nonumber
\end{eqnarray}
\vspace{-1em}
\item<2-> \textbf{训练} - 相对频率估计：$\textrm{P}(\textrm{'现在'$|$'确实'}) = \frac{count(\textrm{'确实 现在'})}{count(\textrm{'确实'})}$
\end{itemize}
\end{frame}
\subsection{统计思想}
\subsection{统计建模实例1：基于全切分的概率分词}
\end{CJK}
\end{document}
Section02-Words-Trees-Probs/section02.tex 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
Section02-Words-Trees-Probs/tikz-qtree-compat.sty 0 → 100644
% tikz-qtree-compat.tex
% Version 1.1, 25 Dec 2009
\RequirePackage{tikz-qtree}
\let\orig@Tree\Tree
\def\Tree{\automath \qtreeprimes@internal \orig@Tree}
\def\qsetw#1{\message{\noexpand\qsetw is not supported, sorry!}}
\def\faketreewidth#1{\message{\noexpand\faketreewidth is not supported, sorry!}}
\def\qbalance{\message{\noexpand\qbalance is not supported, sorry!}}
\def\qframesubtree{\message{\noexpand\qframesubtree is not supported, sorry!}}
% Implement \qroof as a fancy leaf node
\newtoks\@qrooflabel
\def\qroof#1.{%
\begin{tikzpicture}[baseline]
{\pgftransformshift{\pgftree@levelshift}%
\node(qroofbot){#1};}
\@qroof
}
\def\@qroof{\@ifnextchar\egroup{%
% since we are putting the qroof inside a node, we already have an inner sep
% so for the purposes of defining our bounding box, we must eliminate inner sep
\begin{pgfsubpicture}
\node [inner sep=0pt] {\the\@qrooflabel};%
\end{pgfsubpicture}
\pgfpathrectanglecorners{\pgfpoint{\the\pgf@subpicminx}{\the\pgf@subpicminy}}{\pgfpoint{\the\pgf@subpicmaxx}{\the\pgf@subpicmaxy}}%
\pgfusepath{use as bounding box}%
\node (qrooftop) {\the\@qrooflabel};
\draw \roof@edge{qrooftop}{qroofbot};%
\end{tikzpicture}}
{\@@qroof}}
\def\@@qroof#1{\expandafter\@qrooflabel\expandafter{\the\@qrooflabel #1}\@qroof}
%%% This is lifted straight from qtree.sty
% and another odd convenience:
%
% Make _, ^ go into math mode automatically in the scope of \automath
%
{ % Temporarily change catcodes
\catcode`\_=\active
\catcode`\^=\active
\global\def\automath{%
\catcode`\_=\active
\catcode`\^=\active
\def_##1{\@ifnextchar^{\automath@two_{##1}}{\ensuremath{\sb{##1}}}}%
\def^##1{\@ifnextchar_{\automath@two^{##1}}{\ensuremath{\sp{##1}}}}}
}
\def\automath@two#1#2#3#4{\ensuremath{#1{#2}\relax #3{#4}}}
% Restore default catcodes for ^, _
\def\noautomath{\catcode`\_=8 \catcode`\^=7 }
% Let \0, \1, \2 produce ^0, $'$, $''$
% The \rlap results in better centering of the label (ignoring the
% superscript)
\def\qtreeprimes@internal{%
\def\0{\ifmmode ^0\else \rlap{$^0$}\fi}%
\def\1{\ifmmode '\else \rlap{$'$}\fi}%
\def\2{\ifmmode ''\else \rlap{$''$}\fi}%
}
% Same commands, but without the \rlap feature
\def\qtreeprimes{%
\def\0{\ensuremath{^0}}%
\def\1{\ensuremath{'}}%
\def\2{\ensuremath{''}}}
Section02-Words-Trees-Probs/tikz-qtree.sty 0 → 100644
\RequirePackage{tikz}
\RequirePackage{pgftree}
\input{tikz-qtree.tex}
\endinput
Section02-Words-Trees-Probs/tikz-qtree.tex 0 → 100644
% tikz-qtree.tex
% Version 1.1, 25 Dec 2009
% Copyright (C) 2002, 2009 by David Chiang
% This program is free software; you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published by
% the Free Software Foundation; either version 2 of the License, or
% (at your option) any later version.
% This program is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% GNU General Public License for more details.
% You should have received a copy of the GNU General Public License along
% with this program; if not, write to the Free Software Foundation, Inc.,
% 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
% New in version 1.1:
% - sideways trees
%% These macros facilitate building up an object recursively before
%% putting it into the input stream.
\newtoks\@result
\def\@call#1#2{\let\@cont=#2\bgroup\@result={}#1}
\def\@return{\global\@result=\@result\egroup\@cont}
\def\@ifequal#1#2{\edef\testa{#1}\edef\testb{#2}\ifx\testa\testb}
%% scan a tree: this just scans a subtree and then puts it onto the
%% input stream
\def\Tree{\@call\@subtree\@Tree}
\def\@Tree{%
%\showthe\@result %debug
\ifpgfpicture % is there a test for tikzpicture?
\pgftree{\the\@result}%
\else
\tikzpicture[baseline]\pgftree{\the\@result}\endtikzpicture
\fi
}
%% scan a subtree
\newtoks\child@list
\newtoks\root@node
\def\@subtree[{%
\root@node={}%
\pgfutil@ifnextchar.{\@call\@interior\@@subtree}{\@@@subtree}}
\def\@@subtree{%
\root@node=\@result
\@@@subtree
}
\def\@@@subtree{%
\@call\@children\@@@@subtree
}
\def\@@@@subtree]{%
\child@list=\@result
\pgfutil@ifnextchar.{\@call\@interior\@@@@@subtree}{\@@@@@@subtree}}
\def\@@@@@subtree{%
%%% Check for mismatch.
\@ifequal{\the\root@node}{\pgfutil@empty}%
\root@node=\@result
\fi
\@ifequal{\the\root@node}{\the\@result}\else
\message{Warning: mismatched labels, \the\root@node{} and \the\@result.}%
\fi
\@@@@@@subtree
}
\def\@@@@@@subtree{%
\@ifequal{\the\root@node}{\pgfutil@empty}
\edef\act{\noexpand\@result={\noexpand\pgfsubtree{\noexpand\path coordinate (\noexpand\nodename);}{\the\child@list}}}%
\else
\edef\act{\noexpand\@result={\noexpand\pgfsubtree{\the\root@node}{\the\child@list}}}%
\fi
\act
\@return}
%% scan a sequence of subtrees or leaves
\newif\ifscanned@edge
\def\@children{%
\scanned@edgefalse
\child@list{}%
\@@children}
\def\@@children{%
\pgfutil@ifnextchar]{\@result\child@list\@return}{% end of children
\pgfutil@ifnextchar\edge{% explicit edge
\ifscanned@edge
\message{Warning: more than one edge given for a single child}\let\next\@@children % ignore
\else
\scanned@edgetrue\let\next\@@@children
\fi
\@call\@edge\next}{%
% else, a real node is next
\ifscanned@edge\else % no explicit edge, supply default
\expandafter\child@list\expandafter{\the\child@list{\edge@adapter{}}}%
\fi
\scanned@edgefalse
\pgfutil@ifnextchar[{\@call\@subtree\@@@children}% subtree
{\@call\@leaf\@@@children} % leaf
}}}
\def\@@@children{%
% wrap child inside curly braces
\expandafter\@result\expandafter{\expandafter{\the\@result}}
\edef\act{\noexpand\child@list{\the\child@list \the\@result}}\act
\@@children
}
\def\@interior.{\@result{\node[alias=\nodename][every tree node,every internal node]}\@label}
\def\@leaf{\@call\@label\@@leaf}
\def\@@leaf{\edef\act{\noexpand\@result{\noexpand\pgfsubtree{\noexpand\node[alias=\noexpand\nodename][every tree node,every leaf node]\the\@result}{}}}\act\@return}
\def\@edge\edge #1;{%
\@result{\edge@adapter{#1}}%
\@return}
\def\edge@adapter#1{%
\let\tikzparentnode\parentnodename
\let\tikzchildnode\nodename
\path edge from parent #1;%
}
% a label is either text or PGF/TikZ code starting with \node
\def\@label{\pgfutil@ifnextchar\node{\@litlabel}{\@@label}}
\def\@@label#1 {%
\expandafter\@result\expandafter{\the\@result{#1};}%
\@return}
% try to copy \node command into \@result without stripping braces
\def\@litlabel\node{\@@litlabel}
\def\@@litlabel{\pgfutil@ifnextchar\bgroup{\@@@litlabel}{\@@@@litlabel}}
\def\@@@litlabel#1{\expandafter\@result\expandafter{\the\@result {#1}}\@@litlabel}
\def\@@@@litlabel#1;{\expandafter\@result\expandafter{\the\@result #1;}\@return}
% predefined edges
\def\tree@edge#1#2{(#1.\pgftree@parentanchor) -- (#2.\pgftree@childanchor)}
\def\roof@edge#1#2{\csname roof@edge@\leveldirection\endcsname{#1}{#2}}
\def\roof@edge@down#1#2{(#1.south) -- (#2.north west) -- (#2.north east) -- cycle}
\def\roof@edge@up#1#2{(#1.north) -- (#2.south west) -- (#2.south east) -- cycle}
\def\roof@edge@left#1#2{(#1.west) -- (#2.north east) -- (#2.south east) -- cycle}
\def\roof@edge@right#1#2{(#1.east) -- (#2.north west) -- (#2.south west) -- cycle}
%%% Options
\tikzoption{level distance}{\pgfmathsetlength\levelsep{#1}}
\tikzoption{sibling distance}{\pgfmathsetlength\subtreesep{#1}} % different semantics
% I don't really like this scheme
\tikzoption{grow}{\csname grow@#1\endcsname}
\tikzoption{grow'}{\csname growprime@#1\endcsname}
\def\grow@up{\def\leveldirection{up}\def\siblingdirection{left}}
\def\grow@down{\def\leveldirection{down}\def\siblingdirection{right}}
\def\growprime@up{\def\leveldirection{up}\def\siblingdirection{right}}
\def\growprime@down{\def\leveldirection{down}\def\siblingdirection{left}}
\def\grow@left{\def\leveldirection{left}\def\siblingdirection{down}}
\def\grow@right{\def\leveldirection{right}\def\siblingdirection{up}}
\def\growprime@left{\def\leveldirection{left}\def\siblingdirection{up}}
\def\growprime@right{\def\leveldirection{right}\def\siblingdirection{down}}
% defaults appropriate for linguistic trees
\tikzset{every tree node/.style={anchor=base}}
\tikzset{every leaf node/.style={}}
\tikzset{every internal node/.style={}}
\def\tikz@edge@to@parent@path{\tree@edge{\tikzparentnode}{\tikzchildnode}}
% predefined roof style
\tikzset{roof/.style={edge from parent path=\roof@edge{\tikzparentnode}{\tikzchildnode}}}
Section03-Word-Based-Models/section03.test.tex 0 → 100644
% !Mode:: "TeX:GBK"
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{beamer}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\setbeamertemplate{items}[ball]
\usefonttheme[onlymath]{serif} % fout of math
\definecolor{ugreen}{rgb}{0,0.5,0}
\definecolor{lgreen}{rgb}{0.9,1,0.8}
\definecolor{xtgreen1}{rgb}{0.824,0.898,0.8}
\definecolor{xtgreen}{rgb}{0.914,0.945,0.902}
\definecolor{lightgray}{gray}{0.85}
\setbeamercolor{uppercol}{fg=white,bg=ugreen}
\setbeamercolor{lowercol}{fg=black,bg=xtgreen}
%\definecolor{ublue}{rgb}{0,0.298,0.525}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\setbeamercolor{uppercolblue}{fg=white,bg=ublue}
\setbeamercolor{lowercolblue}{fg=black,bg=blue!10}
%\usetheme{default}
%\usetheme{Darmstadt}
%\usetheme{Madrid}
%\usetheme{Frankfurt}
%\usetheme{Dresden}
%\usetheme{Boadilla}
%\usecolortheme{dolphin}
\usefonttheme[onlylarge]{structurebold}
\begin{CJK}{GBK}{song}
\end{CJK}
\setbeamerfont*{frametitle}{size=\large,series=\bfseries}
\setbeamertemplate{navigation symbols}{\begin{CJK}{GBK}{hei} 第三章 基于单词的翻译模型 \hspace*{2em} 肖桐\&朱靖波 \end{CJK} \hspace*{2em} \today \hspace*{2em} \insertframenumber{}/\inserttotalframenumber}
\setbeamertemplate{itemize items}[circle] % if you want a circle
\setbeamertemplate{itemize subitem}[triangle] % if you want a triangle
\setbeamertemplate{itemize subsubitem}[ball] % if you want a ball
\begin{document}
\begin{CJK}{GBK}{you}
\title{\Large{从"词"开始：基于单词的翻译模型}}
\author{\large{\textbf{肖桐\ \ 朱靖波}}}
\institute{
\blue{\url{xiaotong@mail.neu.edu.cn}} \black{} \\
\blue{\url{zhujingbo@mail.neu.edu.cn}} \black{} \\
\vspace{1.0em}
东北大学 自然语言处理实验室 \\
\blue{\underline{\url{http://www.nlplab.com}}} \black{} \\
\vspace{0.2cm}
\hspace{0.1cm} \includegraphics[scale=0.1]{../Figures/logo.pdf}
}
\date{}
\maketitle
\setlength{\leftmargini}{1em}
\setlength{\leftmarginii}{1em}
\section{IBM模型}
\subsection{训练}
%%% outline
\begin{frame}{Landscape}
\vspace{-1em}
\begin{center}
\begin{tikzpicture}
\node [] (history) at (0,0) {\includegraphics[scale=0.29]{./Figures/mt-history.png}};
\visible<2->{
\draw[-,thick] ([xshift=-1.1in]history.south) coordinate (top01) -- ([yshift=-2em]top01);
\draw[-,thick] ([xshift=1in]history.south) coordinate (top02) -- ([yshift=-2em]top02);
\draw[-,thick] ([xshift=-0.3in]history.south east) coordinate (top03) -- ([yshift=-2em]top03);
\draw[<->,thick] ([yshift=-1em]top01) -- ([yshift=-1em]top02) node [pos=0.5,below,align=left,yshift=-0.5em] {\footnotesize{\textbf{Sections 3-4:}}\\\footnotesize{统计机器翻译}};
\draw[<->,thick] ([yshift=-1em]top02) -- ([yshift=-1em]top03) node [pos=0.5,below,align=left,yshift=-0.5em] {\footnotesize{\textbf{Sections 5-6:}}\\\footnotesize{神经机器翻译}};
}
\visible<3->{
\draw[<-,thick] ([xshift=-0.9in,yshift=-0.2in]history.north) coordinate (label01) -- ([yshift=1.5em,xshift=1.5em]label01) coordinate (label02);
\node[anchor=south west,align=left] (label) at ([xshift=-5em]label02) {\textbf{\alert{本章内容：}}\\统计机器翻译开山之作 - IBM模型\\\textit{The Mathematics of Statistical Machine Translation:}\\\textit{Parameter Estimation} by Peter E. Brown et al., 1993.};
}
\end{tikzpicture}
\end{center}
\end{frame}
\end{CJK}
\end{document}
Section03-Word-Based-Models/section03.tex 0 → 100644
This source diff could not be displayed because it is too large. You can view the blob instead.
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论