new update

Section05-Neural-Networks-and-Language-Modeling/section05-test.tex

@@ -20,6 +20,9 @@
 \usepackage{subfigure}
 \usepackage{tikz-3dplot}
 
+\usepackage{tcolorbox}
+\tcbuselibrary{skins}
+
 \usetikzlibrary{matrix}
 \usetikzlibrary{arrows,decorations.pathreplacing}
 \usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
@@ -111,70 +114,34 @@
 \newcounter{mycount3}
 \newcounter{mycount4}
 
-
 %%%------------------------------------------------------------------------------------------------------------
-%%% 张量的单元操作
-\begin{frame}{张量的单元操作}
+%%% 定义XTensor
+\begin{frame}{定义XTensor}
 \begin{itemize}
-\item 神经网络$\textbf{y}=f(\textbf{x}\cdot \textbf{w} + \textbf{b})$也包括一些张量的单元操作（element-wise opertation）
+\item NiuTensor张量由类XTensor定义
     \begin{itemize}
-	\item 加法：$\textbf{s}+\textbf{b}$，其中$\textbf{s} = \textbf{x}\cdot \textbf{w}$
-	\item 激活函数：$f(\cdot)$
-	\end{itemize}
-\item<2-> \textbf{单元加}就是对张量中的每个位置都进行加法
-	\begin{itemize}
-	\item 一般要求两个张量的形状是一样的
-	\item<3-> 不过，这里可以使用加法的\textbf{广播}，重复利用一个张量进行加法，并不要求两个张量形状相同
+    \item \textbf{必须指定}：张量的阶和各个方向维度的大小，关于维度的约定和传统多维数组一样
+    \item \textbf{可以指定}：张量数据类型、稠密程度等等
     \end{itemize}
 \end{itemize}
 
-\vspace{-1.5em}
-\begin{center}
-\begin{tikzpicture}
-\visible<3->{
-\begin{scope}
-\setcounter{mycount1}{1}
-\draw[step=0.5cm,color=orange,thick] (-1,-0.5) grid (1,0.5);
-\foreach \y in {+0.25,-0.25}
-  \foreach \x in {-0.75,-0.25,0.25,0.75}{
-    \node [fill=orange!20,inner sep=0pt,minimum height=0.49cm,minimum width=0.49cm] at (\x,\y) {\number\value{mycount1}};
-    \addtocounter{mycount1}{1};
-  }
-\node [anchor=south] (varlabel) at (0,0.6) {$\textbf{s}$};
-\end{scope}
-\begin{scope}[xshift=1.5in]
-\setcounter{mycount1}{1}
-\draw[step=0.5cm,color=ugreen,thick] (-1,-0) grid (1,0.5);
-\foreach \y in {+0.25}
-  \foreach \x in {-0.75,-0.25,0.25,0.75}{
-    \node [fill=green!20,inner sep=0pt,minimum height=0.49cm,minimum width=0.49cm] at (\x,\y) {1};
-    \addtocounter{mycount1}{1};
-  }
-\node [anchor=center] (plabel) at (-4.5em,0) {\huge{\textbf{$+$}}};
-\node [anchor=south] (varlabel) at (0,0.6) {$\textbf{b}$};
-\end{scope}
-\begin{scope}[xshift=3in]
-\setcounter{mycount1}{2}
-\draw[step=0.5cm,color=orange,thick] (-1,-0.5) grid (1,0.5);
-\foreach \y in {+0.25,-0.25}
-  \foreach \x in {-0.75,-0.25,0.25,0.75}{
-    \node [fill=orange!20,inner sep=0pt,minimum height=0.49cm,minimum width=0.49cm] at (\x,\y) {\number\value{mycount1}};
-    \addtocounter{mycount1}{1};
-  }
-\node [anchor=center] (plabel) at (-4.5em,0) {\huge{\textbf{$=$}}};
-\node [anchor=south] (varlabel) at (0,0.6) {$\textbf{s+b}$};
-\end{scope}
-}
+\begin{tcolorbox}[enhanced,frame engine=empty,boxrule=0.1mm,size=title,colback=blue!10!white]
+\begin{flushleft}
+{\scriptsize
+\begin{tabbing}
 
-\end{tikzpicture}
-\end{center}
+\texttt{XTensor tensor;} \hspace{12em} \= // 声明张量tensor \\
+\texttt{int sizes[3] = \{2,3,4\};} \> // 张量的形状为2*3*4 \\
 
-\begin{itemize}
-\item<4-> 类似的，我们可以对所有位置做加法、乘法等等一系列算数操作，也包括激活函数
-\end{itemize}
+\texttt{InitTensor(\&tensor, 3, sizes, X\_FLOAT);} \> // 定义形状为sizes的三阶张量
+\end{tabbing}
+}
+\end{flushleft}
+\end{tcolorbox}
 
 \end{frame}
 
+
 %%%------------------------------------------------------------------------------------------------------------
 \subsection{参数学习 - 反向传播}
 

Section05-Neural-Networks-and-Language-Modeling/section05.tex

@@ -20,6 +20,9 @@
 \usepackage{subfigure}
 \usepackage{tikz-3dplot}
 
+\usepackage{tcolorbox}
+\tcbuselibrary{skins}
+
 \usetikzlibrary{matrix}
 \usetikzlibrary{arrows,decorations.pathreplacing}
 \usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
@@ -61,6 +64,10 @@
 %\usetheme{Boadilla}
 %\usecolortheme{dolphin}
 
+\newcounter{mycount1}
+\newcounter{mycount2}
+\newcounter{mycount3}
+\newcounter{mycount4}
 
 \usefonttheme[onlylarge]{structurebold}
 
@@ -681,27 +688,70 @@ GPT-2 (Transformer) & Radford et al. & 2019 & \alert{35.7}
 \end{frame}
 
 %%%------------------------------------------------------------------------------------------------------------
+\subsection{人工神经元}
+
+%%%------------------------------------------------------------------------------------------------------------
 %%% outline
-\begin{frame}{入门神经网络(深度学习)的三个基本问题}
+\begin{frame}{入门人工神经网络(深度学习)的三个基本问题}
 
-\begin{center}
-\begin{tikzpicture}
+\vspace{1em}
 
+\begin{tcolorbox}[enhanced,size=normal,left=2mm,right=1mm,colback=red!5!white,colframe=red!75!black,drop fuzzy shadow]
 {\Large
-\node [anchor=west,draw,ublue,very thick,rounded corners=4pt,text width=18em,align=left,fill=white,drop shadow={shadow xshift=0.2em,shadow yshift=-0.2em}] (p1) at (0,0) {\black{\textbf{1. 人工神经网络的数学描述是什么，}}\\\black{\textbf{\hspace{0.9em} 如何实现这种数学模型？}}};
+\textbf{1. 人工神经网络的基本单元是什么,}
 
-\node [anchor=north west,draw,ublue,very thick,rounded corners=4pt,text width=18em,align=left,fill=white,drop shadow={shadow xshift=0.2em,shadow yshift=-0.2em}] (p21) at ([yshift=-1em]p1.south west) {\black{\textbf{2. 如何将简单的网络单元组合成更}}\\\black{\textbf{\hspace{0.9em} 强大的模型？}}};
+\vspace{0.4em}
+\textbf{\hspace{0.9em} 如何组合出更强大的模型？}
+}
+\end{tcolorbox}
+
+\vspace{0.5em}
 
-\node [anchor=north west,draw,ublue,very thick,rounded corners=4pt,text width=18em,align=left,fill=white,drop shadow={shadow xshift=0.2em,shadow yshift=-0.2em}] (p22) at ([yshift=-1em]p21.south west) {\black{\textbf{3. 如何对模型中的参数进行学习，}}\\\black{\textbf{\hspace{0.9em} 之后使用学习到的模型进行推断？}}};
+\begin{tcolorbox}[enhanced,size=normal,left=2mm,right=1mm,colback=red!5!white,colframe=red!75!black,drop fuzzy shadow]
+{\Large
+\textbf{2. 人工神经网络的数学描述是什么,}
+
+\vspace{0.4em}
+\textbf{\hspace{0.9em} 如何编程实现这种数学模型？}
 }
+\end{tcolorbox}
 
-\end{tikzpicture}
-\end{center}
+\vspace{0.5em}
+
+\begin{tcolorbox}[enhanced,size=normal,left=2mm,right=1mm,colback=red!5!white,colframe=red!75!black,drop fuzzy shadow]
+{\Large
+\textbf{3. 如何对模型中的参数进行学习,}
+
+\vspace{0.4em}
+\textbf{\hspace{0.9em} 之后使用学习到的模型进行推断？}
+}
+\end{tcolorbox}
 
 \end{frame}
 
 %%%------------------------------------------------------------------------------------------------------------
-\subsection{人工神经元}
+%%% outline: problem 1
+\begin{frame}{首先}
+
+\vspace{6em}
+
+\begin{tcolorbox}[enhanced,size=normal,left=2mm,right=1mm,colback=red!5!white,colframe=red!75!black,drop fuzzy shadow]
+{\Large
+\textbf{人工神经网络的基本单元是什么,}
+
+\vspace{0.4em}
+\textbf{如何组合出更强大的模型？}
+}
+\end{tcolorbox}
+
+\vspace{2em}
+\begin{center}
+\begin{tikzpicture}
+\node [fill=blue!10] (label) at (0,0) {\Large{$\textbf{y} = ?(\textbf{x})$ }};
+\end{tikzpicture}
+\end{center}
+
+\end{frame}
 
 %%%------------------------------------------------------------------------------------------------------------
 %%% 人工神经元的函数形式
@@ -1600,6 +1650,67 @@ cycle}
 \subsection{神经网络的简单实现：张量计算}
 
 %%%------------------------------------------------------------------------------------------------------------
+%%% outline: problem 2
+\begin{frame}{然后}
+
+\vspace{6em}
+\begin{tcolorbox}[enhanced,size=normal,left=2mm,right=1mm,colback=red!5!white,colframe=red!75!black,drop fuzzy shadow]
+{\Large
+\textbf{人工神经网络的数学描述是什么,}
+
+\vspace{0.4em}
+\textbf{如何编程实现这种数学模型？}
+}
+\end{tcolorbox}
+
+\vspace{1em}
+\begin{center}
+\begin{tikzpicture}
+\begin{scope}[yshift=6.5em,xshift=1em]
+\setcounter{mycount1}{1}
+\draw[step=0.5cm,color=orange,thick] (-1,-1) grid (0.5,0.5);
+\foreach \y in {+0.25,-0.25,-0.75}
+  \foreach \x in {-0.75,-0.25,0.25}{
+    \node [fill=orange!20,inner sep=0pt,minimum height=0.49cm,minimum width=0.49cm] at (\x,\y) {\number\value{mycount1}};
+    \addtocounter{mycount1}{1};
+  }
+\end{scope}
+
+\begin{scope}[yshift=6em,xshift=0.5em]
+\setcounter{mycount2}{2}
+\draw[step=0.5cm,color=blue,thick] (-1,-1) grid (0.5,0.5);
+\foreach \y in {+0.25,-0.25,-0.75}
+  \foreach \x in {-0.75,-0.25,0.25}{
+    \node [fill=blue!20,inner sep=0pt,minimum height=0.49cm,minimum width=0.49cm] at (\x,\y) {\number\value{mycount2}};
+    \addtocounter{mycount2}{1};
+  }
+\end{scope}
+
+\begin{scope}[yshift=5.5em,xshift=0em]
+\setcounter{mycount3}{3}
+\draw[step=0.5cm,color=ugreen,thick] (-1,-1) grid (0.5,0.5);
+\foreach \y in {+0.25,-0.25,-0.75}
+  \foreach \x in {-0.75,-0.25,0.25}{
+    \node [fill=green!20,inner sep=0pt,minimum height=0.49cm,minimum width=0.49cm] at (\x,\y) {\number\value{mycount3}};
+    \addtocounter{mycount3}{1};
+  }
+\end{scope}
+
+\begin{scope}[yshift=5em,xshift=-0.5em]
+\setcounter{mycount4}{4}
+\draw[step=0.5cm,color=red,thick] (-1,-1) grid (0.5,0.5);
+\foreach \y in {+0.25,-0.25,-0.75}
+  \foreach \x in {-0.75,-0.25,0.25}{
+    \node [fill=red!20,inner sep=0pt,minimum height=0.49cm,minimum width=0.49cm] at (\x,\y) {\number\value{mycount4}};
+    \addtocounter{mycount4}{1};
+  }
+\end{scope}
+\end{tikzpicture}
+\end{center}
+
+\end{frame}
+
+%%%------------------------------------------------------------------------------------------------------------
 %%% 张量
 \begin{frame}{如何描述神经网络 - 张量计算}
 \begin{itemize}
@@ -1658,11 +1769,6 @@ cycle}
 
 \end{frame}
 
-\newcounter{mycount1}
-\newcounter{mycount2}
-\newcounter{mycount3}
-\newcounter{mycount4}
-
 %%%------------------------------------------------------------------------------------------------------------
 %%% 张量的简单定义
 \begin{frame}{张量是什么}
@@ -1958,7 +2064,6 @@ cycle}
 \begin{center}
 \begin{tikzpicture}
 
-
 \begin{scope}[yshift=6.5em,xshift=1em]
 \visible<2->{
 \setcounter{mycount1}{1}
@@ -2154,7 +2259,127 @@ cycle}
 \end{frame}
 
 %%%------------------------------------------------------------------------------------------------------------
+%%% 深度学习工具包
+\begin{frame}{如何实现？- 开源张量计算框架}
+\begin{itemize}
+\item 实现神经网络的开源系统很多，最简单好用的一个工具包NumPy \url{https://numpy.org/}
+    \begin{itemize}
+    \item Python接口，多维数组的定义使用方便
+    \item 提供了张量表示和使用的范式
+    \end{itemize}
+\item<2-> 最近，很火的两个框架：TensorFlow和Pytorch
+    \begin{itemize}
+    \item Google和Facebook出品，质量有保证
+    \item 功能强大，用例丰富
+    \item 可以进行大规模部署和应用
+    \item 大量可参考的实例
+    \end{itemize}
+
+    \includegraphics[scale=0.13]{./Figures/tensorflowpytorch.jpg}
+\item<3-> 还有其它还在更新的优秀框架： CNTK、MXNet、PaddlePaddle、Keras、Chainer、 dl4j、NiuTensor等
+\end{itemize}
+\end{frame}
+
+%%%------------------------------------------------------------------------------------------------------------
+%%% NiuTrans.Tensor工具包
+\begin{frame}{NiuTensor}
+\begin{itemize}
+\item 这里使用我们自研的NiuTensor工具包进行教学 \url{http://www.niutrans.com/opensource/niutensor/index.html}
+    \begin{itemize}
+    \item 简单小巧，易于修改
+    \item C++语言编写，代码高度优化
+    \item 同时支持CPU和GPU设备
+    \item 丰富的张量计算接口
+    \end{itemize}
+\end{itemize}
+
+\includegraphics[scale=0.35]{./Figures/niutensor.jpg}
+
+\end{frame}
+
+%%%------------------------------------------------------------------------------------------------------------
+%%% 使用NiuTensor
+\begin{frame}{使用NiuTensor}
+\begin{itemize}
+\item NiuTensor的使用很简单，下面是一个C++例子
+\end{itemize}
+
+\begin{tcolorbox}[enhanced,frame engine=empty,boxrule=0.1mm,size=title,colback=blue!10!white]
+\begin{flushleft}
+{\scriptsize
+\begin{tabbing}
+\texttt{\#include "source/tensor/XTensor.h"} \hspace{4em} \= // 引用XTensor定义的头文件 \\
+
+\texttt{using namespace nts;} \> // 引用nts命名空间 \\
+\ \\
+
+\texttt{int main(int argc, const char ** argv)\{} \\
+\ \ \ \ \texttt{XTensor tensor;} \> // 声明张量tensor \\
+
+\ \ \ \ \texttt{InitTensor2D(\&tensor, 2, 2, X\_FLOAT);} \> // 定义张量为2*2的矩阵 \\
+
+\ \ \ \ \texttt{tensor.SetDataRand();} \> // 用[0,1]的均匀分布初始化张量 \\
+
+\ \ \ \ \texttt{tensor.Dump(stdout);} \> // 输出张量内容 \\
+
+\ \ \ \ \texttt{return 0;}\\
+\texttt{\}}
+
+\end{tabbing}
+}
+\end{flushleft}
+\end{tcolorbox}
+
+\begin{itemize}
+\item<2-> 运行这个程序会看到张量每个元素的值
+\end{itemize}
+
+\visible<2->{
+\begin{tcolorbox}[enhanced,frame engine=empty,boxrule=0.1mm,size=title,colback=black!10!white]
+\begin{flushleft}
+{\scriptsize
+\begin{tabbing}
+\texttt{order=2 dimsize=2,2 dtype=X\_FLOAT dense=1.000000} \\
+\texttt{3.605762e-001 2.992340e-001 1.393780e-001 7.301248e-001}
+\end{tabbing}
+}
+\end{flushleft}
+\end{tcolorbox}
+}
+
+\vspace{-0.5em}
+\begin{itemize}
+\item<2-> 还可以看到：二阶张量(order=2)，形状是$2 \times 2$ (dimsize=2,2)，数据类型是单精度浮点(dtype=X\_FLOAT)，是一个非稀疏张量(dense=1.000)
+\end{itemize}
+
+\end{frame}
+
+
+%%%------------------------------------------------------------------------------------------------------------
 \subsection{参数学习 - 反向传播}
 
+%%%------------------------------------------------------------------------------------------------------------
+%%% outline: problem 3
+\begin{frame}{还有一个问题}
+
+\vspace{6em}
+\begin{tcolorbox}[enhanced,size=normal,left=2mm,right=1mm,colback=red!5!white,colframe=red!75!black,drop fuzzy shadow]
+{\Large
+\textbf{如何对模型中的参数进行学习,}
+
+\vspace{0.4em}
+\textbf{之后使用学习到的模型进行推断？}
+}
+\end{tcolorbox}
+
+\vspace{2em}
+\begin{center}
+\begin{tikzpicture}
+\node [fill=blue!10] (label) at (0,0) {\LARGE{$\frac{\partial \textbf{E}}{\partial \textbf{w}} = $ ? }};
+\end{tikzpicture}
+\end{center}
+
+\end{frame}
+
 \end{CJK}
 \end{document}