MatConNet源代码解读（2）

function [net, info] = cnn_mnist(varargin)
%很多人看到varargin就吓住了，其实可以没有参数的
%CNN_MNIST  Demonstrates MatConvNet on MNIST
%执行vl_setupnn，这么麻烦？
run(fullfile(fileparts(mfilename('fullpath')),...
'..', '..', 'matlab', 'vl_setupnn.m')) ;

opts.batchNormalization = false ;
opts.networkType = 'simplenn' ;
[opts, varargin] = vl_argparse(opts, varargin) ;

%生成实验中途记录文件的名称，每epoch记录一次。记录数据放在data目录下面
sfx = opts.networkType ;
if opts.batchNormalization, sfx = [sfx '-bnorm'] ; end
opts.expDir = fullfile(vl_rootnn, 'data', ['mnist-baseline-' sfx]) ;
[opts, varargin] = vl_argparse(opts, varargin) ;

%图像数据库位置
opts.dataDir = fullfile(vl_rootnn, 'data', 'mnist') ;
opts.imdbPath = fullfile(opts.expDir, 'imdb.mat');
opts.train = struct() ;
opts = vl_argparse(opts, varargin) ;
if ~isfield(opts.train, 'gpus'), opts.train.gpus = []; end;

% --------------------------------------------------------------------
%                                                         Prepare data
% --------------------------------------------------------------------
%网络初始化
net = cnn_mnist_init('batchNormalization', opts.batchNormalization, ...
'networkType', opts.networkType) ;
%如果有mnist数据库就直接加载，没有就从网上下。有没有觉得matlab一下变得好高档
if exist(opts.imdbPath, 'file')
imdb = load(opts.imdbPath) ;
else
imdb = getMnistImdb(opts) ;
mkdir(opts.expDir) ;
save(opts.imdbPath, '-struct', 'imdb') ;
end
%arrayfun以数组的元素作为函数@x的输入，UniformOutput指输出结果的类型是否都相同，为什么要是false呢？没看明白
net.meta.classes.name = arrayfun(@(x)sprintf('%d',x),1:10,'UniformOutput',false) ;

% --------------------------------------------------------------------
%                                                                Train
% --------------------------------------------------------------------
%开始干正事了
switch opts.networkType
case 'simplenn', trainfn = @cnn_train ;
case 'dagnn', trainfn = @cnn_train_dag ;
end
%trainfn就是cnn_train啦，val参数有什么用
[net, info] = trainfn(net, imdb, getBatch(opts), ...
'expDir', opts.expDir, ...
net.meta.trainOpts, ...
opts.train, ...
'val', find(imdb.images.set == 3)) ;

% 取batch数据，不会吧，分割batch还要自己来。仔细看输出是个函数指针，说明实际batch是自动抽取的
%--------------------------------------------------------------------
function fn = getBatch(opts)
% --------------------------------------------------------------------
switch lower(opts.networkType)
case 'simplenn'
fn = @(x,y) getSimpleNNBatch(x,y) ;
case 'dagnn'
bopts = struct('numGpus', numel(opts.train.gpus)) ;
fn = @(x,y) getDagNNBatch(bopts,x,y) ;
end

% --------------------------------------------------------------------
function [images, labels] = getSimpleNNBatch(imdb, batch)
% --------------------------------------------------------------------
images = imdb.images.data(:,:,:,batch) ;
labels = imdb.images.labels(1,batch) ;

% --------------------------------------------------------------------
function inputs = getDagNNBatch(opts, imdb, batch)
% ------------------
4000
--------------------------------------------------
images = imdb.images.data(:,:,:,batch) ;
labels = imdb.images.labels(1,batch) ;
if opts.numGpus > 0
images = gpuArray(images) ;
end
inputs = {'input', images, 'label', labels} ;

%下载MnistImdb，话说Lecun也就是靠这个数据库一战成名，1998年那篇文章到底做了多少乱七八糟的实验啊~
% --------------------------------------------------------------------
function imdb = getMnistImdb(opts)
% --------------------------------------------------------------------
% Preapre the imdb structure, returns image data with mean image subtracted
files = {'train-images-idx3-ubyte', ...
'train-labels-idx1-ubyte', ...
't10k-images-idx3-ubyte', ...
't10k-labels-idx1-ubyte'} ;

if ~exist(opts.dataDir, 'dir')
mkdir(opts.dataDir) ;
end

for i=1:4
if ~exist(fullfile(opts.dataDir, files{i}), 'file')
url = sprintf('http://yann.lecun.com/exdb/mnist/%s.gz',files{i}) ;
fprintf('downloading %s\n', url) ;
gunzip(url, opts.dataDir) ;
end
end

f=fopen(fullfile(opts.dataDir, 'train-images-idx3-ubyte'),'r') ;
x1=fread(f,inf,'uint8');
fclose(f) ;
x1=permute(reshape(x1(17:end),28,28,60e3),[2 1 3]) ;

f=fopen(fullfile(opts.dataDir, 't10k-images-idx3-ubyte'),'r') ;
x2=fread(f,inf,'uint8');
fclose(f) ;
x2=permute(reshape(x2(17:end),28,28,10e3),[2 1 3]) ;

f=fopen(fullfile(opts.dataDir, 'train-labels-idx1-ubyte'),'r') ;
y1=fread(f,inf,'uint8');
fclose(f) ;
y1=double(y1(9:end)')+1 ;

f=fopen(fullfile(opts.dataDir, 't10k-labels-idx1-ubyte'),'r') ;
y2=fread(f,inf,'uint8');
fclose(f) ;
y2=double(y2(9:end)')+1 ;
%训练集为1，测试集为3

set = [ones(1,numel(y1)) 3*ones(1,numel(y2))];
data = single(reshape(cat(3, x1, x2),28,28,1,[]));
dataMean = mean(data(:,:,:,set == 1), 4);

%这个函数牛逼了，可以理解为将dataMean扩展至与data同维数，然后逐点执行minus操作。实际可能分布式计算，好牛叉！

data = bsxfun(@minus, data, dataMean) ;

imdb.images.data = data ;
imdb.images.data_mean = dataMean;
imdb.images.labels = cat(2, y1, y2) ;
imdb.images.set = set ;

%最难的在这里，‘val’是什么意思一直没搞懂！
imdb.meta.sets = {'train', 'val', 'test'} ;
imdb.meta.classes = arrayfun(@(x)sprintf('%d',x),0:9,'uniformoutput',false) ;