如何运行training的源代码

自从看了SRCNN这篇论文之后，特别想自己去亲手实现它，由于之前从来没有接触过DL，所以现在是一脸懵逼0.0，我装好了matcaffe之后，照着源代码中的readingme去执行，
***********************************************************************************************************


Training code for "Learning a Deep Convolutional Network for Image Super-Resolution" (ECCV 2014) 
and "Image Super-Resolution Using Deep Convolutional Networks" (TPAMI 2015)
by Chao Dong (ndc.forward@gmail.com)



***********************************************************************************************************



Usage:


1. Place the "SRCNN" folder into "($Caffe_Dir)/examples/"



2. Open MATLAB and direct to ($Caffe_Dir)/example/SRCNN, run 
"generate_train.m" and "generate_test.m" to generate training and test data.



3. To train our SRCNN, run
./build/tools/caffe train --solver examples/SRCNN/SRCNN_solver.prototxt



4. After training, you can extract parameters from the
 caffe model and save them in the format that can be used in our test package (SRCNN_v1). 
To do this, you need to install mat-caffe first, then open MATLAB and direct to ($Caffe_Dir) and run "saveFilters.m". The "($Caffe_Dir)/examples/SRCNN/x3.mat" will be there for you.

 
执行到 第三步的时候我发现我的路径中没有caffe（./build/tools/caffe train）， 我就自己mkdir了一个.......
之后，再执行命令sh ./build/tools/caffe train --solver examples/SRCNN/SRCNN_solver.prototxt(在caffe 目录下) ， 然后就 完事了？？？ 到第四步的时候运行saveFilters.m的时候说找不到caffe啊.
 
所以我想问..怎么才能正确的去执行readingme， 我这个问题可能是太蠢了，但还是希望请各路大神赐教