21天实战caffe - 运行基于已训练模型的图片分类例程 - 报错：Unable to decode image

今天在《21天实战caffe》中尝试运行一个基于已训练模型的图片分类例程。见《21天实战caffe》第178页，输入如下命令后：

./build/examples/cpp_classification/classification.bin \

models/bvlc_reference_caffenet/deploy.prototxt \

models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel \

data/ilsvrc12/imagenet_mean.binaryproto \

data/ilsvrc12/synset_words.txt \

exmaples/images/cat.jpg

遇到如下报错：

---------- Prediction for /home/yang/caffe/exmaples/images/cat.jpg ----------
F0323 15:35:41.133546 22096 classification.cpp:248] Check failed: !img.empty() Unable to decode image exmaples/images/cat.jpg

*** Check failure stack trace: ***
    @     0x7f0a876f0daa  (unknown)
    @     0x7f0a876f0ce4  (unknown)
    @     0x7f0a876f06e6  (unknown)
    @     0x7f0a876f3687  (unknown)
    @           0x403d8a  main
    @     0x7f0a84f79ec5  (unknown)
    @           0x403f2b  (unknown)
    @              (nil)  (unknown)
已放弃 (核心已转储)

解决办法：

耽误了半天，原来是路径错误，examples写错成exmaples了，好尴尬……以后写路径直接在命令行里敲tab健算了……命令写对，成功解决！

结论：

好好学习，认真仔细！