运行 Caffe 示例时出错：blob.cpp：检查失败：count_ == proto.data_size() (1 vs. 0)

Question

I was trying to run a sample Caffe code from a git repo .

I installed Caffe on my Ubuntu 18.04 with apt-get install caffe-cpu and libcaffe-cpu-dev , and compiled the code and generated DeepVideoSummExample successfully.

For running, this executable asks for googlenet_places205 model files as well as hybridCNN mean file, which I downloaded them online. But when running DeepVideoSummExample with the following example command as stated in the repo, I get an error right away:

./build/DeepVideoSummExample -videoFile Sample_NEW.mp4 -videoSaveFile outdeep.avi -summaryModelSim 0 -simcover 0 -segmentType 0 -summaryAlgo 0 -featureLayer loss3/classifier -network_file Models/googlenet_places205/deploy_places205.protxt -trained_file Models/googlenet_places205/googlelet_places205_train_iter_2400000.caffemodel -mean_file Models/hybridCNN/hybridCNN_mean.binaryproto -label_file Models/googlenet_places205/categoryIndex_places205.csv -budget 30

And error (a lot of outputs....but here is the end):

F0918 20:11:34.079298 17067 blob.cpp:496] Check failed: count_ == proto.data_size() (1 vs. 0)
*** Check failure stack trace: ***

Here are the extended errors.

What's the problem? How to fix it?