images path

[QingkaiLu]

Hi Jos,

Thank you for sharing your script. I am new to caffe and this might be exactly what I need. I want to use caffe to extract features from images, then apply libsvm to do image classification using the extracted features.

I have one question about how to use this script. I want to use a pre-trained caffe model(e.g. models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel ) to extract features from a set of images in the libsvm format. How should I specify the path of my images if I use your script?

Thank you in advance for any help!

[QingkaiLu]

Never mind, I think I figured this out. It is in the *.prototxt: source:”file_list.txt”