Hello. I am trying to follow the instructions you provide for the coneptual-captions feature extraction.

I am having problems with using the caffe.
I also opened an issue in caffe github: BVLC/caffe#7003

The docker installation for caffe works (I didn't succeed in other installations), but now I am missing the _caffe module that I know exists when making make pycaffe (from here). However, when building it from the Docker I don't know where I should make this pycaffe.

Do you know if I am missing something?

Thank you