From the Caffe website:
Caffe is a deep learning framework made with expression, speed, and modularity in mind. It is developed by Berkeley AI Research (BAIR) and by community contributors. Yangqing Jia created the project during his PhD at UC Berkeley. Caffe is released under the BSD 2-Clause license.
Caffe has been compiled at MSI for GPU calculations, so the latest versions are available on Mesabi's k40 queue. The software can be loaded for use in an interactive Linux environment via:
module load caffe
In order to use Caffe's Python bindings with the 1.0 version, you must additionally run the following command to activate the Python environment:
source activate caffe1.0_lite