In my opinion, the author doesn't try very hard to be constructively critical of all the systems. Example: Caffe is given 3/5 in the "Architecture" section bec

In my opinion, the author doesn't try very hard to be constructively critical of all the systems. Example: Caffe is given 3/5 in the "Architecture" section because its layerwise design means you have to implement the forward/backward passes yourself for both CPU and GPU. Torch uses exactly the same design and inherits all of its downsides, but is given 5/5 for no reason. I would also strongly mark Torch down because Lua is difficult for users who are familiar with Python/C/C++. At least, it's difficult for me. I'm still routinely bit by off-by-one bugs due to Lua's lack of 0-indexed arrays. Hardly the kind of thing you should be worried about in a scientific computing framework. This article is a good start, but you should familiarize yourself with each framework and ignore the author's subjective ratings.