neon

I haven't seen CuDNN equivalents (in terms of perf) for common machine learning frameworks from AMD. I don't think they exist, if they did, people would shift to using AMD. For NVidia I have seen some faster kernels than the ones supplied by NVidia - https://github.com/NervanaSystems/neon - though CuDNN introduced Winograd kernels too in their last update

Yes, on their site they post Torch CuDNN v4 as faster than Leaf [0]. Seems exciting for an early release. Can it get much faster than something like Torch? I would think if CuDNN is doing most of the computation time it would be hard to see big improvements. Perhaps go the route of Neon and tune your GPGPU code like crazy [1, 2], or MXNet and think about distributed computing performance [3]. [0] http://autumnai.com/deep-learning-benchmarks [1] https://github.com/soumith/convnet-benchmarks [2] https://github.com/NervanaSystems/neon [3] http://alex.smola.org/talks/NIPS15.pdf

Keras has a low-level compatibility library[1] (`from keras import backend as K`) that people have reported as useful independently. The interface offered by K[2] seems quite similar to that of TensorFuse[3] As the K backend is used for Keras, it also offers proof that K can be used for sizable and complex projects, plus can take advantage of pre-existing testing for Keras. Whilst I love the idea of CGT, it has not yet taken off. I'd be far more interested in seeing a Neon[4] backend considering it has the fastest performance across the board on existing hardware and they're planning to release their own hardware soon. [1]: http://keras.io/backend/ [2]: https://github.com/fchollet/keras/blob/master/keras/backend/... [3]: https://github.com/dementrock/tensorfuse/tree/master/tensorf... [4]: https://github.com/NervanaSystems/neon Edit: Incor

Such projects are nice for educational use but not so much further when you don't have any GPU implementation. It's pointed out in the Readme that this was for education but it still wasn't clear whether it's supposed to have a practical use outside of that (some educational projects develop into something useful). If it has (or maybe even if not), it should have a small comparison to other frameworks (maybe Python frameworks only), e.g. Theano, Blocks, Keras, Brainstorm, Neon, .... I think Brainstorm (https://github.com/IDSIA/brainstorm) or Neon (https://github.com/NervanaSystems/neon) are somewhat close to your framework, because they are not based on Theano and thus do not do automatic symbolic gradient calculation but have explicit backprop code. Also, you really should point out what GPU implementation you have (if any), if it suppor

MJ here - founder of Autumn and creator of Leaf. I can give a quick comparison on Leaf vs. TensorFlow - although this might seem somewhat biased. We like what the Google engineers did on TensorFlow and share a lot of similar ideas on how future Machine Learning should be structured and implemented. Especially that in the end it is just a performant pipeline for numeric information processing. We feel that the biggest difference (besides the different stages of the project) is the language and the ecosystem that it embraces. We are strong believers in Rust and think that it might have good chances to succeed in the long run - which we outline in more detail in the Q&A[1]. Due to your question I would recommend picking up on the basic concepts of machine intelligence and then going with whatever ecosystem fits the task/stack. Might become s